Keating’s control fraud could have taught us the lessons we needed to learn to avoid the ongoing wave of control fraud. Consider Greenspan’s, Benston’s, and Fischel’s evaluations of Lincoln Savings. Greenspan said it “posed no foreseeable risk” to the FSLIC (Mayer 1990, appendix C; U.S. House Banking Committee 1989, 3:603–606).1 Benston said Lincoln Savings should serve as the model for the industry. Fischel proved it was the best S&L in the nation. Three of the nation’s top financial experts took the worst corporation in the nation (perhaps the worst in the world) and pronounced it superb. That is the measure of how successful control frauds are in deceiving experts who do not understand fraud mechanisms and assume that CEOs cannot be crooks.
Our failure to learn the lessons of the S&L debacle led to the new wave of control fraud.
1. FRAUD MATTERS, AND CONTROL FRAUDS POSE UNIQUE RISKS.
Control fraud was one of the largest causes of losses during the debacle, and other frauds were material. Control fraud caused the current wave of financial scandals.
Unfortunately, the conventional economic wisdom failed to learn this lesson. As stated in Easterbrook and Fischel (1991, 285), that wisdom asserts:
[A] rule against fraud is not an essential or even necessarily an important ingredient of securities markets.
2. IT IS IMPORTANT TO UNDERSTAND FRAUD MECHANISMS.
Many of the financial experts’ most embarrassing predictive and analytical failures arose because they did not understand how control frauds operate. For example, Bert Ely, a financial consultant, thought fraud was trivial because there were few cases of CEOs taking money out of the till (Ely 1990). Control frauds were consistently able to fool top economists because they did not understand how CEOs turn accountants and accounting principles from a restraint into a weapon of fraud and a shield against the regulators.
Economists receive no training about fraud risk, incidence, or mechanisms (beyond the conventional wisdom that it was trivial, a distraction during the debacle). Lawyers receive no fraud training. Joe Wells’s Association for Certified Fraud Examiners (ACFE) offers free materials to any business school that will teach a course in fraud examination. Only a small number of business schools took up the offer prior to the Enron scandal. The number even now remains scandalously low. The average new accountant receives no meaningful training about fraud and no training at all about control fraud.
3. CONTROL FRAUD CAN OCCUR IN WAVES.
Control fraud is not random. Criminologists know that some environments produce increased crime. The S&L industry in the first half of the 1980s was a nearly optimal environment for control fraud. The incentive to engage in both reactive and opportunistic control fraud increased sharply. Entry was easy. Opportunistic control frauds flocked to the industry.
The ongoing wave of control fraud, by contrast, arose from the growth and collapse of an enormous financial bubble and the evisceration of regulatory and ethical restraints on fraud. We can identify the environments that produce waves of control fraud and avoid or greatly reduce the problem if we change our approach to regulation and corporate compensation.
Similar waves of control fraud have occurred in many countries (La Porta, Lopez-de-Silanes, and Zamarripa 2003; Johnson, La Porta, Lopez-de-Silanes, and Shleifer 2000). Conventional economists’ efforts to guide the transition of Russia to capitalism were disastrous because they failed to understand both governmental (kleptocratic) and corporate control fraud (Stiglitz 2003, 21).
4. WAVES OF CONTROL FRAUD CAUSE IMMENSE DAMAGE.
Individual cases of control fraud cause severe losses, but waves of control fraud cause systemic injury. The direct financial damage is staggering: many tens of billions of dollars during the debacle and hundreds of billions in the ongoing scandals.
The indirect financial damage is far greater. During the debacle, the indirect financial injury was primarily the inflating of regional real estate bubbles. This deepened the real estate glut and ultimately worsened the severe drop of values that harmed even honest real estate participants. Bubbles waste societal resources through misallocation both when they inflate and when they collapse. Fraud sends inaccurate price signals that move markets ever farther away from efficiency. The S&L control frauds kept real estate prices artificially high by increasing the levels of ADC lending and direct real-estate investments in markets that had vacancy rates seen only in the severest recessions. The fraudulent investments contributed to regional recessions in Texas, Louisiana, and Arizona. Other indirect financial injuries were inflicted on S&L employees, who lost their jobs and pensions when control frauds destroyed the S&L; innocent stockholders; and the victims of ACC’s worthless junk bond sales. The S&L industry was not a major purchaser of junk bonds, but it provided Milken’s most important group of “captives.” They were critical to the overstating of junk bond values, which misallocates and wastes societal resources.
The indirect financial damage during the ongoing scandals has been far greater. They have again created wasteful gluts, particularly in broadband. However, the ongoing scandals’ most harmful indirect economic effect has been to reduce trust. Frauds operate by creating—and abusing—trust. As a result, fraud is the most powerful acid extant for eating away trust. We are now beginning to understand how important trust is to an economy. Economists of all ideological stripes agree that trust is one of the most important resources (Stiglitz 2003, 274; Fukuyama 1995). One way to conceptualize Akerlof’s classic 1970 article on lemons markets is that it shows how widespread control fraud eviscerates trust.2 A broad range of scholars and market participants have concluded that the erosion of trust caused by the ongoing wave of control fraud significantly influenced millions of investors to withdraw from the stock market, which caused a staggering loss of market capitalization (Stiglitz 2003, 274). That loss was $9 trillion, so if the consensus is correct that the erosion of trust contributed materially to that loss at its peak, the indirect losses from control fraud were crushing.3
Current scandals have generated many other indirect economic costs; lost jobs and pensions again predominate. Enron and its fellow conspirators (virtually every major energy trader in America and several electrical generators) caused blackouts in California, raised the price of electrical power dramatically, and bankrupted California’s electrical utilities. (These control frauds were aimed at customers, not creditors.) Similarly, corporate tax fraud surged as some of the nation’s top audit and law firms pushed fraudulent tax shelters and schemes for using tax havens to avoid taxation. Enron was so active in these frauds that it marketed its services as a consultant for eliminating corporate taxes.
The indirect, nonfinancial impacts of the S&L control frauds cannot be quantified. One of these impacts is political scandal. The control frauds’ manipulation of Speaker Wright became widely known. The disclosure weakened Wright while he was in office and contributed to his decision to resign. Keating’s enlistment of the Keating Five greatly damaged Senator McCain’s ambition to become president. He was the Republican Party’s rising star in early 1987 but a political liability by September 1987. He now campaigns for political reform. Is he, and is the nation, better or worse off because Keating convinced him to add his leverage against the Bank Board? The most I can say is that the S&L control frauds powerfully affected our political system. My personal view is that they damaged it.
The social and political effects of waves of control fraud are clear, and in Russia and some of the other former Soviet bloc nations, they are tragic. Life expectancy has fallen dramatically, violence has increased, respect for Western institutions has plummeted, and poverty and disease rates have surged. These social and political effects have made many Russians hostile to the United States, and the effects feed back into economic policies that can cause further damage (Stiglitz 2003).
Effective markets are neither natural nor inevitable. Markets are institutions shaped by law, culture, and morals. They are vulnerable to control frauds. Such frauds cannot be defeated for all time; eternal vigilance is essential. Effective regulation is essential if modern markets are to remain honest and efficient.
5. CONTROL FRAUDS CONVERT CONVENTIONAL RESTRAINTS ON ABUSE INTO AIDS TO FRAUD.
Control frauds do not simply defeat internal and external controls such as outside auditors. They pervert intended controls into allies. We need to listen to Keating. Grogan, his political fixer, warned Keating that hiring Atchison away from Arthur Young just after he gave Lincoln Savings a clean audit opinion (and quadrupling his salary) would discredit Atchison.4 Atchison was critical to ACC/Lincoln Savings’ getting clean audit opinions, recruiting the Keating Five, and making presentations to the Bank Board, so Grogan apparently did not want to repeat Keating’s mistake with Henkel, i.e., ruining such a valuable asset. Grogan describes Keating’s response:
[He] was white hot. He said that spineless lawyers would ruin this company, that he had to have talented accounting people. That lawyers never made a nickel for the company…. And that the accountants and business people made the money for the company. (U.S. Senate Committee 1990–1991a, December 14, 1990, a.m. session, transcript 49)
He meant it literally: the accountants “made the money” (albeit fictitiously). All the S&L control frauds were accounting frauds. All of them were able to get clean audit opinions from top-tier audit firms, typically for many years. No audit firm exposed an S&L control fraud.
Even after two top-tier audit firms resigned from their audit engagement with Lincoln Savings, and after Keating was widely known as a control fraud, Peat, Marwick aggressively “pitched” Keating for the account. It then issued opinions supporting both a scam sale designed to create a massive fictitious profit and a scam reciprocal transaction designed to create a scam sale of Lincoln Savings. Both scams were designed to fend off a regulatory takeover.
Keating showed that control frauds use outside auditors to do far more than bless fraudulent deals and accounting. He used Arthur Andersen to stuff the junk-bond underwriting files in order to deceive the examiners. He used Andersen’s shameful, dishonest resignation letter as well as Atchison’s equally disgraceful letter attacking the FHLBSF examination to help recruit the Keating Five.5 The senators stressed that Andersen’s and Young’s reputations as top-tier audit firms led them to believe Keating’s representations (U.S. House Banking Committee 1989, 3:669–671); Stewart and Dochow were similarly duped. ACC could not have defrauded the widows without Arthur Young’s clean opinion.
Control frauds routinely enlisted top lawyers and academics to aid their frauds. Lawyers drafted the reciprocal deals that covered up the initial frauds. Shopping for appraisers was even easier than shopping for auditors and attorneys. Appraisers’ grossly inflated valuations underlay all the S&L control frauds. The inflated values allowed the auditors to “rely” on the appraisers. This provided the outside auditor and the control fraud important protection against a civil suit.
As Mayer (1991, 285) explained, “Disreputable clients, after all, pay better.” Control frauds pay best: money is no object; they control how much will be spent; and the experts and professionals they hire to defend the firm will actually spend their time defending the CEO while he loots the firm. Zealous advocacy on behalf of the client is perverted into slavish, but lucrative, aid to the CEO looting the firm. The regulators they fought were the ones zealously advocating the interests of the client.
Unfortunately, conventional economic wisdom learned none of these lessons. The legal profession and the accountants, in response to an embarrassing record of the top firms in their fields aiding control frauds, agreed on the primary strategic response: reduced professional liability. Leaders of the bar and the accounting industry testified before the National Commission on Financial Institution Reform, Recovery and Enforcement in 1993. They angrily denounced the OTS and RTC suits that had led, collectively, to over a billion dollars in settlements. They did not offer any apologies, accept any responsibility, or suggest any reforms of their practices.
My original profession, law, was the least professional. It made no reforms until the current wave of control fraud hit, a decade after the S&L debacle. The top law firms in the nation have, again, zealously aided the control frauds in destroying the client. (Shades of “it became necessary to destroy the village in order to save it.”) Even then, they resisted all change, and had to be forced into finally reforming their (unethical) ethics rules barring lawyers from revealing ongoing financial crimes.
The record of the accounting profession has been more mixed. It tried to clean up a number of accounting standards that were abused during the debacle. Overall, though, it thought the debacle had nothing to teach it about auditing. A contemporaneous document compellingly shows this. From an interview with Dan Guy, the head of auditing for the American Institute of Certified Public Accountants (AICPA) in January 1994:
Q. To what extent do you think deficiencies in the audit process contributed to the alleged audit failures in the savings and loan industry? [R]egulators … point out two major deficiencies in audits they have seen: a lack of professional skepticism and a willingness to accept the client’s position.
A. I can’t think of any problem with standards that was identified by regulators looking at the S&L mess. (Craig 1994)
This same complacency existed even among accountants who specialized in fraud detection. A major work, Accountant’s Guide to Fraud Detection and Control (2nd ed.), had the misfortune to be published in March 2000 (before Enron filed for bankruptcy) with a publisher’s press release stating that “[Securities] fraud is being adequately detected by independent auditors (CPAs) in their annual audits.”
Later, however, the profession tried, over great rank-and-file resistance, to get auditors to consider fraud risk when planning audits. The Financial Accounting Standards Board (FASB), to its great credit, tried to require that stock options be expensed. SEC chairman Arthur Levitt strongly supported this initiative. I explain below why this could be important for reducing control fraud. For now, it is enough that the FASB and Levitt knew that the proposal, while clearly correct, would expose them to attack.
The high-tech industries, the top-tier audit firms (under the leadership of Harvey Pitt), Clinton’s Treasury Department, and a bipartisan coalition of legislators led by Senator Lieberman used intense, crude political power to intimidate the FASB and the SEC. Although many were guilty, none paid any political price. Clinton did not back Levitt; Levitt asked the FASB to cave in to the political pressure. He says it was his biggest mistake (Levitt 2002, 10–11, 12).
Accountants led the move to reduce the ability of victims of security fraud to recover their losses. This movement was, cleverly, labeled tort reform. (Everything Congress proposes is a “reform,” no matter how pernicious its effect.)6
Fifteen years ago, when the financial world was rocked by the savings and loan scandal, the accounting industry faced a crisis not unlike the one it faces today. Lawsuits were mounting, millions of dollars were paid out in settlements, and the image of accountants was plummeting….
Jack Henry, a retired managing partner in Andersen’s Phoenix office, said that at the time of the S&L crisis there was a major change in the industry, caused by mounting litigation. “We were tired of getting the crap kicked out of us….”
The nation’s largest accounting firms, including Chicago-based Andersen decided to fight back. They formed a lobbying coalition. They poured millions of dollars into political campaigns….
[T]he top accounting firms successfully lobbied for a federal law that makes it more difficult for investors to sue them. The 1995 law was a major victory for the profession and was achieved only after Congress voted to override a veto by President Bill Clinton—his first ever. (Chicago Tribune, February 13, 2002)
Easterbrook and Fischel (1991, 282) wrote, after Fischel knew from personal experience representing S&L control frauds that the claim was untenable,7 that:
High quality firms must take additional steps to convince investors of their quality. One traditional step is [outside audits]. [The auditor] has a reputational interest—and thus a possible loss—much larger than the gains to be made from slipshod or false certification of a particular firm…. The larger the auditor … relative to the size of an issuer, the more effective these methods of verification are.
The Keating Five made the same argument to us.
DeConcini: Why would AY say these things, they have to guard their credibility too. They put the firm’s neck out with this letter.
Patriarca: They have a client …
DeConcini: You believe they’d prostitute themselves for a client?
Patriarca: Absolutely, it happens all the time.8 (U.S. Senate Committee 1990–1991a, 1:1059)
6. CONFLICTS OF INTEREST MATTER.
The S&L debacle proved that human nature had not changed; conflicts of interest still cause damage. Keating’s actions demonstrate that he believed that maximizing conflicts of interest made it easier for him to suborn auditors and attorneys. He engineered Andersen’s and Arthur Young’s involvement in nonaudit work while they were the auditors. Keating placed the accountants in an advocacy role requiring them to attack the FHLBSF examination and examiners in order to help Keating retain control of Lincoln Savings. Andersen’s resignation letter and Atchison’s letter to the Keating Five were blatant, false advocacy pieces. Auditors are supposed to be independent and objective; they must never become advocates for the client. Here, they became advocates for the individual destroying the client because he was the person who hired and fired them (and in Atchison’s case, quadrupled his salary).
Keating bragged to the FHLB-Seattle that he spent $50 million hiring outside professionals to fight the 1986 FHLBSF examination (U.S. House Banking Committee 1989, 3:776). Some of that was for consultants such as Fischel and Benston, but it went overwhelmingly to Arthur Young and to Keating’s top outside law firms: Kaye, Scholer and Jones, Day. Young’s consulting fees and litigation-assistance fees dwarfed its audit fees. Atchison became Keating’s most effective advocate.
Henkel, serving as Keating’s tax lawyer, business partner, borrower, and Bank Board mole, exemplified the dangers of conflicts of interest, as did his ethics counsel Waxman, was also counsel to Lincoln Savings. A Kaye, Scholer partner received a sweetheart loan from Lincoln Savings. Taggart, the CDSL commissioner who was Keating’s fiercest state regulatory ally, received benefits from Keating shortly before he left office (U.S. House Banking Committee 1989, 2:328–330). Keating, of course, believed that political contributions were essential for recruiting the Keating Five (and many other federal and state politicians). Since Keating backed his beliefs with substantial expenditures and had great expertise and success in manipulating professionals and politicians, there is every reason to conclude from his conduct that conflicts of interest do matter.9
Unfortunately, the conventional economic wisdom taught the opposite. Conflicts became “synergies.” Stiglitz (2003, 10, 133–139) explains how this mindset helped spur scandals. Former SEC chairman Levitt makes the same point in Take on the Street (2002, 114–119).
7. DEPOSIT INSURANCE WAS NOT ESSENTIAL TO S&L CONTROL FRAUDS.
S&L control frauds generally relied on deposit insurance to fund their Ponzi growth. Federal deposit insurance was a key attraction to opportunistic control frauds and the primary reason they clustered in the S&L industry. But that does not mean it was essential. S&L control frauds were consistently able to defraud uninsured private market actors. ACC/Lincoln Savings was able to sell over $250 million in worthless junk bonds—the worst security in America—primarily from three branches in one state. It is true that Keating targeted widows because they generally lacked financial sophistication, but Keating also sold ACC stock to sophisticated investors. He beat the “shorts,” and (worthless) ACC stock sold at a substantial share price after five years of continuous fraud. Milken sold $125 million of (worthless) ACC junk bonds to (purportedly ultrasophisticated) investors in 1984.
S&L control frauds frequently sold subordinated debt because it could count as regulatory “capital.” Subordinated debt is uninsured and inherently risky because the buyer receives nothing until all other creditors are paid. S&L sub debt issued by traditional, healthy S&Ls was far riskier than the norm because S&Ls had minimal capital. The risk from sub debt issued by the high fliers was off the charts. Such S&Ls always defaulted.
ACC is the only exception I recall, and it proves the rule. Milken sold, as I noted, $125 million in ACC sub debt at a high interest rate to the usual subjects: if Milken sold one’s junk bonds, it was understood that one bought junk bonds issued by other Milken clients. The key to Milken’s scheme was to reduce the apparent default rate, so it would not do to let ACC default on its Drexel-issued junk bonds. The situation was as elegant as it was cynical: ACC would sell junk bonds to widows (at a ludicrously low rate of interest) and use the proceeds to retire the Drexel-issued junk bonds sold (at a very high rate of interest) to Milken’s minions.10
This scam simultaneously (1) avoided a default on Drexel-issued junk bonds, (2) considerably reduced ACC’s interest expense, and (3) allowed ACC to book a gain from refinancing its debt at a lower interest rate.
We also should have learned from the debacle that junk bonds actually had fewer debt covenants than less risky debt (which contradicts the theory of private market discipline by creditors). In no case did sub debt holders exercise effective discipline over an S&L. Indeed, I do not recall any case in which they even attempted to impose discipline. All of these facts refute the theory that private creditors exercise effective discipline through debt covenants or similar means.
The other form of private market discipline that failed during the debacle was private deposit insurance. Control frauds caused the failure of private insurance systems for thrifts (state-chartered S&Ls that were not insured by the FSLIC) in Ohio, Maryland, and Utah. In no case did private insurers adopt regulations that conventional economists would consider rational, but that is hardly a defense of the concept of private deposit insurance, for the theory relies on the assumption that they will act rationally. There is no known case where a private thrift insurer successfully stopped a control fraud in time to avoid the collapse of the fund. All the private thrift-insurance funds that did not collapse saw their members convert to FSLIC- or FDIC-insured status because depositors lost confidence in them.
We should have learned from the S&L debacle that private market discipline does not stop control frauds effectively, that sub debt is not equivalent to capital for banks, and that private insurance does not effectively prevent, detect, or limit control fraud. Unfortunately, the conventional economic wisdom learned none of these points. Easterbrook and Fischel (1991, 282), for example, assert that if a firm issues debt, it “(a) forces the managers to pay out the profits, and (b) if there are no profits forces the firm into bankruptcy.” But control frauds follow a third way. They create very large fictitious profits blessed by a top-tier audit firm. They then borrow more money or sell more stock and use part of the proceeds to pay interest on the prior debt.
Because the waves of S&L control frauds and the ongoing wave consistently fooled entities that had the incentive to exercise private market discipline, the efficient-market hypothesis has often been proven false. Control frauds can continue for many years, thereby moving prices further away from their “true” value. The efficient-market hypothesis is the bedrock of modern finance, so the ineffectiveness of markets against waves of control fraud is of major importance.
8. “SYSTEMS CAPACITY” PROBLEMS ARE ENDEMIC AND EXPOSE US TO HUGE DANGER.
Henry Pontell’s systems-capacity theory can help us limit or even prevent future waves of control fraud. Systems capacity refers to the inability of the prosecutorial or regulatory system to cope with crime or misconduct adequately because of resource limitations, deficient authority, or lack of political will. Almost everyone agrees that the Bank Board exemplified each of these severe capacity problems and that the Justice Department was also badly understaffed and wholly unprepared to deal with the wave of control fraud. The Reagan administration bears the primary blame for these capacity limitations.
It is a self-fulfilling prophecy that government will be ineffective if one designs it to be ineffective. The systems-capacity problems of the SEC and the Bank Board are parallel. Both were forbidden to pay salaries competitive with those offered by the banking regulatory agencies (much less private firms). Both experienced rapid turnover and had to staff key operations with inexperienced personnel. Both faced vastly increased supervisory needs in quantity and complexity. The deluges overwhelmed both staffs. The frequency of examinations (or reviews) fell sharply. Government (the OMB during the Reagan administration, the House of Representatives in the case of the SEC) greatly exacerbated the capacity limitations by seeking to reduce the number of staff at a time both agencies were critically understaffed. Neither staff recognized the wave of control fraud until it was too late. If we had learned the lessons of the S&L debacle, we would have supercharged the SEC in the early 1990s, greatly reducing the control frauds.
The combination of our political, ideological, and budgetary mechanisms increase systems-capacity problems. No one in the OMB gets promoted for figuring out that vastly more money should be spent to increase the number of regulators. Again, we did not learn the appropriate lessons from the debacle. The “reinventing government” movement, for example, required every agency to develop formal mission statements and strategic plans, and led the GAO to designate high-risk activities. The SEC’s recent annual reports stress that it is a civil “law enforcement agency.” Nevertheless, the SEC annual reports, prior to Enron’s failure, never identified waves of control frauds as a risk imperiling the success of its mission, and the GAO never designated the SEC’s antifraud activities (or any other SEC activity) as high-risk. (As I write, it has still not done so.) The GAO standards for high-risk activities emphasize fraud risk, but only fraud risk in which the government is directly defrauded. In designating high-risk activities, the GAO should consider the harm that will befall the public if regulators fail to achieve their missions.
One of the great advantages of control frauds is their ability to cause the firm to make political contributions. Audacious control frauds use this ability to help shape their regulatory environment. They seek to undercut effective regulation. There is no “Brotherhood of Burglars” that has apparent respectability and regularly lobbies for restrictions on the quality of door locks or the number of police assigned to neighborhood patrols. The GAO needs to develop a team tasked with looking for critically underregulated areas that put the nation at risk.
9. REGULATORY AND PRESIDENTIAL LEADERSHIP IS VITALLY IMPORTANT.
The S&L debacle should have made both of these points clear. Pratt, Gray, and Wall were more different than similar. Bank Board policies changed radically and quickly when the chairman changed, even though the same president appointed them. These changes, however, did not appear to reflect any change in the Reagan administration’s regulatory policies.
Ryan was very different from Wall, even though President Bush appointed both of them OTS directors. Ryan’s appointment signaled a sharp change in presidential policy concerning regulation, and Ryan changed OTS policy abruptly.
In general then, the debacle was not a case where entrenched bureaucrats drove policies irrespective of presidential or agency leadership. Indeed, the closest example of entrenched bureaucrats making contrary policy was the FHLBSF’s vigorous resistance to Wall’s regulation of Lincoln Savings and American Savings (and Gray’s regulation of American Savings). If Wall had been able to fire the FHLBSF personnel who led this resistance (as he could have done after the creation of the OTS in August 1989), he might have been able to crush that resistance by removing its leaders.
The debacle demonstrated that “the vision thing” is central to regulatory success in a crisis. Pratt was brilliant, charismatic, and had a distinct vision. He also was the regulator who, according to Martin Mayer (1990, 61), would win the prize for most to blame for the debacle. His vision was disastrously wrong, and his brilliance and creativity made the problem worse: he exemplified the phrase “too clever by half.” He believed that some clever fudge (almost always an accounting gimmick) would get him through the latest crisis. He had an uncanny ability to optimize the regulatory environment for control fraud.
Gray got the vision thing right, which is why his critics will never forgive him. Mayer told me that as soon as it became known that he was writing a book about the debacle, many individuals wrote or called him to try to make sure that he attacked Gray. They still believe the control frauds’ very expensive propaganda that tried to paint an intelligent man as stupid.
Gray had many weaknesses, but he had six great strengths that proved decisive in preventing the debacle from becoming a catastrophe. First, he understood, as early as anyone at the agency, that all the high fliers were following the same basic pattern and that their claims of high income and low losses were produced by accounting fraud. Second, he understood that limiting their growth was essential. Third, he understood that the agency was critically understaffed and that he would have to change the structure radically and to openly defy the OMB and the administration. Fourth, he was a very good judge of supervisory talent. He personally recruited Patriarca and Selby to deal with the two districts leading the fight against the control frauds. Fifth, he had a sense of duty that led him to sacrifice his career and defy an astonishing array of powerful opponents in order to defeat the wave of control frauds (which we then called “high fliers”). It also led him to overturn his ideological opposition to reregulation and to be willing to act contrary to the wishes of the president he loved. Sixth, he did not want yes-men or yes-women, and he was comfortable working with people like Selby and Patriarca who had supervisory expertise that far exceeded his own.
Wall remains the enigma with regard to the S&L control frauds. He was such an obsessively secretive man that only a few relevant contemporaneous comments exist on the record. Even a few days before he voted to appoint a conservator for Lincoln Savings, he did not accept the fact that Keating was a fraud. Instead, he continued to praise Keating’s purported business skills. His fellow Bank Board member, Roger Martin, was even more loyal to Keating. Neither of them ever pushed the agency to take action against the high fliers. The nation is fortunate that the growth rule caused each of the remaining high fliers to collapse while Wall was Bank Board chairman. Forbearance delayed these collapses and materially increased the ultimate cost to the taxpayers, but it could not save these Ponzi schemes. In short, Wall and Martin did considerable damage, but they would have done vast damage had they preceded Gray instead of following him.
Both Pratt and Wall demonstrated a recurrent problem with modern regulation. Placing an individual who does not believe in regulation in charge of a regulatory agency can cause great damage, especially if that agency has to deal with control frauds. Deregulatory economists argue that private market mechanisms deal adequately with control fraud. People who believe that claim (and those who have taken one course in law and economics are often its most fervent adherents) believe that financial statements are truthful and that auditors will not aid control frauds. They are taught that public-choice theory has shown that it is naïve to believe that government workers act in the public interest (instead of simply maximizing their personal self-interest). Agencies are routinely captured by the regulatees, according to the economic theory of regulation. Government officials who argue that control fraud is material exemplify an agency problem: they are trying to distract attention from their own failures. Externalities such as pollution are not a valid rationale for regulation; they just represent a failure to adequately assign property rights pursuant to the Coase theorem. Antitrust laws need not, and should not, be enforced, because markets quickly sweep away the rare problems that arise, and antitrust laws are used primarily by unsuccessful competitors to bludgeon rivals they could not outcompete.
The book I use to teach intermediate microeconomics makes most of these points. It is intensely antiregulatory, which means that it presents the conventional neoclassical view. Anyone who believes all (or even most) of the propositions in the above paragraph is going to be personally unprepared to identify or deal with a wave of control frauds. Indeed, such a wave is impossible under their beliefs. Further, if fraud is immaterial, then there is no reason to learn about fraud mechanisms or the means to reduce fraud. If the person who believes such theories is a senior regulatory leader, the agency will be unprepared to deal with a wave of control frauds. Indeed, such a leader would be strongly inclined to believe that any employee who vigorously argued that a corporation reporting strong profits (blessed by a top-tier audit firm) should be closed must have taken leave of his senses or be engaged in a vendetta.
Pratt did not close any control frauds. Therefore, his potential vulnerability to their pleas was untested. Wall appears to be an example of this root vulnerability, which Keating then exploited by demonizing Gray and his lieutenants, threatening personal suits, and bringing political pressure to bear.
None of the four top federal S&L leaders that dealt with the debacle was captured by the industry. The S&L league developed such bad relations with Pratt that they barred him from their meetings. The industry fought against each of Gray’s major regulatory initiatives and the FSLIC recap. After a honeymoon period, the league warred with Wall. Indeed, all three chairmen shared utter contempt for the league. Ryan had no use for it, and it was in any event reduced to a weak group by the time he became OTS director.
Public-choice theory fails to explain the actions of Pratt, Gray, and Wall. Gray is the most obvious case. He acted directly contrary to his self-interest. Wall shows a subtler problem with the theory. Assume for the sake of discussion that Wall chose to appease Keating out of pure self-interest. How could he know that appeasement would best aid his self-interest? His surrender to political pressure and threats of litigation eventually cost him his job and reputation. Perhaps public-sector employees, to maximize their self-interest, act in the public’s interest, because doing so improves their chances of retention and promotion and helps them develop a favorable (and valuable) reputation.
The new wave of control fraud shows that we have again failed to learn the necessary lessons from the S&L debacle. First, Levitt blinked during his ultimate confrontation and gave up on expensing options. His surrender is now his greatest regret. He could have learned from Gray that he would be far happier if he had tried to do the right thing and lost. Note that Levitt (whom I think highly of) is wealthy and that the SEC is considerably more prestigious than the Bank Board. Gray was in a far worse position but persevered. I write this not to disparage Levitt, but to point out that Gray’s courage was unusual.
President George W. Bush appointed Levitt’s successor, Harvey Pitt. Pitt exemplified the danger of appointing regulatory leaders who do not believe that conflicts of interest matter and do not believe that waves of control fraud can occur. Bush made Pitt the SEC chairman because he was the nation’s leading opponent of accounting and audit reform. During a speech Pitt promptly gave to an audience of accountants, he reminded me of Wall and Henkel. He began by expressing his regret that the SEC was not always a “kinder and gentler place” for accountants. He blamed the strained relations between accountants and the SEC on the SEC staff (and implicitly blamed their conduct on Levitt). He eventually resigned after his continued efforts to block the appointment of real reformers (which he did at the request of the White House and Representative Oxley) proved clumsy and embarrassed the administration.
One of the lessons of the S&L debacle is that our nation was very lucky. (I know that a $150 billion debacle does not sound “lucky,” but without Gray’s reregulation and the war against the control frauds, the debacle would have cost over a trillion dollars.) We were thrice lucky in having Gray. First, Reagan appointed him to deregulate. He could not have been appointed if he had supported reregulation. He then underwent a fortunate transformation and became the nation’s leading re-regulator. “Reregulator” was the greatest possible insult in the Reagan administration, and normally would have led to his removal. Gray survived solely because he was a personal friend of the Reagans.
Second, he got it largely right when he reregulated. He got brokered deposits wrong, but he sealed the fate of the control frauds when he adopted the growth rule. Third, he was willing to take on opponents that dwarfed even those that attacked Levitt. We were also lucky because the administration’s effort to give Keating control of the Bank Board failed and because Keating used Henkel so crudely that he could be removed from office before he could destroy reregulation. Henkel would have caused enormous damage if he had remained Keating’s mole.
We cannot afford to rely on luck. We have to take the selection of senior regulators more seriously. That requires us to discuss the role of the president in regulation. President Reagan failed in this role. His administration (and it is important to remember that Vice President Bush was in charge of financial deregulation) took the following actions:
• Insisted on deregulating at a time of mass insolvency
• Insisted on covering up the scope of the crisis
• Barred Pratt from briefing the cabinet finance committee
• Argued in favor of running insolvent S&Ls like Ponzi schemes
• Repeatedly cut the number of examiners
• Fought the agency’s use of the FHLB system to double the number of examiners and supervisors at no cost to the Treasury
• Opposed Gray’s efforts to reregulate
• Refused to allow the FSLIC to obtain any money from Treasury
• Tried to give Keating majority control of the Bank Board
• Appointed Keating’s mole, Henkel, to the Bank Board
• Accepted (through Bush) a $100,000 contribution from Keating even after Senator Riegle had returned his contributions in light of the Keating Five scandal
• Reappointed Henkel to the Bank Board after he tried to immunize Lincoln Savings’ violations
• Tried (through Don Regan) to embarrass Gray into resigning
• Threatened to prosecute Pratt and Gray for closing insolvent S&Ls
• Threatened to prosecute FDIC chairman Seidman for closing banks
• Reached a deal with Speaker Wright to support forbearance and not reappoint Gray
• Conducted a criminal investigation of the FHLBSF at Keating’s request
• Provided no White House support for the FSLIC recap until Gray left office
• Regan testified that while Gray warned of the coming crisis, he, Regan, ignored the warnings
• Not only did President Reagan never request a briefing from Gray about the debacle, but they never discussed it personally after Reagan appointed Gray
• President Bush insisted on appointing Wall as director of the OTS without the advice and consent of the Senate, which was ruled unconstitutional
• Bush appointed Wall OTS director even after he had appeased Keating
10. ETHICS AND SOCIAL FORCES ARE CRITICAL RESTRAINTS ON FRAUD AND ABUSE.
Only a small percentage of traditional S&Ls that did not change ownership in 1981–1984 engaged in control fraud. A CEO who spends her professional career at an S&L working her way to the top generally has substantial loyalty to the institution, staff, and customers. The combination of personal ethics and social ties makes control fraud far more unlikely. Conversely, as I explained earlier, conflicts of interest matter. Real estate developers who bought S&Ls during 1981–1984 were the CEOs most likely to become control frauds. They had serious conflicts of interest and few institutional or personal loyalties.
Whistle-blowers were rare at S&L control frauds. One of the lessons we should have learned from the debacle is that control frauds’ ability to hire and fire personnel makes whistle-blowing an extremely risky act. True whistle-blowers—those who inform the public or the authorities of control fraud—have been rare during the current wave of financial scandals. We cannot rely on whistle-blowers to do the work regulators and the criminal justice system should do against control fraud.
11. DEREGULATION MATTERS AND ASSETS MATTER.
Deregulation can aid control fraud in four ways. It can radically change the environment because we are poor at predicting untested dynamic events. For example, in the S&L debacle, even though an economic theory predicted “competition in laxity,” Pratt did not anticipate California’s reaction to the Garn–St Germain Act of 1982. He did not anticipate how his dozens of regulatory changes interacted to create a perfect environment for control fraud. (I call it “faith-based” deregulation.) Second, deregulation may increase system-capacity problems. ADC loans were often over 100 times more time-consuming to review than single-family home loans. Third, deregulation may allow investment in assets that lack a readily ascertainable market value. One of the keys to accounting fraud is to find such assets, like the large commercial ADC projects in Dallas. They may create guaranteed (though fictitious) income and hide true losses. Fourth, deregulation may provide the authority to enter into reciprocal (fraudulent) transactions used to transmute bad investments into good ones. It may also provide the authority to create an entity that will be used as a straw party.
12. WHY DOESN’T THE SEC HAVE A CHIEF CRIMINOLOGIST?
Virtually every federal agency in America has a chief economist. Most have several chief economists at various subunits. As far as I can tell, no federal agency has a chief criminologist. The federal government does not have a job classification labeled criminologist. All financial regulatory agencies are civil law-enforcement agencies that must concern themselves with control fraud (and many other crimes). They are commonly staffed, however, with people who have received no professional training about fraud. Lawyers, accountants, and finance and economics majors dominate the SEC. None of these disciplines, traditionally, have studied fraud. (A few accounting students now take a class in fraud auditing.)
I do not suggest that the chief criminologist must have a degree in criminology. The goal is to have someone who thinks like a criminologist and is familiar with fraud techniques and indicia of fraud.
We know enough to do much better against control fraud. We can tell what environments are most likely to produce waves of control fraud. We can figure out the fraud schemes that are most prevalent and the best ways to identify such schemes.
There is a basic paradox about trust in the regulatory context, one that the Bank Board under Wall got wrong in their fateful decision to accept Keating’s plea of “trust me.” Trust is vital for efficient commerce, social harmony, and intimacy. But some of us must remain intensely skeptical so that others can continue to trust. The regulators need to be skeptical (accountants and most lawyers also have to rediscover skepticism). That does not mean automatically assuming the worst, but it does mean that one checks hard and skillfully. It also means that if the party being examined interferes with the review, lies, or attempts to intimidate the reviewer, the reviewer’s skepticism should spike and the superiors should support the reviewer.
We can do far better investigations. The S&L debacle is again instructive. Mario Renda was a control fraud who helped loot dozens of S&Ls and banks. A Bank Board enforcement attorney, Hershkowitz, investigated him. Unfortunately, the OE had no experience in complex investigations, and he missed the fraud when he took Renda’s deposition (Binstein and Bowden 1993, 196). Fortunately, Michael Manning of Morrison & Hecker, outside counsel for the FDIC, discovered it (ibid., 214). Manning contacted me, and we (the FSLIC and the FDIC) shared the expense of his investigation that uncovered the widespread fraud. The lesson is that even fairly small numbers of highly experienced, tenacious investigators can greatly increase performance. Too many regulatory agencies refuse to use outside counsel to aid their investigations.
Daisy chains make it harder for examiners looking at one entity to spot reciprocal transactions. But daisy chains are also a weakness if one investigates properly, for they can lead the investigator to identify a large portion of the most dangerous control frauds.
The criminal justice system can do a far better job of investigating waves of control fraud. Daisy chains are highly susceptible to sting operations, as are S&L and bank control frauds. Undercover employees could easily be placed in corrupt financial institutions. FBI and IRS agents often have the necessary expertise. Electronic surveillance could provide direct evidence of conspiracy in reciprocal transactions. None of these techniques was used in any material manner during the S&L debacle. The Dallas Task Force was very effective without these means, but it would have been considerably more effective if it had used them. In our experience, detailing our examiners and supervisors to assist criminal investigations made all the difference. The SEC, of course, is so short of experienced staff that it cannot afford to detail the type of staff that would most aid the criminal investigation.
The SEC also needs to litigate more cases and sign far fewer consent agreements. The SEC negotiates consents disproportionately with failed smaller firms. The commission often gets no meaningful relief in such consents. The greater the system-capacity problems, the greater the tendency to accept a consent agreement rather than litigate. The agency, however, suffers when it relies almost entirely on consents. It does not establish precedents; its enforcement attorneys lack experience and may fear litigation; and the relief granted against serious wrongs will often be inadequate.
Tougher prison sentences in real prisons remain the best hope for deterrence. Keating caused Andersen to stuff the junk bond files in order to deceive the examiners. If the Justice Department had brought a criminal action against Andersen in 1987 for this crime, the accounting profession would have had a powerful incentive to clean up its act. This could have reduced the ongoing wave of accounting scandals.
13. CONTROL FRAUDS DEFEAT CORPORATE GOVERNANCE PROTECTIONS AND REFORMS.
The S&L debacle should have taught us that control frauds were able to defeat a broad range of corporate governance principles. Control frauds control the election of the board of directors. They pick outside directors who are the equivalent of inside directors. Efforts to improve corporate governance may be desirable for other reasons (though we must not lose sight of the cost), but they are unlikely to be effective against control fraud, and we certainly should not rely on their effectiveness. The new Sarbanes-Oxley reform legislation primarily relies on improving corporate governance. Since control frauds have caused the ongoing financial scandals, that is discouraging.
The one hope that the S&L debacle offers in this regard is that S&L control frauds preferred the 100 percent ownership model. They viewed even outside directors picked entirely from business associates as a hindrance. There are virtually no cases in which an S&L board of directors caused any known difficulty for a control fraud. This, however, is still another area in which we have failed to learn the lessons of the debacle. According to Easterbrook and Fischel (1991, 10, 70–75, 133, 282), the ideal form of ownership is the sole owner and CEO because it eliminates agency problems. That may be a disadvantage in the context of control fraud. An agency problem is likely to be either an officer like Mark Sauter, who embezzled from Lincoln Savings, or a whistle-blower who exposes the fraud. Single-ownership will not stop the embezzler, but it may stop the whistle-blower and allow the fraud to grow.
14. STOCK OPTIONS INCREASE LOOTING BY CONTROL FRAUDS.
Control frauds decide how stock options will be structured. They design options to permit easier looting. Typically, this means tying stock options to short-term economic performance, which is easier for the control fraud to distort. Control frauds delight in using seemingly legitimate corporate mechanisms to transfer firm assets to their personal use. According to Easterbrook and Fischel (1991, 282), however, “[With] stock options … If the firm does poorly, the managers lose with the other investors.” By now, this needs no refutation. It should not have needed refutation when written, because S&L control frauds had demonstrated that the CEO could do very well regardless of the fate of the shareholders.
15. WE NEED TO REINVENT THE “REINVENTING GOVERNMENT” MOVEMENT TO DEAL EFFECTIVELY WITH CONTROL FRAUD.
The Clinton-Gore effort to reinvent government did not leave us any better prepared for the current wave of control fraud. Three changes would be helpful. First, the existing strategic plans of few regulatory agencies meaningfully analyze the primary risks threatening the agency’s ability to meet its mission. In plain English, a goal of the SEC should be to prevent waves of control fraud. Second, the GAO should change its definition of high risk to include risks to the public should the agency fail in its mission. It is embarrassing that the GAO, as I write, still does not consider the SEC’s antifraud activities a high-risk function. Third, both the OMB and the GAO should establish divisions to locate units suffering system-capacity problems and to recommend steps to fix the problem. Someone needs to be rewarded and promoted for recommending more regulation and more spending when that is necessary.