About this Issue

If you’ve been staring dolefully at your mutal fund statement and wondering “What the heck happened?!” you are not alone. Perhaps you’re in a position like that of P.J. O’Rourke, whose self-professed understanding of the financial crisis comes to this:

Jim Jerk down the road from me, with all the cars up on blocks in his front yard, falls behind in his mortgage payments, and the economy of Iceland implodes. I’m missing a few pieces of this puzzle myself. 

In this special issue of Cato Unbound, we’ve asked four respected economists, with four very different perspectives, to supply what they think are the missing pieces of the puzzle and to tell us how they all fit together.

Before the meltdown, a few perpetually dour forecasters saw disaster on the horizon, because they always see disaster on the horizon. Yet no one predicted the disaster we got. And the full, true story of what really happened has yet to be told. Of course, this hasn’t kept ideologues from taking the opportunity to malign their enemies and plump for policies they wanted all along. But we cannot place blame until we know what actually happened and why. And we can’t do anything to ensure it won’t happen again until we know what “it” was.

So this month we have brought you Lawrence H. White, the F.A. Hayek Professor of Economic History at the University of Missouri St. Louis; William K. Black, associate professor of economics and law at the University of Missouri, Kansas City and author of The Best Way to Rob a Bank Is to Own One; Casey Mulligan, professor of economics at the University of Chicago; and J. Bradford DeLong, professor of economics at the University of California, Berkeley. Each will provide a short essay laying out his best account of what happened. After all the essays are online, our panelists will then hash out their differences in a lively informal blog chat.

At the end, your mutual fund statement might not look brighter, but it least it will be a bit less maddeningly mysterious. More importantly, with a clearer picture of what happened, we can start deliberating more intelligently about what does, and does not, need to done.

Lead Essay

What Really Happened?

Our ongoing financial turmoil began in the mortgage market. Real-estate loans at commercial banks grew at a remarkable 12.26 percent compound annual rate over the four-year period from the midpoint of 2003 to the midpoint of 2007.[1] The expanded volume of mortgages—notably including an unusually large share of mortgages with “nonprime” ratings—fueled a run-up in condo and house prices to unprecedented heights, followed by a sharp decline. (Was this a “bubble” in prices? Yes, in the sense that the price path was unsustainable; no, in the sense that it was not entirely self-feeding.) Default rates on nonprime mortgages and adjustable-rate mortgages rose to unexpected highs, reducing cash flows to lenders. Financial firms holding securitized mortgage bundles (aka “mortgage-backed securities”) saw the expectation of continuing reductions in cash flows reflected in declining market values for their securities. Uncertainty about future cash flows impaired the liquidity (re-salability) of their securities.

Among the prominent results: the nation’s two largest mortgage finance institutions, the government-sponsored Fannie Mae and Freddie Mac, have gone into bankruptcy-like “conservatorship.” Major investment banks, insurance companies, and commercial banks have gone bankrupt outright (or have been sold for cents on the dollar) because of their heavy exposure to real estate lending. Prices and trading volumes in mortgage-backed securities have shrunk dramatically. Doubts about the value of mortgage-backed securities have led naturally to increased doubts about the solvency of institutions heavily invested in those securities, making it hard for those institutions to borrow at accustomed rates.

So what made the mortgage market boom and bust?

What Didn’t Happen

The housing-finance boom and bust are not the results of a laissez-faire monetary and financial system. We didn’t have one. The boom and bust happened in a system with an unanchored government fiat money and extensive legal restrictions on financial intermediation. Nor have we had banking and financial deregulation since the bipartisan Financial Services Modernization Act (the Gramm-Leach-Bliley Act), signed by President Clinton in 1999. Far from contributing to the recent turmoil, moreover, that act has clearly been a blessing in containing it by allowing acquisitions, such as the acquisition of Bear Stearns by JPMorgan Chase or of Merrill Lynch by Bank of America, that have shielded bondholders from losses.

What Happened

There is no doubt that private miscalculation and imprudence made matters worse for more than a few lending institutions and individual borrowers. (One can’t explain an unusual cluster of errors by citing greed, which is always around, just as one can’t explain a cluster of airplane crashes by citing gravity. Anyway, the greedy aim at profits, not losses.) Such mistakes help to explain which firms have run into the most trouble. But to explain industry-wide errors we need to identify policy distortions capable of having industry-wide effects. The actual causes of our financial troubles were unusual monetary policy moves and novel federal regulatory interventions. Regulatory distortions intensified in the 1990s. Poorly chosen public policies distorted interest rates and asset prices, diverted loanable funds into the wrong investments, and twisted normally robust financial institutions into unsustainable positions.

We can group most of the unfortunate policies under two main headings: (1) Federal Reserve credit expansion that provided the means for unsustainable mortgage financing, and (2) mandates and subsidies to write riskier mortgages. The enumeration of regrettable policies here is by no means exhaustive.

Providing the Funds: Federal Reserve Credit Expansion

In the recession of 2001, the Federal Reserve System under Chairman Alan Greenspan began aggressively expanding the U.S. money supply. Year-over-year growth in the M2 monetary aggregate rose briefly above 10 percent, and remained above 8 percent entering the second half of 2003. The expansion was accompanied by the Fed’s repeatedly lowering its target for the “federal funds” (interbank short-term) interest rate. The fed funds rate began 2001 at 6.25 percent and ended the year at 1.75 percent. It was reduced further in 2002 and 2003, reaching in mid-2003 a record low of one percent, where it stayed for a year. The real Fed funds rate was negative—meaning that nominal rates were lower than the contemporary rate of inflation—for two and a half years.

How do we judge whether the Fed expanded more than it should have? One venerable norm for making fiat central bank policy as neutral as possible toward the financial market is to aim for stability (zero growth) in the volume of nominal expenditure. [2] Second-best would be a predictably low and steady growth rate of nominal expenditure. A useful measure of nominal expenditure is the dollar volume of final sales to domestic purchasers (GDP less net exports and the change in business inventories). During the two years from the start of 2001 to the end of 2002, final sales to domestic purchasers grew at a compound annual rate of 3.6 percent. During 2003, the Fed’s acceleration of credit began to show up: the growth rate jumped to 6.5 percent. For the next two years, from the start 2004 to the end of 2005, the growth rate was even higher at 7.1 percent, nearly a doubling of the initial rate. It then backed off, to 4.3 percent per annum, from the start of 2006 to the start of 2008. But the damage from an unusually rapid expansion of nominal demand had been done. [3]

A more standard way for evaluating whether the Fed was overly expansive is the “Taylor Rule,” the formula devised by economist John Taylor of Stanford University for estimating what federal funds rate would be consistent, conditional on current inflation and real income, with keeping the inflation rate at a chosen target. As calculated by the Federal Reserve Bank of St. Louis, the Fed from early 2001 until late 2006 pushed the actual federal funds rate well below the estimated rate that would have been consistent with targeting a 2 percent inflation rate for the PCE deflator. The gap was especially large—200 basis point or more—from mid-2003 to mid-2005. [4]

The excess credit thus created went heavily into real estate. From mid-2003 to mid-2007, while the dollar volume of final sales of goods and services was growing at a compounded rate of 5.9 percent per annum, real-estate loans at commercial banks were (as already noted) growing at 12.26 percent. [5] Credit-fueled demand both pushed up the sale prices of existing houses and encouraged the construction of new housing on undeveloped land. Because real estate is an especially long-lived asset, its market value is especially boosted by low interest rates. The housing sector thus exhibited a disproportionate share of the price inflation predicted by the Taylor Rule. (House prices are not, however, included in standard measures of price inflation.)

The Fed’s policy of lowering short-term interest rates not only fueled growth in the dollar volume of mortgage lending, but had unintended consequences for the type of mortgages written. By pushing very-short-term interest rates down so dramatically between 2001 and 2004, the Fed lowered short-term rates relative to 30-year rates, making adjustable-rate mortgages (ARMs) increasingly cheap relative to 30-year fixed-rate mortgages. Increasing numbers of new mortgage borrowers were drawn away from mortgages with 30-year rates into ARMs. The share of new mortgages with adjustable rates, only one-fifth in 2001, had more than doubled by 2004. Many borrowers who took out ARMs implicitly counted on the Fed to keep short-term rates low for as long as they held the mortgage. They have faced problems as their monthly payments have adjusted upward. The shift toward ARMs has compounded the mortgage quality problems arising from the regulatory mandates and subsidies described below.

The excess investment in new housing has resulted in an overbuilt housing stock. Assuming that the federal government does not follow proposals (tongue-in-cheek or otherwise) to buy up and then torch excess houses and condos, or to admit a large number of new immigrants, house prices and activity in the U.S. housing construction industry are going to remain depressed for a while. Correspondingly, the book value of existing financial assets based on housing need to be written down and losses recognized to allow a soundly based recovery. No matter how painful the adjustment process, delaying it only delays the economy’s recovery.

Mandates and Subsidies to Write Risky Mortgages

In 2001, the share of existing mortgages classified as “nonprime” (subprime or the intermediate category “Alt-A”) was below 10 percent. By 2006 it had risen to 23 percent. Meanwhile the quality of loans within the nonprime category declined, because a smaller share of nonprime borrowers made 20 percent down payments on their purchases. [6]

The federal government fostered the expansion in risky mortgages to under-qualified borrowers in several ways. It is hard to judge how much each of these contributed, but all worked to loosen lending standards.

First, the Federal Housing Administration over the years progressively loosened its down payment requirements on FHA mortgages. By 2004 the down payment requirement on the FHA’s most popular program had fallen to only 3 percent. [7] Private lenders felt compelled to offer lower down payments on non-FHA loans. Mortgages with very low down payments have had very high default rates.

Second, Congress strengthened the Community Reinvestment Act. The CRA had initially, from its passage in 1977, merely imposed reporting requirements on commercial banks. Amendments in 1995 empowered regulators to deny a bank with a low CRA rating permission to merge with another bank—at a time when the arrival of interstate banking made such approvals especially valuable—or even to open new branches. In response to the new CRA rules, some banks joined into partnerships with community groups to distribute millions in mortgage money to low-income borrowers previously considered non-creditworthy. Other banks took advantage of the newly authorized option to boost their CRA rating by purchasing special “CRA mortgage-backed securities,” that is, packages of disproportionately nonprime loans certified as meeting CRA criteria and securitized by Freddie Mac. Federal Reserve Chairman Ben Bernanke aptly commented in a 2007 speech that “recent problems in mortgage markets illustrate that an underlying assumption of the CRA—that more lending equals better outcomes for local communities may not always hold.”[8]

Third, the Department of Housing and Urban Development pressured lenders for “affordable housing” loans. Beginning in 1993, HUD officials began bringing legal actions against mortgage bankers that declined a higher percentage of minority applicants than white ones. To avoid legal trouble, lenders began relaxing their down-payment and income qualifications.[9]

Fourth and likely most important, implicit taxpayer guarantees allowed the dramatic expansion of the government-sponsored mortgage buyers Fannie Mae and Freddie Mac, at a time when Congress and HUD were pushing Fannie and Freddie to promote “affordable housing” through ever-expanding purchases of non-prime loans to low-income applicants. The two mortage giants grew to hold or guarantee around $5 trillion in mortgages, about half of the entire U.S. market. Institutional investors were willing to lend to the government-sponsored mortgage companies cheaply, despite the risk of default that would normally attach to private firms holding such highly leveraged and poorly diversified portfolios, because they were sure that the Treasury would repay them should Fannie or Freddie be unable. (It turns out that they were right.) Congress pointedly refused to moderate the moral hazard problem of implicit guarantees or otherwise to rein in the hyper-expansion of Fannie and Freddie. Warnings about Fannie and Freddie, and efforts to rein them in, came to naught because the two giants had cultivated powerful friends on Capitol Hill.

A 1992 law, as described by Bernanke, “required the government-sponsored enterprises, Fannie Mae and Freddie Mac, to devote a large percentage of their activities to meeting affordable housing goals.”[10] HUD set numerical goals for Fannie and Freddie: for the year 2000, 50 percent of their financing was to go to below-median-income borrowers; for 2005 the target rose to 52 percent. Additional goals required that increasing shares of lending go to borrowers with income under 60 percent of the median for their areas. [11]

Conclusion

The housing boom and the aftermath of its bust arose from market distortions created by the Federal Reserve, the government backing of Fannie Mae and Freddie Mac, the Department of Housing and Urban Development, and other federal interventions. We are experiencing the unfortunate results of perverse government policies, compounded in some degree by private mistakes.

The remedy for private business mistakes is bankruptcy. The remedy for the mistaken government monetary and regulatory policies that have produced the current financial train wreck is to identify and undo policies that distort housing and financial markets, and dismantle failed agencies whose missions require them to distort markets. We should be guided by recognizing the two chief errors that have been made. Cheap-money policies by the Federal Reserve System do not produce a sustainable prosperity. Hiding the cost of mortgage subsidies off-budget, as by imposing “affordable housing” regulatory mandates on banks and by providing implicit taxpayer guarantees on Fannie Mae and Freddie Mac bonds, does not give us more housing at nobody’s expense.

Lawrence H. White is the F.A. Hayek Professor of Economic History at the University of Missouri, St. Louis and an adjunct scholar at the Cato Institute. This piece summarizes his recent Cato Briefing Paper, “How Did We Get Into This Financial Mess?

Notes

[1] Federal Reserve Bank of St. Louis FRED database, series REALLN. Real Estate Loans at All Commercial Banks, http://research.stlouisfed.org/fred2/series/REALLN?cid=100. My thanks to George Selgin for drawing my attention to this series.

[2] George Selgin, Less Than Zero: The Case for a Falling Price Level in a Growing Economy (London: Institute of Economic Affairs, 1997).

[3] Federal Reserve Bank of St. Louis FRED database, Series FSDP, Final Sales to Domestic Purchasers, http://research.stlouisfed.org/fred2/series/FSDP?cid=106.

[4] Federal Reserve Bank of St. Louis, Monetary Trends, October 2008, p. 10.

[5] Federal Reserve Bank of St. Louis FRED database, series FINSAL and REALLN, year-over-year percentage changes. http://research.stlouisfed.org/fred2/series/FINSAL?cid=106, http://research.stlouisfed.org/fred2/series/REALLN?cid=100.

[6] William R. Emmons, “The Mortgage Crisis: Let Markets Work, But Compensate the Truly Needy,” The Regional Economist (July 2008).

[7] John Berlau, “The Subprime FHA,” The Wall Street Journal, October 15, 2007. 

[8] Ben S. Bernanke, “The Community Reinvestment Act: Its Evolution and New Challenges,” March 30, 2007. 

[9] Dennis Sewell, “Clinton Democrats are to Blame for the Credit Crunch,” The Spectator, October 1, 2008.

[10] Bernanke, op. cit.

[11] Russell Roberts, “How Government Stoked the Mania,” Wall Street Journal, October 3, 2008.

Response Essays

Adam Smith Was Right about Corporate CEOs’ Incentives absent Effective Regulation

Our different views prove that hindsight is often myopic. Larry White’s take is that Clintonian regulations perverted private incentives.

The boom and bust happened in a system with … extensive legal restrictions on financial intermediation. Nor have we had banking and financial deregulation since … 1999.

(One can’t explain an unusual cluster of errors by citing greed, which is always around, just as one can’t explain a cluster of airplane crashes by citing gravity. Anyway, the greedy aim at profits, not losses.) [T]o explain industry-wide errors we need to identify policy distortions capable of having industry-wide effects. The actual causes of our financial troubles were unusual monetary policy moves and novel federal regulatory interventions. Regulatory distortions intensified in the 1990s.

Perverse Compensation Systems are the Key

I disagree with Larry’s theses, but have space to demonstrate only an alternative perverse incentive. What went wrong is that modern compensation systems did not “align” interests, but rather created perverse incentives to engage in accounting “control fraud,” where the CEO uses an apparently legitimate firm as a “weapon” to defraud creditors and shareholders. [1] No regulation forced any lender to make a bad loan. Larry misses the key dynamic: “The greedy” do not “aim at profits, not losses” when compensation schemes are perverse. They maximize short-term accounting “profits” in order to increase their wealth. Making bad loans, growing rapidly, and extreme leverage maximize “profits.” Bad borrowers agree to pay more and it is impossible to grow rapidly via high quality lending. Lending to the uncreditworthy requires the CEO to suborn controls, maximizing “adverse selection.” This produced an “epidemic” of mortgage fraud, particularly in the unregulated nonprime sector. The FBI began warning in September 2004 about the mortgage fraud “epidemic.” [2] Fraudulent loans cause huge direct losses, but the epidemic also hyper-inflated and extended the housing bubble, and eviscerated trust, causing catastrophic indirect losses. When we do not regulate or supervise financial markets we, de facto, decriminalize control fraud. The regulators are the cops on the beat against control fraud—and control fraud causes greater financial losses than all other forms of property crime combined.

The most relevant economic works for understanding these crises are by Akerlof and Romer, Galbraith, and Minsky. Akerlof and Romer explain why “looting” (control fraud) can occur and the fraudulent steps looters take to optimize short-term accounting profits (which destroy the firm). [3] Note that they are writing about a form of a “market for lemons” in which the CEO maximizes information asymmetry. The failure of economists discussing the ongoing crises to cite the work of a Nobel laureate writing in the core of his expertise demonstrates why we have failed to learn the proper lessons from prior financial crises. James Galbraith extends Akerlof and Romer’s analysis to show why the state aids fellow control frauds. [4] Minsky describes the “Ponzi” phase of a crisis and why financial instability reoccurs. [5]

Modern executive compensation systems suborn internal controls. (Control frauds do not “defeat” controls—they turn them into oxymoronic allies.) The Business Roundtable’s spokesman, Franklin Raines, Fannie Mae’s former CEO, explained in a Business Week interview what caused the epidemic of accounting control fraud that became public in 2001 with Enron’s failure.

[Businessweek:We’ve had a terrible scandal on Wall Street. What is your view?

[Raines:] Investment banking is a business that’s so denominated in dollars that the temptations are great, so you have to have very strong rules. My experience is where there is a one-to-one relation between if I do X, money will hit my pocket, you tend to see people doing X a lot. You’ve got to be very careful about that. Don’t just say: “If you hit this revenue number, your bonus is going to be this.” It sets up an incentive that’s overwhelming. You wave enough money in front of people, and good people will do bad things.

Unfortunately, Raines’ insights stemmed from his implementation of just such a system. Raines knew that the unit that should have been most resistant to this “overwhelming” financial incentive, Fannie Mae’s Internal Audit department, had succumbed to it. Mr. Rajappa, its head, instructed his internal auditors in a formal address in 2000 (and provided the text to Raines, who praised it):

By now every one of you must have 6.46 [the earnings per share target] branded in your brains. You must be able to say it in your sleep, you must be able to recite it forwards and backwards, you must have a raging fire in your belly that burns away all doubts, you must live, breath and dream 6.46, you must be obsessed on 6.46…. After all, thanks to Frank [Raines], we all have a lot of money riding on it…. We must do this with a fiery determination, not on some days, not on most days but day in and day out, give it your best, not 50%, not 75%, not 100%, but 150%. Remember, Frank has given us an opportunity to earn not just our salaries, benefits, raises, ESPP, but substantially over and above if we make 6.46. So it is our moral obligation to give well above our 100% and if we do this, we would have made tangible contributions to Frank’s goals [emphasis in original].

Internal audit is the “anti-canary” in the corporate “mines”; by the time it is suborned every other unit is corrupted. The CEO cannot send out a memo urging accounting fraud, but he can safely send the same message through his bonus plan. He does not have to order, or be aware of, the specific frauds—the employees will do whatever is needed to “earn” their top bonus. The CEO simply communicates—by inaction—that he does not care how they meet the target.  

Fannie and Freddie were accounting control frauds that became insolvent because of their private nature. It is naïve to believe that either purchased loans or mortgage-backed securities (MBS) to help poor people. Their senior officers caused Fannie and Freddie to make purchases for the same reason their private peers did: to maximize accounting income. None of these peers had government guarantees, yet “private market discipline” increased their incentive to engage in accounting fraud.

Consider a CFO in 2006 who knows all of this: that there is a housing bubble, that non-prime loans maximize “adverse selection,” that there is an epidemic of mortgage fraud, and that the (declining!) spread on non-prime loans is inadequate. If he does not purchase nonprime paper and lever up, his bank will report far lower earnings than its peers. Firm bonuses and stock appreciation will be lower — sometimes by billions of dollars. The average tenure of a CFO is less than three years. He faces intense pressures to emulate his peers—even if it dooms the firm. This environment creates a “Gresham’s Law” dynamic in which perverse incentives drive good underwriting out of circulation.

The claim that Fannie and Frieddie took excessive risk because of an implicit governmental guarantee stands refuted. They lost market share because they took relatively less exposure to non-prime loans than their peers. Moreover, Fannie and Frieddie’s regulator had power to bar them from purchasing non-prime MBS. The Bush administration did not do so because it favored the expansion of non-prime lending throughout the developing bubble. It expanded FHA’s nonprime loans and opposed regulating nonprime lenders.

The Incidence and Nature of Mortgage Fraud

The defining element of fraud that distinguishes it from other forms of larceny is deceit. Fraud frequently goes undiscovered. Fraud reports understate incidence and are biased. The most competent frauds are least likely to be discovered. Insured depository institutions must file Suspicious Activity Reports (SARs) when they discover credible information of a crime. Many commercial banks and S&Ls, therefore, often filed SARs about mortgage fraud. Mortgage banking firms were essentially unregulated by the federal government and generally did not file SARs when they found fraud. Investment bankers, in the four years during the peak of the epidemic, filed only36 SARs. [7]

Given the fact that mortgage and investment banks were (allegedly) the principal victims of mortgage fraud, why weren’t they the principal SARs filers? One only spots mortgage fraud if one conducts underwriting (and accounting control frauds abhor it), and the last thing a control fraud wants is to invite the FBI’s attention.

In FY 2007, there were 52,868 Suspicious Activity Reports of mortgage fraud—a 40 percent increase over fiscal year 2006. [8] Mortgage fraud has grown rapidly this decade and has overwhelmed the FBI’s resources. [9] Weak regulation and perverse strategic behavior by control frauds led to pervasive underreporting of fraudulent subprime mortgages:

In 2005, 52% of subprime mortgages were originated by companies with no federal supervision, primarily mortgage brokers and stand-alone finance companies. Another 25% were made by finance companies that are units of bank-holding companies and thus indirectly supervised by the Federal Reserve; and 23% by regulated banks and thrifts. [10]

Because insured banks and S&Ls originated only 23 percent of subprime loans in 2005, the most obvious adjustment to using SARs to estimate total subprime mortgage fraud would be to multiply the annual SAR total by five. However, unregulated mortgage lenders made a disproportionate share of the fraudulent loans. The largest mortgage control frauds cause grossly disproportionate losses and represent an enormous percentage of the total incidence of mortgage fraud. According to a Government Accounting Office report:

Of the top 25 originators of subprime and Alt-A loans in 2006 (which accounted for over 90 percent of the dollar volume of all such originations):

  • 21 were nonbank lenders, including 14 independent lenders and 7 nonbank subsidiaries of banks, thrifts, or holding companies.
  • the 21 nonbank lenders accounted for 81 percent of the dollar volume (44 percent was originated by independent lenders and 37 percent by nonbank subsidiaries of banks, thrifts, or holding companies).[11]

The worst mortgage frauds operated primarily in the unregulated sector. The New York Times reported:

43 percent of the cases sampled in the study involved misrepresentation of income, assets or debts. The next-largest category was forged documents, totaling 28 percent of the sampled loans. Mortgage brokers initiated the loans on 64 percent of the reports involving misrepresentation of income, assets or debt…

The FBI erred by partnering with the Mortgage Bankers Association (MBA)—which represented the worst control frauds. The MBA’s priority was blocking regulation of mortgage banking—not stopping mortgage fraud.

SARs underreport nonprime mortgage fraud at insured depositories. Many frauds are not spotted. Nonprime lenders ended verification to make it easy, fast, and cheap for them to approve uncreditworthy borrowers. Verification is the most effective means to deter and identify fraud, so the worst lenders had enormous undiscovered fraud. The Times report continued:

Indeed, according to a report on mortgage fraud released Thursday by the Financial Crimes Enforcement Network, a unit of the Treasury Department, only 31 percent of suspected fraud was detected before loan disbursements in the 12 months ended March 31, 2007. On stated income loans, only 19 percent of the cases of suspected fraud were detected before the loans were financed, versus 33.5 percent on more fully documented loans.

The federally insured control frauds’ incentive was not to file SARs.

The FBI reports that, based on existing investigations, 80 percent of all reported fraud losses arise from fraud for profit schemes that involve industry insiders. [12]

A survey from Fitch Ratings also showed endemic nonprime mortgage fraud.

Characteristics by percentage of the 45 files reviewed included (loans may appear in more than one finding):

  • 66% Occupancy fraud (stated owner occupied — never occupied), based on information provided by borrower or field inspector
  • 51% Property value or condition issues — Materially different from original appraisal, or original appraisal contained conflicting information or items outside of typically accepted parameters
  • 48% First Time Homebuyer — Some applications indicated no other property, but credit report showed mortgage information
  • 44% Payment Shock (defined as greater than 100% increase) — Some greater than 200%
  • 44% Questionable stated income or employment — Often in conflict with information on credit report and indicated to be outside “reasonableness” test
  • 22% Hawk Alert — Fraud alert noted on credit report
  • 18% Credit Report — Questionable ownership of accounts (name or social security numbers do not match)
  • 17% Seller Concessions (outside allowed parameters)
  • 16% Credit Report — Based on “authorized” user accounts
  • 16% Strawbuyer/Flip scheme indicated based on evidence in servicing file
  • 16% Identity theft indicated
  • 10% Signature fraud indicated
  • 6% Non-arms length transaction indicated [13]

Note that Fitch did not conduct any investigation. It identified frauds obvious from a review of the loan files.

SARs filed by federally regulated lenders seriously underreport mortgage fraud. The reported number is enormous. My belief, consistent with fraud incidence found in file reviews, is that it represents roughly 5 percent of the true incidence, which implies roughly one million fraudulent mortgage loans were made in FY 2007. [14]

The testimony of Thomas J. Miller, Attorney General of Iowa, at a 2007 Federal Reserve Board hearing shows why fraud losses are enormous:

Over the last several years, the subprime market has created a race to the bottom in which unethical actors have been handsomely rewarded for their misdeeds and ethical actors have lost market share…. The market incentives rewarded irresponsible lending and made it more difficult for responsible lenders to compete. Strong regulations will create an even playing field in which ethical actors are no longer punished.

Despite the well documented performance struggles of 2006 vintage loans, originators continued to use products with the same characteristics in 2007.

[M]any originators … invent … non-existent occupations or income sources, or simply inflat[e] income totals to support loan applications. A review of 100 stated income loans by one lender found that a shocking 90% of the applications overstated income by 5% or more and almost 60% overstated income by more than 50%. Importantly, our investigations have found that most stated income fraud occurs at the suggestion and direction of the loan originator, not the consumer.

It is no answer to say, “they did not underwrite because they sold their loans.” That model can only work if an ultra-sophisticated entity buys. Investors must stop accounting control fraud if markets are to be efficient. But they do not. Elite investors’ Potemkin models created illusory sophistication. Their only skill was in the intricate footwork required in the Totentanz with their partners at the rating agencies in which they structured toxic nonprime paper into risk-free “AAA.”

As Paul Volcker has concluded, “modern finance” has failed the market test. Its policies optimize the environment for control fraud and create perverse dynamics that create recurrent financial crises.

William K. Black is associate professor of economics and law at the University of Missouri, Kansas City and author of The Best Way to Rob a Bank is to Own One.

Notes

[1] Black, William K., 2005. The Best Way to Rob a Bank is to Own One, (Austin, TX: U. Tex. Press).

[2] “FBI warns of mortgage fraud ‘epidemic’: Seeks to head off ‘next S&L crisis‘” CNN, Sept. 17, 2004.

[3] Akerlof, George, and Paul M. Romer, 1993. “Looting: The Economic Underworld of Bankruptcy for Profit.” Brookings Papers on Economic Activity 2: 1-73.

[4] Galbraith, James, 2008. The Predator State (NY: Simon & Schuster).

[5] Minsky, Hyman, 1982. “The Financial-Instability Hypothesis: Capitalist processes and the behavior of the economy”, in Kindleberger and Laffargue, editors, Financial Crises. See also recent works by Wray, R. and Kregel, J. extending and applying Minsky’s theories to the ongoing crises.

[6] Raines’ observation about the perverse impact of such compensation systems has been confirmed. Lucian Bebchuk & Jesse Fried, Pay Without Performance: The Unfulfilled Promise of Executive Compensation (2004): at 183-85. “Executive Compensation at Fannie Mae: A Case Study of Perverse Incentives, Nonperformance Pay, and Camouflage.” Lucian A. Bebchuk and Jesse M. Fried. Journal of Corporation Law, 2005, Vol. 30, pp. 807-822 (at p. 811).

Even Michael Jensen now warns that they caused endemic accounting and securities fraud. Jensen, Michael. “Putting Integrity Into Finance Theory and Practice: A Positive Approach” (June 9, 2007) (available on SSRN). The Shadow Financial Regulatory Committee has decried rating agencies’ perverse compensation systems. Statement No. 257: “Reliance on Third-Party Credit Ratings” (February 11, 2008).

[7] Gretchen Morgenson, “Fair Game: A Road Not Taken by Lenders” (April 6, 2008). http://www.nytimes.com/2008/04/06/business/06gret.html?scp=2&sq=mortgag…

[8] “Mortgage Loan Fraud: An Update of Trends Based upon an Analysis of Suspicious Activities Reports,” [pdf] Financial Crimes Enforcement Network, 2008. 

[9] “’Mortgage fraud is the fastest-growing white-collar crime affecting the United States today,’ says Karen Spangenberg, chief of the financial crimes section of the FBI’s criminal investigative division.” Mortgage Shakedown,” Mortgage Banking, August 1, 2006. 

[10] Greg Ip & Damiam Paletta. “Regulators Scrutinized In Mortgage Meltdown.” (March 22, 2007). The Wall Street Journal Online

[11] GAO briefing of the Committee on Financial Services House of Representatives. Subject: Information on Recent Default and Foreclosure Trends for Home Mortgages and Associated Economic and Market Developments (October 16, 2007: 54).

[12] “Mortgage Fraud: Strengthening Federal and State Mortgage Fraud Prevention Efforts” (2007). Tenth Periodic Case Report to the Mortgage Bankers Association, produced by MARI.

[13] http://www.fitchratings.com/corporate/reports/report_frame.cfm?rpt_id=356624.

[14] Note that this discussion also refutes two of Larry’s regulatory claims: (1) that the Community Reinvestment Act (CRA) was a material contributor to the crisis and (2) that inadequate regulation did not contribute to the crisis. The principal nonprime lenders were not even subject to the CRA. The principal buyers of nonprime paper were not subject to the Act. The CRA had been in place for decades, yet caused a crisis during a decade when the Bush administration gutted CRA supervision. The regulated lenders that were nonprime specialists (and made the great bulk of nonprime loans in that sector) did not have to do so to comply with the CRA. Lenders made nonprime loans to maximize accounting income. The CRA never requires a lender to make a bad loan.

Larry also misses the entire concept of desupervision. When the OTS Director brings a chainsaw to a press conference to destroy the rules and when he and the head of the FDIC gut the agency staff effective supervision ends. When the guardians don’t guard, rules fail.

Fundamental Origins of the Housing Boom and Bust

 2008 was a landmark year in financial history.  Two major financial institutions — Washington Mutual and Lehman Brothers — failed.  Other major institutions such as the insurer AIG and the mortgage institutions Federal National Mortgage Association (Fannie) and the Federal Home Loan Mortgage Corporation (Freddie) survived only because of government interference.  By autumn, interbank lending and commercial paper markets could not function normally because some of the participants were thought to have failure risks. The malfunctioning of these markets made near-term failures even more likely, which presumably aggravated the malfunctioning.

For good reason, there is concern that these financial events will damage the wider economy.  Predicting the extent of wider damage requires an understanding of the causes of the financial turmoil.  The purpose of this essay is to articulate a couple of hypotheses as to the fundamental origins of the crisis — the tastes, technologies, market structures, and public policies that may have caused these events.  I particularly focus on causes of the cycle in housing prices and construction.

All of the hypotheses have a common chain of logic: something caused a massive housing boom between the years 2000-2006.  The housing boom ended, requiring seismic adjustments by financial institutions, investors, homeowners, and the housing industry.  In some cases, that “something” is tastes and technologies, which means that public policy should not and cannot prevent these events, but can at most affect their incidence and marginally change their magnitude.  In other cases, events were caused or exacerbated by innate market weaknesses, which opens the possibility that public policy might improve matters.

Housing cycle theories differ in a couple of ways.  Some of them focus on changes in tastes, technologies, and public policies during the boom years.  Others focus on the expectations of market participants during those years about events that would occur in the future; those events may pertain to tastes or technologies. Among the expectation theories, some of them are “rational” and others feature bubbles or “irrational exuberance.”  I explain how expectations about tastes differ from expectations about technologies in terms of how the housing bust might affect the economy going forward.

Analytical Framework and the Role of Anticipation

A variety of measures show real housing prices much higher at their peak in 2006 than they were in 2000 or during much of the 1990s.  The OFHEO index shows prices grew 30 percent more 2000-2006 than the PPI for housing construction.  The Case-Shiller index shows 90 percent growth in housing prices relative to PPI over that period. [1]   Moreover, year 2000 real prices were somewhat more than they were during much of the 1990s.  Housing construction activity followed a time pattern like that for prices: increasing through 2006, and then dropping steeply thereafter.  Why did housing prices appreciate so much?  Why did they fall so quickly?

A physical housing structure has value because it helps provide people with flows of shelter, privacy, and convenience. However, landlords and homeowners know that housing services flows are not produced with structures alone: also important are intermediate inputs of brokerage, management and banking services that can match and maintain the physical structures with the families who live in them and the investors who build them.  As shown in the national accounts, banking, real estate brokerages, and others outside the construction industry annually supply hundreds of billions of dollars worth of the intermediate inputs to the housing industry.

Housing occupants pay rent for access to housing services.  In the case of a tenant-landlord relationship, the rental transaction is explicit: the renter writes a monthly check to the landlord.  In the case of owner-occupied housing, the rental transaction is implicit but economically just as real.  The important point here is that the entire rent paid by occupants does not go to the investor in the structure — much of it goes to the suppliers of the intermediate inputs. The purchase price of a house (that is, the structure itself) is the expected present value of just that part of the rent that is left over after intermediate inputs are paid.  The demand for residential structures is a derived demand; it depends not only on the demand for housing services but also the costs of intermediate inputs.  The rental cost of housing services exceeds the capital rental received by the owner of a residential structure by the amount of intermediate input expenditure.

It is easy to show that the housing boom was possible only because of anticipated changes in housing service demand or in the costs of housing’s intermediate inputs.  Specifically, the CPI for housing rent was only about four percent higher relative to the overall CPI during the years 2002-2006, as compared to its value in 2000.  If the fraction of housing rent going to the owner of the structure were constant, and real housing rent was expected to remain four percent higher for the indefinite future, at most it would rationalize only four percent higher prices for the structures. The CPI for housing rent may be imperfect, but it seems clear that the housing boom involved dramatic increases in housing prices relative to housing rent. [2] Moreover, the intermediate inputs compensated by housing rent were pretty constant over these years (Mulligan, 2008b).  Thus, home buyers during the boom years and/or their lenders must have anticipated that (a) housing rents would someday increase (or that a set of new buyers would someday emerge who expect housing rents to increase) or (b) an increasing fraction of housing rents would go to owners of the residential structures rather than suppliers of intermediate inputs.

Some optimism about the costs of intermediate inputs was warranted because applied information technology was rapidly advancing during the 2000s.  Much of the banking and real estate value added relates to information.  Bankers screen borrowers and value heterogeneous collateral.  Real estate brokers match heterogeneous families to heterogeneous properties.  Hall and Woodward (2008) claim

Recent years have seen great improvements in data, especially the introduction of credit scores, which gave lenders new powers to forecast mortgage defaults and to adjust interest rates offered to prospective borrowers. In 1990, credit scores were rare; by 1996, they were standard.

Perhaps lenders also expected to use information technology to better monitor and collect on loans, and therefore put subprime lending programs in place.  Real estate brokers might also expect to benefit from technical progress: Virtual Office Web sites and their technological descendants might significantly reduce the brokerage resources needed to match homes with the persons who value them most. 

Perhaps market participants hoped that information technological advance would eventually reduce the fraction of housing rent going to banks and brokers (increase the fraction going to owners of residential structures), and thereby support a greater equilibrium ratio of structures’ prices to housing rents.  In this view, these expected technological advances (i) increased the equilibrium expected long run housing stock, (ii) increased short run housing prices and housing construction, (iii) increased short run expenditures on the intermediate inputs, and (iv) reduced the equilibrium expected long run housing service rent. [3]

Distinctions between Taste Change, Technological Change, and Bubbles

The observation that housing prices 2000-2006 were high and rising is sometimes taken as evidence of a rational bubble or of “irrational exuberance.”  However, this observation is also consistent with a standard q-theory model in which expectations are rational, investors understood that real housing prices would eventually return to construction cost (and therefore to the levels of the late 1990s), and transversality conditions hold.  Specifically, future technological advances (or future demand shifts — see below) create (v) anticipated capital gains for structures owners for the time prior to the ultimate increase in the fraction of housing rents going to structures owners. When the news arrives that the future technology will be better, structures prices immediately jump up but nothing improves the rental income received by a given structure in the short run.  Those structures therefore earn less rent per dollar of capital for the time prior to the technological advance.  Owners of structures rationally anticipate steady capital gains that compensate them over that period for the low rent per dollar invested. [4] If the technology expectations had been fully realized, housing prices would have peaked when the technological advances were complete, after which time housing prices would slowly fall back to construction cost while the housing stock rose to its long run amount and expenditures on the intermediate inputs were lower.

As with any anticipation, the ultimate results may be better or worse than expected.  In this case, the technological advances were not fully forthcoming, or not forthcoming as rapidly as once anticipated.  The bad news brings down housing prices.  If the news were bad enough that the revised long run housing stock were at or below the actual stock at the time the bad news arrived, then housing prices would fall at least to construction cost.

A related hypothesis is that the housing boom was caused by the anticipation of high future housing demand, and that the crash occurred when it was learned that the high demand would never materialize, or would materialize later than was originally anticipated.  Anticipated taste changes have some of the same effects as anticipated technological progress, because the taste for housing services and the costs of intermediate inputs both can cause shifts of the derived demand for residential structures.  The implications (i) – (iii) for pricing, short run housing construction, and short run intermediate input use are therefore much the same.  For the same reason, the expectation of high demand in the future creates anticipated capital gains for structures owners for the time prior to the ultimate increase in housing service rents, even though market participants understand that housing prices will eventually fall back to construction cost.

Anticipated taste and technical change are different in terms of expected future housing market outcomes and different in terms of how current behavior responds to those expectations.  If and when the time ultimately comes when housing service demand is high, housing service rents are higher until the housing stock fully adjusts. Intermediate input use is higher than it was before the demand increase, merely because of the greater housing stock.  As a result, fewer resources are available for the nonresidential sector.  In contrast, the arrival of a better banking or brokerage technology ultimately reduces housing service rents as the stock of housing accumulates.  Aggregate expenditure on intermediate inputs may ultimately be lower because of the technical progress — leaving more resources available for the nonresidential sector.

Because both the taste and technology hypotheses explain the housing bust as unfulfilled expectations, we cannot directly measure the equilibrium amounts of housing service rent and intermediate input use that would have occurred had the optimistic expectations been fulfilled.  However, the hypotheses differ in terms of how market participants would have behaved during the boom and after the bust because they differ about the effect of the housing events on the amount of resources that would be available for the nonresidential sector.  In other words, anticipated technological progress creates both a housing price increase and an aggregate wealth effect, whereas anticipated high demand for housing services may only increase housing prices without an aggregate wealth effect. [5] Bad news about future technological progress creates both a housing price crash and an adverse aggregate wealth effect, whereas anticipated high demand for housing services may only reduce housing prices without an aggregate wealth effect.

Quantitative empirical work more easily rejects taste and technology theories than theories of bubbles and irrational exuberance.  Rational expectations theory limits how high housing prices can go, based on limits on population growth, housing budget shares, the time until tastes and technologies are expected to change, and the fact that intermediate inputs cannot be negative or be more complementary with structures than fixed proportions.  I have not yet seen quantitative work on the amount housing prices should have increased based on fundamentals. [6]

Easy Money, Lax Regulation, and Other Possible Mortgage Distortions

Another hypothesis is that housing was excessively subsidized for a period of time roughly coinciding with the housing boom, and that the housing price reductions occurred as the end of the subsidy came near.  In one version of this view, housing prices were high because the rental income from housing services — after taxes and subsidies — was higher than normal during the boom.  This view is quite different from the expectations models sketched above, because in theory the (properly measured) rental income received by the owners of structures would be high during the housing boom, even when measured as a ratio to housing prices.  As a result, owners of structures anticipate capital losses — even while housing construction booms — that compensate them for the high short-run ratio of rental income to structures price.  To the extent that the subsidies served to reduce expenditures on intermediate inputs or were concentrated in the owner-occupied sector, housing rents paid by tenants to landlords per unit of housing services should have been low during the boom because rental housing has to compete with owner-occupied housing.

If this version of the subsidy hypothesis has any truth to it, it must be outweighed by the taste, technology, or bubble hypotheses because it is qualitatively inconsistent with two basic facts.  First, housing rental rates were not low when measured per unit of housing services.  Second, structures owners received continual capital gains, not capital losses, throughout the period of rapid housing construction.

Another version of the subsidy hypothesis says that public policy encouraged low mortgage rates, which raised housing prices.  I believe that housing prices would not have gotten so high if mortgage rates had been higher, but low mortgage rates may not explain why 2006 housing prices were so high relative to housing prices in 2003 or 2008.  30-year fixed-mortgage rates were around six percent per year for most of the boom, and continue to be about six percent.  The housing boom witnessed the increased availability of new mortgage products, such as interest-only loans.  However, more work is needed to determine the degree to which the products reflected high prices, rather than product availability raising prices.

Conclusions

The methodology of economics specifies a clear recipe for assigning blame for today’s financial crisis.  First, alternative hypotheses are articulated.  Second, the alternatives are compared on the basis of empirical predictions.  Third, empirical tests are conducted to gauge the relative importance of the alternative hypotheses.  Unfortunately, October and November of 2008 witnessed much confident commentary as to the causes of the crisis, without much open, public application of the economic methodology.

My work attempts to follow these steps as regards the housing cycle from 2000 to the present.  I explain that the housing cycle cannot necessarily be blamed on (supposedly) innate weaknesses of the marketplace, because a well-functioning marketplace that anticipated eventual technical progress in mortgages and real estate brokerage and/or eventually high future housing demand would have produced high and rising prices of housing structures.  If and when future technical progress or demand was learned to be less than expected, housing prices would have crashed even in a well-functioning market.  Of course, these theoretical results do not prove that the housing market was functioning well.  But nor do the housing market observations 2000-2008 necessarily reveal that the market has innate weaknesses, or that its innate weaknesses have by themselves created much social cost.

For similar reasons, it is difficult to know whether the high and rising housing prices prior to 2006 reflected optimism about tastes or technologies.  However, measurement of the existence and direction of wealth effects on behavior can help gauge the relative importance of these two kinds of impulses.  Bad news about future technological progress creates both a housing price crash and an adverse aggregate wealth effect, whereas anticipated high demand for housing services may only reduce housing prices without an aggregate wealth effect.

Space constraints limit the detail with which I can connect housing events to the financial turmoil, but a few basic results can be summarized. [7] First, housing construction and the value of existing nonfinancial businesses is expected to follow a qualitatively similar cycle — rising when housing prices rise and crashing when housing prices crash.  Investment in the nonfinancial nonresidential sector is expected to follow the opposite pattern – low during the housing boom and high since the housing bust.  Second, a variety of risk-sharing arrangements — especially collateralized home mortgages — serve to dissipate housing losses into the wider economy, with the banking sector as the conduit (Arrow, 2008).  Banks’ costs of old mortgages are sunk, so there is no efficiency reason why massive losses by themselves would affect bank operations.  However, financial institutions may merge and some of them lose market share as the financial market dissipates the housing losses and the housing market changes the types of financial services it demands.  Third, the labor market may be harmed in the short run as lenders mean-test their forgiveness of prior mortgage debts (Mulligan 2008a).  Thus, some of the economic and financial adjustments are largely unrelated to the fundamental cause of the housing cycle, and just stem from the fact that housing prices were once high and now are not.

Both optimal public policy and some other economic adjustments depend on the causes of the housing cycle.  In particular, the aggregate wealth effects are different depending on whether the housing cycle was caused by tastes, technology, or public policy.  If the housing bust was an adverse aggregate wealth effect, the economy will have to adjust to less consumption and more long-run employment.  Finally, if housing prices are high and rising because of the advent of technological progress or increased demand, a mis-interpretation of events as “irrational exuberance” might rationalize public policies that retard housing supply, create housing shortages, and  ultimately cause harm.  Empirical testing of carefully articulated alternative hypotheses are needed to make the ultimate judgment.  Until then, we honestly do not know whether market participants’ unregulated planning for the future was part of the problem, or should be an integral part of the solution.

Casey B. Mulligan is professor of economics at the University of Chicago.

Notes

I thank Kevin Murphy for first suggesting to me some of the possible roles of technical change.  I also appreciate the comments and patience of Ed Glaeser, who experienced an even more burdensome exposition of these ideas.  I will provide updates on this matter on my blog www.panic2008.net.

[1] The OFHEO index is considered by Glaeser, Gyourko, and Saiz, (2008).  Mulligan and Threinen (2008) compare growth rates for the OFHEO and Case-Shiller indices 2000-2006.

[2] To the extent that owner-occupied housing is a different good from rental housing, the housing rent CPI could have understated the increase of implicit rental rates for owner-occupied housing during the housing boom.

[3] I assume that intermediate inputs and physical structures are complements in the production of housing services.

[4] This is a modified version of Poterba (1984) and Topel and Rosen (1988).  For the phase diagrams for such a model, see Summers (1981).

[5] Buiter (2008) has a model in which housing booms do not have aggregate wealth effects, but his model has no role for the kinds of technological change I have discussed.  An increase in the size of the economy’s production set has an aggregate wealth effect by any definition.

[6] One obstacle to such work is that the housing price in economic theory (the cost of a structure that would produce one unit of housing services) is different from the housing price measured by Case-Shiller or OFHEO or the price relevant to mortgages (the cost of a given property, including all of its improvements or depreciation).

[7] Some of the additional detail can be found in Mulligan and Threinen (2008).

References

Arrow, Kenneth.  “Risky Business.”  guardian.co.uk.  October 15, 2008.

Buiter, Willem H.  “Housing Wealth Isn’t Wealth.”  NBER working paper no. 14204, July 2008.

Glaeser, Edward L., Joseph Gyourko, and Albert Saiz.  “Housing Supply and Housing Bubbles.”  Forthcoming, Journal of Urban Economics, 2008.

Hall, Robert E. and Susan E. Woodward.  “The Financial Crisis and Recession.”  Manuscript, Stanford University, November 24, 2008. (http://sites.google.com/site/woodwardhall/)

Mulligan, Casey B.  “A Depressing Scenario: Mortgage Debt Becomes Unemployment Insurance.”  NBER working paper no. 14514, November 2008a.

Mulligan, Casey B.  “Housing Wealth and Aggregate Wealth Effects.” Manuscript, University of Chicago, November 2008b.

Mulligan, Casey B. and Luke Threinen.  “Market Responses to the Panic of 2008.”  NBER working paper no. 14446, October 2008.

Poterba, James M.  “Tax Subsidies to Owner-Occupied Housing: An Asset Market Approach.”  Quarterly Journal of Economics.  99, November 1984: 729-52.

Summers, Lawrence.  “Taxation and Corporate Investment: A q Theory Approach.”  Brookings Papers on Economic Activity. 1981(1): 67-127.

Topel, Robert and Sherwin Rosen.  “Housing Investment in the United States.”  Journal of Political Economy.  96(4), August 1988: 718-740.

Liquidity, Default, Risk

Larry White is the best of the Austrians — the most persuasive, the most thoughtful, and the most knowledgeable of the economists working in the Austrian monetary theory tradition, which is an essential part of our collective diversified intellectual portfolio in our age in which economic theory is so underdeveloped that, as John Maynard Keynes wrote in an earlier and somewhat similar episode, “[w]e lack more than usual a coherent scheme…. All the political parties alike have their origins in past ideas and not in new ideas…. It is not necessary to debate the subtleties of what justifies a man in promoting his gospel by force; for no one has a gospel…”

Nevertheless, I think that what Larry White has written misses the big point about what really has happened. So let me try to lay out what the situation looks like to me.

Think of it this way: two years ago we lived in a world in which the wealth of global owners of capital was some $80 trillion — that was the market value of all of their property rights to dividends and contract rights to interest, rent, royalties, options, and bonuses. Now over time the wealth of global capital fluctuates, and it fluctuates for five reasons:

  1. Savings and Investment: Savings that are transformed into investment add to the productive physical — and organizational, and technological, and intellectual — capital stock of the world. This is the first and in the long run the most important source of fluctuations — in this case, growth — in global capital wealth.
  2. News: Good and bad news about resource constraints, technological opportunities, and political arrangements raise or lower expectations of the cash that is going to flow to those with property and contract rights to the fruits of capital in the future. Such news drives changes in expectations that are a second source of fluctuations in global capital wealth.
  3. Default Discount: Not all the deeds and contracts will turn out to be worth what they promise or indeed even the paper that they are written on. Fluctuations in the degree to which future payments will fall short of present commitments are a third source of fluctuations in global capital wealth.
  4. Liquidity Discount: The cash flowing to capital arrives in the present rather than the future, and people prefer — to varying degrees at different times — the bird in the hand to the one in the bush that will arrive in hand next year. Fluctuations in this liquidity discount are yet a fourth source of fluctuations in global capital wealth.
  5. Risk Discount: Even holding constant the expected value and the date at which the cash will arrive, people prefer certainty to uncertainty. A risky cash flow with both upside and downside is worth less than a certain cash flow by an amount that depends on global risk tolerance. Fluctuations in global risk tolerance are the fifth and final source of fluctuations in global capital wealth.

In the past two years the wealth that is the global capital stock has fallen in value from $80 trillion to $60 trillion. Savings has not fallen through the floor. We have had little or no bad news about resource constraints, technological opportunities, or political arrangements. Thus (1) and (2) have not been operating. The action has all been in (3), (4), and (5).

As far as (3) is concerned, the recognition that a lot of people are not going to pay their mortgages and thus that a lot of holders of CDOs, MBSs, and counterparties, creditors, and shareholders of financial institutions with mortgage-related assets has increased the default discount by $2 trillion. And the fact that the financial crisis has brought on a recession has further increased the default discount — bond coupons that won’t be paid and stock dividends that won’t live up to firm promises — by a further $4 trillion. So we have a $6 trillion increase in the magnitude of (3) the default discount. The problem is that we have a $20 trillion decline in market values.

The problem is made bigger by the fact that for (4), the Federal Reserve, the European Central Bank, and the Bank of England have flooded the market with massive amounts of high-quality liquid claims on governments’ treasuries, and so have reduced the liquidity discount — not increased it — by an amount that I estimate to be roughly $3 trillion. Thus (3) and (4) together can only account for a $3 trillion decrease in market value. The rest of that decline in the value of global capital — all $17 trillion of it — thus comes by arithmetic from (5): a rise in the risk discount. There has been a massive crash in the risk tolerance of the globe’s investors.

Thus we have an impulse — a $2 trillion increase in the default discount from the problems in the mortgage market — but the thing deserving attention is the extraordinary financial accelerator that amplified $2 trillion in actual on-the-ground losses in terms of mortgage payments that will not be made into an extra $17 trillion of lost value because global investors now want to hold less risky portfolios than they wanted two years ago.

From my standpoint, the puzzle is multiplied by the fact that we economists have what we regard as pretty good theories about (4) and (5), and yet those theories do not seem to work at all. As far as the liquidity discount (4) is concerned as long as we love our children as ourselves (and most of us do) and as long as we have access to and can credibly pledge collateral for financial transactions (and we can) the magnitude of the liquidity discount should be roughly equal to the technologically and organizationally driven rate of labor productivity growth divided by the intertemporal elasticity of substitution. The technologically and organizationally driven rate of labor productivity growth is a fairly steady 2 percent per year. The intertemporal elasticity of substitution is in the range from 1/2 to 1. The liquidity premium should be in the range of 2% to 4% per year in real terms — and no central bank should be able to drop it to 2% per year by a few open-market operations: big moves in the liquidity premium should require big moves in expected future growth rates of consumption. Perhaps in the old days — back when banknotes and demand deposits backed fractionally by gold or central-bank reserves were the only liquid stores of value, the only means of payment, the only mediums of exchange, or the even older days when the king’s picture on a disc of gold was it, and when the torturers of the Mint and the Tower were standing by — things were different and credit expansion via the use of the seigniorage power could have greater effects. But today the ability of central banks to swing the liquidity discount as they have in the past year and a half is a mystery.

Things are even worse as far as the risk discount is concerned. Our models predict that in normal times, with the ability to diversify portfolios that exists today, the risk discount on assets like corporate equities should be around 1% per year. It is more like 5% per year in normal times — and more like 10% per year today. And our models for why the risk discount has taken such a huge upward leap in the past year and a half are little better than simple handwaving and just-so stories. Our current financial crisis remains largely a mystery: a $2 trillion impulse in lost value of securitized mortgages has set in motion a financial accelerator that we do not understand at any deep level but that has led to ten times the total losses in financial wealth of the impulse.

Thus my dissatisfaction with Larry White’s piece: he talks only about the impulse, while it is the propagation mechanism — the financial accelerator — that is the important part of the story.  $2 trillion shocks to global wealth do, after all, happen every several years, every time there is a recession or a big rise in the prices of natural resources. But financial distress of the magnitude we see today happens once a century. Since the Bank of England developed its lender of last resort doctrine in the 1830s, we have only had two episodes this bad: the Great Depression and today.

Moreover, I do not think that Larry White has gotten the part of the story that he does cover right. I am not convinced of his account of the origins of the trouble in the housing finance market. Larry White blames government subsidies: the implicit government guarantee offered to FNMA and FHLMC, the explicit guarantee to the FHA, the requirements of the CRA, and the subsidy to borrowers provided by the Federal Reserve’s credit expansion — i.e., its open-market operations that bought Treasury bills for cash.

From the start of 2002 to the start of 2006 the Federal Reserve bought $200 billion in Treasury bills for cash. This $200 billion reduction in outstanding bonds and increase in cash surely did lead to an increase in demand for private bonds. But recall the magnitudes here. We have $2 trillion of losses on $8 trillion in face value of mortgages that ex post should not have been made. Are we supposed to believe that $200 billion of open-market purchases by the Fed drives private agents into making $8 trillion of privately unprofitable loans? Not likely. I can see how monetary contraction can make previously profitable loans unprofitable. But I see no way that this amount of monetary expansion can force private agents to make that amount of unprofitable loans. The magnitudes just do not match.

The requirements of the CRA also appear to me to be a red herring. Larry White writes that those who blame the crisis on greed are wrong because “greed… is always around” and you cannot explain a variable result by a constant cause “just as one can’t explain a cluster of airplane crashes by citing gravity.” I say that the same is true of the CRA. It has been around in more-or-less its current form for a generation.

FHA, FNMA, and FHLMC make up the last of the actors to whom Larry White attributes the impulse — the $8 trillion in unwise mortgage loans made over the past five years. Once again the problem is that they have been around for a while. White tried to deal with this by saying that the GSEs changed their policies by cutting severely on down payment requirements — and that the private-sector mortgage lenders had no choice but to match them.

This claim provokes two immediate reactions. First, as your mother says: “If Freddie jumps off a cliff is that a good reason for you to follow him?” The answer to your mother’s question is: “No.” Just because GSEs are leading the market in making stupid money-losing loans did not force private financial companies to follow them and so lose their money too.

Second, Freddie and Fannie and FHA were not the first to jump off the cliff. They lost huge amounts of market share in the mid 2000s. We don’t have a crisis which started when private mortgage lenders losing market share cut back on the quality of the loans they were willing to make. We have a crisis which started when private mortgage lenders cut back on the quality of the loans they were willing to make and so gained market share. The sequence is the opposite of what would have happened if White were correct.

Moreover, if ill-judged loans by the GSEs were the problem, we would expect to see a crisis in which FNMA and FHLMC failed first — which they did not, their troubles coming back in the line well after the Countrywides and the Bear Sternses. And we would expect the failure of FNMA and FHLMC to take the form of them leaking cash as the number of mortgage payments they received crashed with defaults and consequent foreclosures. Instead Fannie and Freddie are still cash-flow positive — as long as they can borrow at nearly the Treasury rate. Fannie and Freddie crashed not because their revenues collapsed but because their borrowing costs ballooned. At it looks right now as though government ownership of 80 percent of Fannie and Freddie will bring money into the Treasury over the next five years.

So why does Larry White’s diagnosis of what is going on differ so much from mine? I think that what is going on is a characteristic weakness of the Austrian tradition: the baseline assumption that all evils must have their origin in some form of government misregulation. If government could be drowned in the bathtub, then an Eden in which people indulged in their natural propensity to truck, barter, and exchange would emerge. And this automatically rules out what I regard as the most likely and fruitful road to walk down to understand this financial crisis: the road that starts from investigating how human psychological limits lead to bad private-sector contract design that then magnifies psychological biases.

I am not happy with the state of such explanations — they seem to involve, at the moment, a great deal of handwaving. But in my judgment it is less handwaving than required to make the case that our current financial crisis is the result of our abandonment of a proper gold standard and our embrace of fractional-reserve banking and government-sponsored mortgage lending enterprises.

J. Bradford DeLong is professor of economics at the University of California, Berkeley and a research associate at the National Bureau of Economic Research.

The Conversation

Fundamentals, Not Accelerators

Professor DeLong made a list of types of factors that determine asset prices.  He went through the list one by one, purporting to show that most of them are negligible when applied to 2008, and then by default assigns causation to the remaining.  The risk of his exercise is that it might miss something fundamental, and thereby assign too much to the residual “financial accelerator.”

We live a world large enough that something fundamental might happen that is not immediately obvious from my desk in Chicago, or even from Professor DeLong’s desk in Berkeley.  But I am surprised that Professor DeLong neglected the role of new capital goods prices and their effect on the quantity and price of capital assets.  After all, Professor DeLong has written one of the most cited papers on this subject.

Just to remind readers of the theory — a higher price for new capital goods raises the price of existing capital assets and reduces the quantity of new investment.  A lower price for new capital goods reduces the price of existing capital assets and increases the quantity of new investment.  This theory has nothing to do with “financial accelerators.”

It seems obvious that new capital goods (esp. new structures) became expensive during the housing boom, or at least slowed down the rate at which they had been getting cheaper over time.  Nonresidential investment had to compete with housing for resources, so in theory this competition reduced nonresidential investment and increased the market value of existing nonresidential capital (e.g., the S&P 500).  In theory, this process reversed itself when housing crashed.  Specifically, in 2007-8, resources were released from housing construction, which reduced new capital goods prices, encouraged nonresidential investment, and reduced the market value of existing nonresidential capital (eg., reduced the S&P 500).

Luke Threinen and I wrote a paper about this in October.  The figures below are from that paper, and from some of my blog entries on the subject.

[Click on figures below to see larger versions.]

The first one shows BEA investment in residential and nonresidential structures, each measured as an index.  The second shows detrended construction spending, each measure as year 2000 dollars deviated from trend.  Clearly the effect I’m talking about is first-order.  Thanks to the end of the housing boom, nonresidential investment can boom again.  This doesn’t happen because the managers of nonresidential businesses are general equilibrium thinkers, but merely because new capital goods are cheap.

The third figure is a time series index for nonresidential investment prices (including structures, equipment, and software).  Economists have watched this series head downward for decades (e.g., see Gordon, 1990); Figure 3 shows how the housing boom may have interrupted that trend.  Because our economic logic is that housing prices affect residential investment which then impacts the price of non-residential capital, it is important to note that nonresidential investment prices peak in the third quarter of 2006.  By this measure, the price of non-residential capital falls by 4 percent from its peak, although less relative to trend.  Given the continued reduction in housing construction, investment goods prices probably have fallen more in recent quarters (we made this figure when 2008 Q2 was the most recent national accounting data) and still have more to fall; an update of this figure could show investment goods prices down five or six percent.

The market value of existing capital can be measured in financial markets by the sum of the market value of debt and equity.  Equity is the residual claimant (that is, stockholders have a leveraged claim on the business assets), so each percentage point change in the market value of capital creates more than a percentage point change in equity values.  So the figure is easily consistent with a 10 percentage point change in equity values.

Professor DeLong knows better than I that true investment goods prices fluctuate more than the measured ones, especially in the short term, in part because of unmeasured price discounts offered by investment goods manufacturers, and in part because waiting times need to be included in the full price of an investment good.  Thus, the true price of new investment goods probably fell more than five percent and the corresponding reduction in the market value of equity should have been more than 10 percent.

In fact, equity prices fell 40 or 50 percent (I will not attempt to provide a precise percentage, because that may change significantly during the time it takes me to type this sentence).  The price of new investment goods cannot explain everything; it is obvious that a changing preference for safety and liquidity has a lot to do with asset price changes and volatility in 2008.  But we have hard evidence that Professor DeLong exaggerated the role of financial accelerators, evidence that was quickly found without moving from my desk in Chicago.  Is it possible that evidence of further exaggeration might lie elsewhere on this Earth?  I expect so.

Sudden, Unanticipated, Discrete Collapse

Casey Mulligan wants to add another to my list of factors that might have been responsible for the steep and unanticipated decline in asset prices over the past year and a half: the pace of capital accumulation — especially the country’s residential capital stock — over the past five years. With every extra piece of capital comes a reduction in the average dividend it will pay and also a reduction in the present worth of that future dividend — so when there is more capital each unit of capital is worth less.

This channel produces a smooth and anticipated decline in the price p of capital goods — as we used to chant while studying for Olivier Blanchard’s midterm in Econ 2410b: “pdot = rpd,” the rate of decline in the price of capital goods is the required ex ante rate of return r times the current price of capital goods p minus the current dividend d

And this is, I think, a problem. For we have to account for not a slow, anticipated, continuous fall in the prices of risky claims to capital but for a sudden, unanticipated, discrete collapse in those values. Capital accumulation won’t do that for us: we need discrete and unexpected changes in resources, technologies, or preferences — and a collapse in risk tolerance, a sudden discrete change in marketwide preference for risk is that one that seems most plausible to me.

The Shock of the News

Professor DeLong is incorrect:  News about current and future supply prices of new capital goods have an immediate and discrete effect on the market value of existing capital.  There is no reason that such news would necessarily hit the marketplace in a smooth, continuous way.

Indeed, the timing of the housing crash hit us by surprise.  Thus, we were equally surprised by the timing of the release of resources from the housing sector.

The discreteness or continuity of actual market price changes will not help distinguish fundamentals from “financial accelerators.”

A Match Can Cause a Forest Fire: A Response to Brad DeLong

My essay on causes of the financial mess focused on trying to identify the initial “impulses” that set the boom-bust cycle in motion because (as this symposium shows) economists have a variety of views about the impulses, and because identifying them correctly is our best hope for avoiding policy mistakes going forward.  Nonetheless I agree with Brad DeLong that the question of the “propagation mechanisms” is also vitally important.  We do face major puzzles in understanding the mechanisms that have turned monetary policy swings of merely large size (combined with regulatory distortions) into a financial mess of freakishly large size. 

In his list of five factors accounting for the decline in the value of global financial wealth, Professor DeLong says little to emphasize the cluster of malinvestments, the rafts of investment projects — particularly in real estate — that have turned out to be wealth-squandering mistakes, leading to the writing down of financial claims that funded them.  I guess they fall under his third heading, “default discount” (although they might go under “savings and investment” with a negative sign).  Clearly, overbuilding has driven the declines in real estate and mortgage markets.  We have a glut of housing overhanging the market.  I was offering an account of how that arose.

I am perplexed that Professor DeLong calls the market’s rate of time-discount (his fourth factor) a “liquidity discount” or “liquidity premium.”  Such a label invites confusion between the premium placed on present-datedness (the height of the real interest rate) and the premium placed on a security’s ready saleability (as indicated by a narrow bid-ask spread).  The time-discount rate is applied based on temporal distance (what Brad calls “the date at which the cash will arrive”), and its determination was famously analyzed by Irving Fisher in The Theory of Interest.  Discounting the future more heavily, as he rightly notes, is not a reason for lower asset values these days:  real interest rates are low.  Heavier discrimination against illiquid securities might be at work, but is not on his list of five factors.

I agree that “the ability of central banks to swing the [real interest rate] as they have in the past year and a half is a mystery.”  But if we recognize that ability as a fact, we recognize that the Fed could have a potent effect in creating the low real rates in 2002-2006.  (It was not only due to a global savings glut, as Alan Greenspan has maintained.)  It is not a mystery how a major drop in real interest rates, if projected to continue, can strongly boost prices of long-lived assets like land:  simply apply the present value formula for a perpetuity.  It is therefore not a mystery how huge amounts of paper wealth can be destroyed when the interest rate rises back to equilibrium.  No doubt lower credit standards and other factors played important roles in exaggerating paper wealth gains and reversals.  But Professor DeLong’s suggestion that monetary policy swings measured by the size of change in the monetary base cannot have led to much larger swings in nominal wealth simply because “the magnitudes just do not match” is, to use Roger Garrison’s analogy, like saying that a tiny match cannot have caused a forest fire.

Professor DeLong ends up blaming the lion’s share of the decline in financial wealth on a change in tastes:  “$17 trillion of lost value because global investors now want to hold less risky portfolios than they wanted two years ago.”  I wonder whether he could say more about how we can distinguish such reduced tastes for given risks from upward revisions in estimates of the sizes of asset risks and their interrelations.  (Many financial firms have of course found that their risk assessment models needed major re-calibrations.)  I agree with him that we don’t have a good theory of changes in risk-aversion tastes.  Nor do we have policy remedies for changes in tastes or in risk assessments.  Efficiency in the face of changing tastes ordinarily calls for letting relative prices and quantities adjust to the new tastes.

A minor correction to what I suppose is a typographical error:  the Bank of England first adopted a lender of last resort policy in the 1890s, in response to the Baring Crisis, not in the 1830s.

Turning from monetary policy to regulatory policy, Professor DeLong says that the Community Reinvestment Act is a “red herring” because it “has been around in more-or-less its current form for a generation.”  My actual argument, however, was that there was an important change in its form:  legislation and regulatory enforcement that gave the CRA teeth for the first time in the mid-1990s.  It may still be the case that the magnitude of direct CRA effects was small, as Randy Kroszner has recently argued in reporting that the CRA-linked share of subprime lending was only about 8%.

I’m not persuaded by the fact that Bear Stearns failed slightly ahead of Fannie Mae and Freddie Mac that Fannie and Freddie and HUD didn’t contribute to the problem of lowered credit standards.

Professor DeLong proposes that we are better off pursing psychological explanations for poor contract design — though he admirably acknowledges that these “seem to involve, at the moment, a great deal of handwaving” — than pursuing the hypothesis “that our current financial crisis is the result of our abandonment of a proper gold standard and our embrace of fractional-reserve banking and government-sponsored mortgage lending enterprises.”  But the second hypothesis is not mine.  We abandoned the gold standard many crises ago, so that can’t explain the timing of this one.  And Professor DeLong has me confused with someone else if he thinks I blame fractional-reserve banking.  Freedom of contract implies the freedom to make fractional-reserve contracts, as well as other innovative financial contracts, and to take risky financial positions.  Financial innovation and risk-taking are, in the long run, good things for economic growth, provided that the system weeds out innovative but poorly designed contracts.  To let it do so we must hold parties to their own contractual commitments to the point of bankruptcy, and let them take their lumps when their counterparties default, rather than bail anyone out.  

The turmoil under our current regime can only strengthen, I would think, the probability we assign to the hypothesis that our fiat money regime and legally restricted financial system is less robust and less efficient than the alternative of a laissez-faire commodity-based monetary and financial system.  But comparative regime analysis is a topic for another day.  My diagnosis of the impulses for the current turmoil supports more immediate policy reforms under our existing regime, namely that the Fed would do better pursuing stable (preferably zero) growth in nominal income, and that, yes, we should end our embrace of government-sponsored mortgage lending enterprises and other housing market distortions.

Surreal Fiddling While Rome Burns

The essays and responsive commentary to date reflect the imperturbable quality of economists and neoclassical economics. The economic world could be burning — in contradiction to their theories and as a result of the policies they had designed — and economists would compete to serve as Concertmaster.

One might think that economists might reconsider whether their models — each of which is premised on efficient markets — should be reexamined in light of the fact that the capital markets have proven extraordinarily efficient at destroying wealth through fraud. Instead, without even discussing whether the models might be wrong, the debate’s primary focus is on the importance of omitted variables in the failed neoclassical models. A policy maker attempting to avoid future crises reading the debate would be mystified by the intricate but monotonous medley, startled by the efforts to minimize the worst financial crisis in our lifetimes, and disappointed that it was bereft of useful praxis.

Casey offers this recipe for how economists should proceed:

The methodology of economics specifies a clear recipe for assigning blame for today’s financial crisis. First, alternative hypotheses are articulated. Second, the alternatives are compared on the basis of empirical predictions. Third, empirical tests are conducted to gauge the relative importance of the alternative hypotheses.

I argue that, overwhelmingly, economists (because of implicit prior assumptions) have not followed this recipe in past crises and are not following it in the current crises. The most critical failure is in the first step — ignoring “alternative hypotheses.” Note that Casey ignores alternative hypotheses such as control fraud. I also argue that the (unacknowledged) omitted variable problem means that econometric techniques produce flawed data that leads economists to fail to learn the proper lessons from prior crises and recommend policies that produce recurrent, intensifying crises.

I begin by discussing omitted variables that have enormous importance for praxis. The current crises are tragic because they could have been avoided had we learned the correct lessons from prior crises, such as the S&L debacle. Neoclassical theory and methodology produce the worst possible policy advice when financial bubbles are inflating. The controversy over the “direct investment” rule provides a good example. A direct investment is ownership of an equity interest in an investment, as opposed to lending funds to an investor. [1] The Federal Home Loan Bank Board (Bank Board) proposed to restrict S&Ls from making direct investments in excess of 10 percent of their total assets.

Charles Keating retained George Benston and Alan Greenspan to attempt to block adoption of the direct investment rule. Benston and Greenspan “studied” the S&Ls that made direct investments in excess of the threshold. They found that such S&Ls reported that they were far more profitable, and had fewer large losses, than other S&Ls. They concluded that the Bank Board should encourage other S&Ls to emulate the 32 S&Ls that exceeded the threshold. They argued that portfolio diversification theory supported making large direct investments.

Within roughly two years all 32 S&Ls had failed — many of them operating in states with robust economic growth. The variable that Benston and Greenspan omitted was control fraud. They assumed that (1) financial markets are efficient, (2) CEOs sought to maximize the firm’s long-term profitability, (3) CEOs added direct investments pursuant to portfolio diversification theory in order to increase that profitability, and (4) CEOs do not commit control fraud. Tellingly, the last assumption was implicit. [2] Unstated assumptions cause the greatest risk of choosing policies that produce financial crises because we are not even aware of the risk.

The obvious question is why direct investments were financial cyanide. “Risk,” at least as we conventionally teach the concept in finance (uncertainty) does not explain the pattern. The two obvious aspects of the pattern are outcomes: record profits followed by certain, unusually costly failure. Accounting control fraud creates certain profits — and looting by the CEO makes failure certain. [3] The S&Ls that Greenspan and Benston lauded didn’t all fail because honest direct investments are moderately riskier than honest traditional mortgages. They all failed because they choose to make large amounts of direct investments because doing so optimized accounting control fraud. Direct investments were one of the two investments S&Ls could make that (1) made it easy to create fraudulent “income” and (2) best hid real losses (Black 2005; Akerlof & Romer 1993).

The less obvious aspect of the pattern, operations, was also consistent with control fraud and inconsistent with rational risk taking. The National Commission on Financial Institution Reform, Recovery and Enforcement (NCFIRRE, 1993: 3-4) described the pattern:

The typical large failure was a stockholder-owned, state-chartered institution in Texas or California where regulation and supervision were most lax. … The failed institution typically had experienced a change of control and was tightly held, dominated by an individual with substantial conflicts of interest. … In the typical large failure, every accounting trick available was used to make the institution look profitable, safe, and solvent. Evidence of fraud was invariably present as was the ability of the operators to “milk” the organization through high dividends and salaries, bonuses, perks and other means. In short, the typical large failure was one in which management exploited virtually all the perverse incentives created by government policy.[4]

The element of the operational pattern that neoclassical economists have had the most difficulty grasping, unfortunately, is the most important element. As James Pierce, NCFIRRE’s Executive Director explained:

Accounting abuses also provided the ultimate perverse incentive: it paid to seek out bad loans because only those who had no intention of repaying would be willing to offer the high loan fees and interest required for the best looting. It was rational for operators to drive their institutions ever deeper into insolvency as they looted them. [5]

Optimizing a lending institution to grow rapidly by making bad loans that will not be repaid also creates a distinctive operational strategy that is rational if one is engaged in accounting control fraud and profoundly irrational if one is running an honest lender. The strategy requires the CEO to suborn internal and external controls and to systematically pervert underwriting. [6] A control fraud that grows extremely rapidly by making bad loans during the expansion phase of a bubble is mathematically guaranteed to report record profits. [7] This explains why econometric studies conducted during the expansion phase produce the worst possible policy recommendations. They will show whatever practices optimize accounting control fraud (which maximizes real losses) will have the strongest association with “income” and (because accounting control frauds fool markets) share price. Econometric studies almost exclusively use accounting “income” or share price as their “performance” measure. Of course, the catastrophic methodological failure — and, in particular, the failure of economists to change their methodology despite repeatedly giving the worst possible policy advice during the expansion phase of bubbles — evinces an underlying failure of theory.

The CEO is in a unique position to optimize the firm as a “weapon.” He can place the firm in business lines that are optimal for inflating accounting values and covering up real losses. He can cause the firm to grow extremely rapidly. He suborns controls by creating perverse financial incentives. Control frauds promote and shower compensation on those who cooperate with management’s goal of ignoring credit quality; those who don’t are punished.

The Bank Board contained the S&L debacle, despite a critical shortage of funding and personnel, and in the face of intense political opposition because it identified the epidemic of accounting control fraud and correctly analyzed how to respond to it. In 1983, the control frauds grew at an average annual rate of over 50 percent. They purported to be the most profitable S&Ls and they had very low (recognized) losses.

The Bank Board, under Bank Board Chairman Ed Gray, figured out three things about the epidemic of S&L control fraud that were critical to praxis. [8] All three aspects were contrary to the economic conventional wisdom. First, they recognized that the frauds relied on accounting as their weapon. Therefore, the claimed record earnings and relatively small amount of realized losses did not demonstrate that the firms were successful and conventional econometric studies would yield perverse results. Second, the agency realized that it could spot the distinctive operational pattern of accounting control frauds and prioritize such S&Ls for examination and closure at early dates when they were still reporting record earnings. Third, the agency realized that growth is the Achilles’ heel of any Ponzi scheme. The agency adopted a rule restricting growth from 1985 onward. The rule caused the remaining control frauds to collapse. [9] It also, in conjunction with the 1986 tax act (which ended the perverse incentives of the 1981 tax act), burst the bubble in Southwest commercial real estate. Had the S&L and bank control frauds been permitted to extend and further hyperinflate that bubble the ultimate losses would have been far greater (NCFIRRE 1993).

Taken together, we can now describe what economists should have learned from the S&L debacle. Akerlof & Romer demonstrated three manners in which epidemics of control fraud caused severe damages. Fraud, of course, causes direct losses to the individual lender. They also demonstrated two indirect fraud losses. Control frauds extend and hyperinflate financial bubbles because they deliberately make bad loans and grow exceptionally rapidly in order to optimize accounting fraud. The losses that occur when a bubble hyperinflates are likely to be nonlinear, so a material increase in the length and size of a bubble is likely to create a very large increase in losses. Akerlof & Romer also explained that control frauds’ (their term was “looters”) exceptional accounting “profits” and eagerness to lend to developers that purchased land for commercial real estate projects at grossly inflated prices sent false price signals to other market participants. The false signals would further extend and hyperinflate the bubble.

Akerlof, in his original article on markets for lemons, explained how a Gresham’s dynamic could spread the incidence of control fraud. (His seminal article discusses anti-consumer control frauds in which the seller misrepresents quality.) Pierce warned about a similar dynamic in S&L accounting frauds (see n. 7).

What has been added since 1993, and has made the criminogenic environment for accounting fraud far more intense, is a Gresham’s dynamic caused by the modern executive compensation system and the extension of the perverse dynamic described by Pierce to the rating agencies.

These changes, plus a very different response by regulators and law enforcement, explain why the current crises are so much more severe than prior epidemics of control fraud during the S&L debacle, the Enron-era frauds, Russian privatization, and the worst of the Washington Consensus in Latin America.

Mortgage bankers have always been the cowboys of the mortgage industry. Without deposit insurance, they could not grow unless they could sell large amounts of their product. They weren’t able to cause enormous losses in the past because they could only sell their product if it was a high quality, conventional product. What changed all this was the rating agencies’ willingness to give AAA ratings to “toxic” mortgage product and investment bankers’ willingness to securitize and sell the toxic paper (and retain a great deal of the most toxic product in their off balance SIVs).

The Gresham’s dynamic arising from the other major change since 1993, perverse compensation systems, explains why investment banking firms (and rating agencies) were eager to inflate the ratings and make these toxic investments. As I noted in my essay, the pay systems do the opposite of “aligning” the interests of shareholders, CEOs, and other managers. [10]

Changes in regulation provided the third major reason that the existing crises are far larger, and created systemic crises. There were no Ed Grays in the current administration. [11] Yes, Edward Gramlich warned Alan Greenspan. Yes, William Donaldson knew there was a developing problem. Like Arthur Levitt (under Clinton), neither was willing to go to the wall to fight for reregulation. Unlike Donaldson, Gray refused to resign and continued to put in place the reregulation and vigorous supervision that prevented the debacle from becoming a catastrophe. (Gray’s career, unsurprisingly, was destroyed. Unlike Donaldson and Levitt he was not wealthy, powerful, and near the end of his career.)

No federal regulator responded effectively when the mortgage bankers could suddenly make (1) loans that were certain to have catastrophic default rates and (2) sell them in vast amounts so that they could grow rapidly. The Fed had unique authority to regulate mortgage bankers — it refused to do so. Greenspan went so far as to praise ARMs (at the precise time they were morphing into “exploding rate” ARMs that the borrowers were incapable of paying after the initial teaser rate. He even praised equity stripping. OFHEO could have banned Fannie and Freddie from buying subprime paper. It declined to do so. The only area in which the federal agencies were vigorous was in preempting state agencies that sought to take action against some of the worst mortgage control frauds.

The Office of Thrift Supervision (OTS) and the FDIC desupervised. The first FDIC Chair under this administration slashed the staff and made it impossible for the FDIC to exercise its “backup” examination authority in any meaningful manner. The OTS Director, as I noted, was far worse. The final straw was when the OTS promoted back to a senior leadership position the (infamous) professional regulator most responsible for caving in to Charles Keating’s demands for unprincipled concessions by the Bank Board. The Director of OTS viewed his “supervision” of WAMU and his concessions to Countrywide so favorably that he appointed him head of the Western Region — which is where the biggest S&L control frauds specializing in nonprime mortgages were located. This is financial desupervision at its most pernicious, and the cost has been catastrophic. The OTS desupervised so completely that like IndyMac, which the FDIC estimates will cost roughly $9 billion to resolve (roughly twice the most expensive prior failure of an insured institution), was not even on the “troubled institutions” list prior to its closure. (The OTS’ effort to blame Charles Schumer for IndyMac’s failure was equally parts dishonest and shameful. He did the taxpayers a service by hastening the closure of an institution that should have been closed five years ago.) [12]

The defining element at law that separates fraud from other forms of larceny is deceit. Deceit is used to cause the victim to trust the perpetrator — who then betrays that trust. As a result, control fraud is a powerful acid to erode market trust. Widespread accounting control fraud creates systemic crises because it makes trust extremely risky. Indeed, it makes it impossible to evaluate risk because everything we use to quantify risk is premised on the assumption that the numbers we input are not fraudulent. The current systemic crises are primarily the product of bankers no longer trusting other bankers. Most of us attend meetings periodically where bottled water is served. How many of us would drink from the bottle if we knew that one in fifty were contaminated? Well before fraud becomes endemic it can cause systemic crises by eroding trust.

In the current crises we have (1) an epidemic of accounting control fraud (arising in the United States primarily, but not exclusively, in nonprime mortgage lending), (2) a complete breakdown in effective regulation, which means that key markets are opaque and even if the industry provides some data they are not verified by a truly independent, competent, and vigorous regulator, and, (3) our most elite private institutions not only failed to exert “private market discipline,” but often also led or aided the accounting fraud.

When we chose not to regulate a financial activity through non-regulation, deregulation, or desupervision we often create adverse consequences that neoclassical economists are unlikely to have understood or intended. First, as I explained immediately above, it frequently makes the market opaque and the data it produces unreliable. This helps allow control fraud to grow to epidemic levels and threatens a systemic crisis of trust. (The data unreliability will typically be masked during the expansion phase of the bubble. This helps explain the problem Brad DeLong has been struggling with — why things (e.g., spreads) are so discontinuous and volatile and why do “mere” trillion dollar fraud losses lead to systemic crises and much larger wealth losses?)

Second, regulators are the equivalent to the “cops on the beat” in the white-collar crime context. No one calls the Houston Police Department and says: “I think there’s a problem with Enron’s accounting. Please check it out.” When we don’t regulate or supervise financial markets we, de facto, decriminalize accounting control fraud. Yes, the SEC and the FBI will still exist and they will investigate some cases on their own. But the truth is that neither can be effective without financial regulators taking the lead. We have to make the criminal and SEC referrals. To be useful, a referral requires extensive work. We then have to train the FBI agents and Assistant U.S. Attorneys (AUSAs) about the industry and how the frauds work. We have to “detail” key staff to the FBI to serve as their internal experts. We provide (free) expert witnesses to the AUSAs for grand jury and trial testimony. We have to fight to get the FBI to investigate our referrals and to get the AUSAs to bring the cases (and prioritize them properly). We are only part of the effort, but we are an essential part of the effort. If we are not there, vigorously, symbolic prosecutions may occur, but there will be no meaningful deterrence to stop a developing epidemic of control fraud. Most of the worst mortgage fraud occurred in the unregulated sectors. Most of the worst mortgage fraud in the regulated sector occurred at S&Ls (and AIG!), which OTS was supposed to regulate. The desupervision is so crippling that mortgage fraud prosecutions have become shambolic. To date, large mortgage control frauds have had more reason to worry about being struck by lightning than being prosecuted.

So, on the subject of omitted variables, I would ask my colleagues what it would take before they would treat accounting control fraud seriously? The incidence of nonprime lender mortgage fraud is exceptionally high. I don’t doubt discussions among economists at Berkeley and Chicago about the crises overwhelmingly omit fraud and the direct and indirect losses it creates. Even though the “typical large [S&L] failure” “invariably” involved control fraud, the conventional economic wisdom about the debacle relegates such frauds to a paragraph (in a book) or a clause (in an article). It is no answer that your models cannot input fraud as a variable and that your theories ignore it.

Here’s the deal. If it’s good criminology, it’s good economics. If it’s bad criminology, it’s bad economics. For example, control fraud does exist. It frequently suborns private market discipline. It causes greater financial losses than all other forms of property crime combined.

Anti-consumer control frauds also maim and kill. China provides topical examples: infant formula, cough syrup, and toys are all contaminated (and the formula also lacks milk and therefore lacks vital nutrients). In each case there is a Gresham’s dynamic because the sellers that cheat gain a cost advantage over honest firms. Another example is school construction. China recently admitted (then retracted) that over 19,000 students died in the Sichuan earthquake when their schools collapsed. The schools appear to have collapsed because contractors purported, but did not in fact, build them in accordance with seismic codes. (Turkey has similar problems with fraud and corruption — which is public sector control fraud). Absent effective regulation and supervision the markets will be deeply inefficient because of this Gresham’s dynamic.

There is no excuse in this debate for ignoring fraud. You have had to read the conclusions of at least some of our research. Ignoring our literature has led economics and economists to recurrent, grave mistakes. Economists have praised the worst frauds in the nation as exemplars. They have repeatedly recommended policies that maximize the criminogenic nature of the environment. [13] They have ignored the successful, real world test of S&L reregulation and resupervision in 1984–87.

More bizarrely, when economists omit fraud as a variable they ignore economics — economic theory that garnered a Nobel. Reread Akerlof’s seminal article before you read Akerlof & Romer. No, Akerlof’s “lemons” article does not use the word fraud. The examples he discusses are frauds. Indeed, they are anti-consumer control frauds. Unfortunately, the profession responded to his article with a literature that focuses overwhelmingly on “the market’s” clever adaptations that (purportedly) solve the problems arising from the information asymmetry.

Information asymmetry, however, is the essence of fraud. Fraud perpetrators manipulate asymmetry. The most elegant frauds use deceit to make the victim feel that the perpetrator is being transparent when he is really presenting only the illusion of transparency. Information asymmetry and accounting control fraud pose a central, existential challenge to efficient capital markets. They are not peripheral topics. They have taken our capital markets to the cleaners and produced the worst global financial crisis of our lifetimes. Our most “sophisticated” specialists and financial elites that have survived many prior challenges are dead or on public life support. Our markets and our elites acted as vectors in this epidemic of control fraud instead of disciplining it. Casey appears to exclude the possibility of deceit. He attempts to explain the bubble as, in large part, the product of improved information.

Much of the banking and real estate value added relates to information. Bankers screen borrowers and value heterogeneous collateral. Real estate brokers match heterogeneous families to heterogeneous properties. Hall and Woodward (2008) claim:

Recent years have seen great improvements in data, especially the introduction of credit scores, which gave lenders new powers to forecast mortgage defaults and to adjust interest rates offered to prospective borrowers. In 1990, credit scores were rare; by 1996, they were standard.

Perhaps lenders also expected to use information technology to better monitor and collect on loans, and therefore put subprime lending programs in place.

I find Hall and Woodward’s claim remarkable. What we have just experienced is that these “great improvements” in information were great degradations. The “new powers to forecast mortgage defaults and to adjust interest rates” is certainly a desirable concept. Unfortunately, reality went in the opposite direction, as they surely knew by the time they wrote their piece. We know the models’ “forecast[ed]” mortgage default rates proved grossly too low. We know that they effectively ignored the risk that there was a bubble. We know that they forecast default rates for nonprime mortgages that were substantially below rates experienced in the past in prior business cycles.

We know that alt-a lending was premised on credit scores and avoiding underwriting and verification. When the market calls something a “liar’s loan” it pays to take it seriously. This is an excellent way to maximize adverse selection and destroy a lender. We have known that for centuries. The passage from Hall and Woodward that Casey cites does not acknowledge any of these reasons why, in fact, the quality of information deteriorated. More importantly, it does not acknowledge that control frauds would engage in each of the errors I have just discussed in order to optimize short-term accounting profits.

Hall and Woodward also should have known that nonprime lenders did not “adjust interest rates offered to prospective borrowers” to reflect their heterogeneous credit risk. They did the opposite. First, they increasingly went to exploding rate ARMs and “qualified” borrowers only at the teaser rate. Those practices do not compensate for credit risk (which is what Casey, and Hall and Woodward, are arguing) — they create it. Indeed, they maximize credit risk while reducing (long-term) return. This is profoundly irrational for an honest lender, but optimizes accounting control fraud by facilitating rapid growth through lending to the uncreditworthy. Second, they increasingly did no real underwriting, so they did not know the credit risk they were taking and they maximized adverse selection. Again, this is inconsistent with Casey’s (and Hall and Woodward’s) assertions, but rational for accounting control fraud. Third, spreads on nonprime loans declined while (1) underwriting declined, (2) the epidemic of mortgage fraud increased rapidly, (3) the quality of the borrowers fell, and (4) delinquencies rose. So, far from using superior information and enhanced underwriting to adjust individual interest rates to compensate the nonprime lenders for the individual risks they were taking, the already grossly inadequate spread fell materially when it should have been surging. This makes no sense under the model Casey supports, but is sensible for accounting control frauds. Note also that this ignores some of the most important factors actually driving those who should have qualified for conventional loans to nonprime loans — the borrower’s race and ethnicity.

Casey understands that underwriting is essential because he says, “Perhaps lenders also expected to use information technology to better monitor and collect on loans, and therefore put subprime lending programs in place.” The neoclassical theory of financial intermediation is premised on the assumption that banks use their superior informational advantage to properly underwrite loans. The problem is reality. Nonprime lending specialists did the opposite. They used their information technology to approve bad loans more quickly. They structured loans, and conducted (limited, perverse) underwriting to approve bad loans (e.g., exploding rate ARMs to borrowers that often did not even have the ability to pay the teaser interest rate). They gutted the internal controls that are supposed to underwrite, to require adequate credit quality, and to monitor credit quality. All of these actions optimize accounting control fraud. The reality, among nonprime lending specialists is the opposite of Casey’s speculation (“perhaps”).

If you think that the ongoing crises were materially caused by changes in housing tastes or informational technology changes then you are necessarily saying that our markets are so fragile that relatively minor changes inevitable in any dynamic nation can send the global economy spinning into a severe recession and gut virtually all our leading financial firms. That means you have to turn to understanding why we are so fragile and what we need to do to reduce dramatically that fragility. (And I don’t want to hear about “creative destruction” — what we’re experiencing is a wealth- and efficiency-destroying potlatch of epic proportions.)

Rome is burning. We need to stop fiddling. It isn’t simply the economy that is in crisis. Neoclassical economic theory, methodology and praxis need fundamental reexamination. The “Imperial Science” has suffered the fate common to arrogant imperialists.

While Becker famously boasted that an economist had no need to know anything about the criminology literature to advise policy makers about appropriate criminological praxis, white-collar criminologists generally value the study of economics. We both study markets and institutions. We find many economic theories persuasive and sound.

White-collar criminologists generally believe, however, that criminological research has falsified the key economics and finance theories underlying modern finance. For example, efficient markets and contracts are, at best, special cases. Some markets are endemically inefficient because of control fraud and corruption. Some markets experience periodic epidemics of control fraud and baseline levels of control fraud inconsistent with the efficient market hypothesis (because fraud almost always creates a systematic bias, e.g., accounting fraud overwhelmingly is used to make firms appear more profitable). We also generally believe that econometric analyses are likely to become perverse whenever control frauds are material. When top neoclassical economists praise the worst firms in the industry as the best (in the case of S&Ls that was a ranking error of over 3000 positions) and recommend investment strategies that are invariably fatal there is a vital problem that requires resolution. If economists were part of the air force their commanders would have long since ordered a profession-wide “stand down” designed to ensure that the profession learned why it was recurrently flying jets into mountains and adopt changes to prevent such crashes from recurring.

White-collar criminologists rarely find literature in which an economist explicitly considers whether the praxis s(he) recommends, e.g., deregulation, will make an environment more criminogenic. Relative to neoclassical economics, we believe our models have demonstrated superior predictive power in explaining financial crises, our research methodologies produce superior results, and our recommendations on praxis are drawn from what worked in the real world and prevented the S&L debacle from becoming catastrophic — as it would have done had the Bank Board continued to follow the advice of financial economists. We may, of course, be wrong. We’re happy to debate economists about the causes of these crises. We’d be even happier to have you be part of a multi-disciplinary effort to develop sound theory, methodology, and praxis designed to reduce these crises.

Notes

[1] Direct investments also included transactions structured nominally as loans in which the lender was, in economic substance, taking an equity risk, i.e., the lender could only be repaid if the project succeeded.

[2] Keating also retained Daniel Fischel to help block the rule’s adoption. The Bank Board’s had cited studies showing that (1) S&L direct investments earned lower returns than traditional mortgages and (2) direct investments were riskier than traditional mortgages. Fischel argued that the efficient market hypothesis made it impossible for both facts to be facts — absent fraud. Fischel then proceeded to ignore the possibility of control fraud. Keating’s Lincoln Savings, of course, was the most notorious S&L control fraud, costing the taxpayers $3.4 billion.

[3] Akerlof, George A., and Paul M. Romer. 1993. “Looting: The Economic Underworld of Bankruptcy for Profit.” Brookings Papers on Economic Activity 2:1-73.

[4] Brad is correct that Austrian economists invariably blame the government, but that bias is common among neoclassical economists. James Pierce, the author of the NCFIRRE report, argued that deposit insurance was the principal cause of the perverse incentives (as did Akerlof & Romer). All three economists believed (then) that the efficient contracts hypothesis required that private market discipline defeat control fraud. Deposit insurance must, therefore, remove the incentive for creditors to exert market discipline. Other NCFIRRE officials challenged Pierce’s view. We argued that uninsured creditors (e.g., subordinated debt holders and S&L counterparties) and shareholders never exercised effective discipline against an S&L control fraud. Enron, WorldCom and their ilk ended any claim that deposit insurance was necessary to an epidemic of control fraud. Akerlof & Romer changed their views in light of that experience; Pierce did not.

[5] Pierce, J. ASSA presentation, 1994: 10-11. See also Robinson, M., 1990. Overdrawn: The Bailout of American Savings. Plume, N.Y.

[6] Note that this is the opposite strategy that one would follow if one were engaged in lawful “gambling for resurrection.” By making underwriting perverse, control frauds maximize adverse selection. The expected value of lending in circumstances that maximize adverse selection is sharply negative. Those engaged in honest, high-risk gambles would optimize through superior underwriting (Black, Calavita & Pontell 1995).

[7] If they can get a clean opinion from a top tier audit firm “blessing” their claimed income. Accounting control frauds have repeatedly shown the ability to achieve such opinions.

Abusive operators of S&Ls sought out compliant and cooperative accountants. The result was a sort of “Gresham’s Law” in which the bad professionals forced out the good. (NCFIRRE 1993: 76)

[8] It recognized that econometric tests produced perverse results and instead relied on in depth review of the causes of failure at the institutions it closed. These “autopsies,” as they were called at the Bank Board, revealed accounting control frauds’ distinctive outcome and operational patterns.

[9] The administration was enraged by Gray’s reregulation. Don Regan made strong efforts to force him from office. Speaker Wright held the “FSLIC Recapitalization” bill hostage to prevent Gray from getting to additional funds to close more Texas control frauds. (Texas and California deregulated and desupervised far more than the federal government did and its S&Ls caused disproportionate losses.) The administration attempted to appoint two members (to the three member Bank Board) chosen by Charles Keating. This would have given Keating majority control of the agency. That would have caused losses to top a trillion dollars. The “Keating Five” (Cranston, DeConcini, Glenn, McCain, and Riegle) successfully pressured the agency, on Keating’s behalf, not to take enforcement action Lincoln Savings’ $615 million violation of the direct investment rule. The administration threatened to prosecute Gray criminally for the “crime” of closing too many S&Ls. The administration’s top priority at all times was covering up the scale of the S&L debacle. The administration’s Treasury spokesman testified in front of Congress that insolvency was irrelevant for S&Ls because they could always meet their liquidity needs by simply growing and using the cash received from new depositors to pay past depositors. While the Treasury’s endorsement of a Ponzi scheme for insolvent S&Ls was remarkable, it promptly exceeded its congressional testimony by testifying on behalf of a failed S&L in its lawsuit against the Bank Board alleging that it was arbitrary and capricious to place it in receivership because it was insolvent. Note that if the owners of the failed S&L had prevailed they would almost certainly have next sued Treasury for damages. The administration did not want even failed S&Ls closed because doing so required them to recognize their massive collective insolvency.

The administration and Speaker Wright reached a secret deal in which the administration agreed not to reappoint Gray to the Bank Board and not to oppose regulatory “forbearance” provisions (pushed by the control frauds’ attorneys) in return for Wright endorsing a $15 billion recapitalization of the FSLIC fund. (Wright, of course, reneged on his end of the deal by sending the word to House Democrats that they should ignore his floor speech on behalf of the $15 billion bill — which they promptly rejected in favor of a farcical $5 billion bill.)

Gray’s successor, who caved in to the Keating Five’s intervention on behalf of Keating, was chosen by the administration because he did not support Gray’s policies. He cut back on Gray’s policies of vigorous examination but the growth rule still proved fatal to the control frauds.

[10] The claim that compensation systems “aligned” the interests is a classic example of “assume a can opener.” The “logic” was that (1) CEOs’ existing compensation misaligns those interests because, de facto, CEOs control their compensation and maximize their self-interest (at the shareholders’ expense) by deliberately misaligning those interests and (2) “therefore” we should encourage CEOs to dramatically increase their compensation because this would cause them to align their interests with the shareholders’ interests. But if CEOs control their compensation, and they commonly act in their self-interest at shareholders’ expense then it follows that if we raise the stakes enormously by offering them annual bonuses so large as to make them vastly wealthy for life, they should have an overwhelming interest in misalignment. It is in their self-interest to make large amounts of money regardless of whether they aid or harm the firm’s long-term interests. Two clear means to maximize their self-interest are causing the corporation to engage in accounting control fraud (which guarantees that they will become extremely wealthy) and backdating/repricing stock options. Both forms of control fraud ensure short-term “success.”

[11] CFTC Chair Brooksley Born was such a hero under Clinton. A bipartisan coalition of anti-regulatory officials, including Summers, Levitt and Rubin, killed her efforts to regulate CDS.

[12] IndyMac specialized in originating for sale “alt-a” loans (aka, “liar’s loans). The FDIC expects to lose roughly $9 billion on a shop that (purportedly) only had $33 billion in assets (apparently it had only $24 billion in assets). This is obviously awful news, but it greatly understates the problem. First, the actual credit losses at IndyMac must have been materially larger than $9 billion because in a receivership the FDIC does not pay the subordinated debt holders. In a receivership loss calculation the FDIC also ignores the loss of roughly $3 billion in GAAP capital. This means IndyMac’s credit losses are likely closer to $12 billion — and that ignores liability claims from purchasers of IndyMac’s alt-a mortgages. Second, economic theory predicts that IndyMac would have taken advantage of asymmetrical information and sold the worst of its alt-a product and held in portfolio the best of its alt-a product. If true, that suggests that the percentage losses to purchasers of its mortgages will be even larger than the (catastrophic) percentage losses on mortgages it held in portfolio. IndyMac served as a “vector” that spread the fraud losses broadly.

[13] Black, W., 2003. Reexaming the Law-and-Economics Theory of Corporate Governance. Challenge 46, No. 2 March/April, 22-40.

Data, Please

The claim has been made that real interest rates were low during the housing boom.  Can someone who believes this claim show us a time series for such real interest rates, and calculate the number of basis points by which those rates were low during the boom?  Then we can consider more carefully how much those rates might have elevated housing prices.

Low Real Interest Rates

Casey Mulligan has asked for a time series for real interest rates so we can judge just how low they were.  Here’s the clearest and most up-to-date picture I could find, from Christopher J. Neely and David E. Rapach, “Real Interest Rate Persistence: Evidence and Implications,” St. Louis Federal Reserve Bank Review 90 (November-December 2008), p. 628 (Figure 2).  It shows that the ex post real interest rate (their measure is the U.S. three-month Treasury bill rate minus the CPI inflation rate) was below 1.82 percent, its post-1989 “regime-specific mean,” roughly from mid-2001 to mid-2006.  It was persistently negative (with one brief exception) for more than three years during this stretch, early 2002 to mid 2005.

 


 

Unfortunately Neely and Rapach do not report the underlying time series from which we could satisfy Professor Mulligan’s request for “the number of basis points by which those rates were low during the boom.”  The raw data on nominal interest rates and CPI levels from which precise ex post real interest rates can be calculated are readily available at the St. Louis Fed’s FRED database.

I Accept Larry White’s Correction…

I accept Larry White’s correction with respect to the word “liquidity.” The conventional language — that we distinguish between situations in which organizations are illiquid and those in which organizations are insolvent — is, I think, confusing at some level. After all, if there is no doubt that some organization is solvent why should it ever be illiquid? In an extreme, its own paper would be good enough. I think that what we usually call “liquidity” is actually made up of two different things — call them “duration” and “information.” Assets that mature in the future are illiquid in that they have unusually low prices today to the extent that there is unusually high time preference — that people really want money now. But assets that mature soon can also be illiquid in the sense of having low prices if you try to sell them when people suspect that you are cherry-picking (or, better, pit-giving) — that there is something wrong with this particular asset that makes it not a good proxy for its class — and the belief that there is something wrong with this asset can be triggered simply by your attempt to sell it.

I think that there is a big difference between me and Larry White in our view of what fundamental values are. You see, I start from something like a utilitarian’s dream — a belief that the rate of (safe) time preference ought to be on the order of 2% per year (2 percent growth in productivity times an elasticity of intertemporal substitution of 1) and that the equity risk premium ought to be on the order of 0.2% per year (the covariance of stock returns with consumption times a coefficient of relative risk aversion of 1). Thus if the market were working well — if we were using our financial markets to mobilize the full risk-bearing capacity of society — expected equity cash flows ought to be discounted at a real rate of 2.2% per year, Treasury bonds ought to yield a real rate of 2% per year, and every intermediate class of debt ought to have an expected value produced by a real discount rate in that extremely narrow range.

Thus practically every risky asset, all the time, sells for much less than its fundamental value — and does so because financial markets do not do a good job of mobilizing the risk-bearing capacity of society. I don’t think we have any prospect of living in a world in which financial markets do their job of risk-tolerance-mobilization well — nobody should trade or invest in anticipation of such a world. But I don’t think that the idea of “overinvestment” makes a lot of sense: the proper public policy response to every situation that White would characterize as one of “overinvestment” is, I think, one in which the government takes steps to mobilize more of society’s risk-bearing capacity rather than letting asset prices collapse and create massive unemployment. 

I have to drop out of this virtual conversation now to take part in a real one in Vienna, but I will be back…

Interest Rates, or Optimism about Interest Rates?

Professor White showed that one-year real interest rates were low during the housing boom.  That’s a good starting point because, if the Fed can affect anything real, it is the short-term interest rate.  Now let’s use that information to demonstrate that the impact on housing prices is minimal.

Since I will demonstrate that the housing-price impact is small, I will assume that the supply of housing is fixed; an elastic supply of housing would only reduce the price impact below what I calculate here.

Each house in place today produces services for a number of years.  To a good approximation, we can assume that each house lasts forever, except that it depreciates exponentially (but slowly).  The market value of the house is the present value of those services.  Low interest rates can raise housing prices (although not much), because future services are discounted less.

Suppose that annual real interest rates were going to be one percentage point (100 basis points) lower for a year.  Then the cost of buying a house, holding it for a year, and then selling it would be essentially one percent less.  The low one-year interest rate would not affect the selling price at the end of the year because, by assumption, the reduction lasted only for a year and the next buyer will be back to normal interest rates.  So the source of benefit from the low rate is that the initial buyer reduces the carrying cost for a year.

A 100-basis-point-lower interest rate for one year would justify paying about $202,000 for a house that would ultimately be worth $200,000.  A 100-basis-point-lower interest rate for two years would raise purchase prices by about two percent. (Actually, it would be less, because of the discounting of the second year, not to mention the supply response.)  A 200-basis-point reduction for two years would raise purchase prices by less than four percent, etc.

Thus, interest-rate reductions for a short horizon do raise housing prices, but not much by the standards of this recent housing boom when housing prices were tens of percentage points higher (according to Case-Shiller, practically 100 percent higher).  A house that would ultimately be worth $200,000 was actually selling for something in the neighborhood of $300,000.

Perhaps Professor White would argue that market participants expected short term interest rates to remain low for much longer than a couple of years.  If so, he is on shaky ground.  First, such a claim is at odds with long-term interest-rate data. As I indicated in my article, long-term mortgage rates were not low during the housing boom.  It’s not hard to find commentary from those years recognizing the low short-term rates were not expected to last.

Second, such a claim gets closer to my hypothesis: that it was optimism that raised housing prices, not much of anything tangible during the boom.  Whether it was optimism about future interest rates, future tastes, or future technology is more of a quibble.

How Do We Explain Housing-Boom Optimism?

If I understand him rightly, I don’t much disagree with Professor Mulligan.  We agree that Federal Reserve policy acted to promote the housing price boom by lowering real interest rates.  The difficult question is:  what share of the boom can we attribute to monetary policy, and what share to other independent sources?  Applying Professor Mulligan’s way of computing the impact of lower real interest rates alone on the present discounted values of houses, correct anticipation in 2002 of real T-bill rates — which were about to go 200 basis points lower for the next three years — can account for only around a six percent rise in house prices.  Thus the milder-discounting effect by itself accounts for only a fraction of the actual run-up in prices observed, assuming correct anticipations.  The present-value calculation is straightforward.

We can get a bit more impact out of lower interest rates by noting that the lowering of mortgage lending standards implied an even larger drop in risk-adjusted mortgage rates than in risk-free Treasury rates.  Market participants did not have any clear basis in historical time series for anticipating that this drop would reverse itself soon.

Still, I agree that the joint hypothesis “real interest rate anticipations were correct and they alone fully explain the rise in house prices” is untenable.  Of course, we already knew that anticipations of house prices could not have been correct, given that nobody would pay $300,000 for a house in 2006 that he knew would be worth only $200,000 two years later. 

Professor Mulligan reasonably proposes to attribute the bulk of the rise in house prices to some kind of ex-post-mistaken (but not necessarily irrationally exuberant) anticipations, offering the hypothesis that “it was optimism that raised housing prices, not much of anything tangible during the boom.  Whether it was optimism about future interest rates, future tastes, or future technology is more of a quibble.”  Optimism about “tastes” here includes optimism about the future growth of demand in particular local housing markets.  Something like that would seem to be required to explain why the house price boom was so highly concentrated in a few states.  We can’t explain such concentration by appealing only to optimism about technology or national economic policy variables.

An appeal to optimism, of course, doesn’t really explain events but simply gives us a reframed question:  how do we explain an increase in optimism?  I suggest that optimism (regarding whatever) during this period was not independent of the rising rate of aggregate nominal income growth that was being fueled by Fed policy.  Expansionary monetary policy may have (at least cyclically) effects on relative prices and real variables, like the real demand for houses, through income channels, not only through its effect on the real interest rate.  I anticipate, and agree, with Professor Mulligan’s likely response that more needs to be done to quantify these other effects.

Does Moral Hazard from the Safety Net Help to Explain Deceptive Accounting?

William K. Black’s interesting contributions to this symposium, and in particular his drawing parallels to the S&L fiasco of the 1980s, prompt me to wonder what role financial safety nets, like the deposit insurance that fostered the S&L fiasco, have played in fostering the current crisis.

Professor Black focuses on the growth of deceptive (or fraudulent) accounting and ratings practices. Supervisors asleep at the switch enabled both. But why did the practices begin to grow? Their main purpose, according to a very informative working paper, “The 2007 Meltdown in Structured Securitization: Searching for Lessons not Scapegoats” by Gerard Caprio, Jr., Aslı Demirgüç-Kunt, and Edward J. Kane (CDK), was to mask overleveraging and excessive risk exposures from creditors and safety-net supervisors. (Note that Ed Kane was the analyst who warned about the developing S&L crisis years before it hit the newspapers.)

If they are right, then a key question becomes: what prompted the overleveraging and excess risk-taking in the first place? CDK argue that an important contributor was “safety-net subsidies,” i.e., moral hazard from implicit or explicit financial guarantees.

CDK write:

During the last forty years financial institutions in the US … became increasingly aware of opportunities to shift the deep downside of their risk exposures onto the safety net. … [In recent years] financial institutions increasingly concentrated on creating instruments and structures with which to exploit loopholes in the regulation and supervision of financial-institution leverage. (p. 6) … Securitization increased these firms’ access to safety-net subsidies not just by increasing their size, complexity or geographic footprint, but also and most importantly by concealing increases in effective leverage. (pp. 6-7)

In tolerating an ongoing decline in transparency, supervisors encouraged the very mispricing of risk whose long-overdue correction triggered the crisis. (p. 9)

Why were the deceptive practices especially keen in mortgage financing? And why were the supervisors asleep at the switch? The Federal Reserve’s cheap-money policy and its positive impact on housing prices contributed. Everything looked fine as long as house prices continued to rise: “Low interest rates and increasing housing prices encouraged an overly friendly regulatory environment both for highly leveraged mortgages and for securitization structures based on them.” (p. 12)

Like the FSLIC that practiced excessive forbearance toward zombie thrifts, hoping for the best, it is clear in retrospect that the FDIC practiced excessive deference toward the AAA risk ratings of mortgage-backed securities. It should have raised deposit insurance premiums or insisted on higher capital requirements for insured institutions heavily invested in MBSs. Likewise the OFHEO practiced forbearance toward Fannie Mae and Freddie Mac when it should have insisted on adequate capitalization for their risky portfolios.

I wonder whether Professor Black finds the CDK analysis complementary or contrary to his own.

Warped Perspective

Professor Black claims that “Rome is burning,” which I take to mean that we are in the midst of an economic disaster, and claims that I ignore it because of a theoretical bias.

2008 has been a disaster for Wall Street.  A couple of years of steeply rising oil prices has been bad for the automobile industry, and a disaster for the makers of gas guzzlers — namely, GM, Chrysler, and Ford.  However, I admit — proclaim — that I do not see 2008 as an economic disaster for the average American.  We will see what 2009 will bring, but my characterization of 2008 is largely based on data.

First let’s look at the employment data through November.  Employment is down about 2 million, which undoubtably creates stress for a couple of million families.  But is that a disaster?  Robert Hall and Susan Woodward have a chart comparing the employment dynamics today to the 1982 recession, and find that, in percentage terms, employment declined more rapidly in 1982.

By 1982 Q1, productivity had fallen 3 of 4 quarters for a cumulative decline of 2.3 percent.  Through 2008 Q3, productivity had risen six consecutive quarters, with an increase of 2.1 percent over the past four.  It is very likely that the U.S. will have the most real GDP per capita in its history in 2008, despite the fact that the entire year will be spent in recession (by the employment definition). [1]

Even the more gloomy forecasts for the next year do not portend disaster.  At quarterly rates, GDP growth, Goldman Sachs says, will fall 5/4 percent, 3/4 percent, and 1/4 percent from 2008Q3-Q4, 2008Q4-2009Q1, and 2009Q2. That’s a cumulative decline of 2.3 percent 2008Q3 – 2009Q2, or $200 billion, or about $700 per person.  Is that a disaster?

Note

[1] Through Q3, 2008 GDP was $8,771 billion (seasonally adjusted by the BEA). 2008 population (July) was 303,824,640, so 2008 produced $28,870 per person already through Q3. That means only $9,338 per capita ($2,837 billion in aggregate) needs to be produced in Q4 to break the 2007 record. In other words, if Q4 is within 3 percent of Q3, we break the record. Even the most pessimistic forecasters admit that Q4 real GDP will be greater than that.

Financial Accelerator vs. Monkey Psychology

Lawrence White writes:

Professor DeLong says little to emphasize the cluster of malinvestments, the rafts of investment projects — particularly in real estate — that have turned out to be wealth-squandering mistakes, leading to the writing down of financial claims that funded them…

I protest.

White misses my point. We have, at most, $2T of losses in securities backed by real estate investments. This is a smaller downward shock as a share of relevant financial wealth than was the collapse of overoptimism about Silicon Valley in 2000-2001, than were the triple shocks of East Asia and Russia and LTCM in 1997-1998, than were the S&Ls at the end of the 1980s, than was the portfolio insurance crash of 1987, et cetera, et cetera, et cetera.

Yet this particular downward shock has triggered a tenfold financial accelerator — a $20T collapse in the value of financial assets worldwide. That there are “malinvestments… wealth-squandering mistakes” some of them obvious ex ante but most of them only obvious ex post (which doesn’t stop people from claiming that they were obvious ex ante) is simply not news.

White’s theory seems to be that our current crisis is deserved and inevitable retribution for financial fecklessness. That theory fails. There have been many past episodes of greater fecklessness. And they have not been followed by any equal or proportionate retribution.

White’s position is, I think, a normal reaction derived from monkey psychology — the band of monkeys hangs together better if monkeys fear that they will be punished if they transgress the rules. But I don’t think it helps us understand what is going on very much.

I Am Not Now, nor Have I Ever Been, a Monkey Psychologist

In response to my observation that he had put little emphasis on the real estate and other malinvestments that kicked off our financial turmoil, Bradford DeLong writes:

That there are “malinvestments… wealth-squandering mistakes” … is simply not news.

I understand that Professor DeLong’s main interest lies elsewhere, in the question of what has made the global decline in measured financial wealth so much larger than in earlier crises.  Nevertheless — news or not — I think the opening phase of the crisis calls for explanation.

DeLong caricatures my account as follows:

White’s theory seems to be that our current crisis is deserved and inevitable retribution for financial fecklessness. … White’s position is, I think, a normal reaction derived from monkey psychology — the band of monkeys hangs together better if monkeys fear that they will be punished if they transgress the rules. But I don’t think it helps us understand what is going on very much.

In fact, I think economic reasoning about cause and effect (not monkey psychology) helps us to understand what is going on.  Economic reasoning does not require the use of, nor have I used, normatively loaded terms like “deserved” or “retribution” or “fecklessness.”  To repeat what I wrote on my blog in September: 

This isn’t a morality play. What we’re seeing are the consequences of monetary-policy distortions of interest rates and regulatory distortions of incentives, amplified in some degree by private imprudence, not the consequences of blackheartedness.

Nor, I would add, need we appeal to “fecklessness,” if that implies some additional moral failing beyond imprudence and miscalculation.

I myself don’t think that Professor DeLong’s “utilitarian’s dream” helps us to understand what is going on very much.  This appears to be a normative yardstick according to which there would be more risk-spreading, more intermediation, and more investment in an ideal world (of zero transactions costs, complete markets, and perfect information?) than in our real world.  By such a standard, “financial markets do not do a good job.” Here we seem to have a variety of what Harold Demsetz used to call the “Nirvana fallacy.”  The real never measures up to the imaginary ideal.

DeLong concludes that there isn’t a lot of sense in talking about “overinvestment” in the real-world economy, apparently because the  ideal world would always have more investment than we have even at the peak of an investment boom.  In my view, by contrast, it is perfectly sensible and useful to talk about whether investment in the real-world economy has been expanded beyond voluntary saving in the real-world economy.  It is sensible and useful to talk about whether investment was misallocated, in particular whether real-estate investment was unsustainably large. 

Professor DeLong writes that “the proper public policy response to every situation that White would characterize as one of ‘overinvestment’ is, I think, one in which the government takes steps to mobilize more of society’s risk-bearing capacity rather than letting asset prices collapse and create massive unemployment.”  But does any decline in asset prices from their level at the peak of the boom constitute a “collapse”?  Does DeLong believe that it is never proper policy to let unsustainably high asset prices return to equilibrium, and to let misemployed capital goods and labor be reallocated to better uses, provided that we take care to avoid a collapse in nominal aggregate demand and thereby the deep depression and massive unemployment that nobody wants?   

Supply Curves Slope Up, Demand Curves Slope Down

There is one huge argument against the claim that the crash in the mortgage market was in some sense the fault of excessively risky lending by the GSEs Fannie Mae and Freddie Mac which pulled the private sector along behind them: it is that Fannie Mae and Freddie Mac lost market share as all the loans that have now gone bad were made.

mortgages

As Milton Friedman always used to say: supply curves slope up, and demand curves slope down. If something is due to a change in supply that causes movement along the demand curve, price will go down and quantity will go up. If something is due to a change in demand that causes movement along a supply curve, price will go down and quantity will go down.

Between 2001 and 2006 the “price” at which Fannie Mae and Freddie Mac sold mortgages — the terms they set — certainly went down. But did the quantity go up? No: Fannie Mae and Freddie Mac together had made a much smaller share of mortgages in 2006 than in 2001: 37 percent as opposed to 48 percent. Price went down, and quantity went down.

This means that the dominant feature of the mortgage market in the 2000s was not an expansion of supply by Fannie Mae and Freddie Mac pushing their implicit government guarantee past the limits of prudence, but was a reduction in demand for Fannie Mae and Freddie Mac’s products as private-sector mortgage lenders aggressively pursued and took away their markets.

At least, this is the dominant feature if you follow Milton Friedman and believe that typically supply curves slope up and demand curves slope down.

Thank God for Anonymous Readers!

I’m back from Utrecht and eager to rejoin the discussion, which has been aided greatly by the contributions of an anonymous reader who has obviously been in the tranches.  Let me first respond to the specific questions my colleagues have posed to me. 

Do I argue that agency problems in the private sector are both an important cause of the crises and exacerbated those crises?  Yes.  My prior essays and long reply explain why, though I will amplify the answer in my next post.

Do I, Larry White asks, agree with Gerard Caprio, Jr., Aslı Demirgüç-Kunt, and Edward J. Kane (CDK)? “[W]hat prompted the overleveraging and excess risk-taking in the first place? CDK argue that an important contributor was ‘safety-net subsidies,’ i.e., moral hazard from implicit or explicit financial guarantees.” 

No.  First, CDK do not provide an explanation of why regulators permitted these activities.  One needs to explain why economists’ endorsement of non-regulation and deregulation led to non-regulation and deregulation.  One also needs to examine the ideology of those who appoint the regulators and those appointed as the top regulators.  As I discussed, when you appoint individuals who believe that regulation cannot succeed, they will prove themselves correct.  You get exactly what we have experienced — desupervision. 

Second, the great bulk of the crisis was not “excess risk-taking.”  Again, my prior pieces provide substantial analysis and data explaining these points.  Mortgage fraud (by lenders) is a “sure thing” (Akerlof & Romer 1993; Black 2005) — it is (in combination with rapid growth, high leverage, and minimal loss reserves) mathematically certain to produce extremely high short-term “profits.”  It is also certain to produce failure — the only question is how soon.  Note that control fraud (rather than “excess risk-taking”) also played the dominant role in prior crises such as the S&L debacle, Enron/WorldCom and their ilk, Russian privatization, and the travails of the “the Washington Consensus” in Latin America.

Third, unregulated (or almost entirely unregulated) lenders committed the great bulk of the mortgage.  These lenders were not covered by deposit insurance.  If CDK were correct then they should have been subject to far more stringent “private market discipline.”  Private market discipline (under their theories) is supposed to ensure adequate capitalization — which is (under their theories) supposed to minimize “moral hazard.”  Thus, the predictive strength of their model has been tested by the subprime crisis — and falsified by it for mortgage fraud (or, if the reader is still in denial about fraud despite the data I provided, which my colleagues don’t even attempt to refute, the “excess risk taking”) was significantly more common in the sector that under their theories had the least moral hazard.   

The Enron/WorldCom, et al., scandals had already shown that accounting control fraud did not require explicit or implicit governmental guarantees.  Fourth, moral hazard theory (as interpreted by CDK) did not explain the S&L debacle even though it took place in an industry with explicit governmental guarantees. They (implicitly) interpret the theory as excluding control fraud.  This is an incorrect interpretation of moral hazard theory, which arose in the insurance context and has long recognized that moral hazard can lead to either excessive risk-taking or fraud by the insured.  For the sake of testing their argument, however, I assume their definition that argues that S&L owners, made insolvent by the interest rate shock of 1979-82, responded by “gambling for resurrection.”  These were honest gambles that would only pay off for the CEO or shareholders if the investments succeeded.  The troubles with this theory include (1) the purported “rational” strategy (as interpreted by economists) was virtually certain to fail, (2) all of the purported “gamblers” economists asserted followed this strategy (the “high fliers”) first reported exceptional profits and then failed, (3) as the national commission (NCFIRRE) found, “fraud was invariably present” at the “typical large failure,” (4) the business practices that the purported gamblers followed were profoundly irrational if they were honest, but were optimal as accounting control frauds, and (5) in fact, there were honest gamblers but they (A) did not act in a manner that economists predicted and (B) they won their bets and substantially reduced the cost of the debacle.

Point 1:  economists assert that the optimal “excess risk-taking” strategy for insured banks that are inadequately capitalized or insolvent is to maximize the value of the put option. [1] The means by which one maximizes the put is to take ultra high risk.  The optimization is straightforward — the greater the risk, the greater the expected value of the put.  The obvious (except to most economists) problem with this concept is that the way to optimize is to take a bet that is so risky that it is virtually certain to fail.  Even economists should be able to understand that financial suicide is an unattractive and oxymoronic form of “optimization.” 

Point 2:  The concept of “optimization” I have just explained could explain part of the pattern of outcomes of the “high fliers” and lenders that specialized in nonprime mortgage loans, i.e., universal failure.  Why S&L subordinated debt holders and other creditors that are not protected from loss by deposit insurance never blocked these suicidal strategies by the “high fliers” is, however, inexplicable under neoclassical economic theory.  (Subordinated debt holders, who economists predicted would be the optimal agents of private market discipline because they had large amounts of at-risk funds and were financially sophisticated, also failed to block control fraud — or “excess risk-taking” — at banks, investment banks, and insurance companies.) 

The concept of optimization, however, cannot explain the first aspect of the pattern of outcomes – uniform high, initial profitability.  There is no reason why hundreds of S&Ls and scores of nonprime lending specialists (A) should have all specialized in the same small number of activities, and (B) should all have been highly profitable while purportedly engaging in ultra high-risk investments. [2]

What does fit the pattern of outcomes is accounting control fraud.  It is a “sure thing.”  It produces certain profits in initial years.  Unless the fraudulent CEO exercises restraint presumably because the company is well-capitalized and likely to generate long-term profits — which was not the case for any S&L in this era — the optimal strategy was “looting” (Akerlof & Romer 1993).  Looting is fatal and typically causes catastrophic losses.  I term the exercise of restraint in converting firm funds to the CEO’s personal use through accounting fraud “grazing.”

Points 3 and 4:  NCFIRRE (1993) found that (A) there was a distinctive pattern to the high fliers’ business practices, (B) the pattern was irrational if they were engaged in honest gambles, (3) the pattern optimized accounting fraud, and, (4) that fraud was “invariably” present at the “typical large failure.”  The nonprime lenders meet those four characteristics, as do Enron and its ilk.

Point 5:  “Traditional” S&Ls did “gamble for resurrection” in 1981-83 by maintaining much of their interest rate risk exposure.  This was not consistent with economists’ beliefs about moral hazard theory.  Every traditional S&L was insolvent on a market value basis in 1982.  Very few of their CEOs (roughly 100 of around 3000) engaged in reactive control fraud due to moral hazard.  This restraint is the most significant reason why the debacle did not grow to catastrophic proportions.  (Roughly 200 “opportunistic” control frauds entered the S&L industry in 1982-84 through changes of control or opening de novo institutions.  The typical opportunist was a real estate developer.  He had large conflicts of interest and no ties of loyalty to the S&L or its shareholders, customers, or employees.  He was more willing to loot than was the typical traditional CEO that had come up through the ranks over the course of two decades.

Moral hazard theory (as taught by neoclassical economists) predicts that traditional S&Ls would dramatically increase their risk exposure through honest gambling in 1982-84.  Instead, traditional S&Ls made modest increases in credit risk and modest-to-moderate reductions in their interest rate risk exposure.  This behavior was inconsistent with the predictions of moral hazard theory.  It was risky.  If interest rates had continued to increase throughout 1982 and beyond the overall industry insolvency (on a market value basis) of $150 billion would have increased.  Instead, the traditional S&Ls were lucky on interest rates after mid-1982, as rates fell significantly and fairly steadily.  Because traditional S&Ls only partially reduced their exposure to interest rate risk, their “underwater” mortgages regained most of their (unrealized) market-value losses.  As a result, “only” about $25 billion of that $150 billion was ultimately “realized.”  (NCFIRRE 1993).  If economists wish to define insolvent S&Ls continuing to take serious, but modestly less, risk as “moral hazard,” then they can conclude that moral hazard significantly reduced the cost of resolving the S&L debacle.

Ultimately, CDK’s logic (if not their conclusions) strongly supports the themes I have developed during this discussion.  Deregulation and desupervision in the financial sphere can have the unintended consequence of optimizing a “criminogenic environment.”  The extension of implicit federal guarantees to every large corporation operating in the United States can exacerbate the problem of control fraud.

Casey reminds me that he has GDP data (emphasis in the original) that, he argues, demonstrate that there is no “crisis” for the “average” American.  Casey is aware that I (and our readers) have data too, and that (unlike lagging GDP numbers) they demonstrate a crisis greater than any of us have experienced in our lifetimes (except for readers who are very long-lived and knew the Great Depression).  We know that there have been unprecedented failures of hundreds of markets — globally.  We know that but for unprecedented intervention by the Fed, Treasury and other central banks it is the considered judgment of the economic powers that be (here and abroad) there situation would have become catastrophic.  We see unprecedented socialism in the United States for our more elite institutions.  We see the failure (or “but for” intervention failures) of many of our most elite financial institutions.  Now, granted, I was born in Detroit and grew up in Dearborn, so the auto industry probably has lubricants running in my veins, but the “but for” collapse of the entire domestic auto industry is a kind of a big deal.  All of this happened not in some “perfect storm” but during the “great moderation.”  It is also getting worse, commercial real estate, the PBGC, SIPC and other shoes are waiting to drop.  All of these facts are data.  They are simply vastly more important data.

Brad argues that:

Thus practically every risky asset, all the time, sells for much less than its fundamental value — and does so because financial markets do not do a good job of mobilizing the risk-bearing capacity of society. I don’t think we have any prospect of living in a world in which financial markets do their job of risk-tolerance-mobilization well — nobody should trade or invest in anticipation of such a world. But I don’t think that the idea of “overinvestment” makes a lot of sense: the proper public policy response to every situation that White would characterize as one of “overinvestment” is, I think, one in which the government takes steps to mobilize more of society’s risk-bearing capacity rather than letting asset prices collapse and create massive unemployment. 

I have a very different view.  Brad believes that financial markets are commonly made deeply inefficient by “lemons markets” problems in a manner that leads to a systematic undervaluing of asset values.  Many “risky assets” trade for much more than their “fundamental value” (assuming that this concept can be made meaningful.) That is precisely why we have epidemics of control fraud.  Spreads on nonprime mortgage loans during the last decade have always been grossly inadequate and spreads narrowed when they should have grown over the course of the decade.  Moreover, the entire concept of an “adequate spread” for this kind of lending was a misnomer because had the loans been properly priced the spread would have been so large that it would have sent the “adverse selection” problem through the roof (pun intended). 

It follows that we should not seek to initiate government action to inflate housing prices to the point that there were minimal credit losses on existing mortgages.  Doing so would misallocate resources further, reward fraud, and make the markets even more inefficient. 

Larry, responding to Brad, urges:

In fact, I think economic reasoning about cause and effect (not monkey psychology) helps us to understand what is going on.  Economic reasoning does not require the use of, nor have I used, normatively loaded terms like “deserved” or “retribution” or “fecklessness.”  To repeat what I wrote on my blog in September: 

“This isn’t a morality play. What we’re seeing are the consequences of monetary-policy distortions of interest rates and regulatory distortions of incentives, amplified in some degree by private imprudence, not the consequences of blackheartedness.”

I will offer a different view.  There is nothing superior or scientific in avoiding the use of “normatively loaded terms” when such terms are accurate and important.  Readers will have discovered that economists have a profound unwillingness to use the “f” word — fraud — to discuss fraud.  You now see an effort to assert that this weakness is a virtue.  Even Brad’s euphemisms are too strong for Larry. 

The crisis we are discussing has its roots in a massive fraud, primarily by lenders.  Frauds this severe greatly extend and inflate financial bubbles.  This directly causes losses that measure in the trillions of dollars.  It indirectly causes systemic risk because (as the anonymous reader agreed) financial systems run on trust and there’s nothing like creating and then betraying trust (which is what fraud is) and causing enormous losses through that betrayal, to destroy trust.  This is the “transmission mechanism” Brad is searching for that produced the broader crisis.

Fraud has moral consequences, and morality is central, not peripheral to market efficiency.  Larry’s unwillingness to take fraud seriously, even after my data and analysis dump, is characteristic of what has gone wrong in economics. 

For the reasons shown in the earlier discussions, Larry’s claim that monetary policy caused or even contributed greatly to the housing bubble fails.  More to the point, if the “solution” were significantly higher short term interest rates the cure would have been far more expensive and harmful than appropriate micro/macro policies (i.e., regulatory restrictions on nonprime lending) that would have killed the bubble at much earlier time with far less collateral damage.

There was no “regulatory” distortion of incentives that caused or even contributed substantially to the bubble. The CEOs that funded these loans did it to create accounting profits.  It is naïve (and unsustainable) to claim they made the loans to comply with rules (that in fact did not require making or purchasing nonprime loans). 

There was, however, plenty of distortion of incentives by private parties.  They distorted those incentives to optimize accounting fraud.  They acted in a manner consistent with past control frauds. 

Economists love to talk about “unintended consequences” and they’re always in the context of social do-gooding gone wrong.  Here are two unintended consequences of non-regulation and/or desupervision.  (The consequences, of course, are intended by the private parties, but absent corruption, not by the legislators.)  First, when you don’t regulate financial activities you de facto decriminalize control fraud because the regulators are the “cops on the beat.”  Second, when you don’t regulate financial activities you make them opaque and you create a situation in which voluntary industry disclosures aren’t verified by a truly independent entity.

So, no it’s not a fictional medieval “morality play.”  It is the real-world economy, the study of which inherently requires trying to develop greater mastery of vital questions such as morals and fraud.  If you are saying that economics and economists are going to continue (1) not to develop a theory of fraud or become cognizant of the findings of other disciplines that specialize in the study of what causes markets to fail profoundly, (2) not to develop a methodology that does not praise accounting control frauds, (3) to recommend praxis that optimizes environments for epidemics of control fraud, (4) to refuse to analyze whether fraud is present, (5) to refuse to call a fraud “a fraud,” and (6) to think that this ignorance and addiction to euphemism is a virtue, then economists need to engage in a fundamental reconsideration of their profession, theories, methodologies, and policy recommendations. 

As constituted, the economics profession endangers the world economy. As even the triumphalist authors of Moral Markets concede (a book announcing the triumph and moral superiority of “free markets” that had the misfortune of being published this year), our business schools and economics programs too often continue to hold up homo economicus as the ideal — but homo economicus is, to quote that volume:  “a sociopath.”  The authors, in essence, charge that our business school and economics programs have become fraud academies.  Many students will have the moral strength to resist that training, but as economists emphasize, the concern is on the margins.    

Notes

[1] Note that this subtly removes the “moral” from “moral hazard” and turns an abusive behavior not simply into a neutral activity, but into a desirable activity.  This is dangerous because it helps “neutralize” abusive behavior, which increases abusive behavior.  As I am about to explain in my discussion of Point 5, “moral” constraints — a broad concept under “moral hazard” theory — can provide some of the most effective constraints against control fraud. 

[2] The markets offer far greater yields for honest risk-taking than honestly underwritten subprime loans or direct investments.  The reason that mortgage lenders specialized in subprime lending or that S&L specialized in “direct investments” or “ADC” loans has is that these types of investments are superb vehicles for accounting fraud, particularly during a “bubble,” not because the yield is spectacular.

The GSE Supply Curve for Nonprime Products Kept Shifting to the Right

Brad DeLong rejects “the claim that the crash in the mortgage market was in some sense the fault of excessively risky lending by the GSEs Fannie Mae and Freddie Mac which pulled the private sector along behind them,” based on the observation that “Fannie Mae and Freddie Mac lost market share as all the loans that have now gone bad were made.”  Presumably “all the loans that have now gone bad” refers to the nonprime and ARM loans made between 2004 and 2007.  He concludes that there must have been “a reduction in demand for Fannie Mae and Freddie Mac’s products” and that “the dominant feature of the mortgage market in the 2000s was not an expansion of supply by Fannie Mae and Freddie Mac pushing their implicit government guarantee past the limits of prudence.”

Professor DeLong presents his argument as an application of the logic of supply and demand.  But supply and demand curves deal with prices and quantities, not prices and market shares.  In a booming market, a declining market share is consistent with a growing contribution to supply (a continuing rightward shift in the firm’s supply curve).  His unlabeled chart of market shares, moreover, appears to depict total mortgage market shares, whereas the claim in question is about “excessively risky lending” rather than total mortgage lending.

Fannie Mae and Freddie Mac were in fact expanding their quantities of nonprime mortgages vigorously from late 2004 to 2007.  After parsing the GSEs’ financial statements, Peter Wallison of the American Enterprise Institute finds that:

From 2005 to 2007, Fannie and Freddie bought approximately $1 trillion in subprime and Alt-A loans, amounting to about 40 percent of their mortgage purchases during that period.

The GSEs thus importantly contributed to the overall supply of nonprime mortgage financing, prompting mortgage brokers to originate more nonprime mortgages.  The increased ability to sell nonprime mortgages to the GSEs and their competitors encouraged mortgage originators to dig deeper into the barrel of applicants to accept more of those previously considered non-creditworthy.

Wallison and Charles Calomiris add that:

Although a large share of the subprime loans now causing a crisis in the international financial markets are so-called private label securities — issued by banks and securitizers other than Fannie Mae and Freddie Mac — the two GSEs became the biggest buyers of the AAA tranches of these subprime pools in 2005-07. Without their commitment to purchase the AAA tranches of these securitizations, it is unlikely that the pools could have been formed and marketed around the world. Accordingly, not only did the GSEs destroy their own financial condition with their excessive purchases of subprime loans in the three-year period from 2005 to 2007, but they also played a major role in weakening or destroying the solvency and stability of other financial institutions and investors in the United States and abroad.

As to total mortgage lending from 2000 to 2005, below is a chart released in 2006 [pdf] by James B. Lockhart III, then Director of the Office of Federal Housing Enterprise Oversight, Fannie and Freddie’s regulator.  It shows the steady expansion in their quantities of mortgage-backed securities outstanding through 2005:

Regarding their total portfolios, Lockhart notes:

Fannie Mae’s mortgage assets grew from about $124 billion in 1990 to $905 billion in 2004, and then declined to about $727 billion last year. That’s equivalent to average annual growth of more than 13 percent over the 15-year period… .  Freddie Mac’s mortgage portfolio grew 26 percent per annum from less than $22 billion at year-end 1990 to $710 billion in 2005. In contrast, the residential mortgage market grew at an average rate of 8.5 percent.

Except for 2004 in the case of Fannie Mae, it is clear that this pattern is not consistent with “a reduction in demand for Fannie Mae and Freddie Mac’s products” dominating over increases in supply.  Quantities supplied increased.  They especially increased for nonprime products.

It is thus reasonable to think that the crash in the mortgage market was “in some sense,” i.e. to some extent, the fault of excessively risky lending by Fannie Mae and Freddie Mac.  Whether or not we call it “the dominant feature of the mortgage market in the 2000s” (emphasis added), it is safe to say that it was an important feature.  Fannie Mae and Freddie Mac did push the lending financed by their implicit government guarantee “past the limits of prudence.”