Most of the victims in the villainous story of the 2008 crash can offer some sort of excuse for their behavior, and even find an occasional defender, but there is one player for whom none of the survivors has a kind word. The ratings agencies acted as advertising copywriters and publicists for bad paper, scattering AAA and AA ratings like balloons at a children’s party – and after the defaults began they went wild, racing around to pop the balloons in bundles.
There was, of course, much to be ashamed of. Coming into 2008, almost 10,000 collateralized debt obligations rated by Standard & Poor’s were held in portfolios around the world. Almost half of them were downgraded in the course of the year. Of the almost 40,000 S&P-rated residential mortgage-backed securities, more than 70% were near or in default by the end of the year. S&P’s only significant competitors in ratings, Moody’s Investors Service and Fitch Ratings, had almost identical records. Nor was this stupidity limited to mortgage securities. Having kept A ratings on Enron Corp. up to the energy giant’s last week on earth, Moody’s did not downgrade its estimate of the worth of Enron’s commercial paper until the day before its collapse.
“The longer the boom lasted, the more the ratings agencies trumpeted the superiority of structured finance over loans to businesses – and the more investors came to rely on these ratings,” economists Roman Frydman of New York University and Michael Goldberg of the University of New Hampshire wrote in “Financial Markets and the State: Long Swings, Risk, and the Scope of Regulation” (Capitalism and Society, volume 4, issue 2, 2009).
Market participants, regulators and their risk models made assumptions based on credit ratings that proved incorrect,” Securities and Exchange Commission chairman Mary Schapiro – who headed the self-regulator Financial Industry Regulatory Authority during the depths of the crisis – told the Financial Crisis Inquiry Commission (FCIC) in January. “Many regulators, including the SEC, assumed that AAA-rated securities would remain liquid. Such reliance was mistaken.”
For the agencies themselves, it was great business. The revenue stream captured from the issuers of debt securities who needed ratings approached $5 billion in 2006. Even in 2010, the Big Three still enjoyed pricing power, as Warren Buffett said rather bitterly.
William Gross, managing director of investment manager PIMCO, said on CNBC that in his opinion, ratings should be abolished; his business would get along fine without them. In his May 2010 Investment Outlook, a withering dismissal of the agencies and their performance, Gross wrote: “Their warnings were more than tardy when it came to the Enrons and the Worldcoms of 10 years past, and most recently their blind faith in sovereign solvency has led to egregious excess in Greece and their southern neighbors. The result has been the foisting of AAA ratings on an unsuspecting (and ignorant) investment public who bought the rating-service Kool-Aid that housing prices could never really go down or that countries don’t go bankrupt. Their quantitative models appeared to have a Mensa-like IQ of at least 160, but their common sense rating was closer to 60, resembling an idiot savant with a full command of the mathematics, but no idea of how to apply them.”
The finance ministers of the European Union now denounce the pretensions of those who would assign simple letter grades to sovereign debt. They have rather grimly exempted Greek debt from the previous European Central Bank rule against discounting paper with less than an investment-grade rating.
Frank Partnoy, a University of San Diego law professor, has noted that the ratings awarded to CDOs tend to track rather than predict the market: The first defaults precede rather than follow downgraded ratings. The June G-20 communiqué from Toronto committed the governments “to reduce reliance on external ratings in rules and regulations.” John Heimann, a former U.S. Comptroller of the Currency and later vice chairman of Merrill Lynch and chairman of the Financial Stability Forum, says that in practice the function of the ratings agency is to go out onto the battlefield after the fighting has stopped, and shoot the wounded.
The Power of AAA
Former Goldman Sachs Group chairman Henry Paulson, looking back on his time as secretary of the Treasury and the growth of the subprime mortgage market, scornfully described ratings as a “crutch” and blamed them for destroying the prudential basis of the traders’ game. “They entered into contracts without the usual margins,” he told the FCIC, “because they had the triple-A.”
“Your basis,” said James Cayne, recalling his days as CEO of Bear Stearns Cos., “was the rating. None of us was equipped to determine whether a Merrill AA was better than a Morgan AA or a Goldman AA. All you saw was the CUSIP [security-identifying] numbers.” Cayne’s successor, Alan Schwartz, remembered “a feeling in the market that the risks were determined by the ratings of the tranches.” One notes in passing that Cayne finds the failures of information not in the disreputable “shadow banking system” that apologists for the industry like to blame, but in packages put together by the biggest names in the business. The SEC has formally accused the biggest name of all, Goldman Sachs, of assembling one CDO in the interests of those who shorted it.
There is a case to be made that the availability of ratings permitted the creation of the trading instruments that were at the heart of the disaster. Jerome Fones, executive vice president of Kroll Bond Ratings, a new venture being organized by noted financial forensics investigator Jules Kroll, argues that “CDOs wouldn’t have existed without the ratings. They played a crucial role in the design.”
As the guarantees and over-collateralization of the traditional bond market were replaced by structures, credit default swaps and tranching, the likelihood of default moved from a remote and unlikely happenstance to a mathematically predicted part of the model. And that was, of course, the function of the ratings agency: to rank debt instruments by the likelihood that they would default.
Michael Milken and Drexel Burnham had preached to the trading world that the market often – indeed, usually – saw the world as riskier than it really was. As the Federal Reserve held down interest rates, more and more investors looked for ways to increase returns at limited risk, and the ratings services offered reassurance that there really was a perpetual motion machine in the housing finance business. The Basel Committee on Banking Supervision okayed procedures that permitted banks to operate with reduced capital if they invested in high-rated paper, and the agencies were off to the races.
All this started as an independent publishing venture, Moody’s Manuals, which John Moody began selling by subscription in 1909. These were guides to railroad bonds, and then to the bonds issued by electricity generators and heavy industry. In those days, the device by which Wall Street assured itself of better information than its customer had was the trust company. Investors – mostly banks and insurance companies – needed guidance through these jungles, and they knew it. Other services imitated Moody’s, and in fairly short order the field shook down to three majors: Moody’s, Standard Statistical (which merged with Poor’s and today is a McGraw-Hill Cos. subsidiary) and Fitch (owned by Paris-based Fimalac).
Note that all were publishing companies. The prospective purchaser of bonds paid for the service by purchasing the manuals. The clerks who prepared them were not taken very seriously. The bar they set was, after all, very low: not whether the bond was a good investment or accurately priced, but whether it was likely to pay out its coupons and principal on schedule.
Irving Kahn, who was an assistant to value investing legend Benjamin Graham before they were founding members of the New York Society of Security Analysts in the 1930s (and who still
comes to work at Kahn Brothers & Co.
in Manhattan at age 104), says that “any
clerk could tell you whether the railroad
was likely to go bust or not. You don’t
need much brains for that.” As late as the
1960s, Sidney Homer, historian of interest
rates and Salomon Brothers’ first research
director, could write his informative
and hilarious The Bond Buyer’s Primer
with only a single reference to ratings.
The Great Depression saw the bankruptcy
of virtually all American railroads
and public utilities, and the decimation
of the bond market. Desperately trying
to rescue the country’s banks and insurance
companies, the Fed and the state
insurance commissioners dropped the
mark-to-market accounting rules for
bonds. Not wishing to encourage the
growth of a schlock ratings industry, the
authorities set conditions on the permit
to carry dubious assets at historic cost:
The new rules could be applied only to
“investment grade” securities – a phrase,
borrowed from the S&P manual, referring
to bonds rated BBB or better.
Eventually, all state insurance commissioners
ruled that the companies they regulated
could invest only in bonds rated at
least BBB. If the ratings agencies reduced
a bond below BBB, the insurer had to sell
it and invest the money elsewhere.
Without anyone intending the result,
these laws and regulations made the ratings
agencies the de facto regulators of
state and municipal securities underwriters
and insurance companies. Former
Comptroller Heimann points out that
they are, in effect, especially since the collapse
of the monoline bond insurers, the
only regulators of the municipals market.
The business has never come to terms
with a sea change that happened quietly around 1970. Instead of selling their
ratings through publications purchased
by the buy side of the bond market, the
ratings agencies began marketing their
services to the sell side. NYU economics
professor Lawrence White thinks the
most likely cause of the change was the
Xerox machine, which enabled non-subscribers
to get copies of ratings reports
without paying the agencies.
Corporations, investment banks, municipal
governments – none could sell
paper without a rating. Indeed, it became
traditional to get separate (almost always
very similar) ratings from two of the Big
Three. And now the banks that underwrote
the paper (and sometimes created
it) could commission those ratings themselves.
The new procedures created an
irremediable conflict of interest.
The analogy is to the authentication
of old masters. For the better part of a
century, this was done in the U.S. by art
history professors serving as consultants
to art dealers. Their fee was a percentage
of a painting’s sales price. If the professor
said it was a Rembrandt, he collected
6% of a fortune. If he said it was journeyman
17th century Dutch artist, he
got scraps from the table. This increased
the number of paintings labeled as Rembrandts
and enhanced the professors’
wealth. Eventually the practice created
enough scandal to make the historians
pass a rule establishing a fixed fee regardless
of a work’s sales price.
Ratings agencies, of course, were not
rewarded with higher fees for AAA ratings
– though there is recent testimony
that at least one charged 4.75 basis
points for rating the top tranches of a
RMBS and only 3.75 basis points for
a subsidiary tranche – but a reputation
for generosity helped draw business. In
theory, fees to the ratings agencies were
due when the ratings were published; in fact, payment was usually held until the
And now the government becomes inextricably
involved. The SEC ruled that
only Nationally Recognized Statistical
Ratings Organizations (NRSRO) could
deliver the benediction of investment
grade. Over most of 30 years when there
were effectively only three members of
this club, the commission failed to establish
or even suggest criteria for new
aspirants. Government not only created
the market for ratings services, it then
shielded the participants in that market
In 2003, Dominion Bond Rating Service,
a Canadian agency with very little
business in the U.S., was admitted to the
club, and two years later the insurance
evaluator A.M. Best was added. Another
NRSRO, Egan-Jones Ratings Co., was
founded in 1994, returned to the buyer pays
model and touts a record of being quicker than the Big Three to detect
the declining fortunes of Countrywide,
Delphi, Enron, Worldcom and others.
(The firm did not answer requests to be
interviewed for this article. Moody’s and
S&P cooperated with information but
declined to be interviewed.)
Morningstar, well known as a source
of ratings for mutual funds, in May spent
some $52 million to acquire Realpoint,
a NRSRO that specializes in structured
finance and has a substantial operation
that does surveillance on borrowers for
the benefit of and at the expense of lenders.
Meanwhile, Jules Kroll is staffing up
his new entrant but is not yet talking publicly
about the details.
The outside organizations that have
meant the most to the ratings agencies
are the banks and pension funds, insurance
companies and hedge funds that
have been willing to offer credit enhancement
for the packages being rated.
Much weight can be assigned by the ratings
agency to these pledges to pay off tranches if the underlying assets do not cut the mustard.
As PIMCO managing director and portfolio manager Paul McCulley puts it, “The friendly faces at the rating agencies, paid by the shadow banker, stood at the ready to provide seals of approval.” Here entered the large AAA guarantors like AIG, Fannie Mae and Freddie Mac, selling old rope to the underwriters for real money, secure in their certainty that the ratings agencies and the bankers themselves had investigated the issuers’ bona fides. But of course they hadn’t.
When I began visiting banks roughly 60 years ago, an old-timer told me that the quickest way to learn whether or not a bank was well run was to walk through the real estate department with downcast eyes. If the shoes you saw were dirty, it meant the lending officers were keeping up with their construction loans; if the shoes gleamed, borrowers could and probably did rip off the bank. At the ratings agencies, all the shoes were clean. And if the bond being rated was simply a set of references to real loans and bonds elsewhere, there weren’t any building sites to be inspected.
Given the depth of their involvement in the trigger moments of the Great Recession, the ratings agencies have gotten off surprisingly easy. People who do business with them every day find them as arrogant as ever. They have interposed between themselves and the wave of lawsuits from pension funds and public agencies the formidable barrier of the First Amendment. And so far the courts have duly thrown out the lawsuits. A rating is someone’s opinion, not a fact, and you can’t keep an American from expressing his opinion. S&P warns at the end of every rating missive that the recipient “should not rely on any credit rating or other opinion contained herein in making any investment decision.”
Senator Carl Levin, Democrat of Michigan and chairman of the Senate Permanent Subcommittee on Investigations, vowed to rating-agency executives that the Senate version of the financial reform bill would do something to rein them in, but the best the Senate could muster was Minnesota Democrat Al Franken’s amendment requiring that issuers pay their fees for ratings to a government repository – and the officers of that repository, not the issuers, would then choose an agency at random to make a rating.
This idea had been barely plausible when raised by Eric Dinallo, the New York State insurance commissioner. As part of the theoretical and legal underpinnings of a federal bureau to supplant what has been exclusively state control of insurance regulation, it was simply outlandish. Nevertheless, it survived in vestigial form in the bill ultimately passed by Congress, which orders the SEC (“after conducting a study and after submission of the report to Congress”) to “create a new mechanism to prevent issuers of asset-backed securities from picking the agency they think will give the highest rating.”
There is no agreement about possible solutions to the problem posed by the misjudgments of the ratings agencies. Warren Buffett points out that opening the business to more competitors could be counter-productive, because the most lenient raters would get the most business.
Transparency is not much help; when the agencies published their criteria, they were quickly gamed by the investment banks. One analyst at Moody’s testified that he always assumed the information fed to him by the bank that had assembled a CDO was “a lie.”
The Money Trail
Some economists believe the problems can be minimized if the agencies are not paid until two or three years after the issuance of the bond, to validate or falsify optimistic projections. Others suggest that ratings should be presented as a range, not as a number or a letter.
Henry Kaufman, whose career as a Wall Street economist goes back far enough to include the time when lenders rather than borrowers accumulated and paid for bond ratings, believes the problem is the business model, and that a return to the old ways is still possible. The Dodd-Frank banking reform bill essentially tells the SEC to take care of the problem. The act establishes an Office of Credit Ratings at the SEC, with an obligation to set educational standards for the analysts who will write the ratings; to perform annual inspection of ratings agencies, with authority to assess fines for dishonest work; and even to “de-register” them if they generate too many wrong ratings.
NYU professor White, who served on the Federal Home Loan Bank Board while the savings and loans were coming apart in a miasma of fraud, believes much of the harm has been done by the laws that require “investment grade” and then leave it to third parties to determine what that is. Bank examiners and insurance inspectors already make their own ratings of loans and investments, and their judgments would substitute for bond rankings in measuring the safety and soundness of an investment portfolio.
Meanwhile, says White, the buyers of bonds “would have a far wider choice as to where and from whom they could seek advice as to the safety of the bonds that they might hold in their portfolios. Some institutions might choose to do the necessary research on bonds themselves, or rely primarily on the information yielded by the credit default swap market. Or they might turn to outside advisers.”
Depending on the attitudes of the judges, the most important force for change created by the banking reform bill could turn out to be its creation of a private cause of action for purchasers of securities “recklessly” rated high by the agencies. So long as the issuers of the paper commission and pay for the ratings, there is a case to be made that the ratings agencies have obligations of diligence to the consumers of the materials they generate.
The issuer-pays model might or might not survive the loss of state laws and Federal Reserve rules requiring “investment grade” as determined by others. The great and necessary change – and not only in the bond ratings business – is away from the idea that risk judgment can be outsourced to third parties. At best, innovations louse up Gaussian distributions; at worst, the computer whimsically creates fat tails while no one is looking. Banking remains a business that earns its keep by maturity transformation, which is profitable only over time. In that world, as much as in Shakespeare, the man who sells the lion’s coat before the beast is dead may be killed in hunting him.