The Poorest of Standards

Apr 10, 2014 · 10 min read

When Standards Are Poor

In December 2001, Richard Reid tried to blow up an American Airlines international flight by igniting explosives in his shoes. The FAA immediately swung into action and in short order all departing passengers had to remove their shoes for security screening.

On Christmas Day 2009, Umar Farouk Abdulmutallab tried to blow up an international flight using explosives packed in his underwear. This did not lead to passengers having to remove their underwear — though it but did give rise to the expression “Don’t touch my junk!” which enjoyed its own fifteen minutes of fame.

No one has since been threatened by shoe bombs or knicker bombs. The authorities would have us feel reassured at what a good job they are doing, while skeptics observe that all we have done is protect ourselves against yesterday’s threats.

In the sorry tale of AAA-rated Collateralized Debt Obligations (CDOs), we are not sure which is worse: the idea that the ratings agencies may have been criminally negligent in assigning AAA ratings to toxic investments, or the fact that the government was nowhere to be seen when the disaster was a-building, but was there with a multi billion-dollar shakedown after the fact. However you view it, the authorities are making S&P go through the equivalent of removing their shoes at security, almost a decade after they failed to curtail the rating firms’ contribution to the biggest financial disaster in a century.

The courts have generally agreed with the ratings agencies’ position that what they publish is independent journalism and protected by the First Amendment. We find this disheartening, as well as incomprehensible rubbish. Meanwhile, the Department of Justice complaint against S&P says financial institutions relied on credit ratings “to identify and compare risks” among various instruments. This is also misleading and we think we know why.

Financial institutions are required by SEC rules to purchase investment-grade rated investments. Portfolio managers are also mindful of the high utility of AAA ratings in defending poor performance. It is less a question of portfolio risk — would that the average portfolio manager could recognize a credit risk if he tripped over it! — than of career risk: the government mandated CYA factor allows portfolio managers to sleep soundly at night, safe from the threat of both regulators and plaintiffs’ counsel, even as the value of the assets they manage dribbles away. The government does not want to out itself as the prime suspect in its own case, which is why the Complaint says the firms used ratings “to identify and compare risks,” rather than for their actual purpose as a government sanction to reduce their cash reserves.

Warren Buffett chuckled when Congress quizzed him about his investment in Moody’s. The government has guaranteed the rating firms a monopoly, he said, why wouldn’t you invest? We are not sure whether Buffett chuckled because he couldn’t believe Congress did not realize they were the very guarantors of that monopoly, or because he was tickled that they were all in on the joke together.

Ratings “agencies”? The SEC is an “agency.” The DEA is an “agency.” The Secret Service and the CIA are “agencies.” Calling the rating firms “agencies” implies they are an arm of the government. Under a decades-old bargain with Wall Street, the SEC designates these firms as NRSROs — Nationally Recognized Statistical Rating Organizations — and requires financial institutions to buy debt that is rated investment grade, giving the NRSROs regulatory standing. The SEC website says ( “A credit rating agency is a firm that provides its opinion…” How can the SEC discipline the raters when the Commission’s own definition accepts the First Amendment argument? This looks like a textbook definition of Regulatory Capture.

John Moody first published basic industry statistics in 1900. In 1909, Moody’s analysis of railroad investments was added to “Moody’s Manual” and an industry was born. Until the 1970s the raters made their money selling subscriptions to investors. The industry converted to the issuer-pays model on the theory that a broad payer base enables firms to take on full-time staff and perform a full modeling on a large number of issues, making it economically more efficient than the subscription model. No one mentioned the potential for conflict of interest.

In 1975, the SEC permitted banks and brokerage firms to reduce their regulatory capital reserve burden. The SEC designated certain raters NRSROs and eased requirements for portfolios holding investment grade paper. As long as most bonds looked the same, most of the raters’ models held up very well. The default rate in investment grade paper was miniscule. The system worked.

Today, even after the massive failure of rating the exotic instruments of the past decade, the biggest value of the major rating firms is their massive databases. Moody’s Investor Services has been gathering and analyzing data on a broad range of industries since 1914. Their bond models are based on a century’s worth of data and modeling experience, and the imposition of the NRSROs’ models as the standard has caused bond issuers to conform to structures the raters find acceptable.

What happened in the recent meltdown was largely attributable to two problems. First, the raters’ standard models did not apply to the new-fangled synthetic instruments being created and traded like so many hotcakes.

Debt analysts work from established models. They determine what kind of debt an issue represents, then plug in the numbers and run it against their historical database. This works very well — except when it doesn’t. In the CDO debacle, the analysts were largely using their standard models, and coming up with results that proved disastrous. Imagine taking your French traveler’s phrase book and trying to talk your way out of a drug bust in Mongolia.

Unlike analysts who write company research for stock investors, bond analysts don’t go out into the field. They rely on information from issuers. The rating firm analysts didn’t determine whether anyone was actually living in the homes that secured the residential mortgage-backed securities (RMBS — one of the categories mentioned in the DoJ complaint.) Their models were not structured to reflect the risk that a pool of mortgages might include million dollar homes bought with no money down by window washers earning $18,000 annually.

The second problem may seem surprising. It appears the competition for high-ticket business drove a frantic race to the bottom among the NRSROs. A recent paper from the London School of Economics ( concludes “relative to monopoly, rating agencies are more likely to inflate ratings under competition, resulting in lower expected welfare.” The authors find the lowering of standards was driven not by the banks pressuring the raters to inflate ratings, but by the rating firms themselves who focused on “the trade-off between maintaining reputation (to increase profits in the future) and inflating ratings today (to increase current profits.)” The rating agencies needed no prodding to act in bad faith. They were induced not by greedy bankers, but by the unfettered operation of the free market. The authors cite a range of academic work that concludes the issuer-pay model leads to inflated ratings in a competitive scenario.

The statistics are shocking — one wonders why no one was shocked at the time. “According to Fitch Ratings (2007), around 60% of all global structured products were AAA-rated, compared to less than 1% for corporate and financial issues.” How can a majority of a category be ranked “superior”? After an unprecedented jump in defaults the rating firms “lowered the credit ratings on structured products widely, indicating that the initial ratings were likely inaccurate.” The Complaint describes the panic that gripped the CDO markets in 2007 once S&P was forced by reality to start issuing downgrades.

The DoJ points to divergent assumptions: the ratings departments consistently were more upbeat on Collateralized Debt Obligations (CDOs) than were the compliance departments of the same rating firms. In the event, even those cautious assessments proved wildly optimistic, which would not have mattered anyway, as “signals from the surveillance department were ignored.”

The DoJ case appears to rest on a novel legal theory. Under the Financial Institutions Reform, Recovery and Enforcement Act of 1989 (FIRREA) Justice accuses S&P of defrauding a federally insured institution, thereby defrauding the taxpayer. The government also accuses S&P of favoring the issuing banks in order to increase market share (the conclusion of the LSE paper) and alleges “S&P falsely represented to investors that its ratings were objective, independent and uninfluenced by conflicts of interest.”

A number of internal S&P emails have been reprinted in the media. “Subprime is boiling over. Bringing down the house,” wrote one S&P-er to a colleague. Another chortled a CDO “could be structured by cows and we would rate it.” But these are mere fluff compared to internal communications the press have not reprinted, such as repeated messages from one executive showing that over 50% of transactions in one AAA product had severely delinquent loans and had already experienced a number of defaults compromising 25% of the credit underpinning the RMBS. This executive expressed frustration that senior S&P officials prevented her from downgrading subprime RMBS because of the negative impact it would have on S&P’s business.

In 2004 one executive objected vehemently to proposed “market insight” and “rating implications” practices, asking what the reaction of the market has to do “with the search for the truth.” “Are you implying,” asked the executive “that we might actually reject or stifle ‘superior analytics’ for market considerations?” The executive references the political fallout when S&P tried to publish revised Predatory Lending Criteria, acknowledging the firm was under significant pressure to be less forthright in announcing the truth, and saying the firm must retain its integrity. The email was never responded to and the new market-friendly procedures were adopted.

The Justice Department seems to argue that S&P bamboozled major financial institutions into issuing toxic mortgage-backed securities, implying that they would not have issued tens of billions of dollars’ worth of bad paper if S&P had alerted them to risks in the portfolios. This appears facetious, but the NRSROs are supposed to act as gatekeepers to the system, and are so designated by the SEC. Warren Buffett testified before Congress in the heat of the financial crisis that he didn’t blame the rating agencies for missing the risks in the market, since everyone else missed them too. Oh really? Would you buy a used AAA rating from this man? Buffett, whose job as an investor is to figure out risks and opportunities no one else sees, was not troubled that the firms whose job is to identify risks that no one else sees were not capable of identifying those risks. Where does the buck stop?

The DoJ Complaint, though sobering reading, may lead nowhere. The DoJ does not seek to prevent the courts from agreeing that the raters are mere journalists voicing an opinion, nor to impose any greater duty of care on portfolio managers and institutions when buying AAA-rated paper, nor to strengthen the SEC and its oversight of the raters. It will require tightened internal standards at the NRSROs, but the real outcome is likely to be an industry-wide ban on ever putting anything in writing, rather than any Road To Damascus rethink of the business model.

The Complaint names a number of S&P executives with significant decision making authority, including the authority to override analyst ratings and slap a higher rating on an instrument. But while they are shown making key calls at the core of the crisis, no redress is sought from these executives. The only entity charged with committing fraud, or any wrongdoing, is “S&P.” This reminds us of the signs waved in protest at the Citizens United Supreme Court decision: “I’ll Believe A Corporation Is A Person When Texas Executes One.”

In settling major fraud cases, regulators routinely take senior management aside and suggest it might be smart to allow such folks to resign. Allowing them to head off to other firms may be the right thing to do, or it may be the equivalent of jailing the rifle while freeing the murderer. Since the Complaint does not charge individuals we may never find out the nature of these executives’ acts or the extent of their influence.

Critics complain that, 2 ½ years after Dodd Frank, the SEC still has not promulgated new rules covering the NRSROs. While we have no excess fondness for the Commission, it has not so much fallen down on the job, as it has been tripped by members of Congress only too happy to see even the feeble SEC regime grind to a halt.

We think the likely outcome is a high-priced settlement, not a conviction. We don’t believe the DoJ wants to go to trial. The presence of those pesky emails probably makes this case juicy bait in their eyes, but they have to be aware of how difficult it will be to prove that S&P corporation acted with prior intent to defraud. We think it is clear that the NRSROs were negligent, in that they used tried and true models for completely new market structures, ones which not even the folks creating them really understood. They added 2 apples and 3 apples and came up with 5 oranges — and never thought to question the result. After all, the numbers worked.

And we accept the obvious power of market forces. Ratings are a commodity. S&P does not have a perceptible edge on Moody’s in terms of database or the brilliance of its analysts. It can only really compete on the basis of offering more for less, which should have been obvious to regulators, Congress, investors, issuers… coulda, woulda, shoulda.

Once this is over, we think S&P will go back to doing what it has always done best: providing decent employment for highly unremarkable business school grads — folks who learn to crunch the numbers but who don’t have what it takes to make it on Wall Street. They will continue to build and to apply their century-old troves of data, though on occasion an analyst may actually go on-site to determine whether assets pledged to secure a bond issue actually exist.

S&P employees will be given rigorous training around the use of various forms of electronic communication. The rating firms will hire additional compliance personnel, not to oversee the process of issuing ratings, but to make sure no one is sending emails about it.

Business will, in short, return to Usual. The DoJ will cash a big check and pat itself generously on the back. And just as we are now safe from shoe bombers on airplanes, we will be protected from the risks in subprime-backed RMBS. We have seen the self correcting process of the Invisible Hand in the financial markets before, and it is gratifying to know it works so well. When was the last time you heard of someone losing their entire fortune in tulip bulbs?

from “Slouching Towards Wall Street…” Notes for the Week Ending Friday, 8 February 2013 — read the book Fixing A Broken Wall Street


    Written by

    A Wall Street compliance professional tells you how it’s really done. How the 1% guarantee they stay that way.

    Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
    Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
    Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade