Credit rating agencies and the subprime crisis

Ratings were/are vital to "private-label" asset-backed securities—such as subprime mortgage-backed securities (MBS), and collateralized debt obligations (CDO), "CDOs squared", and "synthetic CDOs"—whose "financial engineering" make them "harder to understand and to price than individual loans".[9]

Earlier traditional and more simple "prime" mortgage securities were issued and guaranteed by Fannie Mae and Freddie Mac—"enterprises" sponsored by the Federal government. Their safety wasn't questioned by conservative money managers. Non-prime private label mortgage securities were neither made up of loans to borrowers with high credit ratings nor insured by a government enterprise, so issuers used an innovation in securities structure to get higher agency ratings. They pooled debt and then "sliced" the result into "tranches", each with a different priority in the debt repayment stream of income.[10] The most "senior" tranches highest up in priority of revenue—which usually made up most of the pool of debt—received the triple A ratings. This made them eligible for purchase by the pension funds and money market funds restricted to top-rated debt,[11] and for use by banks wanting to reduce costly capital requirements under Basel II.[1][12]

The complexity of analyzing the debt pool mortgages and tranche priority, and the position of the Big Three credit rating agencies "between the issuers and the investors of securities", is what "transformed" the agencies into "key" players in the process, according to the Financial Crisis Inquiry Report.[9] "Participants in the securitization industry realized that they needed to secure favorable credit ratings in order to sell structured products to investors. Investment banks therefore paid handsome fees to the rating agencies to obtain the desired ratings."[13]

According to the CEO of a servicer of the securitization industry, Jim Callahan of PentAlpha, "The rating agencies were important tools to do that because you know the people that we were selling these bonds to had never really had any history in the mortgage business. ... They were looking for an independent party to develop an opinion,"[13]

From 2000 to 2007, Moody's rated nearly 45,000 mortgage-related securities[14]—more than half of those it rated—as triple-A. In contrast only six (private sector) companies in the United States were given that top rating.[15]

By December 2008, there were over $11 trillion structured finance securities outstanding in the US bond market debt.[14]

Rating agencies were even more important in disposing of the MBS tranches that could not be rated triple A. Although these made up a minority of the value of the MBS tranches, unless buyers were found for them, it would not be profitable to make the security in the first place. And because traditional mortgage investors were risk-averse (often because of SEC regulations or restrictions in their charters), these less-safe tranches were the most difficult to sell.[16][17]

To sell these "mezzanine" tranches, investment bankers pooled them to form another security—known as a collateralized debt obligation (CDO). Though the raw material of these "obligations" was made up of BBB, A−, etc. tranches, the CRAs rated 70%[18] to 80%[19] of the new CDO tranches triple A. The 20–30% remaining mezzanine tranches were usually bought up by other CDOs, to make so-called "CDO-Squared" securities which also produced tranches rated mostly triple A by rating agencies.[20] This process was disparaged as a way of transforming "dross into gold"[21] or "ratings laundering"[22] by at least some business journalists.

Trust in rating agencies was particularly important for CDOs for another reason—their contents were subject to change, so CDO managers "didn't always have to disclose what the securities contained". This lack of transparency did not affect demand for the securities. Investors "weren't so much buying a security" as they "were buying a triple-A rating", according to business journalists Bethany McLean and Joe Nocera.[23]

Still another structured product was the "synthetic CDO". Cheaper and easier to create than original "cash" CDOs, these securities did not provide funding for housing. Instead synthetic CDO-buying investors were in effect providing insurance (in the form of "credit default swaps") against mortgage default. Synthetics "referenced" cash CDOs, and rather than providing investors with interest and principal payments from MBS tranches, they paid insurance premium-like payments from credit default swap "insurance". If the referenced CDOs defaulted, investors lost their investment, which was paid out as insurance.[24] Because synthetics referenced another (cash) CDO, more than one—in fact numerous—synthetics could be made to reference the same original. This multiplied the effect if a referenced security defaulted.[25][26]

Here again the giving of triple-A ratings to "large chunks"[27] of synthetics by the rating agencies was crucial to the securities' success. The buyer of synthetic tranches (who often went on to lose his investment) was seldom an analyst "who had investigated the mortgage-backed security", was aware of deteriorating mortgage underwriting standards, or that the payments they would receive were often coming from investors betting against mortgage-backed security solvency. Rather, "it was someone who was buying a rating and thought he couldn't lose money."[28]

By the end of 2009, over half of the collateralized debt obligations by value[29] issued at the end of the housing bubble (from 2005–2007) that rating agencies gave their highest "triple-A" rating to, were "impaired"—that is either written-down to "junk" or suffered a "principal loss" (i.e. not only had they not paid interest but investors would not get back some of the principal they invested).[30][31][32][33][34] The Financial Crisis Inquiry Commission estimates that by April 2010, of all mortgage-backed securities Moody's had rated triple-A in 2006, 73% were downgraded to junk.[35]

Mortgage underwriting standards deteriorated to the point that between 2002 and 2007 an estimated $3.2 trillion in loans were made to homeowners with bad credit and undocumented incomes (e.g., subprime or Alt-A mortgages)[6] and bundled into MBSs and collateralized debt obligations that received high ratings and therefore could be sold to global investors. Higher ratings were believed justified by various credit enhancements including over-collateralization (i.e., pledging collateral in excess of debt issued), credit default insurance, and equity investors willing to bear the first losses. But as of September 2008, bank writedowns and losses on these investments totaled $523 billion.[6][7][36]

MBS Credit Rating Downgrades, By Quarter[37]

Rating agencies lowered the credit ratings on $1.9 trillion in mortgage backed securities from the third fiscal quarter (1 July—30 September) of 2007 to the second quarter (1 April–30 June) of 2008. One institution, Merrill Lynch, sold more than $30 billion of collateralized debt obligations for 22 cents on the dollar in late July 2008.

The net worth of financial institutions owning the newly downgraded securities declined, requiring the institutions to acquire additional capital, to maintain capital ratios, which in turn often lowered the net-worth value of the institutions above and beyond the low of value of the downgraded securities.[37] Adding to the financial chain reaction were regulations—governmental or internal—requiring some institutional investors to carry only investment-grade (e.g., "BBB" and better) assets. A downgrade below that meant forced asset sales and further devaluation.[38]

Dozens of suits involving claims of inaccurate ratings were filed against the rating agencies by investors.[44] Plaintiffs have included by collateralized debt obligation investors (the state of Ohio for losses of $457 million,[45][46] California state employees for $1 billion[47]), the bankrupt investment bank Bear Stearns (for losses of $1.12 billion from alleged "fraudulently issuing inflated ratings for securities"[48]), bond insurers.[49] The U.S. Government is also a plaintiff (suing S&P for $5 billion for "misrepresenting the credit risk of complex financial products").[50][51][52]

On 11 June 2008 the U.S. Securities and Exchange Commission proposed far-reaching rules designed to address perceived conflicts of interest between rating agencies and issuers of structured securities. The proposal would, among other things, prohibit a credit rating agency from issuing a rating on a structured product unless information on assets underlying the product was available, prohibit credit rating agencies from structuring the same products that they rate, and require the public disclosure of the information a credit rating agency uses to determine a rating on a structured product, including information on the underlying assets. The last proposed requirement is designed to facilitate "unsolicited" ratings of structured securities by rating agencies not compensated by issuers.[53]

On 3 December 2008, the SEC approved measures to strengthen oversight of credit rating agencies, following a ten-month investigation that found "significant weaknesses in ratings practices," including conflicts of interest.[54][55]

Structured investment mortgage-related securities were the rating agencies' "golden goose"—in the words of one agency manager.[36] Agencies earned as much as three times more for grading these complex products as for corporate bonds, their traditional business. On top of revenue generated for issuing credit ratings, agencies often earned $300,000–500,000 and as much as $1 million to construct a structured investment vehicle.[47] By 2007 the business accounted for just under half of the total ratings revenue and all of the revenue growth for Moody's—one of the largest agencies.[57] But there was always a danger of losing out on this lucrative business. Issuers played the three big credit agencies off one another, 'shopping' around to find the best ratings.

Richard Michalek, a former vice president and senior credit officer at Moody's, testified to the FCIC that even when they were not realized, "The threat of losing business to a competitor ... absolutely tilted the balance away from an independent arbiter of risk ...." When asked if the investment banks frequently threatened to withdraw their business if they didn't get their desired rating, former Moody team managing director Gary Witt told the FCIC,

“Oh God, are you kidding? All the time. I mean, that's routine. I mean, they would threaten you all of the time. . . . It's like, ‘Well, next time, we're just going to go with Fitch and S&P.'”[58]

At Standard & Poor's rating service one subpoenaed email sent by a security-issuing banker angry over possible revision of residential mortgage-backed security ratings, told an analyst: "Heard your ratings could be 5 notches back of moddys [sic] equivalent, Gonna kill you resi biz. May force us to do moddyfitch [S&P competitors Moody's and Fitch Rating] only ..."[59]

Another email between colleagues at Standard & Poor's written before the bubble burst, suggests awareness of what would happen to the securities they were giving top ratings to: "Rating agencies continue to create and [sic] even bigger monster--the CDO market. Let's hope we are all wealthy and retired by the time this house of cards falters."[60][61]

Critics have claimed there was a conflict of interest for agencies—a conflict between accommodating clients for whom higher ratings of debt mean higher earnings, and accurately rating the debt for the benefit of the debt buyer/investor customers, who provide no revenue to the agencies.[62][63] Being a publicly traded firm intensifies the pressure to grow and increase profits.[64] Of the two biggest agencies Moody's became a public firm in 2001, while Standard & Poor's is part of the publicly traded McGraw-Hill Companies.

One study of "6,500 structured debt ratings" produced by Standard & Poor's, Moody's and Fitch, found ratings by agencies "biased in favour of issuer clients that provide the agencies with more rating business. This result points to a powerful conflict of interest, which goes beyond the occasional disagreement among employees."[51]

While Moody's and other credit rating agencies were quite profitable—Moody's operating margins were consistently over 50%, higher than famously successful Exxon Mobil or Microsoft, and its stock rose 340% between the time it was spun off into a public company and February 2007[65]—salaries and bonuses for non-management were low by Wall Street standards and its employees complained of overwork.

According to journalists McLean and Nocera, "The analysts in structured finance were working 12 to 15 hours a day. They made a fraction of the pay of even a junior investment banker. There were far more deals in the pipeline than they could possibly handle. They were overwhelmed. Moody's top brass ... wouldn't add staff because they didn't want to be stuck with the cost of employees if the revenues slowed down."[66]

Moody's Team manager Gary Witt complained that "penny-pinching" and "stingy" management was reluctant to pay up for experienced employees. "The problem of recruiting and retaining good staff was insoluble. Investment banks often hired away our best people. As far as I can remember, we were never allocated funds to make counter offers. We had almost no ability to do meaningful research."[67] When asked about this by the FCIC, Moody's president Brian Clarkson admitted that investment banks paid more than his agency so retaining employees was "a challenge".[68]

Journalist Michael Lewis argues that the low pay of credit rating agency employees allowed security issuers to game the ratings of their securities. Lewis quotes one Goldman Sachs "trader-turned hedge fund manager" telling him, "guys who can't get a job on Wall Street get a job at Moody's," as Moody's paid much less. This was despite the fact that a job at Moody's—or any of the other rating agencies—gave an analyst the power to upgrade or downgrade a security, whereas the higher paid analysts' recommendations at Wall Street investment banks had no such impact on the market.[69] However, the difference in pay meant that the "smartest" analysts at the credit rating agencies "leave for Wall Street firms where they could use their knowledge (of criteria used to rate securities) to manipulate the companies they used to work for." Consequently, it was widely known on Wall Street that the "inner workings" of the rating models used by the credit rating agencies, while "officially, a secret", "were ripe for exploitation."[70] At least one other investment firm that bet against the agencies' credit ratings with huge success believed "there was a massive amount of gaming going on."[71]

When asked by the FCIC Commission about "the high turnover" and "revolving door that often left raters dealing with their old colleagues, this time as clients", Moody's officials stated their employees were prohibited from rating deals by a bank or issuer while they were interviewing for a job with that particular institution, but notifying management of any such interview was the responsibility of the employee. After getting a job at an investment bank, former employees were barred from interacting with Moody's on "the same series of deals they had rated while in its employ", but not on any other deals with Moody's.[68]

What was it ex-rating agency analysts might show their new employers? According to the hedge fund managers Michael Lewis talked to who had bet against mortgages securities, there were a number of ways to game[72] or "reverse-engineer" the raters' models.[36]

Rating agencies judged creditworthiness of a pool of loans in part by looking at the averages of credit scores of borrowers who made up the security. The agencies used FICO, the "best-known and most widely used credit score model". The average FICO score needed to be about 615 for a pool of loans to meet rating agencies' minimum standard and allow a maximum percentage of triple-A rated tranches.[72]

While using an average was less work than getting a list of the borrowers' individual scores or finding the standard deviation of the pool of scores, it left out useful information. A pool of loans composed of borrowers all of whom had a FICO score of 615 was likely far fewer defaults than a pool of loans with the same average but more dispersion—e.g.composed of borrowers half of whom had FICO scores of 550 and half 680, since someone who had earned a FICO score as low as 550 "was virtually certain to default". Knowing this blind spot, securities issuers who could no longer find high-FICO-scoring families who wanted to take out a mortgage found other ways to raise the average pool score.[72]

One way was to convince immigrants to buy homes. People who had not been in the country long, often had "never failed to repay a debt, because they had never been given a loan". Such people had surprisingly high FICO scores if you ignored the short credit history or "thin file". Rating agencies did, and FICO scores ignored and personal or household income.[72] Thus did low income immigrants increase the percentage of pool of loans that could be declared triple-A"[73] and explain otherwise unlikely sounding press reports of a "Mexican strawberry picker with an income of $14,000 and no English" being "lent every penny he needed to buy a house of $724,000".[73][74]

Other "opportunities" for issuers manipulating credit raters included

• Moody's and S&P favoring of "floating-rate mortgages with low teaser rates over fixed-rate" mortgages;
• their lack of interest in whether "a loan had been made in a booming real estate market or a quiet one";
• whether a lender had a 'silent second',[75] i.e. an undisclosed second mortgages that left the homeowner with no equity in his home and thus no financial incentive not abandon it if real estate prices declined; and
• "the fraud implicit in no doc loans".[72]

"The models used by the rating agencies were riddled with these sorts of opportunities", according to Lewis. "The trick was finding them" before other security issuers did.[72]