Testimony presented to the US Senate from both sides of the debate made interesting reading for both rating agencies and their detractors at the end of September. The agencies have, not surprisingly, been at pains to remind the market exactly what a rating means, and how seriously they take their job. They have an air of righteous indignation – hardly surprising given the mauling they have been given from all sides. "We had not heard a peep from the investor community before about our sub-prime criteria," grumbles one rater. "It is only when they are trying to keep their jobs because they bought sub-prime that they start complaining."
There is a degree of support for their position. "The rating agencies have received a bum rap in all of this," says an observer. "It is not their job to undertake due diligence on underwriting standards in the mortgage market." But certainly the weight of opinion among market participants that Euromoney spoke to in October was against them – or at least in favour of a substantial reassessment of how the relationship between the agencies and the market works.
Point one
The first point on the charge sheet is the most straightforward; they got their assumptions badly wrong on how the sub-prime mortgage pools backing securitized structures would behave. Given what has happened over the past six months, this is more a statement of fact than an accusation. In the space of a couple of weeks in October, Moody’s announced the downgrade of $33 billion of securities backed by US sub-prime home loans, while Standard & Poor’s downgraded $22 billion.
The agencies have made recent efforts to address some of the criticism by revising their loan-loss expectations for the worst loan vintages – 2005 and 2006. But many feel that this is not enough. "One of our biggest concerns on sub-prime RMBS ratings that has yet to be addressed is the questionable utility of avoiding making default projections that do not stem from bottoms-up analysis of the individual loans in RMBS loan pools," noted CreditSights in an October 15 report. The adjustments to their models that the rating agencies have announced seem to be focusing on the deal originator and vintage year as a proxy for the particular characteristics of individual loans. "This flies in the face of traditional credit analysis, which pays little attention to the source of the funds of the credit at the time the credit was made, but focuses instead on the creditworthiness of the borrower and the value of the collateral backing the credit," CreditSights claims.
Another problem is that updated house-price appreciation (HPA) assumptions are not applied to existing deals. If HPA is deemed likely to change significantly this will have an impact on all deals in the market, not just new ones, so all deals should be revisited.
The implications of the misjudged assumptions have been most sorely felt in the ABS CDO sector. In testimony given to the US House of Representatives committee on financial services and sub-committee on capital markets, insurance and government-backed enterprises, Kyle Bass, managing partner at Hayman Advisors, explained why he felt that the market has been hit so hard. "Original correlation assumptions on the assets within these structures [ABS CDOs] were not based on any empirical data but simply reflected the best guess of the NRSROs," he claimed. "The success of these products drove changes to the asset compilations and by late 2003 ABS CDOs were comprised almost entirely of sub-prime mortgages. It is my belief, based on conversations with various structured finance marketing groups at bulge-bracket broker dealers, that NRSROs did not alter their correlation assumptions despite the homogenization of the collateral underpinning ABS CDOs."
Point two
The second charge on the sheet is that the agencies awarded triple-A ratings to securities that did not embody triple-A risk, given how the appetite for these bonds has evaporated and their value plummeted. This oft-repeated criticism is one that the agencies angrily rebut, pointing out that they do not address market risk in their ratings. In a presentation to the committee on banking, housing and urban affairs on September 26, Standard & Poor’s executive vice-president Vickie Tillman stated flatly that "Ratings speak to one topic and one topic only – credit risk. As we have repeatedly made clear in public statements, including statements to the SEC, testimony before congress and innumerable press releases, ratings do not speak to the likely market performance of a security." But certainly for structures such as SIVs and ABCP conduits, there must be some degree of market risk implicit in their ratings? This is indeed the case: "In many cases of structured finance debts, one of the components of credit risk is the market value of something," concedes Ian Bell, head of European structured finance ratings at S&P. But the fact that investors believed that a triple-A rating meant liquidity is, claim the rating agencies, not their fault. And many agree: "The idea that rating agencies are somehow responsible for the liquidity of the bonds that they rate is utterly wrong," says a securitization expert.
 |
|
"Our job is to predict the future, which is not a science" Ian Bell, S&P |
Rating agencies tend to defend their reputation in certain asset classes by pointing to their track record of upgrades versus downgrades over the past 30 years and emphasizing what a rating actually means. "Our job is to predict the future, which is not a science," Bell tells Euromoney. "We have acknowledged that our assumptions and the data on which we based our ratings of recent sub-prime securities have proved no longer as reliable given the unprecedented behaviour of recent sub-prime loans. As a result, we have updated our assumptions and rating opinions on these securities." In other words, we got it wrong. But Bell maintains that it is still too early to tell to what degree. "There is still too much smoke on the battlefield to draw final conclusions on the lessons learnt," he claims.