Last week, Standard & Poor's, the rating agency, was sued by the U.S. Department of Justice (USDoJ) in a Los Angeles federal court for “knowingly and with intent to defraud, devis(ing), particpat(ing) in, and execut(ing) a scheme to defraud investors in (residential property securitisations) and CDOs, including federally insured financial institutions... and to obtain money from these investors by means of material false and fraudulent pretenses, representations, and promises and the concealment of material facts.”
Even to persons legally trained, this is weighty stuff. One of the most amusing ways I know to frighten an unschooled junior lawyer is to sit him or her down in front of a structure diagram of a securitisation, a jumbled mess of agreements, parties, cashflows, security arrangements, and hedging – and then change slides to display a CDO, a securitisation of securitisations, a stacked jumble of jumbles. (Instant fun.) Despite the visuals, however, such transactions are conceptually very simple: one takes assets that throw off a steady stream of income (such as residential mortgages), models the cashflows arising from them, and creates debt instruments which match the payments from the assets with the payments on the notes. Those notes or bonds are then sold, with the seller recouping the capital value of the assets in the present in exchange for investors' acquiring the future flows of income.
As such, securitisation is a remarkably versatile funding tool: one can quite literally put a chicken sandwich into a securitisation, and emerge on the other side with investment-grade debt. This is achieved through a process called “tranching,” where certain classes of debt receive higher payments of interest but absorb losses first, allowing the more senior pieces to be sold as relatively safer assets. So, for example, with a structure that issues £100 million in notes – £40m “B” and £60m “A” – if there is a shortfall of £30 million, the junior “B” tranche is wiped out by 75%, whereas the position of the senior debt is unaffected.
This is where rating agencies come in. Each agency appointed on a particular transaction assesses how these transactions have been structured by the banks and determines, on the basis of a pre-published (and public) set of criteria, their reasoning for arriving at their conclusions. Based on the amount and quality of credit enhancement for a given deal, they issue ratings to the more senior classes of debt which represent the rating agency's opinion as to the probability of default on a given piece – the better the rating, the lower that probability is.
USDoJ asserts that, as the sub-prime crisis began to unfold over the course of 2007, S&P rated 30 CDOs backed, in whole or in part, by non-prime (low-income borrower) RMBS collateral, and that “S&P knowingly disregarded the true extent of the credit risks associated with those non-prime RMBS tranches in issuing and/or confirming ratings for CDOs with exposure to those non-prime RMBS tranches.” One of these, the complaint points out, was “NovaStar ABS CDO I,” a $374 million transaction comprised of securities which were backed by subprime residential loans and issued in 2006 (76%), 2005 (18%), and 2007 (5%). S&P rated $277 million of the notes issued by the deal as AAA, or the lowest probability of default; however, the deal imploded. The result, as put by the Justice Department, was “near total losses to investors,” who were for the most part federally-insured banks.
This is an example of one small part of a wider evil which requires a $4bn lawsuit to be put right. The nitty-gritty of legal arguments – which does not bear repeating here – hinges on whether S&P's statements on these deals were actionable misrepresentations or protected speech under the US Constitution. The more important question, though, is this: is this a case that should have been brought in the first place?
From a jurisprudential standpoint, the only relevant consideration is that the statute upon which the claim is based has never been used before for this purpose (suggesting that this alleged social ill was not what the drafters of FIRREA 1989 had in mind).
In terms of economic substance, a slightly closer look at NovaStar shows that USDoJ falls very wide of the mark. As of late 2006, the company was a reasonably reputable originator of residential mortgage loans: as reported by the New York Times, NovaStar “boasted 430 offices in 39 states... fast becoming one of the top 20 home lenders in the country,” a “Wall Street darling” with “shares trading at $30 (in 2003), up from $9.50 in late 2002.” It would go higher, eventually soaring to $70 per share: “between 2004 and 2007, for instance, the company raised more than $400 million from investors,” eventually attaining a market value of $1.6 billion, with loan origination to sub-prime borrowers reaching $600 million per month.
The secret behind NovaStar's short-lived success was that the firm originated loans which, in hindsight, a reasonably prudent mortgage lender would not have, with low-quality security and to low-quality borowers. “One NovaStar loan on a property in Ohio totalled $77,500 even though the average sales price for the neighbourhood was $31,685, and the same house had been purchased two months earlier for $20,000,” reported the Times, and virtually anyone – “even... a corpse, the joke went” – could obtain one. NovaStar employed accounting methods that “gave the company lots of leeway in how it valued the loans held on its books,” such as one which “allowed it to record immediately all the income that a loan would generate over its life,” even if the mortgage had decades to run until maturity. As a result, virtually the entire market misapprehended NovaStar's business, including its securitisation business, until the subprime bubble was well into the throes of its (widely unexpected and spectacular) collapse.
When one considers that it is not a rating agency's mandate to perform investor due diligence, but rather to look at the stated characteristics of the income flows presented to it and “assess the likelihood, and in some situations the consequences, of default – nothing more or less,” one understands that if garbage data go in, whether by dint of fraud or endemic asset mispricing, garbage conclusions will invariably come out. It is this which should colour USDoJ's claims when being viewed by objective observers.
The credit enhancement involved in obtaining a AAA rating should, according to USDoJ, “on average, be able to withstand economic conditions similar to those of the Great Depression,” a roughly 1% probability of default. Most of the time the rating agencies get this right: the default rate of AAA-rated notes in structured finance transactions has, in the last thirty years, barely exceeded 0.5%, and even then only reached that historically unusual high in 2008.
But sub-prime RMBS was different, and this was not the fault of the rating agencies. Investors everywhere lost money as a result of the sub-prime crisis because they failed to conduct proper due diligence and were caught up in the largest speculative property mania in history – one which was US-government-fuelled at that. The U.S. federal government was left with considerable additional egg on its face after it then had to bail out these banks with these toxic assets on their books. But, if this lawsuit is to be believed, the primary responsibility for these losses should lie with S&P, which has allegedly perpetrated a massive fraud for its own material gain.
Recent rhetoric emerging from the populist left has capitalised on the suit, with financially illiterate commentators asserting that the rating agencies' giving AAA ratings to toxic assets was “tantamount to a massive betrayal of America.” This is plainly absurd. Many people and institutions made bad calls in the run-up to the great recession. It does not follow that it is appropriate to sue organisations which made a profit throughout, simply because they had no skin in the game. Nor will it do anything to protect the American taxpayer from being the guarantor of last resort, a precarious position in which the American people remain: not despite the role of their government, but because of it.