The court ruling that the ratings agencies have been dodging for five years finally arrived this month, courtesy of an Australian court, which found S&P culpable for losses on a credit derivative structure.
While the cost to S&P was small – A $30 million (£19 million) payable to a local authority which lost A$16 million on a ‘grotesquely complicated’ and erroneously AAA-rated structure known as a CPDO, or a constant proportion debt obligation – the potential precedent is huge.
At issue was not, as in previous cases, the question of whether a rating is advice or guidance. The judge instead ruled that the flawed rating process itself was criminally ‘misleading and deceptive’.
‘In the US and elsewhere, the ratings agencies have [previously] very successfully avoided liability by arguing on freedom of speech grounds,’ noted Matthew Barnes, credit manager at ACPI, and previously a member of a collateralised debt obligation (CDO) team at New Bond Street Asset Management.
S&P is to appeal the judgement. The potential ramifications were enough to send shares in S&P parent company McGraw-Hill down as much as 7.1% – its biggest fall since August 2011, according to Bloomberg – with shares in Moody’s also off in an otherwise rising market.
The danger for S&P is the authority of the judgement. Judge Jayne Jagot delivered her verdict over 1,500 pages, and probably became the world’s leading CPDO expert in the process.
‘The explanation unavoidably refers to complex concepts which are likely to be unfamiliar to those without specialist expertise in structured finance,’ she wrote, with considerable understatement.
Essentially the instrument, issued by ABN Amro made a 10-year maturity, 15 times levered investment in the spread available on a five-year investment grade index, rolling over every six months.
If the spread widened over each six-month period, the instrument booked profit; if the spread tightened, it used the leverage to double down on the position and hoped it came good next time.
More important than the actual mechanics of the structure, however, were the assumptions necessary to win the AAA rating – described by Reuters columnist Felix Salmon as ‘bonkers’.
ABN Amro not only recruited former S&P staff to explicitly design a structure which would game the company’s risk systems, S&P then unquestioningly adopted the bank’s extremely optimistic modelling.
If that were not bad enough, S&P then unquestioningly adopted ABN Amro’s range of return assumptions – which to say the least, bore only a passing resemblance to any credible probability.
In the words of judge Jagot, ‘a reasonably competent ratings agency’ would never have come anywhere close to awarding the CPDO, based on the actual maths, anything like its highest award.
Of broader issue than the specific liability of S&P, the case also raises wider questions about the role of ratings agencies in fixed income investment and the usefulness or otherwise of rating systems.
‘One thing we can safely say,’ said Adrian Owens, manager of the £597.8 million GAM Star Global Rates fund. ‘The onus is now on us: any manager these days who is placing much reliability on credit agencies is really not doing their job.’
The issue takes on wider significance once you put it in the context of the extraordinary growth in AAA-rated credit over the last 20 years. Figures from the Bank for International Settlements show that as a proportion of total issue, AAA-grade tripled to 60% between 1992 and 2006.
The proportion dipped in 2006 but then rose back toward its pre-crash high in the years following. The securitised component was largely removed, replaced by a surge of sovereigns, which have risen from 10% of total AAA instruments to almost 30% today, jumping 10% in 2009 alone.
Barnes points out that the same reason that ABN Amro took the CDPO to S&P rather than Moody’s – S&P produces ratings based on the probability of default without reference to the recovery value in default – was the same reason that the agency was alone in downgrading US Treasuries in 2011.
‘The S&P committee looked at the debt-to-GDP ratio and all the unfunded liabilities and said this is a structural deficit and doesn’t have AAA characteristics. In another respect though, the claims payment capacity of the US remains pretty good – there is a lot of demand for secure fixed income, so the US can continue issuing and rolling over its debt with a high likelihood of paying out.
‘That is partially a reflection of central banks being able to buy their own debt and can see why the fundamentals of that make some people upset. I would say there are good reasons to continue considering the US as AAA, but there is a source of disconnect.’
Recent research by JP Morgan suggested half of all AAA-rated sovereign debt was held by central bank-related funds.