S&P's Tony Angel says credit ratings fared well during severe conditions and succeeded as benchmarks against default risk
The financial crisis and economic recession have been the most severe stress test for credit ratings for decades. Contrary to popular belief, they have generally stood up well in the tough conditions and fulfilled their role as benchmarks of default risk.
Much has been said and written about the value of credit ratings over the last two years, not all of it complimentary, but the test of their effectiveness is their overall track record in identifying relative credit risk. This is measured entirely empirically by studies that show the correlation over time between different rating levels and defaults, and the rate at which ratings at different levels change.
The key question is whether ratings are effectively rank ordering default risk. In other words, do highly rated credits generally display lower default rates – and greater credit stability – than credits with lower ratings?
We have acknowledged the performance of ratings in two specific areas – recent US residential mortgage-backed securities (RMBS) and collateralised debt obligations (CDOs) – has been disappointing and we have taken major steps to address that. Elsewhere, however, Standard & Poor’s ratings have generally continued to perform well.
Robust ratings
Our latest global corporate default study spells this out clearly. In a highly stressed environment – with a record number of 264 defaults among rated companies and financial institutions globally in 2009 – the performance of S&P’s corporate ratings remains robust. Investment grade ratings continue to show significantly lower default rates than speculative grade ratings.
Only 0.32% of entities rated investment grade at the start of 2009 defaulted last year, compared with 9.23% of speculative grade corporates. Indeed, 86% of all corporates that defaulted in 2009 had a first (original) rating from S&P of BB- or lower.
The same strong performance is evident in the relative stability of investment grade ratings compared with speculative grade ratings. 84.67% of issuers rated in the A category at the beginning of 2009 were still rated A by year end, compared with only 69.34% of issuers rated in the B category. Despite the extremely tough economic conditions, 64.45% of all corporate ratings were unchanged during 2009.
The 2009 performance of S&P’s global corporate ratings is broadly in line with their strong historic track record. Between 1981 and 2010, the average five-year default rate for investment grade corporate issuers is 1.24%, compared with 17.9% for speculative grade companies. Even in a highly challenging credit environment, the ability of S&P’s corporate ratings to serve as an effective measure of relative credit risk remains firmly intact.
Expectations
A similar picture emerges with European structured finance ratings. While the market valuation of many of these securities may have been depressed over the course of the crisis, their credit performance – which is what our ratings address – has been generally in line with our expectations. Default rates have been relatively modest, even among securities with lower original ratings, and those securities with high investment grade ratings have been relatively stable in credit terms.
Between mid-2007 and the end of 2009, only 0.39% (by value of original issuance) of European structured finance instruments rated by S&P defaulted. Again, generally, the higher the rating, the lower the likelihood of default: 0.36% of investment grade tranches defaulted during this period, compared with 4.17% of speculative grade tranches.
Despite the severity of the recession in Europe and rising delinquencies and defaults in both corporate and consumer loans backing many structured securities, their cushioning against credit losses has meant their ratings have generally stood up well: 88% of European structured finance ratings have been stable or even raised during the crisis, and only 12% (by value of original issuance) have been downgraded.
The ratings performance of European structured securities backed by residential mortgages and other consumer debt – residential mortgage-backed securities, covered bonds and consumer asset backed-securities – has been particularly strong, with an overall default rate of only 0.03% over the course of the crisis and a downgrade rate of only 2.5%.
No guarantees
A high rating, of course, is not a guarantee an issuer or debt issue will not default over time. Creditworthiness can and does change, sometimes as a result of unexpected and unpredictable events. Even a small fraction of credits originally rated ‘AAA’ have defaulted over the years.
However, it remains the case – as historical performance has shown – higher ratings are generally less prone to default and tend to be more stable than lower ratings. That is what investors expect of S&P’s ratings and, with the exception of US RMBS and CDOs in the last three years, that is what we have delivered.
Investors also want ratings to be comparable across asset classes, geography and time. Ratings performance should not diverge wildly for a sustained period in any particular area. That is why we have begun using stress scenarios as tools for calibrating our criteria, in order to improve future ratings comparability.
The scenarios represent hypothetical conditions corresponding to each rating category. We use a scenario of extreme economic stress, on a par with the Great Depression, as the ‘AAA’ calibration case. Our expectation is for ‘AAA’-rated credits to be able to withstand that level of economic stress without defaulting (although, of course, they could be downgraded).
Modifications
Consistent with these stress scenarios, we have updated our criteria for rating US mortgage-backed securities and CDOs. We generally raised credit enhancement levels for these types of securities, and we changed certain assumptions regarding defaults and recoveries on the underlying assets. We have also modified various methodologies and assumptions, including assuming greater correlation among securitised assets, applying more conservative criteria for ratings derived from other ratings and structures dependent on market prices, and using more qualitative analysis.
Overall, we believe these criteria changes will make it more difficult to assign high ratings to securities or issuers in sectors that have displayed poor credit performance in the recent past. These changes are driven by our overarching objective of making ratings comparable across sectors and geographic regions, and over time.
With the exception of US RMBS and CDOs, our ratings have indeed performed in a broadly comparable way during the financial crisis. And we believe the changes we have made to our criteria for the weaker-performing sectors should further enhance the future comparability of all ratings.
Tony Angel is head of Europe, Middle East and Africa at Standard & Poor’s
