The agencies are under attack again!
CreditSights, a New York based Credit Research Firm has launched another attack on its Credit Rating Agency competitors with the release of a report “Distressed CPDOs: We’re Doomed!”. It should be noted that by “Credit Research Firm”, I mean that they are paid by their subscribers; as opposed to “Credit Rating Agencies” which are paid by the issuers.
In other words, to make a living they have to convince the buy-side that the CRA ratings are worthless and that the buy-side should therefore pay them for their analysis. Which is not to say they’re wrong, but it’s always a good idea to follow the money.
There may be one or two people in the galaxy who are unaware of just what is meant by CPDO (it’s not a Star Wars character). The Default Risk site has republished a case study, First Generation CPDO: Case Study on Performance and Ratings and a UBS Primer on the topic.
Basically, the idea is … well, we we all know what a CDS is, right?
consider, on one hand, a portfolio composed of (1) a short position (i.e. selling protection) in the CDS of a company and (2) a long position in a risk free bond. On the other, consider an outright long position in the company’s corporate bond, all with the same maturity and par and notional values of $100. These two investments should provide identical returns, resulting in the CDS spread equaling the corporate bond spread.
Essentially, what a CPDO does is (synthetically) purchase a portfolio of five year bonds (5 years is the standard CDS term), lever the hell out of it (with leverage financing at, essentially, the risk-free rate due to the definition of a CDS) and aim to capture the spread on such a portfolio over a ten year term.
The critical point is that the investment horizon is longer than the term of the assets. To a first approximation, therefore, you don’t really care what happens to spreads, since if they increase then you get to reinvest less money (since there’s a capital loss) at the higher spread.
As the UBS primer points out, the three major CPDO risks are:
- credit events in the underlying portfolio
- costs of exiting the old off-the-run index
- low premium on the new on-the-run index
They conclude, after examining a representative hypothetical product, that the CPDO can withstand 7% annual losses, which will result from 0.8% of the underlying portfolio defaulting with 67% severity every year, while leveraged 13X. Such a loss may also result from the old off-the-run index rising in premium by 12.5 bp per year every year.
A low premium on the new index may result from migration … the CDSs in the index are replaced if the credit rating falls below investment grade. Therefore, the old index may include junk credits at a high premium while the new index will not contain these elements. Therefore, there will be a yield give-up when rolling the index. As Fitch says in their review:
This migration historically shows a downward trend for investment grade assets, which means that they are more likely to get downgraded than upgraded. Every quarter, the negative trend in the migration process leads to an increase in the portfolio spread relative to the underlying driving spread. For a CPDO, this idiosyncratic spread widening will cause MtM losses, which are crystallised on each roll date or following a de-leveraging event. The impact on NAV is significant. For instance, a widening of 5bp every six months in a transaction leveraged 15x on a five-year index with 4.3 years of duration equates to 5 x 4.3 x 15 = 322bp or 3.22% NAV decrease.
This post comes about because of a Bloomberg story: CPDOs Rated AAA May Risk Default, CreditSights Says:
To make matters worse, the CPDOs are likely to earn a lower premium on the new CDX Series 9 index because the credit risk will be lower as the downgraded companies drop out. At least five companies in the CDX and iTraxx indexes have lost investment grade ratings and will have to be replaced, according to Watts. Without the downgraded companies, the new CDX index may be priced 11 basis points tighter than the current benchmark, JPMorgan Chase & Co. analysts led by Eric Beinstein in New York said in a report published this week.
which is discussed in the Fitch paper under the heading “Migration Driven Spread Movements”:
The average migration causes around 2.4% of spread widening over a six-month period or around 2bp for a spread of 80bp. Fitch’s model for migration is not constant but stochastic. It also generates extreme migration scenarios that would cause 20 to 30bp of spread widening over six months. The impact of credit migration is also relative and increases when spreads are high in the model.
So, it’s not as if Fitch didn’t consider this risk, anyway! CreditSights is simply claiming that the risk has been miscalculated.