In defence of CDOs
CDOs (Collateralised Debt Obligations) are being hailed in some quarters as the next split cap catastrophe. As a result of the US sub prime mortgage crisis, some CDOs that are heavily exposed to the US subprime mortgage crisis have fallen dramatically in value and hedge funds that have a big exposure to these CDOs are on the brink of collapse. There are legitimate concerns with some aspects of the CDO market and perhaps the biggest one is that there isn’t a hell of a lot of confidence that the ratings agencies are on the ball (see my previous comment). Mervyn King touched on this subject at a recent speech too (as reported by Richard here). However, a lot of fear and confusion about CDOs comes from a lack of understanding of what they actually are.
A CDO is essentially just like an asset backed security (ABS). The assets that form the collateral for ABS deals are usually pools of credit card loans or car loans, but they could theoretically be anything that throws out a cash stream. David Bowie sold the rights to his songs for $55m in 1997 by issuing a Bowie bond, while Calvin Klein once issued a bond linked to future sales of its perfume products.
In a basic CDO, the collateral is a diverse portfolio of corporate bonds, high yield bonds, asset backed securities, mortgage backed securities and/or leveraged loans. The cash flows from this portfolio are then repackaged into different tranches, each with different risk and return characteristics and these tranches are sold to investors. Cash flows are received by the investors in the lowest risk tranche first, while the highest risk tranche receives cash flows last.
The highest rated tranche forms about 70% of the structure of a CDO and carries an AAA rating because it is the first in the queue and there would have to be a huge number of defaults in the underlying portfolio (about 30%) before any of the AAA rated investors start getting hit. Credit ratings of the tranches deteriorate as you go further down the queue, reflecting the fact that there’s less of a buffer, and right at the bottom is an unrated ‘equity’ tranche which is first in the firing line if things go wrong. Investors in the equity tranche stand to make fantastic returns as long as nothing defaults in the underlying portfolio, but as soon as anything does go bust then these investors start getting hit. And that’s exactly what’s happening now in the US.
Some CDOs have been heavily invested in mortgage backed securities, which themselves were backed by pools of subprime mortgages. These mortgage backed securities may have once been rated BBB, but now that so many US sub prime borrowers are defaulting, the credit ratings of these mortgage backed securities have plummeted. Many investors in CDO equity tranches are hedge funds, who were attracted to the prospect of returns of over 20% per year. Unfortunately, it seems that some investors in CDO equity tranches have also been pension funds, whose gullible trustees clearly didn’t have clue what they were getting into and have failed in their fiduciary duty.
Are all CDOs bad? As long as the underlying collateral is of sound quality, then they’re not bad at all. So called ‘CDO squared’ products, which are CDOs invested in other CDOs (this stuff really is a bit like split caps) are potentially poisonous, but they’re really not that common. CDOs in which the collateral is leveraged loans (these are Collateralised Loan Obligations, or CLOs) are actually very attractive right now and Richard has bought some for the M&G Optimal Income Fund.
In any area that has seen such phenomenal growth, there are likely to be some cases of abuse of the system, but that doesn’t absolve sophisticated investors of not reading the small print, or understanding the investment they are buying. It should also not deflect from how CDOs can offer attractive investment opportunities.
The value of investments will fluctuate, which will cause prices to fall as well as rise and you may not get back the original amount you invested. Past performance is not a guide to future performance.