Jamal Mecklai: AAA-chhoooo!

MARKET MANIAC

]. And while it may take a few weeks (or months) for the terror to subside, it is important that we learn the right lessons from this current (mis)adventure.

To recap, a few weeks ago, Bear Stearns announced that two of its hedge funds had suffered heavy losses as a result of defaults in the sub-prime mortgage market. There had been a few tremors reported earlier, but it seemed like the Bear Stearns announcement focused the market's attention sharply on what was soon to come. And sure enough, more skeletons started tumbling out of the closet, and CDO investment rapidly dried up (from over $40 bn in June to less than $2 bn in July). The sudden increase in risk aversion sent US Treasuries""still the ultimate safe haven""screeching higher and corporate borrowing spreads widened sharply.

The sudden increase in nervousness spilled into equity markets, with a particularly vicious focus on companies that had been labelled or announced take-over targets. With the Dow falling sharply, equities around the world tumbled as well. The yen shot higher as increasing risk aversion led huge unwinding of the carry trade, with the Aussie and the Kiwi taking the biggest beating. The dollar rallied against most currencies (other than the yen), probably linked to the huge buying in US Treasuries.

This terror""an appropriate response to market excess""had its roots in the huge increase in risk appetite over the past few years, driven by the ready availability of very cheap money. This very cheap money was being manufactured in Wall Street factories, making structured products, like collateralised debt obligations (CDOs), by bundling together pools of assets""say, a group of mortgages, some (sometimes all) of which were of low credit quality""and breaking them up into parts with different risk/return characteristics. The very low-risk (AAA) parts were sold to conservative investors while the super high-risk (and super high-return) parts were sold to hedge funds. The magic in all this is that the overall cost of these now-obviously-mispriced CDOs was very low, so banks could use them to provide cheap finance to all manner of companies, notably buy-out firms, whose hectic activity was a key fuel to the rapid rise in equities, particularly in the US.

Now, the niggling nub of it all was that the AAA portions of the CDOs""acknowledged as such by the major rating agencies (for fat fees, I might add)""behaved quite differently than other AAA instruments, like, say, bonds issued by GE. Not only did they respond differently to interest rate movements, but they also responded differently""and sometimes dramatically""to changes in volatility, or, of course, if there was even a single default within the pool of assets underpinning the CDO.

Thus, there were now totally different instruments floating around the market that looked the same from a ratings standpoint. Of course, while markets were calm, they behaved quite similarly""only as differently as, say, AAA sheep and AAA goats. But when markets got volatile, one type (straight AAA bonds) continued to behave like sheep""perhaps simply lifting its head from grazing""while the other""oh, my""turned into something out of a mathematicians' nightmare.

Not only did the prices of these quasi-goats fall sharply, but, since these were very thinly traded, there was no real price discovery available. And, so, if someone had to bail out of a CDO by selling one of the tranches at a distressed price""as many funds, notably the ones managed by Bear Stearns, had to""this new price would, according to accounting standards, have to be used to value the continued holdings, which in turn led to valuation losses, margin calls, and the worst mathematicians' nightmare.

Clearly, the credit rating companies, who so blithely gave away their credibility (yet again), are partly to blame. But, I would ascribe greater blame to regulators who permitted""indeed, encouraged""credit rating companies to continue to run with a fundamentally conflict-ridden business model.

Let's look at it from first principles. If a company gets an AAA rating, it gets to borrow cheaply. The investor who lends to the company depends on the rating to make his investment decision. So, who should pay for the rating? It would seem obvious that it should be the entity that depends on the rating""the investor. Instead, the business model is structured so that the company being rated pays for the rating. This by definition is a conflict of interest, which, sooner or later, is bound to lead to a crisis. Incidentally, the same flaw continues to pervade the audit cost business model, even though the last time it blew up""remember Enron""it took one of the then most respected auditing companies with it.

This structural flaw needs to be addressed immediately. For listed bonds, a simple expedient would be to have the exchange pay for the rating, the cost of which it could recover from trading fees (ultimately from investors). For illiquid bonds, it's much trickier""one idea could be to get investors (in primary issues) to allocate the rating component of their fees to the agency it felt provided the most reliable valuations. It isn't easy, but that's no reason not to do it.

Failing which, of course, the next time the conflict of interest triggers an explosion, we may all come down with much more than a cold. AAAchhoooo!