Sunday, July 29, 2007

The Pricing Fog in the CDO Market

Once upon a time, there were two hedge funds sponsored by Bear Stearns that invested in the mortgage markets. In March 2007, they reportedly had a combined value to investors of around $1.6 billion. In June 2007, delinquencies among mortgage borrowers had impaired the investments of these two hedge funds. The deterioration got to the point where, Merrill Lynch, which had loaned money to one of the funds to finance investments in mortgage-backed securities, reportedly seized about $850 million worth of these investments and sold them off to secure repayment of its loan. Then, with other lenders to the funds hovering closely, Bear Stearns stepped in with a $3.2 billion loan to one fund (apparently, only $1.6 billion was actually loaned). The loan was made to stabilize the situation so that the funds could pay off other lenders in a more orderly manner.

Fastforward to late July 2007. The two hedge funds are now reported to have lost all (in the case of one fund) and over 90% (in the case of another fund) of their investor value. In other words, the investors were left with zero cents on the dollar for the one fund, and less than a dime on the dollar for the other fund. An equivalent drop in the Dow Jones Industrial Average would have been from its approximate average in March 2007 of 12,300 to somewhere around 1,200 or less within four months. If that had happened, sales of the Communist Manifesto and pitchforks for the proletariat would have skyrocketed.

How could the values of these funds go into a tailspin like this? The fact that the funds were heavily leveraged was one important reason. When the value of the funds' investments began dropping, the creditors of the funds started to demand more collateral. If the funds couldn't provide the additional collateral, the lenders wanted their loans to be repaid. If they weren't repaid, the lenders could seize the collateral and sell it to pay down the debt. The speed at which things deteriorated indicates that the funds and the lenders may have overestimated the value of the collateral, particularly the value it could command when put on sale for immediate cash payment.

CDOs don't trade on any active market. They are bought and sold on an ad hoc basis, not in a public market where price quotations are displayed to the world. Instead of using market transactions as a guide to establishing values of CDOs, hedge funds and other market participants use mathematical models. The models tend to be based on expected future performance, and not the cash price the CDO might receive if it were put up for immediate auction.

The use of mathematical models to price derivatives isn't new. The dealers and specialists in markets for standardized stock options have, for some 30 years, used mathematical models to establish their price quotations. One important difference between the stock options markets and the over-the-counter derivatives markets, though, is that the options markets are public. They provide continuously displayed bid and ask quotations to the world, and all trades are reported publicly on the consolidated tape. The open market imposes discipline. If one dealer's mathematical model produces incorrect prices, he gets clobbered early and often by other market participants who trade with him to his disadvantage. He either fixes his model or finds a new line of work.

In the unregulated world of CDOs, there is no continuously trading open market to impose discipline. Mathematical pricing models are based on assumptions, and the numbers they generate will vary depending on the assumptions made. We discussed in our July 27, 2007 blog ( how the CDO market created incentives to write vast quantities of low quality mortgage loans--loans that are now defaulting in many cases and pushing the market down. Was this proclivity toward outsized risk incorporated into pricing models?

Hedge funds and other players probably can exercise some discretion over their valuations. Would they have moved promptly to reflect downturns in values? Well, how are they compensated? 2% of what and 20% of what?

Were there any controls over the valuation issue? Many CDOs were rated by the rating agencies--outfits like Standard & Poors, Moody's and Fitch. Since the Bear Stearns hedge fund debacle, the rating agencies have been downgrading billions of dollars of bonds. Why did they wait until the yogurt hit the fan? The full story may not be out yet, but it seems that the rating agencies, too, were using mathematical models.

The critical problem with mathematical modeling is that, at some point, cash has to enter the picture. When a hedge fund uses leverage, the lender can't rely on a mathematical model to ensure that its loan is repaid. It needs cash. For a collateralized loan, if worse comes to worst, the lender has to be able to convert the collateral into cash. When values deteriorated this past spring from increasing mortgage defaults, the hedge funds' lenders protected themselves by demanding more collateral or selling off the collateral they had. Either option would have involved getting cash (or hard assets readily convertible to cash). This is the point where the rubber hits the road, and mathematical modeling and $5 buys you a cup of coffee.

The CDO market was mispriced. That much is clear. The world of hedge funds and derivatives has been left essentially unregulated--by both Republican and Democratic administrations and Congresses. Absence of regulation facilitates innovation and growth. The plethora of derivatives contracts now available is a bewildering alphabet soup array of financial instruments. The dollar value of transactions in the derivatives markets today probably exceeds the dollar value of stock market transactions. Derivatives are said to diffuse risk and allocate it to market participants that are willing to take it on. All this sounds good.

But the world of CDOs has fostered the creation of risk--a lot of it--as we discussed in our July 27,2007 blog ( And a lot of risk diffused and allocated nevertheless remains a lot of risk. When it materializes, a lot of people will say "ouch." Did the investors in the two Bear Stearns hedge funds that collapsed think there was a reasonable chance they'd lose everything or almost everything?

This isn't just a problem for Wall Street firms, hedge fund operators and the wealthy people and large institutions that invest in hedge funds. The stock and bond market losses of recent weeks affect everyone that has a stake in the financial markets. Every IRA and 401(k) account invested in the stock markets has probably taken a loss due to the mortgage market mess. Every pension fund and university endowment has probably suffered losses. Every small business with a SIMPLE IRA or SEP-IRA plan invested in the stock market has been hit, meaning that the employees of the small business have suffered. A lot of people who have no idea what a CDO is, or if they do, have no interest in them, now have been affected--negatively.

No one knows how this situation will turn out. Prognostications have been made that more mortgage market losses are in the offing. One thing that's clear is that the pricing mechanism in the CDO market needs improvement. We now have a situation where hedge fund managers may be taking compensation based on the valuations of their mathematical models, but investors are taking losses based on cash prices. It's one thing to play "heads I win, tails you lose" with a quarter. But the amounts involved in the CDO mess are much larger than $0.25. These pricing disparities should be reconciled. Since you can't persuade lenders to take repayment in the form of a mathematical model, hedge funds that use leverage (which would be about 100% or so of them) should move toward a cash pricing model. Cash prices may be difficult to establish on a continuing basis, unless CDO and similar products are standardized and quoted on electronic markets. But perhaps these would be good ideas, now that the instruments are proving to be so risky.

Many players in the derivatives markets will recoil at these thoughts, since they would viscerally sense a reduction of profit margins. Nonstandard products traded in an opaque market are likely to be more profitable for dealers than standardized products traded on a transparent market. But the widespread damage that the mortgage market mess is causing will leave the industry with fewer and fewer choices. The perspicacious on Wall Street will get ahead of the curve.

Strange News: a haunted penthouse:

No comments: