On Valuing CDOs

A solid piece, “Collateral value thrust to the fore by woes at Bear Stearns,” by capital markets editor, Gillian Tett of the Financial Times.

It’s inherently difficult to value assets that don’t trade often (think of art), yet dodgy CDO paper has already been valued as colleteral so loans could be made against it. Some of these lenders have discovered how out of date these appraisals are (you wouldn’t see big Wall Street firms wrangling with one of their own for a take out if they could liquidate their collateral at a reasonable price). It’s remarkable that Wall Street’s prime brokerage and other credit functions have allowed themselves to get in this situation. These companies take no prisoners as far as liquid securities are concerned. So why have they chosen to turn a blind eye when they knew values in the subprime sector had to be falling?

Tett doesn’t answer the big question, but she does provide useful detail that may help unravel the larger issues. She highlights the fact that the models used for these instruments function badly in markets under stress.

From Tett:

This week’s news that two hedge funds linked to Bear Stearns have imploded amid hefty losses on subprime mortgage securities has — unsurprisingly – left many credit investors feeling nervous.

But some regulators might also be tempted to utter a small – albeit private – cheer. Never mind the fact that these high-profile woes offer a useful reminder of the dangers posed by excessive leverage and financial innovation.

What is perhaps most important is that they are also concentrating the market’s mind on an important issue that is often ignored – namely the way that banks value the collateral that hedge funds, or other counterparties, typically post when they borrow money or trade.

Take the case of Merrill Lynch, which is one of several Wall Street creditors to these funds. In recent months, it has held a pile of collateral against the loans it has extended to these funds, which have apparently been marked at about $800m in Merrill’s books.

However, these assets include tranches of collateralised debt obligations (or pools of debt assets) which have rarely traded. Thus the real value of the instruments will only emerge after the US bank actually concludes any firesale. While the bank is now trying to conclude this, it is far from clear whether such sales will produce a price close to that on its book. This is important because this sense of uncertainty is echoed across the industry.

Until fairly recently, the value of the collateral that hedge funds (or anyone else) posted was often easy to assess, since it typically related to liquid instruments such as Treasuries. Even today, some collateral remains easily disposable: when the Amaranth hedge fund collapsed last year with $6bn of losses, for example, it sold its $1bn-$2bn portfolio of leveraged loans swiftly, at prices close to book.

But in the last couple of years, funds and banks have increasingly started chucking illiquid instruments into the collateral pot too. And since these cannot be easily priced off the market, banks tend to use financial models to value them instead.

However, these models used by bankers (or almost anyone else) rarely capture properly how these instruments will behave in times of stress. After all, as Ed Shea, a risk management expert, points out, when shocks occur, banks often face nasty operational challenges in unwinding large pools of assets in illiquid markets in a hurry.

Moreover, even without a crunch, one dirty secret of the modern prime brokerage world is that valuation models can vary wildly between banks. One consultant who advises hedge funds on how to select a prime broker, for example, tells me funds these days are increasingly “window-shopping”, to determine which banks will lend them the most, on the fewest assets. “If you don’t like the collateral valuation at one bank, you just go somewhere else,” he says.

Thus far, this does not seem to have created any real headaches for the banking industry: after all no prime broker has admitted to suffering significant losses from misvaluing collateral. But these have been relatively benign times.

Thus let us all hope that the CDO auctions at Merrill Lynch and elsewhere conclude smoothly this week. After all, this might finally shed some light on what these instruments are actually worth – and thus how accurately (or not) they have been valued in the portfolios of the banking and hedge fund world in recent years.*Meanwhile, as Wall Street scours its ranks to uncover the victims of the subprime woes, some European regulators are taking a hard look at some of their local pension funds and insurance companies too.

An analysis by Fitch Ratings earlier this year suggests that there are 129 European CDO tranches exposed to US subprime. That is striking, given that just 66 tranches of US CDOs were similarly exposed.

This may not equate precisely to European investor losses (after all, it just covers Fitch-rated products and the ultimate investors may not be European). But there are rumours that some Europeans institutions are holding particularly toxic tranches. One more reminder – if any was needed – that the subprime saga is not just a Wall Street tale.