Merrill should take its own advice

In July, Merrill Lynch (MER:NYSE) advised a nervous market that its exposure to the frothy subprime mortgage business was “limited” and “contained”. According to the WSJ, Merrill’s exposure was probably US$25 billion at the peak. If that’s someone’s idea of “limited exposure”, they need a remedial briefing from the SEC’s Fair Disclosure SWAT team.

Yesterday, the Thundering Herd announced a US$5.5 billion writedown, US$4.5 billion of which is the result of depressed values in the Collaterialized Debt Obligations and subprime mortgage portfolio. For the third quarter, what analysts thought would be a US$1.24/share profit became a US$0.50 loss.

For all the handwringing about bulge bracket firms sitting of a pile of private equity bridge loans, Merrill only wound up eating $967 million of losses there, which was even easier to swallow once the fees earned on those deals were used as an offset: voila, a US$967 writedown turns into a US$463 million kiss.

Subprime and CDOs generated 10x the pain for Merrill shareholders as compared to the constipated corporate loan book.

For months, the point has been made here that fair-market value accounting was fraught with subjectivity (see post “Accountants are failing investors with “fair value” accounting“, August 6-07). Particularly in the world of illiquid — or non-trading — assets. This from Merrill:

These valuation adjustments reflect in part significant dislocations in the highest-rated tranches of these securities which were affected by an unprecedented move in credit spreads and a lack of market liquidity in these securities, which intensified during the third quarter. During the quarter, the company significantly reduced its overall exposure to these asset classes.

That Merrill Lynch has discovered their subprime book isn’t trading at par isn’t a shock. Except perhaps to those equity research analysts who took management’s word for it and believed the exposure was “limited”, and projected MER to throw off a US$1 billion profit for the quarter vs. the US$450 million loss announced yesterday.

From the press release, you get the idea that at least some of the losses came when Merrill actually exited positions, crystalizing the losses. After all, the only way to “significantly reduce” exposure is to ditch that exposure on to someone else’s books. Makes one wonder how much of the US$4.5 billion represent actual cash losses versus the mark-to-market kind.

Senior Fixed Income managers are dismissed, the balance sheet is cleaned us, the CEO tugs on his forelock, and the US$65B stock rises 2.5% on the relief that the announcement is now out of the way.

Not so fast.

Bulge bracket firms value their assets each nano-second. Each evening, vast server farms collect and process the input from Merrill’s various operations around the world. Risk managers review trading and valuation data. The head of Risk Management would know exactly where the firm is at all times, just in case the CEO stuck his head in the door after a call from one the the firm’s directors, or the Federal Reserve.

Dozens of people would have access to the values of Merrill’s CDO and subprime book, yet the timing of yesterday’s release would sugget that US$5.5 billion in lost value was only discovered during the internal process leading up to the quarterly results. Unlikely. Given that a US$1 billion profit was expected by the analysts, Merrill would know that a US$450 million loss is a material EPS miss. By a Country(wide) mile.

Where was the profit warning at the time? To pretend that no one in authority knew until this week that the firm’s CDO assets were wildly over-valued isn’t credible.

Perhaps one of our readers can cite the specific SEC rule for U.S.-listed firms, but in our neck of the woods, if you are going to miss profit forecasts by more than 10%, you put out a release. Merrill’s own research analysts will be quite familiar with this tradition, as many firms in their coverage universe will occasionally suffer the ignominity of “stubbing their toe”. But, unlike a software firm that doesn’t know what its revenue is until the end of the quarter, Merrill must have known weeks ago that their subprime and CDO book was badly underwater, making the Q3 earnings forecasts anything but achievable.

Back in August, BNP Paribas concluded that the tumult in the subprime market made it impossible to price ~US$2 billion of subprime assets in their fund management arm (see post “BNP Paribas plays the canary role“, August 9-07). Their stock got hammered by nervous investors.

Two months passes, and Merrill concludes that BNP was right, inasmuch as subprime assets were no longer worth anything close to face value; unlike BNP, MER delayed the news until quarter end and trades higher as a result. Merrill’s honesty came on the heels of similar ephinanies from Citibank and Deutsche Bank. A CNBC commentator said this week that being in the woodshed isn’t so bad when you have lots of company.

If only BNP had followed that PR advice.

Bulge bracket firms aren’t software companies, where much of the revenue comes in during the last 3 days of the quarter. In the modern day world of portfolio risk management tools, senior Merrill exectuives must have known weeks ago that their CDO portfolio was down a material amount. If fair market value accounting is to serve investors well, bank executives can’t put their heads in the sand when bad news bubbles up to the executive floor. Come clean when your portfolios are deeply underwater, and the quarter is going to be a miss.

Your own research analysts demand no less from the rest of The Street.

MRM