Are investment banks really worth anything at all?

Bear Stearns's share price went from $145 a year ago to $57 last week to $30 Friday to $3.82 this afternoon. That's still $1.82 above the $2 a share JP Morgan Chase has offered to buy Bear (hope springs eternal, I guess), but it ain't much. Now the FT's Alphaville blog reports that "spurious" rumors are devastating the share price of brokerage MF Global (spun off last year from hedge fund giant Man Group):

A firm that was worth $3.6bn at the end of February, was worth $2bn on Friday and then $600m on Monday. Without any news…

Investment banks/brokerages tend to be big agglomerations of debt piled upon little bitty capital bases (the same goes for many hedge funds). They can report huge profits when markets are with them, but it's always possible that those profits are a mirage--the product of taking risks that will eventually bankrupt the firm. The classic example of this, which has happily been getting some blog ink lately, is MIT finance professor Andy Lo's hypothetical Capital Decimation Partners, a hedge fund that does nothing but sell out-of-the-money put options on the S&P 500 index. That is, it sells people the option to sell it the S&P at a price below the current price.

As long as the S&P rises, or drops only slightly, this strategy is a steady money maker. CDP pockets the money from selling the puts, and never actually has to honor any of them. Between 1992 and 1999, Lo reports, CDP would have been up 2721% to the S&P's 367%, with only 6 down months over that stretch compared to the S&P's 36. It's obvious if you think about it that in a sustained bear market all those put owners would come to Capital Decimation Partners and demand that it buy the S&P from them at a price above its current price, the fund would quickly go belly up. But any statistical examination of its record from 1992 to 1999 would rank it is as a low-risk, high-return endeavor.

Writes Lo:

[G]iven the secrecy surrounding most hedge-fund strategies, and the broad discretion that managers are given by the typical hedge-fund offering memorandum, it is difficult for investors to detect this type of behavior without resorting to more sophisticated risk analytics, analytics that can capture dynamic risk exposures.

I think the same argument applies to investment banks like Bear Stearns, Lehman and Goldman Sachs. It's not that I really think they're following strategies as simplemindedly idiotic as that of Capital Decimation Partners. It's just that it's awfully hard for outsiders to tell for sure whether they are or they aren't. It may even be hard for insiders. Bear Stearns' record of turning a profit every year since 1923 would seem to be an indication that, for most of its existence, it was balancing risk and reward pretty well. But then, poof!

Update: My Fortune buddy Shawn Tully makes more or less the same point:

Wall Street firms can't resist taking big positions in currencies, mortgage backed securities, oil futures, junk bonds or other speculative vehicles, especially when they're hot. It happened in the late 1990s with the Asian Contagion, the LTCM meltdown, and the Russian debt crisis, and again with the tech bubble. When risk premiums on those assets keep falling, as they did consistently from 2002 until the middle of last year, the Wall Street mavens look like geniuses.

But when the risk spreads expand suddenly and dramatically, the story of the last 9 months, the prices of those assets drop sharply. Wall Street leaders from Stan O'Neal, former Merrill Lynch CEO, to John Mack, chief of Morgan Stanley, boasted that their firms' skills in risk management would prevent massive losses from proprietary trading, part B, as we call it.

But the enormous profits on the way up was a danger signal that Wall Street was playing on the edge. The losses on the way down are proof that either Wall Street isn't as good at hedging as it claims, or more likely, that too much hedging would spoil the huge profits it craves when markets are roaring.