In 1990, when high-yield bonds had the brush with difficulty… (meaning spreads widened to 1,100 basis points, and a law was passed that required S&Ls to reflect price declines on their balance sheets), I was asked to brief the board of TCW on the risks. I presented a parable about a regulated financial institution that went bankrupt under the weight of mark-to-market accounting. I joked with Bill Spencer, who was president of Citibank when I worked there, that in the 1980s, that could have been Citibank if it was required to recognize mark-to-market losses on real estate loans.
Guess what: today that's the rule.
This raises one of my favorite questions: what's an asset's price?
I don't have the answer. Mainly because there is no answer. In short, an asset doesn't have ?a price.? It has many possible prices, and no one can say which is the right one. The ads for a jeweler here in Los Angeles lead with a great headline: ?guaranteed to appraise for more.? In other words, either (a) he sells jewelry for less than it's worth (and, if so, why?), or (b) he sells things for what they're worth but guarantees they'll appraise for more, which makes you wonder about the appraisals. The way I see it, the appraisals he touts are just as meaningless as many of the ?market prices? being used today to price assets at banks, hedge funds, CDOs and CLOs.
A view has begun to be expressed that mark-to-market accounting ? in conjunction with the vicious circle that prevails today ? is causing asset values to be understated, writeoffs to be overstated, and the credit crisis to be exaggerated. Certainly there's every reason to believe that:
It's hard to believe these are really the bases on which financial institutions should value their trillion-dollar balance sheets. But we're stuck for now with mark-to-market accounting. At minimum, you should expect it to contribute extensively to continued volatility. Believe me, it already has.