In Which I Soften on Mark to Market

Here's a quick summary of all debates in accounting: how much truth can you stand? Translated: it's had to remember any major controversy over accounting rules that didn't involve (at least) one side trying to keep something out of the financials, because investors or creditors might not like it. Contingent pension liabilities? Options? Oh, dear god, don't book 'em; end of Western Civilization as we know it. None of this has ever made much sense to me: if the truth is too awful to bear, than that is just the sort of truth the users should have.

My attitude to mark-to-market has been pretty much in character. Of course you should mark to market. Not marking to market is like (I read this somewhere) the doctor not telling the patient he is sick.

I'm beginning to soften: wiser heads than my own are persuading me that maark-to-market may not work when there isn't any market--as pretty clearly seems to have been the case lately in, oh I dunno, short-term money, real estate, whatever. Marking to market assumes a market; if there is an exogenous collapse in the market structure, then mark to market doesn't make a lot of sense.

And so? Well, it does not follow that we just default to book value. We might be able to cook up some sort of red-flag system that will warn the reader that the usual bets are off. Or maybe we could do some kind of "price smoothing," where we assimilate the market meltdown over time--providing an opportunity for the crisis to pass.

By corollary, I don't see any need for regulators to hew blindly to mark-to-market decisions in situations where the market simply isn't working.

But the core point remains: accounting is about information, and information includes bad news as well as good. Every debtor always thinks his assets are undervalued and if only he gets a little time blah blah something will turn up. God love 'em, some unknowabale number of these straitened debtors may be right. But there is really no way of telling for sure, and in default of a confident judgment, the market's judgment is likely to be the best judgment we have.

A couple of useful links: here and here.