There has been a great deal of critism of the Mark to Market principle.
My observation is that Mark to Market is a good idea. The problem is that, to make mark to market work, one would have to redesign accounting and regulatory principles from the ground up to make Mark to Market work.
The problem we have is that currently regulatory regime piles rigid liabilities on firms. Having a system where a company must mark its assets to market creates a situation where a company must report a brittle set of liabilities and a fluid set of assets. This system where liabilities are fixed and assets liquid create a economic where businesses are not able to sway within the turmoil of the market.
Mark to Market in the present system simple creates a system where one side of the books is brittle and the other side fluid. Market turmoils causes businesses in such a climate to snap.
Mark to Market is a good idea. For the system to work, both the liabilities and assets must be designed to fluctuate with economic winds. For example, in a fluid system, the payroll might automatically contract in a down cycle. There also has to be a realization that the contraction of liabilities often trail the contraction of assets.
For example, sales commissions contract in a down cycle. However, since the payment of sales commissions tend to follow sales by a month or so, the actual contracting of commissions might show up on the next report.
Speaking of tools. I just added a page on redesigning financial tools. The article claims that much of the current instability is the direct result of the tools on the market.