T Much of the current mess has come from the "mark to model" method of valuing "assets".
i agree with this for the most part - trouble is: how do value something that is not very liquid? such as many of the instruments that are heading way south these days. you have no option other than mark to model - they are illiquid instruments (i.e. their is no readily accessible market to sell these assets)
presumably these instruments and their valuation models are based on some expected future cash flows versus the financial instrument's unsystematic risk - so, in theory that is pretty simply to value.
so now lets assume that delinquency rates on subprime mortgages suddenly go up & you as a CDO investor are highly leveraged & your business model is essentially buying CDOs with s-t borrowing and making money from the yield spread. now all of sudden since delinquencies have risen your lender is not so happy with your CDO as collateral for the s-t borrowing your accustomed to getting in the commercial paper market. & you were accustomed to using these s-t borrowings to not only fund existing debt you own but to buy more CDOs (& gain more revenues via the yield spread). so now your cash flow stream has diminished
& your banker wants some of his money back - so where do you come up with cash? you are forced to liquidate assets to repay your indebtedness. concurrently your buddies you play golf with are doing the same thing & consequently their is a heck of a lot of CDOs available for purchase & since all pricing decisions are essentially based on supply & demand a dramatic increase in supply suggests price must go down. the increase in supply & the perceived increase in risk have in this case caused the asset prices to go way down.
consequently the valuation models have failed - they become irrelevant in the short-term b/c after all something is only worth what another individual will pay.
however, i would presume that everyone is pretty confident in the U.S. economy over the mid to long-term & in the longer-term (say 5 years) these models will again be closer to an accurate valuation.
so, i give this schpill only to say that these models cannot deal with s-t market disruptions like we have today (they don't deal well with uncertainty) - when mortgage delinquencies get back to a more reasonable level then the models will again prevail.
truth be told...if you can build a better mousetrap (in this case the model) you will be poised to make a
ton of money in the near future...this whole mess has fasb, the fed, aicpa, treasury, etc. looking at how to value assets - will be a very interesting future!!