Understanding the Mark to Model Concept

The current credit market situation would be laughable if it wasn’t so serious. It is playing out like a bad daytime soap opera. The latest buzzword that is entering the everyday lexicon is mark-to-model.

All bond traders to some degree price securities using models, even one that trades a benchmark security like Treasuries. The simplest model would be a straight yield spread to Treasuries. As time marched on and increasing amounts of data could easily be processed, models became ever more complex. Synthetic benchmarks were used with increasing frequency, adding to the mix. It seemed that no amount of variables could stop some mathematical genius from developing a model to calculate the exact value of X. However, there are three variables that could never accurately be taken into account by the models. Those three variables are fear, greed, and liquidity.

Fear and greed are the two that create outsized moves in the market, either ends of the pendulum swing, as it were. Greed usually takes effect more slowly, building up to level of overvaluation. That is the point that the credit market reached in June over a buildup of several years. Then, fear took over, as investors sold and exited markets in an attempt to lock in what they had gained. Fear takes hold very quickly, like being hit over the head with a sledgehammer. It was at this point when traders were reporting “10,000 standard deviation” market moves. Finally, a second kind of fear kicked in, observed in the market as illiquidity. This is when the market players said “No mas” and brought the CP market, especially the asset backed CP market, to a standstill. With models used to calculate values starved of accurate input data, marking or pricing to model became a best guess scenario.

The faith that Wall Street and the investment community has placed in these models has been shaken. Almost daily now, new “best guesses” as to the carrying value (as it doesn’t look like many of them have been sold or unwound) of these assets are released. It is one thing to interpolate the spread of a GE 6yr maturity bond, when the value of the GE 5yr and 10yr is known. However, in the current environment, it is near impossible to determine what value an asset has when the underlying securities have been repackaged four times and have uncertain cash flows going forward. To state it bluntly, Wall Street needs to mark-to-market (or mark-to-worst case scenario) already.