How much for toxic assets?

ABX indices (based on credit derivatives written on MBS backed by subprime mortgage loans) track the price of credit default insurance on a basket of such deals. Since the start of the CC liquidity squeeze (summer '07), the ABX indices are the barometer of the collapsing price/ valuations in the US subprime mortgage market, at the centre of the credit market turnmoil. ABX data is also used by banks and other
investors for hedging and trading and risk pricing of mortgage-backed portfolios generally.
TARP and similar schemes (e.g. swaps of RMBS securities at central bank windows when sub-triple-A tranches or pools are accepted - as BoE are now starting to do) are based on the belief that observed ABX mark-to-market prices are significantly less that their 'true fair value' where the value is based on model projections of what % of defaults in the underlying pool of subprime MBS will over the cycle turn into a net economic loss, after recoveries etc. Understanding the specific factors driving the variation of ABX prices is important for policymakers as changes in the weight of credit- and non-credit elements have different implications. For instance, indications of changes in risk appetite and the liquidity squeeze context may help explain the discrepancies between ABX market prices and estimated 'hold-to-maturity' values. What a BIS published study by Ingo Fender and Martin Scheicher show is that confidence factors and the effect of a high LIBOR over time explain about 30% of the fall in MBS prices. Therefore, central banks, US Treasury, and others could in all conscience offer at least 30% above current mark to market prices, even before the more complex modeling of hold-to-maturity valuations?
Empirical work on these important matters has rarely been made public (so far as I the BIS study and its references may be the only ones, though the US Fed or Treasury must publish something similar very soon). I know banks and hedge funds and others have such models (or are trying to build them) but may see them as a valuable competitive tool to be kept confidential whereby they might confidently buy valuable assets at fire-sale prices.
For those who wish to learn more I recommend pages 71-81 of the BIS Quarterly Stability Review, Sept. '08 - or contact me directly. The models employed in the BIS study are not the last word on the matter as a more complete economic model (rather than economic driver indicators treated individually) might improve the rubustness and extend the scope to all RMBS and CBMS.