Comparing the present counter-party risk induced lending squeeze to the classical "lemon problem" in the used car market, Daniel Gros analyzes the "bad bank" model of cleaning up the balance sheets of banks and thereby restoring normalcy in the credit markets. The market today views the sellers of securities with these assets as the under-lying, or securitized debt, as being risky and therefore boycots them.
A "bad bank" would take over all those assets which either do not have a market or whose market is very thin. The purpose, as he writes is that, "Those banks left with insufficient capital once their 'troubled' assets have been transferred to the bad bank would of course have to be closed or recapitalized, but the healthy ones could start a new life with a clean balance sheet and could thus resume normal lending."
However, for this "bad bank" model to succeed and to get normalcy restored in the credit markets, he claims that "the bad bank should be big, and banks should be forced to transfer their entire portfolio of toxic assets". Since it is difficult to distinguish between toxic and clean assets, he suggests that "all financial institutions should be forced to surrender all assets in certain categories to the 'bad bank'". Further, such assets would have to include credit card, commercial, student and auto loans and leveraged loans, besides the residential mortgages. He values all these troubled securitized assets in the US at an $2-3 trillion, in addition to whatever has already been purchased by the US Treasury. Incidentally, Goldman Sachs estimates the total value of troubled US bank assets to be $5.7 trillion.
The problem with the "bad bank" model lies in identifying the distressed assets and then valuing them. He claims that the prices of toxic assets could be determined by the government on the basis of "a mixture of market prices, models and other valuation approaches". He also makes the interesting point that the anomalies in any price paid by the government would be largely nullified by the fact that if the government pays a lower price (and thereby stand a greater chance of making profits when exiting), it would need to inject a larger amount to recapitalize the bank, and vice-versa. But it is also possible that the government may end up paying a higher price, without even spending anything on recapitalization of the "toxic asset free" bank, and then having to exit at a much lower price or be left with worthless assets.
If a universally valid valuation model which weights market prices and specific valuation models, is applied one can ensure that the prices paid by the government are consistent and avoid problems of selective pricing. But the problem with this approach is the extreme uncertainty surrounding the quality of many securitized debt assets, which makes application of any valuation model meaningless.
In any case, there is a clear case in the argument that a selective "bad bank", which buys only certain categories of bad assets and that too from certain banks, will in most likelihood fail since the "lemons" continue to exist in the bank balance sheets. Therefore if governments decide to go with the "bad bank" model, it should go the full hog, covering all baks and all bad assets. There cannot be any half measures. Or else the alternative is to nationalize them!
Paul Romer inverses the "bad bank" model and pumps for the "good bank" model, where the banks will transfer all their assets and liabilities "that are easy to value using market prices" into their respective newly created good banks, which will then leverage the government bailout money, to restore some form of normalcy in the credit markets by removing counter-party uncertainties. But it still leaves the distressed assets to be lodged somewhere.