Numerous commentators, including Willem Buiter, have emphasized the potential for adverse selection in the Geithner plan auctions of troubled assets. Join the party, dudes. This was the graveman of my skepticism of the original TARP auction scheme.
But, I think that moral hazard is the key problem with the Treasury plan. That is, the plan defers Judgment Day for the banks. We need to separate the Saved from the Damned, but instead Geithner lets dodgy banks continue to operate. They have the option as to whether to participate in the sale of troubled assets, and will exercise that option to maximize their value. Unfortunately, maximizing this option value means that the banks will do what it takes to defer Judgment Day, and in so doing, they have very perverse incentives. Moreover, inasmuch as there is arguably an externality from the continued operation of insolvent banks (as it calls into question the solvency of the entire banking system, and poses the ongoing risk of another systemic event), deferring Judgment could be very costly indeed.
This is not to say that adverse selection is important. It is, but in a way different from what Buiter and others have emphasized. In particular, adverse selection arising from the fact that sellers of the toxic assets are likely to have better information about their value than buyers means that, to protect against the “winner’s curse”, potential buyers will low-ball their bids. That is, their bids will be at a discount (on average) from the true value of the assets.
This represents a further deterrent to banks to participate. Indeed, if the information asymmetry is sufficiently severe, trading activity in this market may be de minimus. Thus, low ball bids will represent an even further disincentive for truly bad banks to participate in the sale of toxic assets. And they will be able to make a credible case that these prices do not reflect true values.
This problem is a reason why, in the Fall, I proposed Humpty Dumpty. By aggregating bad assets, and selling equity claims on the pool of these assets, asymmetric information/adverse selection problems are mitigated. Substantially. This selling stuff a piece at a time is just nuts, and maximizes the potential for adverse selection.
Like Casey Stengel once said: “Doesn’t anybody know how to play this game?”
One interesting thing in the news today. New media darling Sheila Bair supposedly recognizes that bad banks have little incentive to participate in the sales of toxic assets, and is reported to be planning to exert pressure on the banks to sell. Moreover, the government will supposedly use the results of stress tests (remember them?–funny how Geithner didn’t mention them in his most recent announcement) push the banks to sell.
I guess that’s an improvement, inasmuch as it at least indicates that there is a glimmer of understanding of the problem at the root of the Treasury plan. But why do so in such an opaque, indirect, and potentially ad hoc and arbitrary way that could be undermined by influence activities? (Who decides who gets pressured, and how much? How is compliance going to be measured and monitored? How will political pressure and lobbying affect the process?) Why not do it in a straightforward, public, mandatory, non-discretionary, non-arbitrary way? This is, after all, allegedly the era of transparency in finance. Why is the government being so un-transparent in its handling of this? Could it be that they don’t have the political will to do things openly and honestly, or that they are skeptical that such an approach will work? Could it be that they prefer such ill defined and imprecise approaches because it maximizes their residual control rights?
So, Sheila, I say: If you think that the moral hazard involved in allowing banks to choose their degree of participation in the sale of toxic assets is a bad thing, come up with a credible, public statement of a way to address that problem. Delphic utterances and leaks that create uncertainty are not what we need right now. We need to resolve uncertainty, not layer it on.