How to buy toxic waste

Lawrence Ausubel and Peter Cramton, two well-known auction theorists at the University of Maryland, are trying to talk Geithner into using an auction scheme to buy some of the so-called "troubled assets" from banks.

In this scheme, assets would be purchased using a descending clock auction. The idea is that the government decides to buy some portion of the outstanding troubled securities, say 50%. They start with a high price, say the face value of the bonds. All potential sellers announce how much they are willing to sell at that price. Presumably, at the initial price, all banks would want to sell all their bad assets, so this would facilitate transparency about what the banks actually own. Since supply would surely exceed demand, the price would be slowly lowered. As the price descends, sellers will begin to pull their assets. The price is lowered until supply equals the fixed quantity demanded.

Ausubel and Crampton have thought through many of the details and are rumored to be in discussions with the Treasury (see link above). They say there may be ways to invite private investors to the table--this part is a little vague.

I would suggest that the government act as the sole purchaser in this scheme and then immediately attempt to sell these assets using standard auctions the Treasury knows well, but with a reservation price equal to their purchase price. There may be no buyers. But I wouldn't be surprised if there were. This way the government might be able to immediately make a little cash for taxpayers.

Odds are this will not recapitalize the banks. But I think it would be a cost-effective and necessary first step.

Update: Yes, I realize this was kicked around awhile back under the first round of TARP spending. I just had reason to believe it might be coming back into play, and so that's why I mention it. Also, some folks around here may not know about it.


  1. Due to mark-to-market, banks will have an incentive to collude (or even without collusion, simply by selling only part of what they would like to sell). They would rationally hold assets back from sale to a) get a higher price on what they do sell, and b) allow them to mark-to-market and hide their losses until later.

    In other words, it's hard to see how this will result in "rational" pricing or government paying the market-clearing price.

    At very high prices, banks will sell - but as they approach "real" price, they will start having incentives to hold back, particularly if they can collude.

    Also, think of effect if everyone knows government will only buy e.g. 20% of an issue; we can't all get out, and we don't want to have to show losses; so we "agree" to hold back to get higher price, delaying the pain until later.

    This still doesn't resolve the fact that without regulators pushing them hard, they have incentives to hide losses as long as they can.

  2. Thanks for your comment.

    Collusion is a common concern. This is what auction theorists spend a lot of time worrying about. Ausubel and Cramton have consulted for many large scale auctions with few players. If you read their short article you would see some of the ideas they have for discouraging collusion. They can (and have) tested this and other auction schemes in a laboratory. I've done a little bit of this myself on a much more limited scale, which draws my interest to the subject. Anyhow, I think significant collusion is unlikely.

    And yes, it could be that there will be lots more to learn about the banks' finances. I do think this would be a good start and not much of a net cost to taxpayers since they should get a decent price for their toxic waste.

  3. The discussion of auctions is useful, but still misconceives the incentives operating in large financial firms, specifically the gap between the executive (agent) and the shareholder (the principal).

    If we realize that the CEO has no interest in advancing the wealth of the shareholder, then the solution becomes clear. And the expected cost of the buyout to the taxpayer becomes insignificant.

    Rather that mandating pay caps in the range of $500,000 the Treasury plan should align the interests of the CEO with those of the taxpayer. Specifically, CEO's at the nations top 15 banks -- those now considered too big to fail -- should be given incentive bonus contracts under TARP. The banks will sell the Treasury their really toxic assets (anything worth less than 40% of book value). The incentive bonuses will be based on the appreciation of those assets over three years versus the price paid by the Treasury.

    With payout bonus potential in the range of, say, $50 - 100 million depending on the size of the bank and 15 banks in the game, the total initial cost to the Treasury should be less than $1 billion dollars. With CEO payouts tied to the eventual value of the assets, we can be certain that all packages acquired in this auction will be worth much more than the prices assigned by the CEO's.

    Furthermore, the CEO's will no longer have an incentive to fear the "mark-to-market" accounting effects since they will have their money in cash, effectively as a golden parachute. In keeping with contemporary executive compensation practice, the Treasury would "gross up" the bonuses, i.e., offer to pay all income taxes due on these packages.

    A further advantage of this scheme is the continuity of bank operations during the coming depression. Other schemes have proposed using TARP funds for the creation of "good banks" -- leaving the toxic assets in the existing banks and letting them fail. This laudable mechanism of retributive justice fails the test of feasibility as the new banks would have to hire staff, acquire and consolidate branch networks, and meld Red Team and Blue Team computer systems.

    Industry observers have witnessed the repeated inability of banking mergers to accomplish these tasks even in boom times and when led by the best and the brightest. It would be dubious public policy to entrust the nation's re-born banking system to the relative morons in the regional banks who did not have the intelligence to profit from the merger, sub-prime and securitization trends when there was ample money to be made.

    To summarize: we can successfully extract all the toxic assets at a cost of less than $1 billion -- with a guaranteed profit to the taxpayer -- if we give the CEO's the right incentives and let the invisible hand of the market work in the way Adam Smith foresaw.

    John Carragee
    Harvard Business School
    MBA 1980

  4. Hi,

    We have just added your latest post "How to buy toxic waste " to our Directory of Environment . You can check the inclusion of the post here . We are delighted to invite you to submit all your future posts to the directory for getting a huge base of visitors to your website and gaining a valuable backlink to your site.

    Warm Regards Team


Post a Comment

Popular posts from this blog

Nonlinear Temperature Effects Indicate Severe Damages to U.S. Crop Yields Under Climate Change

Commodity Prices and the Fed