Inman

Will Treasury’s purchase plan work?

This article was written the day after the House approved the revised version of the Treasury/Federal Reserve bailout plan, which had already been approved by the Senate.

Much the most striking thing about the plan is that it represents a completely new approach to dealing with financial crisis. To my knowledge, no government has ever done anything like it before. That is not a conclusive argument against it if there are compelling reasons for believing that it will work. But — and this is the stupefying part — no such reasons have been given! A search of the Federal Reserve Board and Treasury Web sites comes up empty.

The alternative is the lender-of-last-resort (LLR) approach, under which the central bank lends at a penalty interest rate to solvent firms in distress. The LLR approach, which has been used until now to deal with the crisis, has behind it a long intellectual history as well as substantial operating experience. The new approach has no intellectual foundation at all.

The proposed plan is to have government buy substantial amounts of mortgages and mortgage securities. The alternative is an expanded LLR program that would provide those who own mortgages and mortgage securities new and less restrictive ways to use them as collateral for loans, possibly supplemented with equity participation.

The distinction between the two approaches is akin to a distinction between fighting individual forest fires as they arise, and wetting down the forest in the hope that that will prevent any fires from breaking out. Where we know how to fight individual fires, we have no experience in dampening the forest and no blueprint for doing it has been offered. About all that we have been told is that some kind of reverse auction will be used where sellers would bid the lowest price they would accept for their mortgage assets. Government presumably would array the bids with the lowest-price bid at the top, and accept all those that add up to the total amount committed to the auction.

The most damning argument against the asset purchase plan is that it won’t work. The logistics of arranging auctions involving this most heterogeneous of all assets are mind-boggling. To get any semblance of uniformity in the assets offered in any one auction will require many hundreds of auctions. If the number of auctions was reduced to make the process manageable, the variability in asset values would be enormous and the government would be purchasing those with the lowest value.

In each auction, the government would have to allocate some amount of funds for purchase, scaled to their expectations of bid volume. If their estimates of bid volume are too low, they will pay too little, and vice versa. Because of this, along with the quality variations within the assets allowed in the auction, the distribution of benefits among bidders will vary greatly and arbitrarily.

Worse, this process may not put money in the hands of the firms that need it the most. They may not have the types of assets covered by the auction, or they may not be able to wait until the auction is scheduled, and then they can’t be certain that their bid will be successful. In contrast, the LLR approach of extending loan facilities directly targets those with the greatest cash needs.

There are no existing government facilities to implement an asset purchase program. Because the book has to be written as it is executed, the potential for waste, favoritism and fraud is enormous. In contrast, a loan program would be only a modification of well-developed LLR principles, which could be implemented through the existing Federal Home Loan Bank System, acting as agent of the Treasury or Federal Reserve.

In an earlier comment, I argued that the government would make significant profits on an LLR plan because it would charge borrowers a penalty interest rate. Sober second thought suggests that to make the LLR program work, the government would have to substantially liberalize its rules for valuing collateral and the percent of value that can be borrowed, which will result in losses in cases where the borrowers fail. This could turn the profit into net losses, but these would be trivial compared to the costs of the asset purchase plan.

Assuming the asset purchase plan is enacted, here is my prediction: An important player who runs into a cash squeeze and cannot wait for an auction will petition the government to do a negotiated purchase. With no objective way to establish the value of the assets offered, the government will turn down the request but offer to lend the needed funds against the same assets as collateral. The government may also require that it be given an ownership interest in the firm. This will quickly be followed by similar cases, and a new LLR program will evolve before the asset purchase program is off the ground. It may never get off the ground.

The writer is professor of finance emeritus at the Wharton School of the University of Pennsylvania. Comments and questions can be left at www.mtgprofessor.com.

***

What’s your opinion? Leave your comments below or send a letter to the editor. To contact the writer, click the byline at the top of the story.