I indicated earlier that there's been some debate about this question. Karl Denninger wrote up a proposal for how gaming the system might work, and Tyler Cowen seconded it. How would this work? Denninger's plan has three steps:

--I become a "bidder" and "bid" on my own assets at [an inflated price]

--I am providing 5 or 10% of the money.  The rest is covered by Treasury, The Fed and the FDIC via guaranteed bond issuance.

--The loan, ex my contribution, is non-recourse.  That is, I can lose 5 or 10% of the total portfolio purchased, but nothing more.

But this probably won't work. The problem lies in step one: You can't bid on your own assets in the manner Denninger describes. The Treasury guidlines (pdf) aren't totally clear on this point, in part because the sentence restricting bidding on one's own assets looks like it as written by William Faulkner. But here is the restriction:

Private Investors may not participate in any [public private investment fund (PPIF)]  that purchases assets from sellers that are affiliates of such investors or that represent 10% or more of the aggregate private capital in the PPIF.

Right then. One problem with this sentence (there are many) is that the first and second halves appear to be contradictory: The first part implies that you can't bid on your own assets, but the second part implies that you can bid on your assets so long as you are less than 10% of the investment fund. Which interpretation is correct? A Treasury official clarifies by email:

They cannot bid on their own assets, except under very specific circumstances in which they are less than 10% investor in PPIF, and even then, they would have to be cleared through our anti-fraud provisions to ensure that just such behavior is not occurring.

So the second part of the original restriction carries the day: You can bid on the assets only if you are less than 10% of the bidding partnership. (And even then you would have to be cleared by some anti-fraud provision.) I think this makes the risk of cheating pretty unlikely.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.