Bum Assets Auction Idea

Steven Landsburg at Slate floats an interesting idea for the Treasury TARP auctions:

Treasury Secretary Paulson has committed $250 billion of his $700 billion war chest to purchasing bank equity but that leaves plenty (if he receives full funding) to buy troubled assets directly.  However...

The big question about those distressed assets is: What price should Treasury pay for them?

You might reasonably say that the fair price for an asset nobody wants is zero. But bailout proponents tell us that these assets are plenty valuable; it's just that nobody's stepping up to buy them because it's hard to borrow right now. So how do you set a fair price for an asset that nobody else is bidding on?

Last week I couldn't have answered that question. Today, thanks to a conversation with my University of Rochester colleague Mark Bils, I can. Bils proposes to adapt an auction scheme designed (in an entirely different, currently irrelevant context) by Harvard professor Michael Kremer.

Here's (roughly) how a "Bils-Kremer" auction would work: First, put 10 similar distressed assets (such as a series of collateralized debt obligations) up for auction. At the close of the auction, the Treasury pays the winning bids for nine of these properties. The 10^th property (chosen randomly) gets sold to the winning bidder.

The advantage of a Bils-Kremer auction is that the Treasury buys assets and recapitalizes the firms holding those assets while paying only what some private bidder thought each property was worth. Now repeat with 10 more properties. And so on. Under this plan, nine-tenths of the liquidity comes from the Treasury, but ten-tenths of the price setting comes from the assessments of private investors with the incentive to bid judiciously. In other words, the prices can reasonably be considered fair.

Hmm.  I would propose one modest tweak - have the Treasury commit to keep its assets off the market for, say, five years.  A slight reduction in supply might help prices as well.

I also have a large question about the meaning of "fair" in this context.  At current Treasury rates Paulson could fund his purchases for ten yars at a rate below 4%.  That is absurdly lower than the hurdle rates for debt an dcapital that any private investor could be contemplating.  However, low hurdle rate or not, it means that Paulson could (attempt) to estimate the expected cash flows of these assets, discount them at four percent, and pay that price.  Over the next ten years Treasury will then break even on a cash flow basis.  Bid a bit lower, et voila!  Treasury makes a profit.

Now, that is not a "profit" in the sense that it reflects proper risk adjustments.  That said, as a practical matter even if every investor came to agreement about the expected cash flows of an asset, no other investor would be able to match Treasury's discount rate.  Consequently, Treasury can always "afford" to bid more; the question is, why would it make sense to pay more than the next highest bidder?

On the other hand, this is the Treasury bidding, with a public duty to bail out the economy; they are not simply some profit-seeking private capital pool looking for an attractive return on their capital.

Under the proposed bidding scheme Treasury will end up earning private capital rates of return on assets funded at Treasury rates.  That is a great deal for the taxpayer, but may not represent the salvaton of the financial system.

Comments

[Charlie (Colorado)]

I suspect anything that gets the MBS and CDS stuf off the market long enough to allow for fair evaluation and price discovery without rendering eight or ten top financials insolvent will save the financial system.

[Joe the Spellchecker]

$250B to purchase bank "quity"? Paulson should have ponied up another $50B and pick up the "e" too.

[glasater]

Interesting article at portfolio.com. Goodbye and f You

From a hedge fund manager.

[Rick Ballard]

However, low hurdle rate or not, it means that Paulson could (attempt) to estimate the expected cash flows of these assets, discount them at four percent, and pay that price. Over the next ten years Treasury will then break even on a cash flow basis.

Not really. That would just create a two tier market the day after the $700 billion is all spent. A clean, transparent estimate of expected cash flows (perhaps done with a range of default estimates) discounted at average market rate for a selected period would provide a justifiable economic rationale for the purchases. I really don't see pricing the offer to purchase correctly as being much of a problem. Extinguishing the CDS 'web of such complexity that it is inexplainable' seems far more difficult to me. Blocking enforcement of collection on any non-registered CDS might be a decent first step towards clearing that minefield. Give CDS parties one month in which to register, no registry thereafter, no standing higher than any other unsecured debtor in a bankruptcy court.

It's the "there's something in the bushes and nobody knows what it is" that is preventing a quick return of confidence. The CDS was sold as a form of insurance, the fine print revealed that on certain days and at certain times it also functions as a hand grenade. The precise date and time are only revealed by detonation.

How many would you like?

[spongeworthy]

I buy distressed assets in a different market and I've given a lot of thought to this problem. The auction you describe is brilliant, certainly, and far better than any idea I had.

But finding that 10th bid will be tougher than you might think. If there were a buyer out there willing to pay anywhere near a deeply-discounted, risk-adjusted price, they'd be up to there ears in sellers in no time.

Still, being only on the hook for one-tenth of the auction while the other 9 disappear off the market for 5 years, that is tempting. And after the first auction or two, buyers will see a market forming and might rush to get in.

I'm sold.

[steve sturm]

The reason nobody's rushing to buy is NOT because "it's hard to borrow now". It may be hard to borrow, but there are plenty of players with cash. They're not buying for the same reason nobody is buying: nobody has a clue as to what those assets are worth. And by 'worth' I don't mean what they are trading for, I mean a value based on the PV of the expected cash flow. And the reason for that is because it's darn near impossible to gauge cash flow.. and that is because (1) nobody is sure what is in each bucket, what with all the slicing and dicing of mortgage debt (is package A full of 30 year paper with good credit and nice downpayments? Or is it full of junk, overvalued assessments and borrowers who lied about their income? Who knows?) and (2) the political climate keeps anyone from being able to estimate what the risk of loss is (how do investors figure out whether delinquent borrowers will be foreclosed on? Again, who knows?) Without a clear understanding of what one is buying, and Paulson isn't going to be any better able to predict cash flow than me (Ok, bad example), the only prudent thing is to value them at close to zero.

[Charlie (Colorado)]

Steve, you just succinctly made the case for Paulson's purchase: we know perfectly well that the assets in aggregate are worth a lot; we just don't know how to value an individual tranche. What's the answer: someone buys them until price discovery can work. Why the US Treasury? You broke it, you bought it.

[steve sturm]

Charlie: I agree that the cash value of these instruments (in total) is higher than their current trading value; while I've objected to aspects of Paulson's plan on philosophical grounds, I've long realized that we the taxpayer are likely to make ourselves a nice profit.

Since nobody has a clue as to what the earn-out value of this debt is, putting in a bid is analogous to bidding on boxes of spoiled cereal, some of which contain diamonds. Without the diamonds (i.e., performing mortgages), the cereal is worthless. But nobody knows how many diamonds there are, nor does anybody know the size and quality of the diamonds that are in the boxes. Given this, how does anybody justify putting in anything but a low ball bid? (and how could one be sure they were making a low bid? The face value of the tranche offers no clue).

While this would ensure that the federal government didn't overpay, faced with getting only lowball bids, why would holders of this debt want to sell? Wouldn't it make more sense to try and hold on to maturity, or at least long enough to better gauge the cash flow?

That's why I think any rescue plan has to involve a concerted effort to open up the boxes and figure out just what is inside; everything else is treating the symptoms and not addressing the underlying cause of the problems (lack of transparency). Those holding boxes with no diamonds are going to suffer a 100% write-off and will probably be liquidated, but at least the others will be able to finally establish the value of their holdings. And with market prices so low now, the firms holdings diamonds will be able to recapture some of the excess write-offs they've been forced to take, rebuilding their capital. Any ideas on how to open up the boxes?

[clarice the plumber]

I asked a long time ago whether since there was software created to make up these instruments there couldn't be software created to unbundle it. (In essence you'd be x-raying the cereal boxes to find out where the diamonds are). Cathyf said it certainly could be done. I can't imagine why some smarties aren't working on that right now. Create it, sell it and let everyone figure if they want open private auctions or what once we know what's there.

With that and pay per view executions of the evildoers which Krauthammer suggested, the fed treasury might soon be awash with gold again.

[Pofarmer]

If I'm one of the companies that has these for sale, and I can buy 1 while selling 9, you better bet your ass that somebody is gonna run them up for me.

[steve sturm]

Of course, the same people who put it together are the ones who are going to unwind them... how confident does that make us?

[clarice the plumber]

I'm talking about the creators of software that shows what's in the boxes not the guys who thought buying and selling gold wrapped in trash bags with a week's worth of detritus was a neat idea,steve.No one has yet suggested to my knowledge that there were errors in creating the bundles.

[Rick Ballard]

"Any ideas on how to open up the boxes?"

They're open. Individual tranche performance isn't a secret. The Fed and the OCC have precise">http://www.occ.treas.gov/mortgage_report/2008/Jan_Jun/executive_summary.htm">precise knowledge concerning loans and the tranches with which they are associated via regulated servicers - "The combined servicing portfolio constituted more than 90 percent of all mortgages serviced by national banks and thrifts, and approximately 60 percent of all mortgages outstanding in the United States."

The company servicing an MBS deal has the tranches mapped out very tightly. I still haven't heard any details as to how the Fed proposes to detach the Stinky Stuff tranches from specific MBS. Will the detachment constitute a trigger event for the CDS? I can see the mechanics of calling in a tranche - it would be relatively easy to notify all owners that CUSIP 123XYZ was going to be called effective 11-15-08 at $670 per $1,000 face par. That's just nuts and bolts. The impact upon the CDS is a lot less obvious to me.

It would be sorta nice if B & B Enterprises would toss a model out for comment. It might even restore a little more confidence.

[Rick Ballard]

The OCC is really brilliant - they can even go back in time.

That one is actually a very good move - it strengthens the regulatory capital base of all the regional and local banks which were strong armed into ownership of FM preferred.

[Gmax]

From the link above:

The combined servicing portfolio constituted more than 90 percent of all mortgages serviced by national banks and thrifts, and approximately 60 percent of all mortgages outstanding in the United States.

And from the footnote to the link:

1 The nine banks are Bank of America, Citibank, First Horizon, HSBC, JPMorgan Chase, National City, USBank, Wachovia, and Wells Fargo.
2 The five thrifts include Countrywide, IndyMac, Merrill Lynch, Wachovia FSB, and Washington Mutual.

Bank of America, Citi, JPM Chase, Wachovia and Chase already got injections of capital from Herr Paulson. Ditto Countrywide ( via B of A ) Merill Lynch ( Via B of A ) and Wachovia FSB ( via Wells Fargo ).

IndyMac was closed by OTS and WaMU was sold to JPM Chase.

That leaves only First Horizon, HSBC, Nat City and US Bank as being not yet addressed.

US Bank is widely thought of as the best capitalized second tier institution in the country. I would guess in their Minnesota nice way, they will tell Paulson to go pound sand.

Nat City needs the help, HSBC and First Horizon may or may not.

But we are about done with capital injection for all intents and purposes I would say.

[Intors]

Bush and Buffet say we'll make money on the loans. Joe's loans are written off every year and it's only 80 billion for one type. We get the payment of knowing we done good for poor foreigners. Soon we can help our own the same. Why shouldn't we have free money too?

[Rick Ballard]

GMax,

Did you read the hedge hog arsonist's good-bye note that glasater linked above? I won't be surprised if he spends quality time with SEC investigators explaining how naked short selling isn't really illegal as long as your Soros inspired heart says it's the right thing to do. I wonder how fast Obama will be to issue that first pardon?

[Gmax]

An ode to the great and beneficial George Soros? You would have to be an idiot to give your funds to such a man. Sadly apparently a lot of folks did.

Meanwhile, Highland Capital here in Dallas is on a race to see how fast they can turn $8 billion to something significantly smaller. They are causing some market dislocations for sure.

[Rick Ballard]

"Sadly apparently a lot of folks did."

Yeah - and he pegs them pretty well. People smart enough to be smug about having profited by giving an arsonist the money necessary to buy gasoline and matches. They got a nice return - unless their "safe" investments were in Citi stock.

I'm wondering if a lot of this folderol is about covering Citi's butt. That list of 14 is reduced now to 8 and JPM, BAC and WFC will emerge (after a while) as the strong players. WFC has an edge now everywhere in the mortgage business with BAC running second (with a better originator in Countrywide than WFC picked up with the Wachovia buy). WFC services so many mortgages that they can drill data to the point where they can probably match BAC on writing clean mortgages within six months, should they choose to do so. Now that I understand the background on HSBC, I'd say the correct ranking in safety going forward would be WFC, HSBC, BAC and then JPM. I'd like to find a mutual with those holdings - plus a few big home builders. Housing cannot continue to stay below it's natural growth level for very long without rents being driven to the point where not buying a house is foolish.

[Pofarmer]

Housing cannot continue to stay below it's natural growth level for very long without rents being driven to the point where not buying a house is foolish.

Depends on the rest of the economy. Plus, how many units have been opened up by illegals jumping ship?? Plus?????

Let's say we divvy these CDS up into regional packages, maybe even by county or principality or whatever. Make the chunks small enough that a whole lot of players could digest them. One of the problems we've gotten into here is that too much power and money got concentrated at the top. Let's let the market spread it back out. Make the MBS transparent, then let anybody with the means bid on it. Take it down to the local level so you give access to as many bidders as possible. Otherwise, all you are going to do is shuffle the portfolios around and reconcentrate the problem at the govt level. Let's spread it around and let a few other folks make a buck. There's some pretty good ole cow town bankers out there that might surprise a few folks.

[Rick Ballard]

Pofarmer,

I can run numbers on housing stock until your eyes will not just glaze over but actually melt. The illegals don't offset the necessary increase. The basic equation is simple (Births - Deaths) + (Immigrants - Emigrants) divided by a upickem average per household. The total has to be increased by the number of units which are destroyed or demolished or rendered useless by falling population counts in dying areas (vide Buffalo, NY). It does not depend upon "the rest of the economy" because shelter is not a "want", it's a "need". It's kinda like corn - you can find a substitute for a while but the market will equalize within a season or two.

I know that all the idiots who lied their way into loans and then handed in their keys when appreciation stopped are right back in the rental market. So are all the kids that come along every year who are just starting on their own and need a place to live. The concept that you bust and then rent a place equal to that which you busted on for less than your mortgage payment will work for all of two years (which it has). Then, the landlords will take advantage of tighter vacancy factors and, by golly, your rent just went up 10%.

I sort of agree with you on the auctions but you're ignoring the CDS factor. The regionals don't really play heavily in that market and without a secure registration system I'm not sure that it's a good bet. If a local or a regional could pick up a Stinky Stuff tranche that was for their area, that would be great. Those don't exist, unfortunately, because the geniuses who designed the MBS in the first place used geographic distribution as a primary consideration in the design of the product. A very logical and very stupid response to the fact that housing isn't national - it's regional. The concept works great until there is a national setback and then it doesn't work at all. There is no risk offset.

Don't worry though, the finest minds in the US are working this evening to come up with a product that will be even more stupid (and much more complex) than the current failed model. Just thinking about all those brilliant, well educated and well trained minds searching for the perfect new solution to a problem that has been successfully dealt with for centuries should give you great comfort.

Do you think that hand loads are so much better than mass produced to be worth the effort when you're going to be using a great deal of them over a short period?

[kim]

"Doesn't cause barfights or wife beatings". Or trips to the Emergency Room.
==================================

[Pofarmer]

sort of agree with you on the auctions but you're ignoring the CDS factor.

Yeah, that's really the fly in the ointment. Without that, it's pretty simple, with that, you don't know when the whole thing is going to go up in flames.

How do you get rid of the accompanying CDS?? Just wait it out?

Don't worry though, the finest minds in the US are working this evening

See, that's precisely what worries me. All these very smart people are trying to get themselves and their buddies out of this without the ass kicking they so richly deserve. Unfortunately, that involves me and a whole bunch of other peons taking the load on our shoulders for a while. No thanks.

Any other operation in the world and you would auction off the carcass and be done with it. They managed to booby trap this thing.

[Thomas Esmond Knox]

Dear Rick Ballard,

If you had Paulsen's money, which allowed assets would you buy?

I'd buy the best quality at the lowest price.

In that order of preference.

[Rick Ballard]

Thomas Esmond Knox,

As a private investor I would follow your suggestion. Paulson is not a private investor and best return with safety is not the criterion used to hire firemen to put out a fire. You cannot hire firemen properly without ascertaining which of them have purchased insurance policies upon the structure in flames. Paulson must deal in an arsonist's market with clever thieves who will steal your fillings, should you open your mouth.

Restoring public confidence in an institution replete with whores, liars, thieves and cheats is a very difficult task and Paulson must make the actions taken appear to be open and accessible. Otherwise the suckers will never entrust another nickle to the frauds and mountebanks who have managed to sustain the illusion that "models" have anything whatsoever to do with reality.

The trick here is to come up with the proper illusion - not because the income stream from a thousand mortgages is indeterminable, for that requires aught but a simple computer program. The trick is to make the answer seem almost incomprehensible because otherwise the streets of Manhattan would have lampposts festooned with the bodies of investment bankers. I do not dispute the fact that justice would be served should that happy day occur but justice has no place in the equation. Maintenance of the system has more importance because the collapse of the charade would entail a greater degree of suffering than its maintenance.

I would cheerfully see every board member of every investment bank involved in this fiasco hanged next to every politician whose purchase assured its failure tomorrow morning if that were assuredly an outcome which would cause the least distress to the most people. Each of them has, without question, earned the denigration which I offer for betrayal of fiduciary and public trust and very few will pay a penalty.

I cannot countenance the failure of the system as a response to the moral turpitude of those entrusted with its operation. The likelihood of a man on horseback arising from the ashes rather than a Phoenix is simply too high.

Better to begin the war with 'delenda est Citi' than to pull the system to pieces in order to see what happens next.

[Tom Maguire]

Paulson should have ponied up another $50B and pick up the "e" too.

He is holding out - he hears vowels are cheaper on Wheel of Fortune.