« China fact of the day | Main | China re-estimate of the day -- whoops! »
Game theory and balance sheets
What's really the problem these days?
The single best thing that could happen would be for the true magnitude of the losses suffered by banks and other exposed parties to be revealed and put in the P&L. Until what happens, fear of getting stuck with the hot potato makes banks unnaturally unwilling to extend credit against the kind of collateral that they would not have thought about twice accepting at the beginning of the year.
Here's the source. This is an important game to understand. Think of bank managers as being collectively averse to admitting a loss, if only because they might be blamed for that loss. So at first no one admits losses. Even though the market knows the losses are there, the market just doesn't know exactly where. But then the market has no accurate valuations for some asset classes. Those asset classes can't serve as collateral because just about any result -- relative to an unevaluated base -- could count as a loss and we've already seen that managers are loss-averse. Plus dealing with hard-to-value assets is difficult in any case.
The simple lesson is that bad news can be good news. "Predictability of environment" is a public good across managers. Once managers admit their losses, market liquidity can spring back to life and we can avoid a credit crunch.
Once managers admit their losses, that is.
Posted by Tyler Cowen on November 15, 2007 at 08:07 AM in Economics | Permalink
Comments
So does this mean we need an Ansari X Prize for the first manager to report his losses? Sir Richard Branson, where are you?
Posted by: Nate at Nov 15, 2007 8:53:50 AM
I think of it a lot like diseases that eventually become symptomatic, given enough time everyone will know, but until then, lots of people don't want to get tested (and aren't publicising the results of their test. So most information travels on rumor.
Posted by: nelsonal at Nov 15, 2007 9:22:55 AM
This was a major part of Japan's problem in the 1990s. They refused to write down the bank loses and left the banks to limp along rather than putting their problems behind them and resuming business as normal.
Posted by: spencer at Nov 15, 2007 9:28:11 AM
The Great Mea Culpa is already underway, though. The big money-center and investment banks have all sucked up middlin'-to-large writeoffs on both MBS and CDOs. Concurrently, The Terror has been instituted at the CEO level, though the results may not be what your average moralist would be hoping for.
Posted by: Bernard Guerrero at Nov 15, 2007 10:14:28 AM
The US accomplished a quick turnaround from the real estate blowup of the 1980s by nationalizing the
losses through the RTC and FDIC. Most of the exposed institutions (savings and loans comprised the
bulk of the US housing finance market back then) were driven insolvent, all the holders lost their
investments and most the people lost their jobs, and the Treasury threw an extra half-trillion in for
good measure. The effect was to drive real estate values down so far so fast that real estate went from
a questionable asset to a clear winner in very little time.
I do think this was beneficial to the US economy vs. Japan, where the effect of the 1980s bubble still
linger (and we've been through two more bubbles since!) But it really sucked for those involved in
the business, plus requiring a large Federal bailout whose proceeds went to enriching some astute fellows.
Fortunately the current mess is not as large, no one is really thinking the government needs to take over
Citi and Merrill. Much of the problem now is finding out who really bears the loss - a dollar of loss on a mortgage
is a default on a mortgage security, is a default on a CDO, is a default on a CDO-squared, is a default on
a SIV, which causes a money market fund to break the buck, leading to the fund's sponsor bailing out
the fund and finally taking that dollar of loss. So we see all these securities defaulting, but it's
still just one finite loss being passed around the table.
Posted by: mkl at Nov 15, 2007 10:16:13 AM
Part of the problem is that losing money is punished, when in a more rational environment what ought to be discouraged are poor risk/reward decisions.
If you can get someone to pay you $100 for picking the ace of hearts from a deck of cards, but you only have to pay $1 if you pick something else, then that is clearly a bet you ought to take*. This despite the fact that you have a greater than 98% chance of losing money.
It works the other way around too. If you get $1 for whent he other guy misses picking the ace of hearts, but only have to pay him off $25 when he succeeds, then the vast bulk of the time you'll make money, but occasionally you'll lose a big sum of money all at once. Despite this big loss, on a risk/reward basis you clearly made the right decision for your firm, if you get fired for your decision then your remaining colleagues will be far less likely to make optimum risk/reward decisions, to the detriment of the firm.
I'm not saying that this is what is going on here, they may well have made horrible risk/reward decisions, but I don't hear any talk in the media of risk/reward, only risk. If this mirrors what is going on during board meetings then these companies and the American people will be worse off in the long run than they otherwise would have been.
*Assuming you feel confident you'll get paid off, but that's outside the scope of my analogy.
Posted by: happyjuggler0 at Nov 15, 2007 10:32:48 AM
We need a coordination mechanism here, a group confession or amnesty. Perhaps a big drum circle of bank managers wasted on PBR? The incentives must reward cooperation and punish delay. In some ways, forcing resignations is a good way: the old guy takes the blame, the new guy sets the base as low as possible to get the growth jump he can claim as success. Hard to fire all those managers, but what if they all rotated by one company?
Posted by: David Zetland at Nov 15, 2007 11:08:09 AM
It's just the old "sell on the rumor, buy on the news"...
Posted by: martin at Nov 15, 2007 11:24:12 AM
"Think of bank managers as being collectively averse to admitting a loss, if only because they might be blamed for that loss."
Admitting a loss on individual assets/securities is only a piece of this game. Increasing the expected default risk on these asset classes, especially those that were part of a "rating transformation", can have cascading effects on regulatory capital calculations and borrowing costs (especially for loans collateralized by these assets) I'm not defending anyone that would hide losses to avoid painful writedowns, but there are forces more powerful than pride that would make a bank manager take a deep breath before running losses through the P&L.
Posted by: dk at Nov 15, 2007 11:45:37 AM
On Wall St. when the CEO loses his job, it is a very cleansing experience. The new guy wants to come in and clean house, be it people or the balance sheet. Expect MER and C (when they hire a new CEO) to be very conservative in their Q4 marks. MER will most likely take a write down of its First Franklin mortgage biz purchased last year at absolutely the wrong time. Also, expect all the banks to take further impairments on CDOs, leveraged loans, RMBS and now it appears CMBS in Q4 and Q1 '08. Part of the problem with immediately recognizing the losses, versus stringing them out over a few quarters, is capital adequacy. If they take the losses all in one quarter, their capital ratios will significantly deteriorate. If they recognize the losses over 3 or 4 quarters, they get the benefit of net income from all their other operations to offset the losses incurred, bolstering their capital base and not running afoul of regulator requirements.
Posted by: joe at Nov 15, 2007 12:14:12 PM
Here's another analogy: you work for an investment firm, under an arrangement that gives you 10% of any profits from your dealings, but no corresponding liability for losses. Under the circumstances, you are happy to make an arrangement whereby your firm gets $10M (of which $1M goes to you) every time some other party misses drawing the ace of hearts, but your firm pays $1B if the ace of hearts comes up. The other party is pretty happy with this too; in fact, he's willing to play this game every month. For months and months things keep going your way, and now you're a superstar at the firm. Eventually the ace of hearts is drawn; your firm holds a press conference, announces a big writedown, and fires you. You walk away with your millions, and the firm hires somebody else to do what you did. Everybody's happy, right?
Posted by: y at Nov 15, 2007 3:03:44 PM
Felix Salmon gets specific on this very point.
Posted by: mobile at Nov 15, 2007 6:11:53 PM
They would do if it they could. Assuming that the writedown wouldn't threaten solvency, there is absolutely no advantage for management to take writedowns over an extended period. Especially if you are brought in as a new manager.
There isn't a 'number'. There are estimates and models and not much of a market. There is a lot of stuff that doesn't have a price. I think the blame goes to financial engineering. No one really knows how bad things are, but they do know how complex they are. A worst case wipes you out, and isn't realistic. There is really no market, no history, and inadequate accounting and regulation for this stuff.
Banks would be better off if they could undo the securitization and simply evaluate mortgage portfolios. Banks are good at this, regulation and accounting is straightforward, and the underlying mortgages may not be that bad.
As an example, wfc has $83 billion in home equity loans, and 1.3% of them are not performing. They have taken some hits and raised reserves, but not enough to be much of a problem. I woud bet if they securitized these loans they would need to take much higher writedowns, and might have severe problems. Even if they sold off the lower tranches the remaining securities would take a much bigger percentage hit then unbundled loans. This, of course, makes no sense.
The argument is really that the problem is lack of transparency. For a while, they were actually paid for complexity and the sum was sold or booked as more then the parts. Now we are in the opposite environment where the parts, if you could disaggregate them, would be worth more then the securities.
Everyone has been burned. The regulators, the rating agencies, the accountants, the customers they sold this stuff to. No one is going to cut them any slack, and they are left with a problem that is, in principal, unquantifiable.
Posted by: Zigurrat at Nov 15, 2007 11:02:26 PM
SWG Credits for all star wars players~~
Posted by: swg credits at Nov 16, 2007 3:59:02 AM
The classic prisoners dilemma, but with the twist the first to come clean goes bankrupt. This is what is happening in the markets today. People here are assuming that the actual numbers faced by the banks are not that large, but they are due to implicit leverage and the discovery that some puts were buried in the deals. Goldman estimates a trillion, but if we have a 40 to 50% retracement in real estate prices nationwide, the losses are going to be much much higher than $1T.
If you think this is impossible, take a look at the blogs out there following Sacremento real estate prices. There are already 45% and greater losses out there, and we are nowhere near the bottom. Case in point: House sold in 2006 for 657,500 listed for 339,000 and on the market for 161 days or almost 6 months. Even this price is clearly too high. You could knock off 20% from the current price before you might get a buyer, and we are not at the peak in supply - that should be sometime in q3 - q4 2009.
http://flippersintrouble.blogspot.com/
One major investment bank will be allowed to fail, the rest will be bailed out. They are too big to fail as a group, but not individually, and we are facing a crisis where they might fail collectively. So the solution: Let one fail to appease the masses and put the fear of god where it should be, and then bail out the rest.
So there is a vested interest in not being the first to admit they are bankrupt, because the first to admit bankruptcy will be the IB forced to liquidate before the bailout comes.
Posted by: mickslam at Nov 18, 2007 7:33:44 PM
Posted by: 翻译公司 at Feb 25, 2008 9:44:18 AM
Posted by: Alii at Apr 3, 2008 9:14:37 PM
aion gold
aion money
cheap aion gold
cheap aion money
buy aion gold
Mabinogi online gold
Mabinogi gold
buy Mabinogi gold
cheap Mabinogi gold
Mabinogi money
2moons dil
2moons gold
buy 2moons dil
2moons dil
cheap 2moons dil
flyff gold
flyff penya
flyff money
buy flyff penya
cheap flyff penya
cheap flyff gold
Dofus kamas
buy Dofus kamas
cheap kamas
Dofus kama
Dofus gold
Dofus money
Knight online gold
Knight Gold
Knight Noah
Knight online Noah
Posted by: aion at Jul 14, 2009 2:38:19 AM
花蓮旅遊景點,花蓮,花東,租車,花蓮 租車,花東旅遊,花蓮租車公司,花蓮旅行社,花蓮租車網,花蓮行程,花蓮地圖,花蓮租車公司,租車,花蓮租車旅遊網,花蓮 租車,花蓮旅遊租車,花蓮租車資訊網,旅遊景點,租車旅遊,租車公司,租車網,花蓮包車,花蓮旅遊,租車 花蓮,花蓮旅遊網,賞鯨,花蓮旅遊租車,租車,花蓮租車旅遊,花蓮租車公司,花蓮租車旅遊資訊網,花蓮租車旅遊,旅遊,花蓮旅遊景點,花東旅遊,花蓮泛舟,租車旅遊,花蓮租車,賞鯨,花蓮泛舟,花蓮溯溪,花蓮旅遊,租車,花蓮租車,花蓮租車,花蓮租車,花蓮租車,花蓮一日遊,花蓮租車公司,花蓮租車,花蓮租車,花蓮一日遊,花蓮租車,花蓮旅遊地圖,花蓮租車,花蓮行程,花蓮租車
Posted by: 花蓮租車 at Aug 12, 2009 5:38:52 AM
花蓮租車公司,花蓮旅遊租車,花蓮,租車,花蓮租車公司,花蓮旅遊,花蓮租車公司,行易花蓮租車,租車網,花蓮旅遊租車,花蓮租車,花蓮,花蓮,租車,花蓮 租車,泛舟,花蓮租車,泛舟,花蓮溯溪,花蓮泛舟行程,花蓮溯溪旅遊,租車,花蓮一日遊,租車,花蓮旅遊,花蓮租車網,花蓮租車旅遊網,花蓮一日遊,花蓮租車公司,花蓮 旅遊 租車,花蓮 租車,花蓮,花蓮租車網,租車,花蓮 租車,花蓮租車,花蓮美食,花蓮租車,花蓮租車,花蓮租車,花蓮旅遊,花蓮租車,花蓮租車,租車,花蓮租車,花蓮租車,花蓮租車,花蓮旅遊,花蓮租車,花蓮租車,花蓮 租車,旅遊,花蓮,租車,花蓮旅遊,花蓮,租車,旅遊,地圖,花東,租車
Posted by: 花蓮租車 at Aug 12, 2009 5:39:13 AM
花蓮吉安慶修院,花蓮景點,花蓮旅遊,花蓮,鯉魚潭,花蓮太魯閣,砂卡礑,合歡山,九曲洞,七星柴魚博物館,花蓮碼頭,瑞穗牧場,台東景點,安通溫泉,花蓮旅遊景點,立川漁場,太魯閣,太魯閣豁然亭,台東旅遊景點,七星潭
Posted by: 花蓮租車 at Aug 12, 2009 5:39:29 AM