« The European collective response | Main | Why is the Fed Paying Interest on Excess Reserves? »
What caused the financial crisis?
Forget about particular details for a moment, in conceptual terms what led so many financial institutions to take so much excess risk? Bob Frank addresses that question and here is my list of major factors:
1. Collective stupidity: A lot of Greeks believed in Zeus and a lot of people in 1938 thought Hitler would be good for Germany. They were just plain, flat out wrong. I'll also put "model error" under this heading. The relevant stupidity concerned both the fate of home prices and the degree of acceptable leverage.
2. Writing the naked put: This is Bob Frank's main explanation, noting that he uses different terminology and adds a relative status dash to the argument. If you don't know options theory, just imagine betting against the Washington Wizards to win the NBA title every year. For a lot of years you'll earn super-normal returns, but one year (not anytime soon, I can assure you) you'll be wiped out. That is essentially the strategy the banks were playing. They were going "short on volatility," so to speak. In the meantime they reaped high returns and some amazing perks for private life. It's hard to just call the party to an end, even if you have a relatively long time horizon.
3. The neutering of debtors. This is the sophisticated form of the moral hazard argument. Bailouts mean that debtors and depositors don't have enough incentive to keep safe the firms they give their money to. Note that #3, as a corollary, suggests that equity holders do not on their provide adequate safeguards against a crash.
Evaluation: You can pin most of the blame on #3 provided you think that a) our government really could let these firms default on their debts ex post and b) society is willing to live with significantly less liquidity transformation up front and also lower returns for depositors. I reject this mix for reasons of time inconsistency, namely that ex post the bailout is always on its way so this is simply something we have to live with.
You're left with #1 and #2 but it is hard to assign relative weights because they work together. The people earning money under #2 won't work terribly hard to disillusion the fools and frauds operating under #1.
At times I am tempted to add #4 to the mix:
4. The increasing value of human capital: Bankruptcy is no longer so painful for the wealthy. You can always get another high-paying job plus you have $10 million squirreled away somewhere in Switzerland. You could end up working for the guv'ment for $130K a year and your life still is pretty good once you get over the shock of adjustment. So why not take lots of risk and try to get ahead of the other guy?
The full story then involves additional resources being put on the table -- for possible risky investment -- as a result of easy monetary policy, pro-housing government policies, the global savings glut, and simple bad luck. I'll cover those factors in more detail soon. And I'll also have more to say about some of the details of mortgage-backed securities and accounting practices and regulation; those were factors too, although not at the level of generality I am covering here.
Addendum: Here's Mark Thoma and Barry Ritzholz. In the comments Robert Feinman is square on, read him.
Second addendum: Megan McArdle adds quite a bit.
Posted by Tyler Cowen on October 6, 2008 at 06:23 AM in Economics | Permalink
Comments
Mr. Cowen,
What about the scope creep of the Federal Government, where the idea that anything improving the "general welfare" goes,
and let the 10th Amendment be damned?
Once could argue that the New Deal opened pandora's box of moral hazards, and "The Paul Wellstone Mental Health and Addiction Equity Act of 2007" is just another in the sequence of foul things pouring forth from the precedent.
Cheers,
Chris
Posted by: C Smith at Oct 6, 2008 8:05:20 AM
"The Financial Stability Forum (FSF) was convened in April 1999 to promote international financial stability through information exchange and international co-operation in financial supervision and surveillance."
great success on the financial stability front led to an increase in leverage which led to....etc....read minsky for more information.
the gnomes did it.
Posted by: jck at Oct 6, 2008 8:08:24 AM
How about this for an incredibly simple-minded summary:
The Fed has understood its job to keep rates as low as possible (and growth and employment as high as possible) so long as inflation remained under control. The ultra-low rates did not cause a big spike in the CPI, so the Fed didn't feel the need to raise them, but cheap money did cause (or contribute to) the bubble.
So, going forward, the Fed will be charged with preventing bubbles as well as inflation and should never again accept the argument that "this time it's different".
Posted by: Slocum at Oct 6, 2008 8:21:34 AM
#4 is the newest and most interesting thing i've read about the financial crisis in quite some time...
Posted by: paul at Oct 6, 2008 8:32:30 AM
there are so many things that fall under the general stupidity category.
i will add to the list the culture of "you must own a house." and the tax code that encourages it, and the way people get so emotionally attached to housing. I think housing should be thought of as a durable good that appreciates in value rather than this major goal that you must accomplish or your life is incomplete.
Feinman is right on in the comments though. This mess would've been way more contained if it weren't for the derivative products--and their lack of regulation. For that reason it bothers me a lot that there are still many people who place the blame for the crisis squarely on the shoulders of stupid decisions on the part of homeowners or predatory lenders.
Posted by: pants at Oct 6, 2008 8:33:47 AM
So can anyone explain how $700 billion is going to unwind $65 trillion in CDOs? That's 93:1.
Posted by: David at Oct 6, 2008 8:47:21 AM
#4 is true and it isn't entirely new. I was a junior trader at a hedge fund in '95, and it was well known that a failed trader would just got a job somewhere else, often for more pay! The result is that bonuses are a call option on one year of trading performance, and good risk managers/internal controls are necessary to keep gambling in check.
I would argue though that this is a good thing! The point of hedge funds, VC, PE, and traditional financial intermediaries is to take on risk, which on average increases society's returns/wealth even if some desks and some years turn out badly. We are better off now than we were in the early '90's, and even a severe recession won't wipe out 15 years of growth.
The problem with the mortgage market, IMHO, is that in too many firms sales and trading escaped risk management's control. And that too much of risk management was based on historical projections and not enough consideration of unprecedented risks like a nationwide home price decline.
Posted by: DK at Oct 6, 2008 8:49:55 AM
Isn't the global savings glut a big factor here? The Chinese et al have a ton of money to invest, much more than apparently we had good investments to put it in.
Posted by: Sam TH at Oct 6, 2008 8:59:15 AM
Global savings glut: So, reduced demand caused a bubble eh? Not only did they give up all their loot and didn't use our bucks to invest in things that would yield high returns, they got hammered. Suckers.
We all know why Greenspan has been AWOL lately. But what about the Fed on the back end? There is still a lot of value in the houses. Had the GDP kept growing at 3%/yr the value would have caught up with price if the market had had a soft landing. Bernanke held interest rates high for a long time. Which came first, foreclosures, housing values dropping, or increasing rates? There was shift in perceived costs versus values, but when did it hit the tipping point? Who cares?
I am thankful to live in a country that can have a rip-roaring bubble from time to time. We found the point of diminishing returns on securitizing real-estate and put some folks into houses along the way. So, now these people who tried to make money leveraging other people's property will have to get real jobs where they invest real money into real productivity. Yay!
Posted by: Andrew at Oct 6, 2008 9:01:08 AM
It seems to me that #1 and #2 are symptoms not causes. WHY are people suddenly stupid? Isn't it true that people are equally stupid all the time? And equally greedy, for that matter? And the underlying willingness to risk their livelihoods is also not going to change.
The idea that people are richer, have more human capital and so on, does not make people more likely to risk their livelihoods, but it does give them -- combined with how we define bankruptcy and so on -- a fall-back in case they lose their shirt. They are risking less because they know they can recover -- hence, moral hazard.
That could be a small contributing factor, but people aren't going to suddenly want to bet everything on something stupid, just because they can recover. There is obviously something else going on.
#3 and the little side notes in the paragraph starting "The full story..." begin to get to the core of it. It isn't just more resources on the table due to easy monetary policy, it is tricking people into thinking that they can make easy money: even if its not real and is ultimately inflation, how much can I make right now? So long as housing prices are going up and up TODAY, I don't need to worry about the ultimate crash. And so on. Combine that with the government ultimately promising that housing will keep booming, with its pro-housing policy and bailouts, and you have a recipe.
Posted by: liberty at Oct 6, 2008 9:03:11 AM
Re #4: But when wasn't this the case? Bankruptcy has been a part of the high-risk finance playbook since Jay Gould, at least.
Posted by: Cyrus at Oct 6, 2008 9:04:23 AM
David:
That's easy: there aren't $65 trillion in CDOs. There are $65 trillion (give or take) in CDS, which are a totally different thing, and the vast majority of that nominal value arises because the CDS market is so illiquid that when you want to get rid of a CDS you're carrying it's easier to just write an offsetting CDS than sell the original one.
So instead we just have to deal with $2 or 3 trillion in face value of mortgage backed CDOs and god knows how much net CDS exposure, plus a peak-to-trough loss of $5 trillion+ in value on housing, plus every other asset that was overvalued by abundant liquidity and excessive leverage compared to our current situation of frozen credit markets and rapid deleveraging. $700 billion is far from obviously up to the task, but it's not quite as bad as you would make it out to be.
Posted by: Dave H at Oct 6, 2008 9:09:21 AM
As phrased here, number two sounds as if the participants were completely aware of the game they were playing but might be even more potent shrouded in uncertainty. How can it be separated from the collective madness of number one and the moral hazard and agency problems of number four?
But what is the solution to this problem? It seems pretty deep rooted. How should banks decide when to stop? Is there any feedback mechanism between pre-emptive self-denial and market crash? Without a good one the severity of the current crisis may be a function of the increased concentration of the financial services sector. That concentration is only increasing so does that mean the next crash will only come to a head with an existential crisis for the new big three?
Proper, end-user, investors didn't buy enough of the mortgage backed securities which is why so much was left on on bank balance sheets. In other words lenders did turn down many of the problem investments, it's the intermediaries who did not.
Posted by: bunbury at Oct 6, 2008 9:14:05 AM
At the end of Frank's essay he gives recommendations:
quote/
=============
Where do we go from here?
Many people advocate greater transparency in the market for poorly understood derivative securities. More stringent disclosure rules would be good but would not prevent future crises, any more than disclosing the relevant health risks would prevent athletes from taking steroids.
The only effective remedy is to change people’s incentives. In sports, that means drug rules backed by strict enforcement. In financial markets, asset bubbles cause real trouble when investors can borrow freely to expand their holdings. To prevent such bubbles, we must limit the amounts that people can invest with borrowed money.
=============
/unquote
Is he right?
Posted by: John B. Chilton at Oct 6, 2008 9:14:22 AM
Andrew Lo also stressed #2, the naked put, a few years ago although I think he stressed a "deep out-of-the-money naked put"
Posted by: jack at Oct 6, 2008 9:24:05 AM
I'm not sure I see why having to live with #3 means we can't blame it?
Posted by: conchis at Oct 6, 2008 9:33:40 AM
Slocum--
The Fed was trying to keep rates as low as possible without inflation--- ok. But, the metric they were using for inflation had a huge flaw. They included rent but not house prices in the housing component of inflation. That meant that they really were causing inflation; they just didn't know it.
"Preventing bubbles" is, in general, not possible. Encouraging bubbles-- as the fed and freddie and fannie did here-- is all to easy.
Posted by: Joe In Morgantown at Oct 6, 2008 9:44:29 AM
There's no Zeus?
Posted by: Jayson at Oct 6, 2008 9:46:38 AM
I'm willing to concede that this may have been a small factor overall, but would contend it was a large factor at least for AIG: the role the government plays in skewing investments towards assets with a higher return in exchange for a low risk of calamitous outcome. When you can get a AAA-rating by protecting against the first 10% downside in a CDO (by segmenting that 10% out into a separate tranch) it becomes the preferred investment vehicle, given insurance company requirements to go for such-rated instruments.
However, I wish I could find the cite, I remember reading Mises on the dangers of incentive pay -- that when you reward a manager for upside only without a corresponding liability to share in downside you get inherently risky decision-making.
Posted by: Gary Leff at Oct 6, 2008 9:48:11 AM
So much for Rational Expectations, I guess?
Posted by: Howard at Oct 6, 2008 9:57:16 AM
I would add
#5 the ability to distribute failure.
Mortgage brokers didn't care because so long as minimal standards were met, they could sell to banks. Banks could sell to investors. Intermediaries such as Fannie/Freddie developed Congress as a cushion. [etc.]
Financial intermediaries assumed liquidity would continue to grow, not shrink, and they'd never be stuck with things they couldn't sell.
But #1 is my favorite -- anyone who's served on committees knows that occasionally the committee will recommend something that's dumber than anything any single member of the committee would recommend.
Posted by: ZBicyclist at Oct 6, 2008 10:00:51 AM
Tyler,
What happened to the phrase "accounting fraud"? People who say that the problem was lack of oversight simply don't understand the game. There is plenty of communication between people in finance and the people who run government. The people running government certainly weren't fooled by this collapse -- they just didn't know exactly when it would happen. I have been warning friends about it for nearly a decade.
Everything that has happened recently has been understood since the 1990s or earlier. While I was working as a fixed income trader, everyone I talked with seemed to believe that there would one day be a massive mortgage debt crisis. The government seemed to be doing everything it could to write regulation that allowed for the packaging and repackaging of debt in a way that allowed GSEs, banks, and hedge funds to play the game of pass-the-hot-potato.
I was unfortunately close to the LTCM collapse, and traders from the firm I worked for met with Fannie Mae officials to find out that their leverage was far greater than the 80:1 number I've seen quoted. Scary. They were more highly leverage it seemed than D.E. Shaw, whom was attacked in the market due to fears that it was too much like LTCM. Though perhaps David Shaw never mentioned the absurdity of Fannie Mae's Sharpe Ratio when he was hosting Al Gore for lunch?
Why indeed would Andrew Cuomo pick such an atmosphere to start jamming low-quality loans into the Fannie Mae umbrella?? Lack of regulation indeed!
Economists and pundits need to stop saying "oh goodness, this is too complex" and start waking up to the fact that there was most certainly massive conspiracy by some people in the market to make a lot of money and pass the buck, and that the bigger part of the bubble begins there. Accounting! People screamed that word for years, but ears are deaf and memories short. Money appearing out of nowhere! It made government look so good for awhile. When hedge fund owners go to lunch with the most powerful people in government, what do you suppose they talk about? Everything except how the government will handle the bailout?
Posted by: infopractical at Oct 6, 2008 10:02:47 AM
Clawback the bonuses, make people accountable with prison terms, and you'll see this problem go away quickly.
Posted by: meter at Oct 6, 2008 10:15:44 AM
This mess would've been way more contained if it weren't for the derivative products--and their lack of regulation.
Perhaps -- but derivatives wouldn't have been a problem without the bubble. And in non-bubble situations, derivatives are useful. That is a housing bubble without securitization and swaps would still have been very bad. But securitzation and swaps without a bubble would have been harmless / beneficial.
"Preventing bubbles" is, in general, not possible.
But is that true? Have we ever really tried? If people knew the Fed was going to jack rates if housing prices kept rising at very high rates, would anybody have been willing to speculate on continued increases?
Posted by: Slocum at Oct 6, 2008 10:16:11 AM
You can't stop a bubble if you don't see one. The Fed didn't see a bubble.
There were clues; they should have been in the inflation rate, properly measured. But they didn't see it:
2005, Bernanke: "There is no housing bubble"
http://www.washingtonpost.com/wp-dyn/content/article/2005/10/26/AR2005102602255.html
I guess that argues for option 1.--- stupidity. And Bernanke is one of the smart guys.
Posted by: Joe in Morgantown at Oct 6, 2008 10:35:47 AM
