« Are you Brian Donnelly? | Main | Wikipedia on reverse auctions »
Regulation, a dialogue with Warren Buffett
QUICK: If you imagine where things will go with Fannie and Freddie, and you think about the regulators, where were the regulators for what was happening, and can something like this be prevented from happening again?
Mr. BUFFETT: Well, it's really an incredible case study in regulation
because something called OFHEO was set up in 1992 by Congress, and the sole job of OFHEO was to watch over Fannie and Freddie, someone to watch over them. And they were there to evaluate the soundness and the accounting and all of that. Two companies were all they had to regulate. OFHEO has over 200 employees now. They have a budget now that's $65 million a year, and all they have to do is look at two companies. I mean, you know, I look at more than two companies.QUICK: Mm-hmm.
Mr. BUFFETT: And they sat there, made reports to the Congress, you can get them on the Internet, every year. And, in fact, they reported to Sarbanes and Oxley every year. And they went--wrote 100 page reports, and they said, 'We've looked at these people and their standards are fine and their directors are fine and everything was fine.' And then all of a sudden you had two of the greatest accounting misstatements in history. You had all kinds of management malfeasance, and it all came out. And, of course, the classic thing was that after it all came out, OFHEO wrote a 350--340 page report examining what went wrong, and they blamed the management, they blamed the directors, they blamed the audit committee. They didn't have a word in there about themselves, and they're the ones that 200 people were going to work every day with just two companies to think about. It just shows the problems of regulation.
QUICK: That sounds like an argument against regulation, though. Is that what you're saying?
Mr. BUFFETT: It's an argument explaining--it's an argument that managing complex financial institutions where the management wants to deceive you can be very, very difficult.
Here is a good article on what the mortgage agencies have been up to.
Posted by Tyler Cowen on October 5, 2008 at 05:46 AM in Law | Permalink
Comments
that reminds me of 1984 book by Perrow "normal accidents" where he argued that you cannot make nuclear energy save, because the stuff is simply too complex.
In the same vein Zizek in interview recently said this idea of more regulation is a way of not thinking if this financial mess could be normal functioning of the system (freely paraphrased by me).
Posted by: Dainius Blynas at Oct 5, 2008 8:43:26 AM
A lot of my friends are repeating the Washington D.C. line: "We need stronger/more regulation." I'd like to answer them with specifics, about how regulation can be a problem, but I don't know how exactly.
The best explaination I've heard comes from Yves Smith:
Magan Mcardle interviews Yves Smith
What Yves is suggesting is that the government credit rating agencies, the SEC, the entire regulatory structure is part of the problem, and as is, it can’t “fix” anything because it’s part of the problem. Worse, because the “mixed” system (i.e., the credit markets and government regulation) is a ‘tightly-coupled’ complex system, the unintended negative consequences of any government intervention overtake and overwhelm any “good” the intervention and regulation was designed to do.
Buffett seems to be saying something similar, but it has to do with the market co-opting and/or fooling the regulators. They can do this because the system is so complex.
But -- doesn't regulation assume that only the government cares about risk and honesty? It's like the assumption is that the rest of us, market players, have no real incentive to manage risk, so government must do it. In the absence of regulation, there simply would be no evolved system to isolate, manage, and measure risk.
But is that really the case? It is true that government can regulate the market better than the market can regulate itself?
I think Yves is correct -- the market is too complex for the regulators, and on top that, the market can co-opt regulators and politicians, which is think is what really happened with Fannie & Freddy -- see Alan Kling "Financial Apocalypse Explained"
http://libertydesirebelief.thechartersofdreams.com/2008/09/financial-apocalypse-explained.html
Financial Apocalypse Explained
Regulation and intervention seems to almost always do more harm than good.
Posted by: The Charters Of Dreams at Oct 5, 2008 9:29:17 AM
"More regulation" is a variation on "Hope springs eternal."
Posted by: at Oct 5, 2008 9:48:41 AM
"More regulation" is a variation on "Hope springs eternal."
Or
"What's a heaven for but a man's reach exceeds his grasp."
Posted by: at Oct 5, 2008 9:49:39 AM
So in the end Buffett has no answers either.
Posted by: meter at Oct 5, 2008 10:23:56 AM
Buffett has no answer because he can't bring himself to admit regulation won't work. His answer would be to stay within your circle of competence through self-regulation.
In the end, the institutions who bought the toxic stuff are the ones getting hurt. Now, they are selling the idea that we'll all be hurt if we won't spare a dime. So, we're compassionate folks, and we are in no mood for a depression today, so we comply.
We can also bluster and make believe that if we spend some millions of dollars we can prevent the next instance if it makes us feel better. Maybe we can put a regulator next to every desk in every financial institution. That's what we could do with all the newly unemployed financiers.
Posted by: Andrew at Oct 5, 2008 10:58:26 AM
So, if these things are by definition too complex to regulate, does the current bailout then essentially encourage complexity and risk taking ?
Posted by: Affe at Oct 5, 2008 11:09:17 AM
I grant that the way regulation has been implemented hasn't worked. That much is obvious.
How about this: regulate, but when it becomes apparent that the entities being regulated are gaming the system*, change the regulation scheme.
* anyone with an ounce of logic (or readers of Calculated Risk, for example) saw the real estate meltdown coming 2 years ago. Goldman Sachs saw that CDOs would be a problem in the long term and so avoided them. Bernanke should have been pushing for new regulations of mortgage terms and investment vehicles in the same way he's been pushing for a $700B blank check. Neither are in his job description per se, but that never stopped Greenspan from pumping up the real estate bubble.
http://www.businessweek.com/bwdaily/dnflash/mar2005/nf20050315_1569_db093.htm
Posted by: meter at Oct 5, 2008 11:11:46 AM
The human body is complex and probably not more understood than the financial markets. Does that mean that doctors should not try to treat patients?
Posted by: sort_of_knowledgable at Oct 5, 2008 11:37:34 AM
George Will just has a good comment. Moments earlier the liberal commentator had blamed the credit crisis on the free market and said Barack Obama had a historic opportunity to fight for more regulation. In a followup segment, the liberal commentator said that the electoral system isn't going to be prepared for the turnout.
George said "Yet again, the government is taken by surprise by their job."
"regulate, but when it becomes apparent that the entities being regulated are gaming the system*, change the regulation scheme.
* anyone with an ounce of logic"
Still waiting for evidence.
Posted by: Andrew at Oct 5, 2008 11:47:20 AM
"Does that mean that doctors should not try to treat patients?"
When they are going to make it worse, yes, that's what it means. In fact, there's an old saying that doctors used to say regarding this phenomena.
Posted by: Andrew at Oct 5, 2008 11:48:51 AM
Capitalist as the bubble inflates: "It's different this time!"
Beaurocrat as the bubble deflates: "It's different next time!"
Posted by: Andrew at Oct 5, 2008 11:52:03 AM
Andrew prefers the "on a wing and a prayer" school of governance.
Does that mean that doctors should not try to treat patients?"
When they are going to make it worse, yes, that's what it means."
Your after-the-fact opinion of 'make it worse' is no more an answer than current regulation schemes.
Posted by: meter at Oct 5, 2008 12:05:37 PM
The NYT has a typical NYT screed today blaming "deregulation" of the ibanks' capital/debt ratios in 2004. Bear levered up to 33 to 1, up from 15 or so, etc.
They neglect to note (as if they knew in the first place) that Easy Al had lowered the Fed-funds rate to 1%, which was well under the natural rate, where it stayed for over a year. This was the cause of the recent events. If the Fed-funds rate had magically stayed in line with the natural rate (which balances the supply and demand for loanable funds, or more broadly investible resources), the banks wouldn't have had the incentive to lever up to the extent they did, even with the 2004 change. Their calculations wouldn't have been thrown off the way they were by the government's manipulation of money and credit.
And there certainly wouldn't have been any contagion to other institutions and markets.
As I said before, would someone do a citizen's arrest of Easy Al? State governments are very jealous of their court n' cop monopoly, so check to make sure your state allows them. People's Republic of New York does not. They are legal in Connecticut.
Posted by: Bill Stepp at Oct 5, 2008 12:40:39 PM
Regulators should be paid for the violations they catch and penalized for the violations they don't catch.
Posted by: Gary Sugar at Oct 5, 2008 1:57:22 PM
The human body is complex and probably not more understood than the financial markets. Does that mean that doctors should not try to treat patients?
Exactly... the human body would not run much more efficiently if every heartbeat, every breath, and the intricate functions of every organ was only allowed to take place after a detailed analysis from a doctor. It would pretty much be a guarantee that the patient would die.
Most doctors don't regulate human bodies, they leave it to the body to regulate itself. Most doctors simply check on a few details of the patient every 6 months (heart rate, blood pressure, a few blood tests, breathing, etc.), and only if they detect a big problem to they take any further action. Even then, the action doesn't tend to be strict control, but rather some small change (i.e. taking a medicine) that helps the body regulate itself.
ow about this: regulate, but when it becomes apparent that the entities being regulated are gaming the system*, change the regulation scheme.
Because regulations are a critical part of gaming the system. Do you imagine that the regulated institutions don't have any political influence and can't effect the bodies responsible for regulating them? Who is going to have greater influence over the political authorities who decide on regulations, the mortgage lenders being regulated, or the typical homeowner with a mortgage?
Posted by: Rex Rhino at Oct 5, 2008 6:41:43 PM
Tyler, the only problem with Buffett's comments on OFHEO is that OFHEO seems to have been the result of a Congressional debate between Jim Leach (R-Iowa) and Barney Frank (D-Mass) in 1992 over whether to have a strong or weak regulator over Fan and Fred. Frank won, of course. The point is that it was precisely the purpose of OFHEO to provide weak oversight, so that the politicization of the mortgage market that Frank and friends aimed to achieve could proceed unabated. Wapo reported on all this not too long ago.
Posted by: Voltaire in '08 at Oct 5, 2008 7:50:40 PM
Tanta at Calculated Risk would
Posted by: eRobin at Oct 5, 2008 8:51:18 PM
I must have goofed up the html. Here's the quote from Tanta:
I think we can give Fannie and Freddie their due share of responsibility for the mess we're in, while acknowledging that they were nowhere near the biggest culprits in the recent credit bubble. They may finance most of the home loans in America, but most of the home loans in America aren't the problem; the problem is that very substantial slice of home loans that went outside the Fannie and Freddie box. But Krugman is right to focus on the fact that it was the regulatory and charter constraints of the GSEs that kept that box closed. In the schizoid reality of the GSEs, when they had their "shareholder-owned private company" hats on they did plenty of envelope-pushing. When they had their "affordable housing" hats on, they rationalized dubious theories of credit quality--like the fervent belief that low or no down payment can be fully offset by a pretty FICO score--to beef up their affordable housing goals, often at the expense not of the poor put-upon "private sector" but of FHA, whose traditional borrower pool they pretty thoroughly cherry-picked. Nonetheless, the immovable objects of the conforming loan limits and the charter limitation of taking only loans with a maximum LTV of 80% unless a well-capitalized mortgage insurer took the first loss position, plus all their other regulatory strictures, managed fairly well against the irresistible force of "innovation." If there has ever been an argument for serious regulation of the mortgage markets, the GSEs are it.
at this url: http://calculatedrisk.blogspot.com/2008/07/krugman-on-gses.html
Posted by: eRobin at Oct 5, 2008 8:53:09 PM
You're always going to have guys trying to run Ponzi schemes. With no regulation, a lot of them will get away with it. With regulation, sometimes they'll still get away with it.
The accounting scandals at Arthur Andersen etc have shown that private regulation doesn't work reliably either. Companies whose very existence depend on public belief in their honesty still faked data for their paying customers.
http://en.wikipedia.org/wiki/Accounting_scandals
It appears that fraud has tremendous economy of scale. Other things equal, the bigger the corporation the more complex the accounting and the easier to hide stuff.
In general, we are better off with short feedback loops and simple transactions. How to arrange that is another story.
Posted by: J Thomas at Oct 6, 2008 12:47:02 AM
"Your after-the-fact opinion of 'make it worse' is no more an answer than current regulation schemes."
Huh? What's the question again? So, regulation is a brand new idea. This is the first time someone has said "Eureka! They shall be regulated!" After the fact regulation is exactly like treating the last disease.
"Andrew prefers the "on a wing and a prayer" school of governance."
Nope. I just think all the people claiming we have an unbridled banking and mortgage orgy and all this would have been prevented had the benevolent regulators been in charge are clowns. They show up after the main show and stumble out of their clown car honging their horns and pulling their noses.
Most of what Buffett says about governance in general I would like to see happen. I'd like to see stronger shareholder activism. Maybe that will come if enough of them get wiped out by the government. Which brings me to another silly statement the statists keep parroting, that the people who caused this are getting away unscathed.
In the end, regulation has to come from inside the system. Come on. Do people really believe that some government desk jockey is going to see these things coming before the people who got creamed by them do? Seriously? A government regulator does the same thing as a regulator on an engine. It just slows it down, it can't make right choices. So, that's your choice. If you want to slow things down, fine for you, but you don't get to choose the magical best of all worlds.
Posted by: Andrew at Oct 6, 2008 6:28:41 AM
The human body is complex and probably not more understood than the financial markets. Does that mean that doctors should not try to treat patients?
Well firstly we can do randomised double-blind tests on thousands of patients. So we can test a medical intervention in ways that we can't test financial market regulation. To be more specific, there are only a few hundred countries, and some of them don't have financial markets, it's basically impossible to get many of them to agree to experiments, and double-blind experiments on financial market regulation are impossible.
Secondly, doctors are expected to avoid doing harm, which includes not doing treatments if they have no good reason to think them effective (and aren't testing if they are effective).
Posted by: Tracy W at Oct 6, 2008 6:28:44 AM
How about this: regulate, but when it becomes apparent that the entities being regulated are gaming the system*, change the regulation scheme.
But 'gaming the system' involves amassing political power and connections sufficient to prevent government from 'changing the regulation scheme'.
Many people did know that Freddie and Fannie were in trouble, and there were serious efforts to 'change the regulation scheme', but Freddie and Fannie's political allies prevented it.
Posted by: Slocum at Oct 6, 2008 7:35:03 AM
It should be obvious that self-regulation fails pretty consistently.
And it should be obvious that government regulation fails sometimes too.
The central problem is that some of the guys who design ponzi schemes are real smart, and to catch them you have to be even smarter and also honest.
This time there's the extra problem that the US government may have been central to the scam. We needed lots of foreign money and we had no better way to get it. So we maybe-on-purpose encouraged these financial transactions that were real complicated and nobody particularly understood. In recent years they've been one of our major exports. Now that they've blown up we can say "Well, nobody could have forseen this, heh heh, let's just pick up the pieces and go on, anybody want some T-bonds?".
I'm not saying definitely that the US government participated in the scam, but if we had stopped it in 2002 what would we have done for foreign exchange? Unless we found something else we would have gone into a deep recession then, deeper than the one we were just coming out of.
I think self-regulation works best when the details of how things happen haven't changed for at least five human generations. People do things the way their fathers did, and the way they do them works. Introduce something new and they aren't interested. When you have that set up you have something that works, and people know it works, and they don't want to fiddle with it. But change, including technological change, makes that fail.
And government regulation works best when you have smart dedicated people doing it, who are insulated from politics. That fails when the dedication goes.
Posted by: J Thomas at Oct 6, 2008 7:50:42 AM
"The human body is complex and probably not more understood than the financial markets. Does that mean that doctors should not try to treat patients?"
The difference in complexity is stability.
In the finacial markets intelligent beings are actively trying to circumvent your regulations. The only corrolary to medicine would be a rapidly mutating disease that changes form with each treatment and moves from organ to organ.
That is what makes regulating finacial markets so hard.
Posted by: eccdogg at Oct 6, 2008 10:20:16 AM
