March 17, 2008
We will never have a perfect model of risk
By Alan Greenspan
The current financial crisis in the US is likely to be judged in retrospect as the most wrenching since the end of the second world war. It will end eventually when home prices stabilise and with them the value of equity in homes supporting troubled mortgage securities.
Home price stabilisation will restore much-needed clarity to the marketplace because losses will be realised rather than prospective. The major source of contagion will be removed. Financial institutions will then recapitalise or go out of business. Trust in the solvency of remaining counterparties will be gradually restored and issuance of loans and securities will slowly return to normal. Although inventories of vacant single-family homes - those belonging to builders and investors - have recently peaked, until liquidation of these inventories proceeds in earnest, the level at which home prices will stabilise remains problematic.
The remainder of this column can be read here. Comment from our expert panel and guest members can be read below.
Brad DeLong: 130 years ago we had unregulated organizations whose use of leverage created systemic risk. So we regulated them - to curb the amount of leverage and thus the risks they ran, and to give the lender of last resort the information necessary to make good decisions.
Today we have unregulated organizations, et cetera. And we need to regulate them: if the Fed is going to issue puts on Bear Stearns’s assets, it needs to know what they are and to be able to put bounds on the bets that the Bears of the world make.
Meanwhile, Alan Blinder - whose judgment I trust - says that best option right now is the legislative vehicle that Barney Frank and Chris Dodd have to guarantee mortgages through a greatly expanded FHA.
All public-spirited economists who want to play a positive role in public policy should get behind Frank and Dodd, and quibble about the details and the implementation later.
Posted by: Brad DeLong | March 17th, 2008 at 2:25 pm | Report this commentAlice Rivlin: Greenspan is right, of course, that we will never have a perfect model of risk in a complex economy. But the culprit was not imperfect models. It was failure to ask common sense questions:
Q: Will houses prices keep going up forever?
A: Not likely.
Q: What will happen to the value of mortgage-backed securities when housing prices stop rising or fall?
A: They will go down.
We didn’t need fancy models to answer questions like that. We just needed to ask them.
Posted by: Alice M Rivlin | March 17th, 2008 at 2:32 pm | Report this commentDavid Heigham (guest): When working as a British Government economist, I answered both Alice Rivlin’s questions correctly before the British housing price crash at the end of the 1980s. The problem was that the macro-economic models guided policy, and there was no way of incorporating the answers consistently into the models. Plus ca change…
Nowadays there is a macro-economic model which does reliably incorporate roughly right answers in advance, even though it is vague on when its predictions will come to pass: Wynne Godley’s at www.levy.org . Mainsteam economists (myself included) are chary of this model because some of its formulation does not square with the way we think. Non-economists may care to ignore our doubts and get useful guidance.
Posted by: David Heigham | March 17th, 2008 at 3:56 pm | Report this commentMary s Schranz (guest): The former chairman of the US Federal Reserve is correct in noting that we can learn much from the current crisis as well as from past crises. However, his commentary ignores the reality that the US Federal Reserve deserves much of the blame for the current crisis.
By maintaining a low real interest rate environment between 2003 and 2007, the central bank encouraged reckless borrowing and lending. Poor regulatory oversight allowed financial institutions to lend to borrowers who were not creditworthy at very low interest rates. The traditional relationship between creditworthiness, risk and the rate charged on loans was broken. Now that the lenders are reaping the losses that any prudent banker could have predicted, the former chairman is blaming the complexity of modern securities and elaborate financial models for this plight. The collapse in the credit markets is no surprise to students of bank management and credit market bubbles. It is the result of misaligned incentives and poor management.
It is astonishing that the former chairman would write “Those of us who look to the self-interest of lending institutions to protect sharholder equity have to be in a state of shocked disbelief.” Those of us who have studied the incentives of managers of financial institutions are not shocked. The incentive structure in the financial services industry rewards risk-taking with managers and employees bearing little or no cost from their actions. The bailouts engineered by the US Federal Reserve in the past 6 months only reinforce this poor decision-making. Those who blow bubbles should not be surprised when the bubble bursts. The tragedy is that residents of the US face higher inflation in the future, a weakened currency, and no regulatory action that addresses the true source of the problem.
Posted by: Mary S. Schranz | March 17th, 2008 at 7:40 pm | Report this commentPaul De Grauwe: So the reason why we have this financial crisis is that we can’t forecast the future. What a deep insight. We have never been able to forecast the future, but this did not necessarily lead to a financial crisis.
Greenspan’s article is a smokescreen to hide his own responsibility in making the financial crisis possible. There are at least two reasons why Greenspan’s Federal Reserve bears a heavy responsibility for the present crisis.
First, as has been argued by John Taylor, the Fed kept the interest rate at very low levels for too long after the recession of 2001. The Fed waited until the middle of 2004 to start raising the interest rate from its historically low level of 1%. The point is not that in 2001 the Fed reduced the interest rate too much when it cut it from more than 6% to less than 2% in less than a year. This was probably the right thing to do in a recession. The problem is that it kept the rate there for too long, when the economy showed signs of recovery. This laid the groundwork for a massive credit and liquidity expansion which in turn created an asset bubble in the housing market.
Second, the Fed failed to take up its mandated responsibility to supervise and to regulate the financial institutions. Why did this happen? The root cause is the religious belief of Greenspan in the benevolence of markets and perniciousness of government interventions. At the moment financiers were increasing their exposure to liquidity risk and made fantastic profits doing so, in the knowledge that the Fed was insuring them freely against such a risk, Greenspan stood by and marveled at the creativity of markets. In his wonderful book “The Age of Turbulence” he discusses the new financial instruments and he concludes with a beautiful metaphor: “Why do we wish to inhibit the pollinating bees of Wall Street” (p 372). The problem is that the financiers of Wall Street were mostly pollinating themselves.
The corollary of this religious optimism in the wondrous workings of the financial markets is the belief that governments and the Fed can do nothing to regulate these new financial instruments. On p 489 of the same book Greenspan writes: “Only belatedly did I, and I suspect many of my colleagues, come to realize that the power to regulate administratively was fading; We increasingly judged that we would have to rely on counterparty surveillance to do the heavy lifting”. So he did nothing and decided that self regulation is the answer. It can now be seen by everybody except the blind, that financial self-regulation does not work and has never worked. Yet in his FT column Greenspan continues to hope that financial self-regulation will be the “fundamental balance mechanism for global finance”.
Religion quite often stands in the way of rational analysis. Greenspan, who was at the helm of the most important monetary institution in the world, failed to take his responsibility to supervise the financial markets blinded as he, and his colleagues, were by a belief that markets and bankers know better than governments.
Posted by: Paul De Grauwe | March 19th, 2008 at 11:53 am | Report this commentMartin Wolf: I have to say that I find myself in agreement with the critics of Mr Greenspan’s article, the ideology he displays and some of the policy decisions made at the Fed in the 2000s.
I have laid out some of my thoughts on the lessons of the crisis in my contribution to a superb conference held at the Banque de France. I recommend a careful reading of the proceedings, particularly papers by Kenneth Rogoff, John Taylor and Bill White. I also strongly recommend a short presentation by Helene Rey. The web site is here.
I would like to add nine points.
First, Mr Greenspan ignores the now overwhelming evidence of malfeasance and gross incompetence in the chain of agents, from mortgage origination to the ultimate holders, including rating agencies, banks, investment banks, and so forth. This is not just about poor risk management. It is far worse than that. This was a huge failure of regulation.
Second, we do have to look very carefully at the incentives at work inside the financial system - a subject Mr Greenspan ignores. It is not just about ignorance. It is about wilful ignorance.
Third, it is now clear that the regulatory net has to be both more effective and broader. If Bear Sterns is to be rescued, then the liquidity position of all investment banks becomes a regulatory concern of the highest order. I wonder now how long it will be possible to ignore hedge funds.
Fourth, the reason the regulatory net has to become broader is that, in the end, it is the state that is expected to save the day. Social insurance has its price. The financial industry cannot be allowed to privatise profits and socialise losses on any scale it wishes.
Fifth, the case for treating huge asset price surges as prima facie indicators of excessively loose monetary policy is also overwhelming. Central banks cannot eliminate the bubbles, but they can surely lean against the wind. Cleaning up, afterwards, is far more difficult than some had previously thought, particularly where lending against real estate is involved.
Sixth, we must also ask whether the “risk-management” approach to monetary policy, followed by the Fed under Mr Greenspan and his successor, is not dangerously asymmetrical, since it seems to mean far more energetic responses to downside risks than to upside ones.
Seventh, far too little attention is still being paid to the role of the global imbalances (excess savings outside the US) in shaping the environment for US monetary policy. I believe this partily explains why US monetary policy had to be so loose. (I have a short book on this coming out in the summer.)
Eighth, maybe there must be a US government rescue of mortgagees, as Brad de Long argues. But I would admamantly oppose such an idea for the UK. It is wrong to bail out people who decided to borrow far more money than they could afford in the belief that the relative price of houses can rise without limit (see Alice Rivlin above, on this). I do understand the need to keep the core financial system in being and sustain demand in the economy (though recessions may be unavoidable, if not necessary, from time to time). But public guarantees of mortgage losses are very dangerous.
Finally, nobody should head a central bank for more than 8 years, or so. Mr Greenspan is a brilliant and fascinating man. But the dominance he achieved at the Fed was unhealthy. Nobody is perfect. What is needed is a debate among divergent viewpoints. The position Mr Greenspan had achieved by the late 1990s made this very difficult, if not impossible, to achieve.
Posted by: Martin Wolf | March 19th, 2008 at 2:48 pm | Report this commentThe following letters responding to Alan Greenspan’s article were received by the Financial Times:
We should reject this extremism of Greenspan’s
From Jérôme Guillet, Editor, European Tribune
“It was oddly fitting that Alan Greenspan should argue in favour of continued self-regulation of the financial sector on the very day that self-regulation demonstrated its absolute failure.”
Read the full letter
The problem with the models is their opacity
From Paul Munton
“Alan Greenspan surely misses the point when he says that ‘both risk models and econometric models, as complex as they have become, are still too simple . . .” The problem is not complexity, it is opacity.”
Read the full letter
Those elusive ‘animal spirits’
From Leander Schneider, Concordia University
“With regard to forecasting, do not blame the model, blame the modeller. More to the point: with regard to system behaviour, blame the regulator.”
Read the full letter
No remorse for letting the party get out of hand?
From Edward Gottesman
“The central banker’s job description is to “take away the punchbowl” before the party gets out of hand. Is a small sign of remorse too much to ask?”
Read the full letter
Crisis in the making was there for anyone to see
From Jean Barnard
“When home prices return to 2000 prices the market will stabilise and the economy will heal itself. No amount of artificial “stimulus packages” or manipulation will override that fact.”
Read the full letter
No apology?
From Richard Guthrie.
“I’ve read Alan Greenspan’s article over and over, but still can’t find any reference to an apology for the fact that the financial apocalypse he describes is in the most part down to the actions/inactions of the US Federal Reserve under his watch.”
Models can never replace basic, prudent banking
From Tim Keese
“It would be nice if Mr Greenspan would admit his mistake and stop talking about models that didn’t work. It was the regulators who didn’t work.”
Read the full letter
Silence speaks louder than words
Posted by: Damian Carrington | March 20th, 2008 at 2:55 pm | Report this commentFrom Marcus Miller, University of Warwick
“On the issue of “moral hazard” in banking, Mr Greenspan is silent. Sometimes, silence speaks louder than words.”
Read the full letter
Robert McDowell: As a banker who specialises in risk and as an economist, I can say that we don’t need, nor should we want, perfect models of risk. But,the lack of models and good analysis has not been a principal cause of the credit crunch; a bigger problem is the deafness with which bankers listen to economists during periods of market euphoria (Greenspan’s term). We do need economists to be more involved in building global models that incorporate the role of banking and finance, and more economics awareness among our top bankers. But, banks have genuine problems of considerable complexity to relate their own experience to the underlying economy, so-called.Central banks and economic research groups and departments of universities have not stood idly by, but they do presume that they are read and listened to and understood by bankers, and that bankers have not lost sight of classical banking precepts concerning balancing the funding qualities of assets and liabilities alongside the ever-present concerns that should attend matters of morality, trust, reputation and confidence. There is no doubt that setting risk limits constrains asset growth, and that much of banks’ four times higher profitabilty per person employed compared to all other business sectors taken together has been gained by gaming long established risk precepts. In large part this has I suspect been a consequence of too many banking officers being persuaded to take early retirement in the past decade and a half, thereby leaving the field of play to a less corporatively responsible generation of traders and financiers.
Posted by: Robert McDowell, Edinburgh | April 8th, 2008 at 1:42 pm | Report this commentMost economists agree that price controls never work, so why do so many implicitly accept that controlling the price of money is acceptable? If we accept Greenspan’s missing “explanatory variable” of human responses to euphoria and fear, it is difficult to conceive of any model that will work with a fiat currency. In ‘another life’, Greenspan appeared to understand the problem when he said in 1966: “In the absence of the gold standard, there is no way to protect savings from confiscation through inflation.”
Posted by: Peter Bayley, Oregon State University | April 11th, 2008 at 7:59 am | Report this commentMartin Wolf: The obvious response to Mr Bayley is that the government has to control the price of money, since it can create it without limit. His answer is: go back to the gold standard. I wish him luck with that argument. We know that faced with the choice between a depression and going off gold, governments will always choose the latter. The argument that we should go back onto gold is as futile, therefore, as the argument that divorce should be abolished. There is little point, other than a purely academic one, in making arguments for something that is never going to happen. The interesting point, after all, is that any country in the world could now go onto gold. Name one that has.
Incidentally, there were plenty of financial crises under gold, too.
Posted by: Martin Wolf | April 11th, 2008 at 10:23 am | Report this comment