This week's "Economics Focus" in the Economist joins the pack piling on to Alan Greenspan for not deflating America's stock market bubble earlier:
Economist.com: ...There may be no painless way to deflate bubbles. Yet the correct test is not whether a bubble can be deflated without some loss of output. Rather, it is whether the early pricking of a bubble causes less pain than letting it grow only to burst later. The longer a bubble is allowed to inflate, the more it encourages the build-up of other imbalances, such as too much borrowing and investment, which have the power to turn a mild downturn into something nastier. If the Fed had let some air out of the bubble earlier, America's economy might now be better placed for future growth...
Admittedly, for the Fed to justify an increase in interest rates when inflation was low would have been hard—but not impossible. It could, for instance, have argued that raising rates and so containing financial imbalances would avoid future economic instability and hence a large undershoot in future inflation.
Central bankers do not have a political mandate to respond to asset prices. Even so, Mr Greenspan could still have done more to warn investors about their irrational exuberance (which he talked of as long ago as December 1996). At least he could have refrained from talking up share prices, if unintentionally, through his enthusiasm for the new economy. Mr Greenspan repeatedly expressed confidence that America's productivity growth had risen significantly, encouraging investors to form unrealistic profit expectations. Estimates of productivity growth have since been cut. Goldman Sachs has shaved its estimate of trend productivity growth to 2%. Two years ago, most economists had their sights on 3% or more...
The first thing that must be said in Greenspan's defense is something that he knows well, but that the Economist has forgotten. Raising interest rates to prick an asset market bubble--reducing demand, causing unemployment, and throwing people out of work when there is no reason to think that the current state of the labor market is inconsistent with price stability--is not a policy to be entered into unadvisedly or lightly; but only reverently, discreetly, advisedly, soberly, and in the fear of God. "Sufficient unto the day is the evil thereof"--in short, deal with the bubble when the serious problems it threatens to cause are visible and approaching, not simply because there might be serious problems in the future.
The second thing to say is that Greenspan did warn of the dangers of "irrational exuberance," received considerable political flack when he did so, and found that such warnings had little effect. Taken in total, without selecting sentences on one side or the other, Greenspan's beliefs that on the one hand the market might be suffering from irrational exuberance but on the other hand look at these remarkable productivity growth numbers still seem well-founded.
The third thing to say is that Alan Greenspan's judgment is not perfect, but it is a damned sight better than mine. I can think of a number of times that I thought that Greenspan had made a significant monetary policy mistake:
By my count, the first five times Greenspan was right: I was wrong. The sixth time I was right. And, although the jury is still out on the seventh, I think I am right. Nevertheless, the score is still Greenspan 5, DeLong 2. That's an impressive record, which we should not lose sight of. He knows things about how to make good monetary policy that I clearly do not.
EVERY August central bankers and economists gather in the Rocky Mountain resort of Jackson Hole, Wyoming, for the annual symposium of the Federal Reserve Bank of Kansas City. This year Alan Greenspan, chairman of the Federal Reserve Board, used the opportunity to give his fullest defence yet against charges that he should have raised interest rates in the late 1990s enough to prick the stockmarket bubble before it got too big. The Fed, like other central banks, takes account of rising asset prices (shares or property) to the extent that they boost spending and hence future inflation. Yet a financial bubble can inflate even when inflation in goods and services remains low. And when a bubble bursts, it may cause severe balance-sheet strains—of the kind now showing in America.
Mr Greenspan offers two defences for failing to respond to the bubble. First, he argues that it was impossible to be certain that the rise in share prices in the late 1990s really was a bubble until after it had burst. Second, even if a central bank can detect a bubble, it is not clear what it can do. A small rise in interest rates might not work; by increasing confidence in the central bank's powers, it might even boost share prices further. On the other hand, a sharp increase in rates could trigger a recession—the very outcome central bankers would be seeking to avoid, says Mr Greenspan.
Detecting and pricking bubbles are both difficult, but that is not a justification for doing nothing. Monetary policy always deals with uncertainty. Judging whether a rise in share prices is justified by an increase in productivity growth is surely not that different from deciding whether the potential rate of growth has increased or decreased. Central banks have to do that to estimate the gap between actual and potential output—itself an important input for forecasting inflation. A central bank does not need to be completely certain to act. Unrealistic profit expectations built into share prices in the late 1990s pointed to the strong probability of a bubble.
Supporters of Mr Greenspan argue that central bankers are unlikely to have more information or to make better judgments about share prices than markets do. Yet central bankers have longer time horizons and different incentives from the private sector. In other words, in many circumstances they may respond differently to the same information.
What of Mr Greenspan's second claim, that a small rise in interest rates might prove counterproductive? He cites three years—1989, 1994 and 1999—when share prices continued to rise even as the Federal Reserve raised rates. All the same, the impact of higher rates might be different if the Fed were actually to state that its aim was to cool the stockmarket.
There may be no painless way to deflate bubbles. Yet the correct test is not whether a bubble can be deflated without some loss of output. Rather, it is whether the early pricking of a bubble causes less pain than letting it grow only to burst later. The longer a bubble is allowed to inflate, the more it encourages the build-up of other imbalances, such as too much borrowing and investment, which have the power to turn a mild downturn into something nastier. If the Fed had let some air out of the bubble earlier, America's economy might now be better placed for future growth.
A recent paper* by Claudio Borio and Philip Lowe at the Bank for International Settlements addresses the problem of identifying bubbles. The authors argue that the focus on asset-price bubbles alone is wrong. It is only when a boom in share prices or house prices is combined with a big increase in debt and overinvestment by firms that economic and financial stability is threatened. From a study of 34 countries since 1960, Mr Borio and Mr Lowe conclude that a simultaneous surge in both credit and asset prices gives a pretty reliable warning of financial problems ahead. The case for a rise in interest rates is therefore stronger when asset-price rises go hand-in-hand with rapid growth in credit—as in America in the late 1990s.
Admittedly, for the Fed to justify an increase in interest rates when inflation was low would have been hard—but not impossible. It could, for instance, have argued that raising rates and so containing financial imbalances would avoid future economic instability and hence a large undershoot in future inflation.
Central bankers do not have a political mandate to respond to asset prices. Even so, Mr Greenspan could still have done more to warn investors about their irrational exuberance (which he talked of as long ago as December 1996). At least he could have refrained from talking up share prices, if unintentionally, through his enthusiasm for the new economy. Mr Greenspan repeatedly expressed confidence that America's productivity growth had risen significantly, encouraging investors to form unrealistic profit expectations. Estimates of productivity growth have since been cut. Goldman Sachs has shaved its estimate of trend productivity growth to 2%. Two years ago, most economists had their sights on 3% or more.
One positive sign: in his speech Mr Greenspan did at least accept that the Fed should try to identify bubbles and to incorporate them into economic models. But for the moment he doubts that central banks can do anything about them.
If the American economy recovers fast, then Mr Greenspan's policies will be vindicated—if bubbles left to burst of their own accord result only in a mild recession, central banks do not need to prick them. If America suffers several years of slow growth as its financial imbalances unwind, however, the verdict must be that the Fed got it wrong.
Professor DeLong, to your credit, you should add that He chairs the biggest employer of economists in the US... He has means no other economist - be they even from Goldman-Sachs and company - do not have. His is an expensive christal ball... We are right to demand results.
In my darkest hours, I sometime ponder about what political incentives an (ex?-)Randist like chairman Greenspan, is subject to, even perhaps unconsciously. In anycase, I hope he is subject to a lot of influences in a way.
So, my question is: what was he thinking re:the bubble as presidential elections were looming ever closer... Personally, I ended up thinking that a man like this must surely want to leave the best possible record in history.
Here is my hunch: his incentive was to try to shoot for a partial punctiuring of the bubble. Some sort of Salomon's justice. He would take away some of the shine from Clintonian macro-policy - more or less proportionally to his take on how much credit he thinks the Clinton administration deserved for the productivity boom of the late 90s - but hand the hot potatoe to the successor.
And if it were a Democrat, well fully deserved, and it were an Republican, well, they typically kind of like to begin the polical-business cycle with a recession anyway...
Does anybody has more than conjectures about this?
Posted by: Jean-Philippe Stijns on September 6, 2002 05:46 PMGreenspan's biggest mistake by far was counting on a federal surplus so large it would become a problem, so Bush's tax cuts were in order.
Since Greenspan told the Fed Board in November 1996 that a rise in margin requirements would limit a market move to a bubble, why not have raised margin requirements in late 1997 or early 1998 or the middle of 1999? When was the danger of continued market speculation apparent enough?
Posted by: on September 7, 2002 11:25 AMLooks like a lot of people are joining Gigot and the WSJ editorial page out there in the Gamma Quadrant -- The Economist, Krugman, Akerlof, etc. And the WSJ only sent in a question, the rest are shooting in torpedoes.
Still, it all looks like low-grade second guessing to me, and I'm going to stay on Alan's side until someone says...
"Bubbles should be controlled and here's my spelled-out formula for doing so. It shows...
"[] How to avoid deflating six out of three bubbles.
"[] How to know if a bubble in one sector of the economy will affect other sectors, or won't affect them.
"[] The expected magnitude of the effect of any bubble on the rest of the economy -- so we know with *foresight* how many millions of people we should unemploy currently to come out ahead in the future, measuring the trade-off prospectively.
"[] The means for determining the timing and scale of our bubble-deflating actions.
"Thus, I do not say 'Greenspan shoulda done something' but show by analysis that interest rates should have been raised by X% in the X quarter of 199X, to reduce employment by X million and GDP by X% cumulatively over the following X years, which would leave us with net benefit of X today."
But I don't see anything like that. Where's all the rigorous analysis economists are supposed to be famous for? There's nothing easier and less rigorous than saying somebody else should have done something different, without putting yourself on the line by saying what different, and what its cost and net result actually would have been. Now that the NFL season's started, we can get plenty of second guessing of the kind I'm seeing every Monday.
It's not going to be a happy job for the central banker of the future who is called upon to say, "I know there's no inflation, but I'm going to unemploy a million people today to drive down the stock market and housing values anyhow, because of what might happen in the future". Even if he's right and actually keeps a bubble from bursting he'll never be able to prove the negative to show he was right. And there will always be plenty of people who say he was wrong who won't be able to be disproved either.
Jim: We're not completely clueless when it comes to assessing bubles. Price : earnings ratio are one rough indicator, and I bet one can be a lot more sophisticated than this. If an asset is not the discounted sum of its stream of dividants, it's a bubble, it's that simple.
The difficulty is in forecasting this stream of revenue from the little information we have at time t. I think from now on, our basic assumption should be that no technological revolution can justify any "new economy" (at least in a financial sense) denomination anymore than it can lift the law of gravity. My guess is that we'll get caught again, though.
In macroeconomic policy everything is difficult to predict, bubbles and other things. "Monetary policy is like driving by looking in your side mirors." But that's no excuse for inaction, that's a rationale for prudence. Recall that Greenspan knew there was "irrational exuberance". These are his words after all.
I like the anonymous point about letting surpluses swell too much, though. Personally, I don't think big surpluses, everything else being equal, are a problem (but I recall Prof. DeLong thinks different from me on this one. I am nostaligic about his old-fashioned mailing-list ;-)
But I can see, how large surpluses can create in the eye of the public and of some politians the illusion (and greed) that we can dispense with these surpluses and need to give tax breaks. Had the surpluses been more modest, they may have attracted less attention...
It's not clear to me, what Greenspan's role was in setting the size of the surplus though. Seems to me that it was %50 the result of political gridlock in DC and %50 due to a combination of good economic policy advisors talking to a Clinton who had decided to change his strategy. Of course, hadn't monetary policy been lax, surpluses wouldn't have been politically feasible. I don't know which caused which.
Posted by: Jean-Philippe Stijns on September 7, 2002 01:59 PM>> Jim: We're not completely clueless when it comes to assessing bubles. <<
Being "not completely clueless" is a long way from being suffiently informed about the results of alternative policy options. If there are basic concepts to economics, one is that you have to compare alternatives in some sort of quantifiable manner. Deciding, "I know this is bad, therefore something else has to be better" is a very old and frequently very costly logical fallacy.
For instance, many apparent bubbles never create effects that move out of their own sector the economy into the economy at large. If you impose a price on the general economy to deflate them, you lose. But if you are sufficiently clued in to tell the difference with surety, and to clarify the other items I posted, I will glady defer to you on policy.
>> If an asset is not the discounted sum of its stream of dividants, it's a bubble, it's that simple. <<
If that's your definition of a bubble you'll be deflating 10 out of 2 of them for sure.
>> Since Greenspan told the Fed Board in November 1996 that a rise in margin requirements would limit a market move to a bubble, why not have raised margin requirements [later]? <<
Greenspan's comments, when not quoted misleadingly in out-of-context snippets as they have been in so many news reports and op-eds, make it clear that:
(1) He did *not* believe raising margin requirements per se would deflate the bubble (as only about 1% of stock is owned on margin and there are many easy substitutes for margin financing), but that doing so would act as a signal that seriously higher interest rates were coming, and concern over the prospect of *them* affecting the real economy would deflate the bubble. To quote him:
"If a change in margin requirements were taken by investors as a signal that the central bank would soon tighten monetary policy enough to burst a bubble, then there might be the appearance of a causal effect. But it is the prospect of monetary policy action, not the margin increase, that should be viewed as the trigger."
(2) He was very concerned about the damage to the real economy that would result from such a raise in rates, and worried about having no way to measure the cost of that real damage against whatever benefits might result from bringing down the market.
E.g., What Greenspan actually said in 1996 was:
"What is really needed to keep stock market bubbles from occurring is a lot of product price inflation, which historically has tended to undercut stock markets almost everywhere. There is a clear tradeoff.
"Now, unless we have the capability of playing in between and managing to know exactly when to push a little here and to pull a little there, it is not obvious to me that there is a simple set of monetary policy solutions that will deflate the bubble.
"We have the possibility of raising major concerns by increasing margin requirements. I guarantee that if you want to get rid of the bubble, whatever it is, that will do it. My concern is that I am not sure what else it will do." [Note that last sentence.]
Unfortunately, we are getting very low quality- to-deceptive journalism on this, enabling the journalists to have the fun of second-guessing, bashing, and looking superior after the fact.
For instance, a NY Times Op-ed this past week stated that Greenspan not only knew of the bubble in 1996 but also "had a solution for it" then.
In support of this contention, the writer quotes Greenspan's own words: "We do have the possibility of ... increasing margin requirements. I guarantee that if you want to get rid of the bubble, whatever it is, that will do it."
The writer then shakes his head over how Greenspan inexplicably failed to do anything -- but feels free to surmise it was due to lack of character on Greenspan's part, "he made investors furious, and lost his nerve."
But look at the careful selection and editing of Greenspan's 1996 words that have gone into this presentation:
(1) All the preceding words, especially, "it is not obvious to me that there is a simple set of monetary policy solutions that will deflate the bubble" are cut. Of course they are. They flatly contradict the writer's claim that Greenspan's *did* have a simple solution to the bubble in 1996.
(2) Ellipses [...] are used to delete the words "of raising major concerns" in the quoted sentence. This completely changes its meaning from Greenspan's 'major concerns caused by the signal of a margin increase will deflate the bubble' to the writers' 'a margin increase, a simple thing, will deflate the bubble'.
(3) And, significantly, note the deletion of the following sentence: "My concern is that I am not sure what else it will do".
This editing, ellipses and all, is too careful to be accidental, and completely reverses the picture of what was in Greenspan's head in 1996 -- from Greenspan's not having a simple answer to the bubble and being worried about harming the real economy, to Greenspan's having a simple answer and just not having the personal character to apply it when investors got mad at him.
I mean, is that Orwellian or what?
I do hope this writer doesn't spend a lot of time going around accusing others of being deceptive.
Why does the discussion only touch upon interest rate targeting? The Fed could also follow
a quantity rule. If the Fed is operating on a quantity rule, trying to manage the supply of money in the system, it has to buy bonds from the banks when the "M" target is higher than the recorded amount. And it has to sell bonds when the "M" target is being exceeded. Liquidity is drained.
Of course, there are other possible indicators
to follow - the dollar price of gold, e.g.
Why is the supply of liquidity - via the selling and buying of bonds - not given any consideration
in the discussion of bubble prevention?
Joerg Wenck
Posted by: Joerg Wenck on September 8, 2002 09:45 AMJim Glass
Nice argument. Then, I find no need to criticize the Fed for failing to limit the stock market rise. The Fed really could not have been sure there was a bubble, for so few economists were. Raising margin requirements, might well have proven a dangerous precedent for future market interference. Besides, was margin the problem?
My only quibble, was support of the decade long tax cut by Greenspan.
Posted by: on September 8, 2002 10:25 AM'as only about 1% of stock is owned on margin and there are many easy substitutes for margin financing'
I've seen scuttlebutt in other blogs about how a few heavily connected players were taking enormous margin positions. Anyone know anything about this?
Posted by: Jason McCullough on September 8, 2002 12:51 PMI believe you unfairly critiqued the economist article.
The Economist consistently tok the view that rates should be raised to prick the bubble believing the assymetries caused by an inflated stock market and the inevitable bursting of the bubble would result in greater pain in the future then the pain caused by pricking the bubble sooner.
The article does credit Greenspan for the "irrational exuberance" comment. It then goes on to point out from 1995 to 2002 Greenspan opinioned morphed into a strong belief that productivity improvements justified market valuations. Bychanging his opinion Greenspan, in his position as expert, led credence to the investors belief the rally would continue.
I would never try to keep score against the Fed. They get their info way before I getit.
Otherwise I really like this site. Keep it up.