Free Exchange, Round two
It’s always an honor to be noticed by The Economist. Unlike Eugene Fama, I don’t plan to cancel my subscription to the best magazine in the world, even if they say this about me:
Mr Sumner disagrees. He seems to think it’s funny that The Economists pent much of the last decade warning that, globally, home prices were rising in a troubling manner. Contrarianism is fun and all, but this strikes me as an odd way to process the experiences that led us to this point.
A few comments:
1. I would note that Free Exchange seemed to enjoy making fun of Fama’s views.
2. I do think it funny that Free Exchange would claim that The Economist was exactly right in its housing bubble predictions. Then demonstrate this by linking to an excerpt from The Economist that was 180 degrees off. That predicted American housing prices would fall 10% over the 4 year period after May 2003, when they actually rose about 30%.
3. I do admit to forgetting that The Economist is real people, as they don’t tend to list names of reporters. So perhaps that made me more sarcastic than I should be.
4. I use humor to wake people up, to try to make points more accessible.
5. The crisis of the past year has consumed all my time. I’d much rather be reading books or seeing movies, and my school doesn’t compensate me for this. I am deadly serious about trying to wake people up to the incredible harm done by tight money in 2008. Unfortunately I seem to recall The Economist advocating tighter money, on the grounds it was needed to squeeze the excesses out of the US economy. It looks like they got their way, and it looks like that monetary policy made the housing crash prediction look even more impressive than otherwise.
I don’t want to be dogmatic about this. I’ve never denied that the EMH is merely an approximation of reality. Or that astute observers do sometimes have insights that the markets miss. But that doesn’t make Fama wrong. Social science is not about the exceptions, it is about broad principles. I presume that “the market” does gain some value from the information in The Economist. But here is the problem; that fact doesn’t really give the anti-EMH position any real traction. Useful insights in The Economist are just as available to “the market” as they are to “the regulators” or to me. If I invest based on advice from The Economist, I won’t consistently beat index funds. And if regulators try to predict bubbles based on predictions from The Economist, they’ll do no better than they would by throwing darts at a stock page.
Here is one accurate prediction I did make. In my previous post I showed that the 2003 Economist prediction that everyone seems to think was amazingly accurate was actually almost completely wrong. I said I expected comments that somehow it was actually correct. They are already coming in. People are saying something like “yes, they explicitly predicted a 10% fall over 4 years. And yes, prices actually rose 30%, but they rose 35% in three years and then fell 5% in the 4th. So they were right.” Well not in those exact words, but that seems to be the general view out there. Another argument is that they were wrong in the 4 year window, but then spectacularly right in the 5th.
Yes, but if we don’t have standards, if we aren’t going to hold people to their words, then what do we really have? Suppose I said; “I predict a major 20% to 30% drop in the S&P500 within the next 4 years. And if it doesn’t happen, but happens sometime later, I should still get credit.” Would you take me seriously? People don’t seem to understand that unless a prediction is both accurate and timely, it really isn’t of much value.
The Economist seems to have wanted the Fed to do something in 2003. But what if the Fed had tightened, perhaps we would have gone into a Japanese style deflation/zero rate trap. Then what? I will concede that The Economist was probably right about monetary policy in 2004 and 2005, in retrospect a bit more restraint would have helped. But May 2003? That is still very much an open question.
We all interact with our fellow humans. I am sure that you observe the same things I do, our acquaintances are absurdly overconfident in their predictions in sports, politics, economics, everything. It’s human nature. But we all go along to be polite. I am the same. Obviously it’s a bit absurd for me to think that a professor at Bentley has the answer to the world’s financial crisis. But I think I do. I act as if I do. I am as arrogant as you expect such a person to be.
You might ask why then do I want the Fed to adopt my policies? The answer is that I don’t expect them to do so merely on my say so. I think Bernanke would be foolish to adopt my ideas unless other more distinguished economists have also come on board. The goal of my blog is to start a conversation among economists about the possibilities for unconventional monetary stimulus. Being a libertarian with slight monetarist tendencies, I have been surprised that in recent months the conversation in this direction has been more noticeable in liberal/Keynesian blogs such as Krugman/DeLong/Yglesias. But any news is good news.
Free Exchange says we can have confidence that The Economist’s prediction was not luck, because they gave lots of good reasons for that belief. I agree that helps. But consider this analogy. Suppose you have a friend who predicts his football team will lose in the first round of the playoffs. He cites two reasons, the quarterback is inexperienced in big pressure games, and the star running back is injured. Now suppose they win all their playoff games and go to the Super Bowl. The prediction was wrong. But suppose they lose in the Super Bowl because they have no running game and their quarterback looks very shaky under pressure. I’ll bet the average fan would say “Aha! I knew that was going to happen.” We remember our successes and forget our failures. Even worse, we shape the game’s narrative to fit our predictions. The Economist could have written articles saying:
1. A 2007 crackdown on immigration sharply slows US population growth in the Southwest. Housing starts to slump. High energy prices hurt commuters with long commutes. Home prices slump in the exurbs.
2. Tight money by the Fed in 2008 causes NGDP growth to slump 8% below trend. Unemployment soars. Nominal incomes fall at the fastest rate since 1938. This further weakens an already shaky housing market.
But that’s not what most journalists wrote. They wrote a narrative that fit their preconceived ideas about what happened. Am I claiming that those other factors didn’t play a role? Of course not. But wouldn’t it be equally absurd to write a narrative that didn’t give prominent mention to the factors I just cited? Especially since the fall in NGDP closely correlates to the housing crash spreading nationwide? I think you’ll agree that it would be negligent. But that doesn’t stop most financial reporters from ignoring those other factors.
[BTW, before you say “immigration doesn’t explain the other countries” consider that those crashes tended to occur later than the US. And were correlated with falls in NGDP. And the one country where NGDP didn’t fall? Oh that’s Australia, the one country where The Economist’s prediction wasn’t even right in the long run. They had no housing crash at all according to my commenter from down under. I wonder how The Economist would explain that.]
One final comment. When I ask people why a mutual fund run on their insights about fundamentals would not beat an index fund, they often dodge the question by talking about the difficulty of selling stocks short. But again, if you really can know the fundamentals, there are plenty of profitable long positions in various markets around the world, and in the brief period when there might not be, just park the money in index funds. I still say that type of mutual fund wouldn’t reliably beat an index fund. And I still say the implication of the bubble theorists is that it would. I can’t quite tell what the view is at Free Exchange:
I feel like this is the sort of critique that sounds lovely so long as one remains comfortably in the realm of abstract intellectualism. The price-to-income ratio has risen above its long-term trend, but how can we know that it’s a bubble? Fundamentals? Well, perhaps they’ve shifted. And if you’re so confident, why aren’t you ringing up your trader and telling him to short housing?
Yes, that is my question. And the answer is that if you had shorted housing builders in May 2003 you would have gone bankrupt. But then he seems to want to have it both ways:
Some people managed to make money off their models of the bubble despite the significant logistical difficulty in betting against housing markets.
Those would be people who ignored The Economist in 2003 and went short in 2007. I do accept one part of this argument. Housing is harder to short than stocks, bonds, commodities and currencies. But that really doesn’t help Free Exchange’s argument. You see The Economist is equally opinionated about these other assets. Which does imply that an “Economist mutual fund” should reliably outperform index funds. I say it wouldn’t. Robin Hanson suggests we need to test this. I don’t think so. It’s like astrology. As soon as you start suggesting to an astrologer that we do tests of whether personality types correlate with birth month, their whole attitude changes. “It’s more complicated than that.” There will never be the sort of test Hanson wants because no one has enough confidence in the musing of The Economist on fundamental values of stocks, bond, commodities and currencies to even lift a finger setting up such a mutual fund.
If this has been too negative, how about ending on this note. What would make the world a better place is not people investing or regulating on the basis of The Economist’s asset price predictions, but rather governments around the world switching to the sort of high saving neoliberal economic model that The Economist advocates. That’s what makes them the best and most valuable magazine in the world—their influence on policymakers.
Tags: bubbles
15. January 2010 at 14:37
If prices are unpredictable, then rational behavior is an impossibility. If rational behavior is possible, they must be at least somewhat predictable. They may be more predictable in the long term than the short, but without predictability no coherent action is possible.
15. January 2010 at 15:23
Lord – I don’t think anyone is saying prices are unpredictable. Just that to the extent they are predictable everyone else is also trying to predict them.
15. January 2010 at 16:12
Is Ryan Avent the only person who writes free exchange?
15. January 2010 at 16:37
Australia experienced a dip in house prices as seen by the ABS’s house price index stats. Drop began in Jun 08 quarter and continued until the end of Mar 09 quarter. Drop was a miserly 5.5% or so. And since Mar 09, prices have climbed 8.5%. So go figure. We’re weird down under. Would be interesting to see if your hypothesis backtested well on Australia data.
ABS stats:
http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/6416.0Sep%202009?OpenDocument
15. January 2010 at 17:39
Equating prices to value eschews any independent assessment of value. Without any fundamental value, prediction can only be extrapolation, and extrapolation leads more or less directly to bubbles. Price is not always value. Often it becomes an assessment of the value of others, an assessment subject to wild speculation in the absence of any fundamentals. Stocks have earnings and housing has rents and while these are subject to change, they aren’t totally unpredictable, but once people focus on price and price change, they are no longer investing but speculating. Speculation may be self correcting eventually, but it is not necessarily in the short term. Speculation can feed on itself until the ponzi scheme runs low on entrants or encounters a sufficient number of deserters. Prices then collapse to or below fundamentals.
15. January 2010 at 18:48
Dr. Sumner,
Lump all the sarcasm on me you want. We all deserve it sometimes and I’m not nearly as sensitive as some might be at the Economist.
15. January 2010 at 18:49
Lord, I agree with Simon (as usual)
Niklas, He writes most of them. But why not sign your name? I guess it’s The Economist’s policy.
Manny, You aren’t weird, you are smarter than the rest of the world. You don’t let NGDP drop. We should have learned form the 1930s that was a bad idea.
Lord, Each person makes an individual prediction. The market aggregates those predictions into a price. It is hard to predict better than the market. If 100 people guess the number of jelly beans in a jar, it is very hard to outguess the average.
15. January 2010 at 19:43
If only that were true. But it isn’t. It is precisely when people give up making their own prediction and rely instead on the predictions of others that trouble arises. They are no longer interested in the prediction, but what others believe and predict, what others perception of value is. They can then extrapolate that to absurdity.
16. January 2010 at 02:14
“And if regulators try to predict bubbles based on predictions from The Economist, they’ll do no better than they would by throwing darts at a stock page.”
1. The key problem is here: “if regulators try to predict bubbles”. Regulators have always inflated bubbles. I know only one case when regulators have prevented a bubble – in 15th or 16th century China successfully prevented malinvestments in shipbuilding industry.
2. I’d be verry happy if regulators tried to improve micro regulation of sectors where Economist says bubble exists. I see no chance of that happening.
16. January 2010 at 07:26
“Obviously it’s a bit absurd for me to think that a professor at Bentley has the answer to the world’s financial crisis. But I think I do. I act as if I do. I am as arrogant as you expect such a person to be.”
In reality, this passage displays your modesty and humility. You know perfectly well that you might be wrong, and your readers, if at all perceptive, know you know it. After all, anybody who ever publishes an opinion might be wrong, and many (not all!) are fully aware of that. But it would be intolerably tedious for you to begin every sentence with: “Of course, I am only a professor at Bentley, and I’m not 100% certain about any of the following, but . . . .” You may as well just continue to say what’s on your mind straightforwardly–this is not “arrogant”–and we’ll take the disclaimer as understood background.
And don’t overdo the humility and modesty; in my opinion you are on your way to eventual enshrinement in the Blogger’s Hall of Fame!
16. January 2010 at 12:29
This Q&A with Fama and French provides, I think, a clear and concise answer to the (it´s obvious it was a) bubble argument:
http://www.dimensional.com/famafrench/2009/08/qa-nasdaq-at-5000-a-mistake.html
The NASDAQ climbed by a factor of 2.5 between mid 98 and early 2000. The DOW and S&P went up by a (realtively measly) 1.2.
Was it “irrational” (or inefficient) to have such high expectations about tech stock profitability relative to the other sectors at the time? Note: Those expectations have to be formed ex ante. No monday morning quartebacking allowed.
17. January 2010 at 10:25
Lord, I didn’t say people should give up making their own predictions. I am probably better off with index funds, but people who work on Wall Street do make an effect to estimate values, and thank God they do. Where would we be without human overconfidence?
123, Good points.
Philo, Thanks, and I do understand the point you’re making. I suppose I meant that I often forget that I have that self-awareness. Like most people, I am literally of two minds.
Marcus, I think they are right. But I also think the question they are addressing is in some sense unanswerable. What is the difference between an irrational prediction, and one that was rational but very mistaken? I think we all have some idea, but it’s pretty hard to draw the line. My next post addresses this from a different angle
17. January 2010 at 15:02
“1. The key problem is here: “if regulators try to predict bubbles”. Regulators have always inflated bubbles. I know only one case when regulators have prevented a bubble – in 15th or 16th century China successfully prevented malinvestments in shipbuilding industry.”
Bubbles aren’t necessarily bad. A point Greg often tries to make here. If china had not tried to prevent the malinvestments, they may have reached the new world and began colonizing. Instead the new world was european not chinese.
18. January 2010 at 13:27
Doc Merlin, Would America be better off Chinese or European?
19. January 2010 at 10:46
If you mean would America have been better off being founded with 16th and 17th century British institutions or with 16th and 17th century Chinese institutions, I believe it would have been better off with the British ones. The Chinese, despite their early technological lead, didn’t evolve the institutions necessary for a widespread free society to develop. By the 17th century their technological lead had completely vanished, and so had their economic lead, because of the institutional deficiency.
19. January 2010 at 19:19
Doc, Good point. (But we’d probably have a higher saving rate.)
20. January 2010 at 20:35
Wouldn’t a decent Hansonian test of the Econonomist’s past predictions (rather than their claims of their past predictions) be this:
look at a sample of Economist subscribers in what turned out to be bubble housing markets in 2000 and a control of non-Economist subscribers with the same income residing in the same zip and post codes – and measure their housing buying-and-selling patterns over the next seven years.
Do you think that Economist readers in Boston and Los Angeles and London did better in their personal real estate transactions and beginning/concluding MVs?
My bet would be that they will have done worse than the readers, say, of COMMENTARY – because they would share the magazine’s attitudinal arrogance and self-confidence more than they absorb the magazine’s wisdom (if indeed it existed “in real time”).
23. January 2010 at 06:45
Sam, Nice idea, but there is a huge omitted variable problem, unless you included all sorts of other characterisits in the regression.
I agree that reading the Economist probabaly doesn’t make you a smarter investor in housing.