Image: Illustration by Juliette Borda
In December 2010 I appeared on John Stossel’s television special on skepticism on Fox Business News, during which I debunked numerous pseudoscientific beliefs. Stossel added his own skepticism of possible financial pseudoscience in the form of active investment fund managers who claim that they can consistently beat the market. In a dramatic visual demonstration, Stossel threw 30 darts into a page of stocks and compared their performance since January 1, 2010, with stock picks of the 10 largest managed funds. Results: Dartboard, a 31 percent increase; managed funds, a 9.5 percent increase.
Admitting that he got lucky because of his limited sample size, Stossel explained that had he thrown enough darts to fully represent the market he would have generated a 12 percent increase—the market average—a full 2.5 percentage points higher than the 10 largest managed funds average increase. As Princeton University economist Burton G. Malkiel elaborated on the show, over the past decade “more than two thirds of actively managed funds were beaten by a simple low-cost indexed fund [for example, a mutual fund invested in a large number of stocks], and the active funds that win in one period aren’t the same ones who win in the next period.”
Stossel cited a study in the journal Economics and Portfolio Strategy that tracked 452 managed funds from 1990 to 2009, finding that only 13 beat the market average. Equating managed fund directors to “snake-oil salesmen,” Malkiel said that Wall Street is selling Main Street on the belief that experts can consistently time the market and make accurate predictions of when to buy and sell. They can’t. No one can. Not even professional economists and not even for large-scale market indicators. As economics Nobel laureate Paul Samuelson long ago noted in a 1966 Newsweek column: “Commentators quote economic studies alleging that market downturns predicted four out of the last five recessions. That is an understatement. Wall Street indexes predicted nine out of the last five recessions!”
Even in a given tech area, where you might expect a greater level of specific expertise, economic forecasters fumble. On December 22, 2010, for example, the Wall Street Journal ran a piece on how the great hedge fund financier T. Boone Pickens (chair of BP Capital Management) just abandoned his “Pickens Plan” of investing in wind energy. Pickens invested $2 billion based on his prediction that the price of natural gas would stay high. It didn’t, plummeting as the drilling industry’s ability to unlock methane from shale beds improved, a turn of events even an expert such as Pickens failed to see.
Why are experts (along with us nonexperts) so bad at making predictions? The world is a messy, complex and contingent place with countless intervening variables and confounding factors, which our brains are not equipped to evaluate. We evolved the capacity to make snap decisions based on short-term predictions, not rational analysis about long-term investments, and so we deceive ourselves into thinking that experts can foresee the future. This self-deception among professional prognosticators was investigated by University of California, Berkeley, professor Philip E. Tetlock, as reported in his 2005 book Expert Political Judgment. After testing 284 experts in political science, economics, history and journalism in a staggering 82,361 predictions about the future, Tetlock concluded that they did little better than “a dart-throwing chimpanzee.”
There was one significant factor in greater prediction success, however, and that was cognitive style: “foxes” who know a little about many things do better than “hedgehogs” who know a lot about one area of expertise. Low scorers, Tetlock wrote, were “thinkers who ‘know one big thing,’ aggressively extend the explanatory reach of that one big thing into new domains, display bristly impatience with those who ‘do not get it,’ and express considerable confidence that they are already pretty proficient forecasters.” High scorers in the study were “thinkers who know many small things (tricks of their trade), are skeptical of grand schemes, see explanation and prediction not as deductive exercises but rather as exercises in flexible ‘ad hocery’ that require stitching together diverse sources of information, and are rather diffident about their own forecasting prowess.”
Rights & Permissions
See what we're tweeting about
26 Comments
Add CommentIs anthropomorphic CO2 hypothesis just another example of "reducing complex phenomena into one overarching scheme"?
Reply | Report Abuse | Link to thisI was struck that one could switch the word "economic" to "climate" and the insights equally apply.
Maybe its time that we should reconsider making predictions to complex systems and avoid the cognitive traps highlighted in this article.
... especially by experts who are "deeply knowledgeable on one subject that narrows focus and increases confidence but also blurs the value of dissenting views and transforms data collection into belief confirmation."
"This type of cognitive trap is why I don't make predictions and why I never will"
Reply | Report Abuse | Link to this'nuff said?
I'm a big fan of Michael Shermer, but I can't help pointing out a small discrepancy in this article. The last sentence reads "This type of cognitive trap is why I don’t make predictions and why I never will." Ah, but he just did.
Reply | Report Abuse | Link to thisOne big problem with this entire article is the fact that stock and portfolio management selection isn't done by people anymore, its done by computers.
Reply | Report Abuse | Link to thisThis is certainly more true of short time-frame transactions than long-term holdings, but the fact is that people are less and less a part of the equation on Wall Street. Everything is increasingly going to predictive analytics and trading algorithms, so people's cognitive biases are becoming increasingly irrelevant.
As systems like IBM's self-learning Watson become more pervasive, the role of computers in trading will certainly increase even more.
@dj_allcock, @milt95033
Reply | Report Abuse | Link to thisYou think perhaps that might have been a teensy weensy little joke?
First, I'm certain that Dr. Shermer knew what he was writing at the end. It's called humor.
Reply | Report Abuse | Link to thisSecond, there's a difference between continuous and discontinuous systems. Climate is continuous in that changes happen (relatively) slowly whereas markets are wildly discontinuous. Temperature outside my house doesn't (and can't) fluctuate from -40 to +40 in the span of minutes but markets can certainly go from fantastic to abysmal and back again in that time. So one can have more confidence when predicting in continuous conditions than one can in the markets. That's why you can't substitute climate for economics here.
Hmmm......
Reply | Report Abuse | Link to this'Is anthropomorphic CO2 hypothesis just another example of "reducing complex phenomena into one overarching scheme"? '
The short answer is NO. There is just too much evidence from diverse sources supporting the idea.
On the other hand, the climate cynics are guilty of this (guilty, guilty, guilty!)
Examples of climate cynics overarching (and over reaching) schemes include:
(Geology) geological phenomena (such as the weathering of the Himalaya mountains) have a much larger impact on atmospheric CO2 than human activities, therefore the recent CO2 spike must from natural geologic causes (it's only a heck of a coincidence that atmospheric CO2 tracks human contributions so closely)
(Astronomy) fluctuations in solar output drive climate changes. (which begs the question of how a periodic +-0.1% change in insolation can drive a steady +4% increase in temperature)
(Politics) Al Gore is a socialist, therefore climate change is a socialist plot (believe it or not, I have actually met people who argued this. Never mind that Al Gore is not a socialist, nor that 1000's of scientific papers do not invoke politics to show the effects of CO2--unlike the cynics who ONLY invoke politics to deny the effect, and that Al Gore does not have his name on any scientific articles--at least, nothing hardcore)
Ice core data suggest climate changes abruptly not gradually:
Reply | Report Abuse | Link to this"Millennial and sub-millennial scale climatic variations recorded in polar ice cores over the last glacial period"
Capron et al, Clim. Past Discuss., 6, 135–183, 2010
http://www.clim-past-discuss.net/6/135/2010/cpd-6-135-2010.pdf
A huge flaw I saw in this article is: How can anyone believe or trust in anything the Fox news stations say since they have proven themselves to be a carnival side show of lies and deception. It would be more believing if the author of this article had spoken to one of MSNBC's business professionals.
Reply | Report Abuse | Link to thisGood point -- the recent increase in CO2 may largely be caused by fossil fuel consumption based on our current understanding of sources and sinks.
Reply | Report Abuse | Link to thisHowever, the hypothesized CO2 relationship to global warming still qualifies as a reducing complex phenomena into one overarching scheme.
The Capron et al article (among others) provides a good overview of the complexity and the multitude of factors that require "flexible 'ad hocery' that require stitching together diverse sources information"
And of course, the A in AGW stands for anthropogenic, not anthropomorphic. Caused by human, not looking like one.
Reply | Report Abuse | Link to this"One big problem with this entire article is the fact that stock and portfolio management selection isn't done by people anymore, its done by computers."
Reply | Report Abuse | Link to thisIs this intended as a serious response? Firstly, computer algorithms are written by people, based on the expert knowledge these people have. Secondly, taking intuition and broad interconnected knowledge out of the decision-making process would, according to Mr. Shermer's thesis, make the management by computers less effective, not more.
This author is absolutely incorrect. I don't know where he's getting his data. I just ran a screen on morningstar showing me how many actively managed funds that have the the Russell 1000 as their bench mark and outperformed it. I got 145 fund. Check out Franklin Rising Dividends, Credit Suisse Large Cap Blend, Eaton Vance Tax Mgd Growth, Gamco Westwood Equity, GE Core Value, Lord Abbett Classic Stock, Prudential Jennison Value, MFS Core Growth, Calamos Growth. There are a lot more examples if you go with total return funds ones that will go into any asset class. The funds beat after expenses and after having to hold some in cash for redemptions.
Reply | Report Abuse | Link to thisBasically this article should not be published, this is a scientific magazine.
Reply | Report Abuse | Link to thisEconomics like Psychology is not a science, its more a religion.
To be classified as a science, its proponents must present a theory that predicts things that can be measured, the test results must concur with the theory no matter who conducts the test or where. Psychology talks of repressed memories, no one has ever proved them, charlatans hypnotise people and make them believe in such nonsense, no different from you are born in sin.
Dr. Shermer,
Reply | Report Abuse | Link to thisBurton Malkiel is the proponent of the Random Walk theory of the stock market. It says that stock price movements are random resembling the random walk function. It is not entirely true. Yes, it is random in the short term and impossible to predict accurately. But in the long term say 10 yrs or more, it follows economic fundamentals and can be evaluated with fundamental analysis.
The empirical evidence of this 'fundamental theory' are fundamental investors like Buffett who can consistently outperform the market over long periods of time. The managed funds cannot consistently beat the market bec. they are so large and so diversified that they mimic the market.
Are you begging a market tip of me?
Reply | Report Abuse | Link to this1) Recently, U.S. money supply is growing exponentially and steeply.
2) Recently, Canadian money supply is growing much less steeply and has even seen negative growth.
3) Yet for months, the US$ and C$ have been at par.
4) When the inbred population finds out this is true, we can expect inflation in the U.S. relative to Canadian inflation and a relatively stronger C$ is my current prediction.
When the population is a mass inbred, it thinks similar things in response to similar issues. Sometime the issues can be imagined and forecast in advance supplying useful risk/reward ratios. If you come to the same conclusion the majority comes to first, you can beat the market.
The prediction appears duplicated if it does not come to pass right away and you ask me again later. It is not in my power to know when the above information will be accounted for properly by the market.
Until it is, I do not accept US$ in Canada at the going exchange rate.
This article contains a few sentences that constitute a “weasel" section. It also ignores the fact that some active money managers outperform the market over long periods.
Reply | Report Abuse | Link to thisThe weasel section is "Malkiel said that Wall Street is selling Main Street on the belief that experts can consistently time the market and make accurate predictions of when to buy and sell. They can’t. No one can." "Consistently" is not defined. Is it 100%, 51%, or somewhere between the two? "Accurate" is not defined. Must the predictions be 100% accurate, 51% accurate, or somewhere between the two. Since neither of these terms has a functional definition, the assertion is meaningless mush. Mr. Shermer, of all people, should know this. He spends plenty of time debunking other people’s arguments.
While Wall Street is making that pitch, and few (not no) "experts" can do it, it's not necessary for a prognosticator to "consistently [whatever that means] time the market and make accurate [whatever that means] predictions of when to buy and sell" in order to outperform the market. If a manager's correct calls result in great positive results and a manager's incorrect calls result in small negative effects, the net would be the manager outperformed the market. This is what risk management is about; cut your losses quickly and let your winners run. There's no reason to expect that the effects of the correct and incorrect calls would be symmetrical.
There are money managers who actively trade and outperform the market over long periods. Their performance is several standard deviations beyond what would be expected. Among them are Peter Lynch, Steven Cohen, George Soros, Jim Rogers, and John Henry. During the first ten years of the Quantum Fund, the fund did 4200% while the S&P advanced about 47%. How many standard deviations is that from random?
Reply | Report Abuse | Link to thisComparing the returns of funds of today with average return of the market is biased because the underperforming funds of yesterday are closed down. Therefore, it is not surprising that a 145 actively managed funds, that are currently open, outperformed the market.
However, in order to avoid "survivor" bias, one needs compare future returns of the actively managed funds against the market index. The authors appeared to have completed the correct analysis and the comment that suggests the analysis is incorrect in not accurate.
The data concerning financial advisors [being unable to "beat the market"] was well known 40 years ago. Nothing startling about it.
Reply | Report Abuse | Link to thisThe dartboard test is (of course) flawed, but testing funds with random portfolios can be made rigorous. When they are specialized to the particular fund, they can have a lot of power to tell if the fund outperforms or not. In broad brush roughly half of funds have to underperform -- not counting costs. Including costs we can expect most funds to underperform. However, there do seem to be a few funds that consistently outperform.
Reply | Report Abuse | Link to thisjmaki,
Reply | Report Abuse | Link to thisThe biases of the anthropomorphic CO2 crowd go far deeper than just "reducing complex phenomena into one overarching scheme".
Just as with Hedge Fund managers who convince their clients to pay them billions for their supposed financial acumen, the Global Warming/Global Chaos (and, if we get a couple of years of moderate weather "Global Calming") scientists are also reaping millions plus fame and celebrity that other scientists could never even dream of. And, they get to tell women they’re “saving the world” – talk about a pick up line. No nerd ever had it so good. But it gets worse. “Players” such as Al Gove are becoming “green billionaires”. Third World countries are seeking Trillions. And, Big Government statists whose instinct is that they alone have the expertise to run all aspects of our society and economy are using CO2 to regulate every aspect of our lives. He who controls carbon controls everything.
These people can’t even accurately tell you the global mean temperate right now! Yet, they claim to be able to tell us global mean temperatures from thousands of years ago within 1/10th of a degree. Give me a break! Of course, the fame, fortune and power they’re getting has nothing to do with their confidence.
Global Warming/Global Chaos/Global Scamming is an unholy alliance of greed, ego, power, politics and control.
To Alchemguy,
Reply | Report Abuse | Link to thisThe biggest 1 day stock market change was 20% in 1987. Just last week parts of the country experienced a 100 degree change in a 24-hour period.
Whether can easily be more radical than the stock market. Your counter argument is FALSE.
Denswei,
Reply | Report Abuse | Link to thisYou said "There is just too much evidence from diverse sources supporting the idea" but this is FALSE.
Yes there are lots of difference intrepations of the data, but that data only comes from 2 sources, one of which, East Anglia University, is marred in controversy for "cooking the books" on the data.
Al Gore may not have his name on any scientific articles but he is the worlds first "green billionaire" and has his name of movies, books, articles, etc.
Experts can't beat the fees they charge, so go with low cost index funds.
Reply | Report Abuse | Link to thisAs I read this article, I predicted that there would be commentary that invoked the climate change issue.
Reply | Report Abuse | Link to thisIt's always amazed me how some people are so ready to religiously advocate modeling schemes vis a vis economics (and religion as well it seems), and here I go beyond the subject of the article and refer more to large scale economic ideas, like Reaganomics for instance, yet feel that any failure of climate modeling to measure up precisely is a reason to throw the baby out with the bath water, so to speak. A complete "refudiation" indeed, to hear them speak.
Michael Sherman was quoting the English footballer David Beckham when he predicted that he would never make a prediction.
Reply | Report Abuse | Link to this(The footballer had been asked to predict the outcome of a game.)
It was obviously a witty demonstration of a "cognitive trat".