Predicting the future has never been easy, but the standard today is the same as in ancient times: does the prediction come true? As an ancient Hebrew author wrote, "When a prophet speaketh, ... if the thing follow not, nor come to pass, ... the prophet hath spoken it presumptuously: thou shalt not be afraid of him." [Deut. 18:22].
Time is running out for 2014 market predictors. So how have they fared? Are prophets good at making profits?
A rough year for hedge funds
At least one sector of the finance world has a disappointing record so far: the hedge fund industry. While some highly skilled hedge funds have done well, others have not. In fact, according to a Bloomberg report, 461 hedge funds closed in the first half of 2014, and, if the pace of closures continues, the total for 2014 may approach or exceed that of 2009.
The average return on all hedge funds is up only 3.9% for 2014 (as of 2 December 2014). Needless to say, this 3.9% average return compares rather unfavorably with a simple index-based, buy-and-hold strategy in almost any equity market class. The S&P500 index is up 13.9% so far for 2014; the DJIA index is up 10.2%; the Nasdaq Composite Index is up 15.2%; and the Tokyo TPX index is up 11.7%. In this context, the 3.9% average hedge fund performance is not very impressive.
The 3.9% hedge fund performance is based on net gain. Even so, it should be noted that for most hedge funds (and mutual funds as well), the annual fee is assessed irrespective of the fund's performance. The rationale for the fees is to fund the analysis resulting in superior performance; but performance is not always superior. Some investors are now asking that hedge fund managers beat benchmarks before charging fees.
Along this line, about six years ago famed investor Warren Buffett "bet" a New York hedge fund that money invested in a S&P500 index exchange-traded fund would beat the hedge fund over a ten-year test period. Six years into the bet, it appears that Buffet will win handily.
Prophets of doom2014 has also been a hard year for prophets of doom. Even a casual Internet search yields dozens of such predictions (NOTE that in most of these cases, the author of the blog or article is not the specific party making the prediction):
- [28 Dec 2013] Doomsday poll: still a 98% risk of 2014 stock crash
- [23 Jan 2014] How about a 50% crash?
- [11 Feb 2014] Scary 1929 market chart gains traction
- [19 Feb 2014] Crash of 2014: Like 1929, you'll never hear it coming
- [15 Mar 2014] New doomsday poll: 99.9% risk of 2014 crash
- [27 Mar 2014] Is this the correction, or a coming crash?
- [09 May 2014] 10 peaking megabubbles signal impending stock crash
- [29 Jun 2014] New Doomsday poll: 98% risk of 2014 stock crash
- [23 Jul 2014] For stocks, Dow 20,000--then a crash?
- [05 Aug 2014] How you'll know if it's time for a market crash
- [11 Aug 2014] Here"s my obligatory column that says stocks will crash
- [21 Aug 2014] How to protect your retirement from the next big crash
- [10 Sep 2014] Opinion: Republican-run Senate could speed stock crash
- [24 Oct 2014] Opinion: 15 Big Oil sell signals that warn of a 50% stock crash
- [10 Nov 2014] 4 near-term catalysts for the next stock market crash
- [20 Nov 2014] The man who called the last stock crash is already blaming the Fed for the next
But statistically speaking, what is the substance of such a prediction when it is repeated over and over again? Even a stopped clock is precisely correct twice a day. Highlighting correct predictions, and ignoring numerous incorrect predictions, is a classic "selection bias" statistical error, one that afflicts numerous fields of modern science and technology. For example, pervasive selection bias problems in the pharmaceutical industry have led to a call for all trials, successful or not, to be made public.
Is 2014 just a bad year?
While 2014 has been a particularly bad year for market forecasters, they have not done much better in previous years. For example, Mark Hulbert reported that of the 51 advisors (out of over 200) who beat the market over the ten-year period ending 30 April 2012 (as measured by the Wilshire 5000 total market index), just 11 outperformed the market in the next 12-month period. In a separate analysis, the CXO Advisory Group evaluated the performance of 68 market "gurus," based on published market forecasts. While a handful of "gurus" did quite well, the average performance was only slightly better than chance.
With regards to crash predictions, it is instructive to consider that although economist Roger Babson famously predicted the 1929 crash, he had predicted a crash for years; only in 1929 did it finally materialize.
Can markets be predicted?
These results should not surprise researchers who have studied market behavior. After all, markets by definition encapsulate the collective wisdom of tens of thousands of highly trained market analysts, assisted by some of the most sophisticated mathematical algorithms and some of the most powerful supercomputers on the planet. The net result of these competing analyses is to squeeze out any potential intelligence, leaving nothing but noise (a "random walk") and perhaps a very low-intensity signal that relatively unsophisticated analyses have little chance of even seeing, much less profiting from.
Markets are not perfectly predicted by these analyses, and they are always subject to vagaries of market psychology. But let's not be surprised when market prophets do poorly -- after all, they are betting against a (nearly) random walk.