The tell-tale signs of unreliable stock market forecasters

A general rule of thumb: the most confident are the least accurate, according to research


Market forecasting is a tricky business, with countless studies indicating we should be sceptical towards those who purport to know what the future holds for investors.

Nevertheless, investors "still have to plan for the future by making educated approximations about how things will play out in the coming years and decades", writes US money manager and A Wealth of Common Sense author Ben Carlson. So how should investors sift through the myriad of conflicting narratives out there? How do we decide who is worth listening to and who isn't?

Don't get caught up in the specifics of an argument, advises Barclays' behavioural finance expert, Dr Greg Davies. In a recent paper, he cautions investors to look out for "tell-tale signs" that suggest a particular forecast may have less merit than it appears to have.

Betting against the house

Warren Buffett

READ MORE

once noted the Dow Jones rose from 66 to 11,497 during the 20th century, despite enduring two world wars, the Depression, “a dozen or so recessions and financial panics” as well as numerous other traumas.

Most of the time, says Davies, economies grow and stocks rise, so bearish forecasters are “betting against the house”. Ignoring what happens most of the time is known as base rate neglect, he says – “placing too much evidence on your immediate short-term story, and too little on the underlying statistical ‘base rate’ from observed history”. Forecasters often find their own narrative “so compelling that they ignore the long-term lessons of history”.

Of course, an expert may argue current circumstances are “sufficiently compelling to override what happens most of the time,” says Davies. “But we should be very suspicious of any article that even fails to mention or address this point – it is evidence of muddy thinking.”

Beware extreme forecasts

Veteran forecaster

Robert Prechter

recently warned US stocks were “at high risk of sharp collapse”. It wasn’t his first warning: at the bottom of the 2000-02 bear market, he warned the Dow Jones would eventually fall below 1,000. He reiterated this forecast in 2010, telling the

New York Times

the biggest bear market of the past 300 years was nigh. Stocks, he said, would fall by more than 90 per cent over the next five or six years.

Indices have almost doubled since 2010; the Dow has spent most of 2015 hovering around the 18,000 level.

It’s easy to laugh, but extreme predictions result in valuable media exposure for forecasters. Research indicates many people are swayed by them, assuming the forecaster must have great confidence to assert such unpopular opinions. In reality, extreme forecasts are generally money-losing propositions.

Moderation isn’t always a good thing

If extreme forecasts are to be avoided, then moderate forecasts must be a good thing – right?

Not necessarily. Most years, strategists predict indices will gain around 7 to 10 per cent. This makes them appear like a reasonable bunch – bullish but not too bullish.

However, such forecasts are unlikely to be accurate. Yes, stocks have enjoyed average annual gains of more than 9 per cent over the past century. Historically, however, annual returns have been in the 0 to 10 per cent range in just one year in six. Rollercoaster rides are more common – since 1871, stocks either gained or fell by more than 20 per cent in more than 40 per cent of all years.

An evidence-based forecast is always preferable to a forecast that is moderate for the sake of appearing moderate. The intellectually honest approach would be to admit that while long-term annual returns may well be in the vicinity of 7 to 10 per cent, stocks are likely to swing around in highly unpredictable fashion in any given year. Pretending otherwise leaves investors ill-prepared for inevitable volatility.

Too specific

At the end of 2015, the Dow Jones will be above 20,000 while both the FTSE 100 and the Iseq will rise above the 7,000 level. However, it won’t be plain sailing – expect an autumnal pullback that will slice 5 to 7 per cent off the major indices before stocks roar back in the dying months of the year.

Actually, I just made that up – I don’t have a crystal ball. Unfortunately, many forecasters pretend they do, making precise predictions that are at best useless, at worst dangerous.

Even if a forecaster’s reasoning is correct, the market may be slow to agree, which is why Greg Davies advises investors to beware of “overly specific” predictions. They should also not concern themselves with precise forecasts irrelevant to their investment needs, such as when “the advice is to exit a well-diversified portfolio to escape some predicted short-term dip”.

Davies's advice echoes that of iconic investor and former Fidelity fund manager Peter Lynch. "Far more money has been lost by investors preparing for corrections, or trying to anticipate corrections, than has been lost in corrections themselves," he cautions.

Overconfidence

“As a general rule of thumb”, says Davies, “the more confidence with which someone tells you they know what is going to happen in the markets, the less you should believe it.”

Research confirms this point. Political psychologist Philip Tetlock examined some 27,000 predictions made by political experts over a 16-year period. The most confident were the least accurate – those who said they were 80 per cent sure they were right were accurate on just 45 per cent of occasions.

Modest forecasters will recognise the wisdom in Donald Rumsfeld's oft-quoted line about "unknown unknowns – the ones we don't know we don't know". Renowned investor Howard Marks makes a similar point, albeit in less clumsy fashion: "You can't predict. You can prepare."

Ulterior motives

Ironically, Tetlock found overconfident experts were, despite their inaccuracy, most likely to be invited onto TV shows. Many are deliberately “entertaining” and “flattering the prejudices of their base audience,” he says; “accuracy is a side constraint.”

Davies makes the same point, opining that “provocative opinions that gloss over the fundamental uncertainty and complexity of the real world sell much better” than more “circumspect” observations.

There are other obvious cases where a pundit may have an ulterior motive. Bond managers may say to buy bonds, equity managers may say it’s time to buy equities, someone selling commodities may say . . . well, you get the picture.

Devil’s advocate?

“Forecasting articles are generally written to provide a convincing narrative, rather than a thoughtful analysis,” says Davies. Better analysts play devil’s advocate to “challenge their own thinking”.

Ben Carlson agrees, noting how easy it is to cherry-pick data to fit your narrative. A gold sceptic might note the metal is down 40 per cent since 2011 and that it has barely kept up with inflation since 1980, says Carlson. A gold bull might say prices have quadrupled since 2000 or that annual returns have averaged 7 per cent since the 1970s.

You could say US stocks are overvalued by pointing to the S&P 500 having tripled since March 2009.

Alternatively, you might argue the opposite by noting the S&P 500 has only returned around 4.3 per cent annually over the past 15 years.

Forecasters often say a bear market (a 20 per cent decline) or correction (a 10 per cent decline) is overdue, seeing as there have been none for six and four years respectively.

But what about declines of 19.4 per cent in 2011 and 9.9 per cent in 2014? Should they be counted?

Harry Truman famously asked for a one-armed economist, instead of having to listen to advisers talking about "on the one hand . . . on the other". Investors might share Truman's frustration, but they should remember two hands are better than one – especially when it comes to market forecasts.