Forecasting is an integral part of the human experience. Almost every key decision we make in life involves some element of guessing at the future; from the suburb in which we buy a property, the studies we undertake and those we might encourage our children to pursue, the career direction we take and, of course, to the investment decisions we make.
We want our financial futures to be as secure as possible, and so it is only to be expected that anyone claiming investment foresight (real or otherwise) will find a willing ear.
We therefore listen attentively to the forecasts of those only too happy to prognosticate on television, the radio or online. And there is, at present, no shortage of investment sages proffering views on everything from the re-emergence of US inflation to the future of fiat currencies.
But how do we, the investing public, determine if any of these predictions are in any way superior to simple guesses? Well, help is at hand.
The Professor of Prescience
Enter Philip E Tetlock. Tetlock is a Professor at the University of Pennsylvania, with appointments across the Wharton Business School, psychology and political science. In 1984 Professor Tetlock embarked on a multi-year research project to understand the nature of expert judgement, studying some 28,000 forecasts made by hundreds of experts across a variety of fields including politics, the economy and finance. His is among the longest, most detailed academic studies of whether humans can actually predict the future with any statistically significant degree of accuracy.
Tetlock pitted the expert forecasts against a range of algorithmically derived forecasts (so-called ‘Clinical’ versus ‘Actuarial’ forecasts). His conclusions at the end of almost 20 years of investigation? “Surveying these scores across regions, time periods, and outcome variables, we find support for one of the strongest debunking predictions: it is impossible to find any domain in which humans clearly outperformed crude extrapolation algorithms, less still sophisticated statistical ones.”
Forecasts made by the experts were statistically indistinguishable from random guesses. This finding subsequently morphed, somewhat unflatteringly, into a metaphor of the average finance guru being no more accurate than a ‘dart-throwing chimp’ in guessing future investment outcomes.
Are investment professionals better forecasters?
OK, so geo-political analysts can’t predict the location and timing of the next military coup. But surely seasoned investment professionals, with all that financial market data at their disposal, can outperform random guesses? One might be tempted to think so.
Unfortunately, the evidence is to the contrary. In a 2015 study of investment analyst forecasts, researchers Marcus Spiwoks, Zulia Gubaydullina and Oliver Hein found that none of the predictions for share markets, interest rates and exchange rates for four countries were able to beat simple forecasts based on a simple ‘random walk’ assumption (that the best guess as to the future state is the current state).
Delving specifically into the forecasting of interest rates, they studied 158,022 4- and 13-month rate predictions across 12 countries only to find that over 98% of the forecasts better reflected rates at the time of making the forecast than at the future date. Hence the name of their paper, “Trapped in the Here and Now – New Insights into Financial Market Analyst Behavior”. In short, there was no noticeable evidence of investment foresight present.
Separating forecasting skill from luck
So how does one determine if an expert’s forecast might suggest a degree of foresight rather than being just a product of luck? You certainly wouldn’t want to ask for examples of his or her forecasting track-record. Why? Because if someone makes enough forecasts, some of them are bound to be proved right. These correct ‘calls’ can then be used as proof of foresight, while the unsuccessful ones go unmentioned in the hope of a quick burial without the possibility of exhumation and further analysis.
Just as a broken clock can still be right twice a day, it can be difficult to know where forecasting skill ends, and luck begins. Without a way of separating the two we’ll never know if a forecaster is lucky but without skill or skilful but unlucky.
Thankfully one such way does exist, and it is remarkably low tech. It involves committing a forecast to paper in an investment diary. But, and this is key, the forecaster must provide both the forecast and an explanation as to why it will eventuate. In this way the forecaster commits to providing a prediction of both cause and effect.
As simple and effective as this technique is, it is roundly ignored by the investment industry. Few investment manager(s) or financial planners are prepared to back their forecasts in such a manner. Fewer clients are prepared to demand they do. That’s a pity as keeping an investment diary is one of the most powerful ways to determine if one is right for the wrong reasons (lucky rather than skilful) or wrong but for the right reasons (skilful but without luck).
If it all sounds a little pedestrian and something that an investment manager or adviser would deem to be beneath them, consider this. One of the world’s most astute investors, hedge fund billionaire George Soros, recalls a period when he recorded all his investment decisions and the reasons behind them in an investment diary. In his words: “I kept a diary in which I recorded the thoughts that went into my investment decisions on a real-time basis. …The experiment was a roaring success in financial terms – my fund never did better. It also produced a surprising result: I came out of the experiment with quite different expectations about the future.”
Learning to sit with irreducible uncertainty
Danish physicist and Nobel prize-winner Niels Bohr is reputed to have once said “Prediction is very difficult, especially if it’s about the future.” What holds true in the sub-atomic world of matter holds doubly true in the world of investing, where the unpredictability of human behaviour, unlike the predictable laws of quantum mechanics, have an outsized effect on outcomes.
View the forecasts of investment gurus and talking heads for what they are, guesstimates. Undoubtedly intelligent and well informed, but guesstimates of an essentially unknown and unknowable future all the same. T’was ever thus.
Post-script: Those interested in delving further may wish to avail themselves of a copy of Philip Tetlock and Dan Gardner’s “Superforecasting: The Art and Science of Prediction” Crown Publishers (New York), 2015.
Harry Chemay has more than two decades of experience across both wealth management and institutional asset consulting. An active participant within the wealth and superannuation space, Harry is a regular contributor in Australia and overseas, writing on topics across investing, retirement planning and the provision of financial advice.
Harry was a co-founder in one of Australia’s earliest digital advice services, Clover.com.au, and has more recently been appointed an Australian Ambassador to the Transparency Task Force, a UK-led initiative to bring greater transparency and accountability to financial services.