Andrew Page
Andrew Page

At a recent trivia night the host offered free drinks to the person who could outlast everyone in a game of heads or tails. Everyone was asked to stand up and place their hands either on their head or on the backside. The host would flip a coin; those that guessed correctly would go on to the next round, the others were eliminated. As it turned out, a winner was selected from about 100 people in about 8 flips of the coin.

From a probability standpoint, the winner had managed to achieve a rather remarkable feat – the odds of guessing correctly 8 times in a row are 1 in 256 (or a 0.39% chance). However not a single person at the event was that impressed, and why should they be? After all, at each toss of the coin there was a mix of expectations, and it was inevitable that someone would be proven right. At the end of the day the winner just got lucky. Very few people would consider the winner to have been successful because of skill or intelligence.

Contrast this situation with those that make economic forecasts. At any given point in time there will be a wide range of forecasts, each provided by well respected experts and supported by solid arguments. Presently, economic forecasters are broadly divided into two camps: those that feel that the worst is over, and those that feel that this is simply the calm before the storm. Within each group there will be those that provide quite specific forecasts in terms of figures and dates.

As with the coin tossing game, we will discover that only a small minority of players will be shown to be right. Once again, it is inevitable. The problem is that the market takes this as ‘proof’ that these soothsayers have the uncanny ability to accurately predict the future, and they will do the usual round of interviews and presentations to adoring crowds who are eager to discover what the next premonition will be.

If you need evidence of this just consider the success of those economists who are reputed to have predicted the credit crisis and resulting GFC. Most were all but unknown prior to the event, and most will be forgotten within a few years. Furthermore, many that did ‘predict’ the correction were calling for it to happen many years prior to the actual event. The fact is that if you continually call for a correction after a prolonged bull market, it’s only a matter of time before you are proven right.

I don’t mean to belittle economists, after all the industry demands forecasts and they simply do the best they can at what is an amazingly difficult task. Rather, I want to caution investors to always take forecasts with a grain of salt and not to confuse them for immutable facts. Also, just because a certain economist got it right last time, does not mean they have any greater likelihood in being correct the next time (a view that is supported by contemporary research findings).

The ‘heads or tails’ phenomenon is in fact something that is exploited by con artists. Consider this well known con: a person sets up a stock market advisory service and purchases a large database of contact details, let’s say 100,000 people. To half you send a free report recommending you buy a stock because it is about to go up. To the other half you send the opposite advise, that is that the stock is about to go down. Regardless of what the stock actually does, you are guaranteed to be right in the eyes of 50,000 people. To these 50,000 you do the same thing a second time, that is, send 25,000 people a bullish call, and the other 25,000 a bearish call. The process is repeated a number of times.

You can see what’s going to happen here. After 10 rounds of this you will be left with about 100 people who have received a newsletter which accurately predicted the stock market 10 times in a row. At this stage, you request a high fee for your newsletter, and you will find that most will pay anything to continue to receive this amazingly accurate advice. Let’s face it, even those that saw a correct prediction most of the time, say 7 out of 10 times, will most likely consider your newsletter to be worth many times its weight in gold. The beauty of this con is that you are guaranteed to impress a large number of people, regardless of what the market actually does!

The reality is that for any predictive ability to validated, we should demand rigorous empirical evidence. In other words, you need to be able to verify whether or not chance alone can account for the observed accuracy of the forecasts. Unfortunately, the market rarely demands such validation. But let’s just be pragmatic about it all. Forecasts are useful in broad terms if we treat them as simply a gauge of general market expectations. They also force us to consider possible scenarios and allow us to put in place some contingency planning. The main thing to remember is that forecasts are just guesses, albeit educated ones. Those that treat them as fact will more often than not be disappointed.

Make the markets work for you

Andrew Page