PsyFi Search

Thursday 26 March 2009

You Can't Trust the Experts with Your Investments

Investment Experts Make You Poor

In a complex world we increasingly have to rely on experts such as lawyers, doctors and scientists to guide us. We expect them to get things right most of the time and we sue the hell out of them if they don’t. In the world of investment, though, we get something different. They get it wrong most of the time, are never held accountable and generally argue that they actually didn’t get it wrong anyway and, if they did, it wasn't their fault.

Meanwhile people listen to these “experts”, take their advice, almost hero worship them at times and mostly get poorer while they make money at our expense. Who are the smart ones in this relationship?

Political Misjudgement

Remarkably there don’t seem to be any studies carried out on misjudgement by financial experts but there is a set of research covering a similar area. It’s been undertaken over two decades by Philip Tetlock (1) on the subject of political misjudgement.

His results demonstrate a set of pervasive human errors that won’t come as any surprise to behavioural psychologists and are uncannily similar to those we see in investment circles. Predictions in politics and investment are closely related anyway – both are complex areas, with no easily predictable outcomes, and are subject to unexpected events.

So, just as the pundits were predicting South African apartheid to be unassailable it collapsed. Just as investment analysts were extrapolating dotcom earnings to the stars the sky fell in. Just as the hegemony of the Soviet Union over Eastern Europe appeared unshakable the Berlin Wall came down. Just as collateralised subprime mortgage lenders were forecasting ever rising earnings the world’s financial system imploded.

It’s like any system with human beings as part of its wiring – short-term prediction isn’t so much difficult as impossible. We’re the grit in the gears.

Systematically Wrong Expertise

What interested Tetlock wasn’t, necessarily, that the experts got things wrong. It was that they got things wrong a lot and that they got things wrong in a systematic and predictable fashion. Do you find any of his conclusions familiar?

1. Experts weren’t any better at predicting outcomes than non-experts . If anything they were worse. It turns out that a little extra knowledge might improve forecasting but lots make it worse. Back to the monkey with its dartboard.

2. The more famous the forecaster the more wrong their predictions. High profile commentators are expected to provide headlines – consensus views are of no interest, so it’s in their interests to be controversial rather than right.

3. Experts assessed new information differently depending on whether it supported their views or not. They were quick to accept new data that backed up their theories but applied much higher levels of proof to anything which opposed them. So not very expert at all, really.

4. Mostly, experts failed to remember their predictions – they claimed, with hindsight, to predict much more successfully than they actually did. When right crow loudly, when wrong deny everything.

The experts came up with lots of explanations for why they were so bad at prediction including that they were nearly right, but for unforeseen last-minute circumstances that couldn’t possibly have been predicted. Which sort of suggests that prediction is impossible and we shouldn't bother but, strangely, none of them pointed that out.

Best of all they sometimes simply denied they’d got things wrong and then got very cross indeed when shown that they had.

Consistency and Commitment Tendency

Many of these mistakes were predictable. In his speech on The Psychology of Human Misjudgement to Harvard Law School in 1995 Charlie Munger covered a whole range of such issues. High on his list was bias from consistency and commitment tendency:

... the human mind is a lot like the human egg, and the human egg has a shut-off device. When one sperm gets in, it shuts down so the next one can’t get in. The human mind has a big tendency of the same sort.
Applied to investment experts these traits ought to be familiar to us. They’re dealing with a chaotic world in which short-term events are simply unpredictable. Just as political experts can’t really know what the world will throw at them, neither can investment experts. The variables are too great, the possibilities too difficult to judge, the complexities too mindbogglingly complicated to predict.

Judgement Driven Experts Aren’t Very Expert

Unfortunately we’re trained to listen and believe in experts. If you’ve got a gut ache you go to the doctor. If you need to sue someone you go to a lawyer. And if we want to invest we go to an investment advisor. Only an investment advisor isn’t a proof driven expert like a doctor or a lawyer. They’re a judgement driven expert like a political advisor.

Let’s simplify this. They’re guessing. With our money.

Political advice is important mainly when given to politicians who, heaven help us, we hope are well enough trained to tell the difference between fact and judgement. Well, some of the time, anyway. Hopefully.

Investment advice, though, is doled out to us, the masses, who aren’t trained at all. Mostly we can’t tell the difference between someone who really knows what they’re talking about and someone who doesn’t. Usually the experts can’t either.

Cautious Experts are Best but the Right Experts may be Wrong

Tetlock also showed that the best judgement was exhibited by the experts who were least certain. They drew on lots of different data sources, weighed their evidence carefully and came to cautious conclusions. Unfortunately these are likely to be the least popular experts in public circles. The media doesn’t like caution, it deals in certainty and the fact that yesterday’s certainty is different from today’s matters not at all. So the most prominent investment commentators are the least cautious – and likely to be the most wrong.

Even more difficult is that experts may be wrong today and right tomorrow. Tetlock has shown that what appear at one stage to be incorrect judgements may turn out to be correct – so, for example, an expert who suggested that the Soviet Union would implode by 1985 looked stupid in 1986 but brilliant in 1990. Similarly you’d have looked an idiot for recommending Amazon in 2000, but pretty smart in 2005.

In reverse, Tetlock makes the point in a historical context:

We should recall that the same Winston Churchill who, as early as 1933, was uncannily correct about Hitler’s aspirations attributed almost equally malign motives to Gandhi and his independence movement. In short, good judgement may sometimes be the product of fortuitous coincidences of slowly changing preconceptions in a rapidly changing world.
Or, to paraphrase, even a stopped clock is right twice a day.

Trust the Crowd Not the Expert

We can’t rely on the experts who advise us on our investments. It’s not that they don’t know, it’s that they can’t know. We can’t distinguish between the rare real thing who’ll make us rich and the rest who won’t. Meanwhile today’s investment superstar may be tomorrow’s subprime black hole.

Almost the entire world of mass market investment is built on experts but if we can’t rely on their judgements then we need to operate differently. If you stood a better than one in two chance of your doctor getting a diagnosis wrong you wouldn’t take the first opinion offered. Yet as far back as 1906 Francis Galton noted that the averaged guesses of a crowd estimating the weight of an ox gave a result uncannily close to the correct one even though the majority of guesses were way off.

For most of us, in the random world of investment, the wisdom of crowds may be as good as it gets.

(1) Tetlock, Philip: Good Judgement in International Politics: Correspondence and Coherence: Indicators of Good Judgement in World Politics, Thinking: Psychological Perspectives on Reasoning, Judgement and Decision Making, Harman, D. and Macchi, L., John Wiley & Sons, 2003


  1. It's pretty funny that people actually pay active fund managers to do worse than trackers (in aggregate). If the costs were similar, it might be a more testing intellectual battle.

    That said, you and I both pick stocks with some of our money, so we not only trust the experts - we're foolish enough to believe on occasion that we ARE the experts (even if we'd NEVER put it that way to ourselves or our friends).

  2. This article began well but then ended with the "wisdom of crowds". If crowds were wise, we wouldn't have an economic cycle.

  3. This article began well but then ended with the "wisdom of crowds". If crowds were wise, we wouldn't have an economic cycle.

    Fair point: the wisdom of crowds only works when people aren't inherently biased by the same information. So, for instance, when everyone simultaneously decides that there's no price not worth paying for tulips, or dotcom stocks, or whatever, then it breaks down. I touched on this in Contrarianism.

    But you're right, it's a bit of a weak ending.