PsyFi Search

Sunday 18 December 2011

HONTI #6: Distrust the Experts

Rule #6: Don't take experts at face value: check their knowledge and their results.

Trust Needs To Be Earned, Not Assumed

In a complex world we often have no choice but to rely on experts to help guide us. If we’re going to lean on such people, though, we really ought to make sure that they know what they’re talking about. This is certainly true for financial experts, but is equally true in other areas as well: relying on the first medical opinion you get isn’t generally a smart move, either.

It seems there are certain areas in which we actually default to trusting our advisors rather than distrusting them - doctors, teachers, attorneys and so on.  Often this trust is justified, but sometimes it isn't - and when this happens we lose out. It turns out that a default of not trusting our advisors is the safest approach, even if that feels ethically dubious1. We are, on the whole, inclined to trust people, but mistaking an advisor for a friend is a risky strategy.

When we start looking at areas characterised by uncertainty the experts are particularly good at making predictions and equally fine at getting them wrong. In fact, the more fond they are of publicly pronouncing on the future the more likely they are to be in error. Famously Philip Tetlock looked at expert political judgement and determined that most of the experts get things wrong slightly more often than the average punter in the street2. This is probably because the experts feel that they can’t sit on the fence when decisions are particularly tricky, whereas most of us are quite happy to admit we’re not sure some of the time.

On the other hand, it may be because the experts are unwarrantedly optimistic about their abilities. As we saw when looking at confirmation bias3, experts are less inclined to seek out disconfirming expertise, and so are less likely to spot problems. Either way it doesn’t give you much confidence that financial experts have any real expertise to offer.  In fact the more popular the expert the less reliable may be the results, because professionals may be trading on the affect of the experts' pronunciations on naive investors. Studies of the so-called Cramer bounce – the tendency of stocks to rise after the eponymous pundit has recommended them on TV – shows that these shares offer average returns, and suggests that smart people are exploiting dumb money, as the price reverses after an immediate rush of interest4.

You might think the people carrying out these studies would be better placed to provide expert advice.  Well, they're quite good at offering advice, but it turns out they're not much good at following it themselves. Economics professors may understand all the issues associated with investing, and the low level of probability that they can beat the markets, but it doesn’t stop them trying. They are just as irrational and overconfident about their personal abilities as the rest of us: more knowledge doesn’t make them better investors5.

Paying for independent financial advice isn't a panacea either.  Financial advisors are  known to be compromised by the differential rewards on offer from various parties anxious to secure their recommendations. The recommended approach to dealing with this is to make them declare any conflicts of interest, but this only results in the recommendations being even more biased. Presumably the public statement of personal conflict makes them feel they no longer need to exercise their own judgement. Of course, given that most people are paying them for their expertise we might question that approach, but most likely this is unconscious, they’re not even aware of it6.

An alternative to using a financial advisor or investing for ourselves is to use a fund manager. Active fund management is a way of getting access to people steeped in the markets, who claim to be able to provide excess returns. The evidence, however, shows that the majority of them can’t do this, and most fail to beat the market. This is partly due to fees, which immediately reduce investor returns, but most active funds have poor long-term records when it comes to outperforming7. Mean reversion effects loom high in this area – funds that have done well over the last few years tend to attract investor money and subsequently perform poorly: finding ways of investing the new funds is increasingly difficult. The tendency of na├»ve investors is to chase top-performing funds, which strongly implies that they’re relying on historical figures on fund returns – which, in turn, encourages funds to find ways of artificially inflating returns8.

You might think that having the fund managers themselves managed by professionals would help – over 50% of US investment monies are controlled by institutions such as pension schemes, who hire and fire fund managers with whom they place their money. Of course, as you might expect by now, they’re very bad at this, basing decisions on past performance, which usually mean reverts after they’ve made a decision. So poorly performing funds get ditched and then outperform and top-rated funds get selected and promptly crash and burn9.

The standard answer to these problems, of course, is to invest in index funds, which simply aim to track the markets. If we can’t beat the markets ourselves then we can at least ensure we don’t lose money against them. A slightly cleverer strategy is to invest across a range of index funds – this will ensure that we don’t get too badly hurt if one of the markets we invest in goes badly wrong. Similarly, we should spread our investments across asset classes, so that if stockmarkets swoon we can at least hope that  government bonds, or commodities, or real estate, or something, holds up10.

This isn’t a bad answer, as it stands, but it’s not entirely complete either, because not all index funds are the same. There's no real regulated definition of an index trackers so a whole industry has arisen to create new indexes to track against and to finding new ways of tracking the indexes. The latter problem is particularly severe, because so-called synthetic index trackers may actually not track the index at all, but rely on some complex set of derivatives: and when markets go badly wrong you can’t always rely on these to pay out11.

Of course, we could learn to become expert investors ourselves. Unfortunately most of us aren’t very good at investing for all the reasons we’ve previously discussed in this series. We have a tendency to believe we’re above average and the normal rules don’t apply to us. However, when we looked at people who are genuinely above average in intelligence it turned out they weren’t smart enough to pick the cheapest index tracker out of a set of four12. In fact a rule of thumb is that it takes about 10,000 hours of concentrated practice to become an expert in any given domain. Putting this into the context of investing, if we can devote an hour a day to it, in between all the other tasks making up a busy life, then it’d take about 28 years to achieve expert status. Whichever way you cut it, that’s a lot of our lives lost in balance sheets and stress13.

So, in the end, we need to take some short-cuts: we may not be able to attain expert status but we need to take steps to ensure we’re sufficiently clued up to challenge what we’re being told. If we’re told we’re suffering from a potentially fatal medical condition the natural reaction is to defer to the expert in front of us. Most people, when asked if they want a second opinion, refuse it. This is a mistake. Most people, when asked for a decision about what treatment to have make it instantly. This is also a mistake.  The correct policy is to take a deep breath, go home and do some research, and then go and get a second opinion. And a third, if necessary. And be prepared to challenge what you're being told, however much against the grain it may be to dispute with experts.  It's your life, money, children or liberty at stake.

This isn’t going to require 10,000 hours of practise, because there’s no attempt here to become an expert. We simply need to do a good enough job and then go and get on with the rest of our lives. Of course, if we absolutely have to invest our own money then we should expect to put a lot more effort in. The fact that it’s easy to invest in shares doesn’t mean that investing in shares is easy. If it were then no one would need to work for a living.

Notes to the article:
  1. See Deciding To Distrust by Iris Bohnet and Stephan Meier.  As they point out there were significant concerns about corporations like Worldcom and Enron before they went bust - but people and regulators preferred to trust rather than distrust.  Quite a lot of behavioral biases are implicated in this, including framing and over-optimism.
  2. As discussed in You Can't Trust the Experts with Your Investments.  Tetlock draws on Isaiah Berlin's essay The Hedgehog and the Fox.  The hedgehog is the expert, who knows one big thing, the fox is the thinker who knows lots of little things.  The fox is adaptable, the hedgehog is not. We're on the side of the foxes. Tetlock wrote a book about this Expert Political Judgment: How Good Is It? How Can We Know? which sounds like a bore, and is anything but.
  3. As we saw in Disconfirm, Disconfirm, Disconfirm.  
  4. We covered this in Money Matters, Your Opinion Doesn't. The research quoted contains one of my favourite examples of where rational researchers run into irrational investors and stagger off, unable to comprehend the reality of the world: "“We also find little evidence that Cramer has skill in selecting stocks, so it is unclear why the market responds at all to his recommendations.”
  5. This striking evidence that knowing about behavioral bias doesn't stop you falling prey to it was covered in Intelligence Can Seriously Damage Your Wealth.
  6. The problems with disclosure as a mechanism for removing bias were discussed in Disclosure Won't Stop a Conflicted Advisor. The really puzzling question is why anyone ever thought this was a reasonable approach in the first place: what's the point is finding an advisor if all they're going to do is tell you to make your own mind up? 
  7. That's when the underperforming active funds don't mysteriously vanish: see Jack Bogle and the Bogleheads, or Survivorship Bias in Magical Mutual Funds.
  8. The research by Friesen and Sapp for instance, quoted in Buy and Hold: The Least Worst Option, which suggests that investors drop about 1.5% in returns annually through market timing efforts.  More startlingly, the best performing funds have the worst performing investors, and we can't blame the experts for that.
  9. This research, discussed in Seeking Alpha, Finding Omega, is perhaps even more extraordinary than what's preceded it. Basically the nation's money is being managed by people who don't have any skill in doing so whatsoever .
  10. Read A Random Walk Down Wall Street by Burton Malkeil.  Even if you don't intend to invest in index trackers.
  11. Not all index trackers are what they appear to be: Exotic ETFs are Toxic ETFs.
  12. This research, by Choi, Laibson and Madrian shows that some of the smartest people in America can't figure out that simple index trackers should all ... track an index: also discussed in  Intelligence Can Seriously Damage Your Wealth.
  13. Original research on the 10,000 hour rule came from Anders Ericsson and colleagues in The Making of An Expert. It's been popularised by Malcolm Gladwell's Outliers: The Story of Success although Ericsson's own book on this, The Road to Excellence, covers the specific point in much more detail.

1 comment:

  1. I genuinely believe one of the virtues of my own blog is that I think most people should invest 100% in index funds, including me, and I put my hands up and admit I don't.

    The truth shines!

    Thanks for your great posts in 2012.