Complex systems and an Oscar-worthy f*** up

Reading Time: 2 minutes

And the winner is … La La Land!!

You might remember this one. Oscars night in 2017. Award for best picture. Warren Beatty and Faye Dunnaway presenting. The entertainment world waiting with baited breath as Beatty hesitated -playing with the audience- as the winner of the best picture was to be announced …

But it all ended in a terrible farce as the La la land was incorrectly called out (the actual winner was Moonlight, announced moments later to excruciating embrassement all round).

But why am I bringing this up? Well the reason this screw up occurred had to do with complexity (this is brilliantly explained by Tim Harford in his excellent podcast series Cautionary Tales) .

One problem with complex systems is that they are more likely to go wrong, and to do so in unexpected ways.

The system of envelopes for the Oscars might not seem complex, but the way it ran in practice actually involved having TWO copies of each winner’s envelope in circulation, plus a complicated series of hand-offs. Suddenly it’s not so hard to see how an error occurred. This was compounded by the unnecessary complexity of what was written on the card. There were a lot of words, but arguably the most important fact – what category the card referred to – was only mentioned in very small print. Which resulted in the error of Faye Dunnaway erroneously reading from the best actress card.

We have a strange relationship with complexity in today’s world. In one sense, many day-to-day items that we take for granted: cars, planes, even the humble toilet contain far more complexity than we can realistically understand fully. We develop ways of being able to trust these things with, in some cases, our lives, or our life savings.

And these complex things have in many, many ways dramatically improved our lives, of course. If we insisted on interacting only with things we could understand fully, we’d lead a pretty limited and boring life, whatever our capacity for understanding difficult things.

But do we go to far? Does our confort with complexity morph into a fetishisation of complexity? A fascination that complex must be better, that something needs to feel complex to be worth paying for, that smart people with long reports and weighty slide-presentations surely must know more than us?

Investing sits at a fascinating point on this issue. There is no one solution to investing and there are complex solutions and there are simpler ones. A lot of time is spent by investors and advisers like me in figuring out where complexity adds something meaningful and where it doesn’t.

I think the bar needs to be high for complexity in investing. As illustrated by the Oscars story, complex solutions go wrong in hard to foresee, and hard-to-diagnose ways. That isn’t to say that simple solutions can’t go wrong, of course they can. But this fact is a good reason why we propose an investment equivalent of occum’s razor – a simpler solution should be preferred to a more complex one if both can deliver the same outcome.


Leave a Reply