The uses of irrationality

Robin van Persie and Paul Sagar have something in common. Both have reminded me that cognitive biases can be useful.
Asked if he thought Arsenal could win something this season, RvP replied: “We have to believe. If you don’t believe you might as well stop playing football.”
This draws attention to the fact that overconfidence has beneficial effects. The player who over-rates his chances of success is more likely to train well than one who thinks he is fighting a lost cause - and he is, therefore, more likely to succeed.
A similar thing happens with entrepreneurs. Only people who over-rate their chances of success set up in business. Without such overconfidence, we’d have a sclerotic economy.
In the same vein, Paul points out that the halo effect is useful because it creates social capital.
I’ll throw in two other examples of the uses of irrationality.  Mental accounting - thinking of money as being non-fungible - can be a way of increasing our self-discipline and helping us save.
And the availability heuristic can be a useful way of assessing probabilities. If we want to know what are the chances of West Ham winning a game, the fact that their many defeats spring readily to mind tells us these are slim.
This is why cognitive biases are not always mere simple errors. In fact, they are often correct (as with the availability heuristic) or even (as with overconfidence) self-fulfilling. And if they don't directly benefit the individual, they might have group benefits, as Paul suggests.
This, though, means that the Econ 101 view is plain wrong. This (am I hitting a straw man here?) says that irrationality will eventually die out, as it loses the evolutionary race to smarter folk.
But it ain’t necessarily so. Evolution might even select in favour of some cognitive biases - I’m thinking especially of overconfidence. This is precisely why they are so persistent and pervasive. Which is why we should at least be alert to them.
Another thing: RvP seems to be hinting that self-deception also has its merits, which in turn is hinting at a multiple self conception. Anyone who thinks such sophisticated reasoning is improbable in a footballer is committing the category error of mistaking a Dutch player for an English one. 

Comments

[Luis Enrique]

asserting "arsenal has a non-zero probability of winning something this season" isn't what I call evidence of bias

[Luis Enrique]

does even the most unadorned Econ 101 say irrationality will eventually die out?

in the sense of A>B B>C but C>A irrationality? no, if people had that sort of irrationality, econ does not predict their fate.

In the sense that competitive firms making bad decisions will be beaten by firms making better decisions. Maybe, but that hardly says much and doesn't say anything about rationality as such. And isn't 'plain wrong' either.

does econ 101 say anything about the survival prospects of individuals exhibiting 'irrational' traits? No. Few people fail to reproduce because they slightly over estimate their chances of succeeding in certain projects. What is 'plain wrong' exactly?

[Phil Ruse]

Ha - brilliant last line!

[Niklas Smith]

"This, though, means that the Econ 101 view is plain wrong. This (am I hitting a straw man here?) says that irrationality will eventually die out, as it loses the evolutionary race to smarter folk."

Yes, you are hitting a straw man!

The "Econ 101" view (if you mean introductory undergraduate economics) is that economic actors cannot be assumed to be rational, but that modelling them as rational decisionmakers USUALLY produces correct (or at least useful) results. And I've never heard evolution referred to throughout my economics diploma course!

Otherwise a good post. Nice to see that evolution has equipped us with useful cognitive shortcuts (which is basically what these biases are - they save us the time and energy required to make an exhaustive rational deliberation before every decision).

[john Terry's Mum]

This idea is major plank of the Matthew Syed book "Bounce". Syed shows many examples of top performers having to consciously visualise and convince themselves of unrealistic outcomes in order to create these outcomes.

RvP got the idea of believing in belief from Wenger who has mentioned it explicitly many times. However, Wenger himself shows how easily this tendency can go too far and tip into delusion bordering on mental illness; he clearly believes that many of his players are better than they are and this irrationality causes him to neglect dull but practical steps such as getting a proper GK, CB and DM.

5 years of trophyless reality has still not dented his consciously created belief that the intermittent quality of Arsenal's attacking play can overcome the fact that he won't/can't teach them to defend.

He reminds me of someone who has read "The Secret".

[Sean]

See Robert Frank's textbook "Microeconomics and Behavior". Lots of talk about biases, altruism--i.e. non-homo-economicus stuff. And no apparent implication that it will be eliminated.

It's quite popular, too. LSE uses it for intro econ.

[rogerh]

Perhaps today's university students suffer from overconfidence - that they will get a job paying enough for the loan and a home.

Were I an international employer I think I would be looking to buy brainpower on the global market. The upper quartile being best and plentiful if I look at the global population.

[stephen]

From a French general in the Peninsular war, reporting back to Napoleon:

"At this point, by the brilliance of our manoeuvres we had in effect won the battle. Unfortunately the English were too stupid to understand this, and remained in positions from which they should logically have withdrawn, and from which we were not able to remove them."

Moral: there is more than one kind of overconfidence.