Sunday 30 March 2008

Predictably Irrational review

Because I'm lazy I'm just reposting the review I've bunged on Amazon -

I bought this having seen Dan Ariely speak at LSE recently. He was an engaging speaker and his research sounded interesting.

Having read the book I was left a bit underwhelmed, because I found that I was already familar with both some of the research and a number of the concepts, and was tempted to give it 3 stars. However on reflection that's probably a bit unfair. This is actually a good book for people interested in learning about the field of behavioural economics. It's nicely written with a chatty style, and some of Ariely's research is very interesting.

Just a few snapshots to give you an idea of what this book covers. He looked at subscription packages for The Economist and found that and obviously bad deal led people to choose an option that was like it but obviously better (because it gave them a way to measure the options). In contrast when there were two options that were different but hard to compare they tended to just go for the cheap option.

In a maths test where subjects were given a cash reward based on the number of problems solved and were given an opportunity to cheat, he found that asking them to recall the Ten Commandments ahead of the test appeared to make them less likely to be dishonest.

And in taste tests people prefer Pepsi to Coke when tasting blind, but prefer Coke to Pepsi when they know in advance when they know what they are going to drink. This suggests that we prime ourselves to enjoy something we expect to enjoy.

If this all sounds 'obvious' to you, to some extent you are right (although there are many examples in this field that are counterintuitive). But people like Ariely make the point that although in a 'common sense' way we know that we are easily influenced and 'irrational', policy is often still made with the assumption in mind that we behave as rational self-maximising economic agents.

One point that can be, and often is, made in return is that although behavioural economics is good at describing seemingly irrational behaviour, it is yet to prove itself as a useful resource for designing better policy (although opt-out, rather than opt-in, approaches to both pension saving and organ donation are arguably influenced by behavioural insights). And in fact Ariely's book is at its weakest when he tries to suggest ways that his research findings might inform policy (I'm not surprised that the bank didn't call him back about his credit card idea!).

So for someone such as me interested in policy the book is enjoyable, but a bit limited in value. However if you are new to this field and interested in finding out more this is a good starting place, and you may well find yourself surprised by some of the findings.

No comments: