Tuesday, 19 October 2010

Misunderstanding nudge

This article from Suzy Dean of the Institute of Ideas utterly misunderstands - or maybe understands, but thoroughly misrepresents - the ideas, application, scope and definition of behavioural economics.

Some choice quotes:
If we consider at the growing influence of ‘nudge theory’ within political circles, the tendency to explain social ills through the individual and widespread hostility towards lower class drinking, we can see that price fixing is symptomatic of the contempt politicians and other elites have for the public and their behaviour.
...Thaler and Sunstein argue the state can prompt people to behave a certain way by creating a ‘choice architecture’ which enables people to make the right decisions. The government’s ‘Nudge Unit’ has been tasked with public health issues including obesity and alcohol intake.
Yet the problem with nudge is evident in the decision to place a minimum cost per unit on alcohol. It skips the debate about whether it is the government’s business in the first place to prevent the public from adopting what are considered bad habits. Nudge assumes we are drinking too much and that a reduction in alcohol consumption is desirable. Based on the idea that government knows best, policy focuses on modifying the behaviour of the citizens without them knowing.
Where do I start?
  1. "Nudge theory" is not an attempt to explain social ills by blaming lower class drinking. It is a set of experimental results showing certain specific, statistically predictable, patterns of human behaviour and proposing some simple ways to respond to those patterns. IoI writers do, however, often enjoy projecting their own concerns onto other people's motives.
  2. Nudge does not skip the debate about the government's business. Nudge is a tool - just like making laws, raising taxes or imposing price regulations - which can be used after the government (or the people, democratically) decide what the government's business is. We've just had an election which was largely won by parties promising to reduce alcohol use. The winners are surely entitled to use what they see as the best policy tools to achieve that goal.
  3. Nudge policy might indeed modify the behaviour of citizens without them knowing. But minimum alcohol prices are the exact opposite of that. How can anyone's behaviour be modified by higher prices if they don't know about the higher prices?
But the most important mistake is this: a minimum alcohol price is not remotely related to behavioural economics. Not even close. This policy comes from a completely different field of economics. Called "economics".

Price changes are the most fundamental tool of conventional economics. They are nothing to do with the behavioural field at all. In fact, this is possibly the least behavioural government policy proposed in recent years. All of the reasons why minimum alcohol prices are bad (and there are many) arise directly from conventional economics. Behavioural has simply nothing to do with this policy and Dean's article is one of the weirdest bits of analysis I have seen in a long time.

The logic of the argument appears to be something like this:

  • Government meddling is bad
  • Governments sometimes use nudge policies
  • Therefore, nudge is also bad
  • Therefore, all government meddling is nudging
Fortunately the IoI is a big fan of classical education, so no doubt they will be sending Suzy on a course in basic formal logic soon.

No comments: