Thursday, 19 May 2011

Behavioural economics is not economics (yet)

Economics is useful not when it makes broad, one-sided assertions (markets are good; externalities are bad) but when it uses a model to give a firm, quantitative answer to a specific question.

What difference will it make to the price in the milk market if one new supplier enters, producing 3 million litres a year? If I buy this piece of land and build a gas station, will I make a profit? What's the optimal price for me to sell my consulting services if I value my free time at £40/hour?

Standard economics gives us models which - if we can find the right data to calibrate them - will answer all these questions for us.

Behavioural economics does not.

It can show us the existence of certain phenomena: hyperbolic discounting, framing and priming biases, misperception of risk, anchoring.

But can it answer questions like these?

  • How high a price should I post in my shop window to maximise the benefits of anchoring and minimise the number of customers who are put off from entering the shop in the first place? Or more precisely, to maximise my profits when these two factors are considered.
  • What win probability should I offer on a lottery to get the highest profit?
  • What will be the effect on productivity of my staff if I replace their individual performance bonuses with a team-level bonus?
At the moment, it can't. This is the direction in which the discipline needs to travel. Until it can - until the quantitative models are developed and empirically tested to allow quantitative predictions to be made - behavioural economics will remain a branch of psychology, not a branch of economics.

1 comment:

PunditusMaximus said...

Behavioral economics won't be real economics until it provides a transparent cover for financial elites to loot the system. :)