Behavioural economics in public policy

Earlier this year Raj Chetty gave the keynote address at the annual AEA meeting. He discussed the role of behavioural economics for public policy, giving examples of successful nudges such as a change in defaults for retirement saving. Unusually, he took the goal of policy as given and spent his lecture talking about how behavioural economics can help achieve those goals.

Discussions of behavioural work among economists usually revolve around whether the discovered heuristic or deviation from ‘perfect rationality’ is policy relevant. They argue over whether it requires a policy intervention and whether an intervention would increase welfare from the individual’s perspective. Chetty bypasses that discussion to demonstrate how understanding of people’s biases can help the Government to control their actions.

For example, when he talks about changing defaults to encourage saving he takes it as given that it is a worthwhile objective of policy. He’d have no argument from me that improving outcomes for the least cost is worthwhile, but assuming that saving rates are too low begs the question. Why is it that people are saving too little? Are their behavioural biases truly a policy-relevant problem?

By decoupling the normative questions from the positive interventions, Chetty makes economics a tool for effective intervention rather than a tool for improving outcomes and individual welfare. I’m sure that’s very helpful for government analysts who know what they want but it is not the road to better policy. We’ve often talked on this blog about recognising normative assumptions implicit in policy decisions. But we say that to encourage debate about the right framework for considering them, not to sidestep them altogether. Chetty’s approach is undoubtedly useful and may be pragmatic but I can’t help feeling very uncomfortable about its amoral view of the world.