Incentives: Do they work?

​Read online​

I've been critical of incentives in the past, citing examples where they backfire.

But this¹ incredible 2016 study altered my opinion.

Across 63 schools, 10,649 pupils in their final year of GCSEs took part in an experiment.

The students were split into three groups:

1. Control group: these students received no financial incentives
2. Financial rewards: would receive to £320 for attendance, behaviour, classwork and homework
3. Non-financial rewards: rewarded with the chance to attend events that were chosen by student representatives

Classic economists would assume the financial rewards would work for all students.

Behavioural scientists might argue that the non-financial rewards would work best.

But that's not what happened.

It turned out that the incentives' impact depended on the circumstances of the pupils.

The rewards had next to no impact on those pupils who were already expected to do well. It seemed that these pupils didn't need an extra incentive.

However, for 50% of pupils, the effects of both the financial and the non-financial rewards were significant (with slightly bigger effects for financial rewards).

The effects were especially big for those from lower-income groups.

As Owain Service writes in Think Small²:

Financial incentives do work, but only if they're targeted at those that'll benefit most from the intervention.

Don't ignore the power of financial incentives (like I did) because they won't work all the time; instead, accept that they work if targeted correctly.

Interested in more behavioural science?

The fantastic Richard Shotton has started the AstroHacks newsletter, taking one behavioural bias, explaining it with studies and showcasing it with an example. It's one of the few newsletters I open every time.

​Subscribe today.​

What do you think? — Phill​