Skip to content

Nudges and Reasoning

Back in what now seems like a previous age, when David Cameron was prime minister, there was quite a lot of attention paid to his so-called ‘nudge unit’. Nudges, named after Thaler and Sunstein’s well-known book, are ways of getting people to make better choices by making these options more salient or less effortful for them. For example, you can (apparently) nudge people to save more for retirement by changing the default option for retirement plans: when the default is a higher proportion of income people save more than when it is lower. Similarly, you can increase the proportion of organ donors by making the system opt out rather than opt in, and you can nudge people to eat healthier by ensuring that fruit, and not crisps or chocolate, is at eye level in the queue for the register in the lunch room.

Sunstein and Thaler promote nudges as both respectful of individual autonomy and as welfare promoting. They advocate ‘libertarian paternalism’: paternalism, because nudges are ways of making it more likely that people act in their own interests, in cases in which they would not otherwise do so, but libertarian because nudges don’t prevent people from acting as they like. If you want to buy crisps and not fruit, go right ahead (the crisps are there if you look). If you want to opt out of being an organ donor, all you have to do is say so. No real constraints are imposed by nudges after all. But nudges are controversial nevertheless.

Many critics don’t accept that nudges respect autonomy; roughly, our capacity to make rational choices. The reason why is succinctly stated by Tamsin Shaw in a recent article in The New York Review of Books:

The behavioral techniques that are being employed by governments and private corporations do not appeal to our reason; they do not seek to persuade us consciously with information and argument. Rather, these techniques change behavior by appealing to our nonrational motivations, our emotional triggers and unconscious biases.

Contrast two ways of getting your friend to save more for her retirement. You might sit down with her and do the maths; showing her how the magic of compound interest will lead to a much more comfortable retirement if she saves just a little more every month. Or you might nudge her, by changing the default option on her form, relying on the disposition she shares with most other human beings to simply accept the default. The first is giving her reasons and is addressed to her as a thinking being. The second bypasses her reason, taking advantage of non-rational dispositions instead. The first therefore respects her autonomy and the second disrespects it. That, at least, has been the worry many critics have expressed.

Thaler and Sunstein’s response to this charge has been to insist that if nudges bypass reasoning, they don’t increase the degree to which our reasoning is bypassed. Something has to be at eye level: if we leave it to the retailers, they will put the stuff on which there are the biggest profit margins rather than the stuff that is best for us. Our reasoning will be bypassed and we will choose worse than we would have. Since our reasoning is going to be bypassed anyway, why not bypass it in a way that promotes our welfare?

I think there is a stronger response available: I think we should deny that (many) nudges bypass reasoning at all. They may bypass conscious reasoning, but conscious reasoning is one small (and often unimportant) aspect of reasoning.

We have the dispositions nudges target for good evolutionary reasons: because they are adaptive responses to recurrent challenges. They evolved to allow us to respond to information. That’s what reasoning does. In fact, conscious reasoning not only works the same way as unconscious reasoning; it is also pervasively dependent on unconscious reasoning (why does a good inference strike you as good? Chances are you can’t articulate why).

Why, for instance, are we disposed to accept the default option? Because being the default carries information. Usually, something is the default because most people find it good enough, or because experts have set it as the default. It makes sense to rely on defaults and not to expend effort trying to second guess the wisdom of the crowds or of experts (in the jargon: it is rational to be satisficers). Of course, sometimes our evolved dispositions lead us astray. But that doesn’t show that they are reasoning mechanisms. All reasoning mechanisms go wrong in some situations, include conscious deliberation.

There may be all sorts of reasons to object to nudges. But we shouldn’t object to them on the grounds that they bypass our reasoning. They don’t, or at least many of them don’t. We do better to focus on who is nudging us and whether we ought to trust them than on the fact of nudging.

Share on