Best Selling Book of 2008 push, written by Richard Thaler and Cass Sunstein, helped inspire experimentally tested, psychologically based political work around the world, often developed by “behavioral analysis groups” in or near government. Now, two leading behavioral scientists, Nick Chater and George Loewenstein, have published an academic working paper saying the movement has gone astray. Professors Chater and Loewenstein are scientific advisers to the British Behavior Study Group and blame themselves as much as anyone else for what they now see as mistakes. It is worth paying attention to what they say.
But first, remember the 1971 advertising campaign called “The Crying Indian.” In this powerful television commercial, a Native American swims down a river that is becoming increasingly loaded with debris.
“Some people have a deep respect for the natural beauty that this country once was,” says the voice-over. “And some people don’t. People start pollution. People can stop it.”
The Indian turns to the camera, a single tear rolling down his cheek. But the message wasn’t what it seemed (and not just because the actor’s parents were actually Italian): it was funded by some of the leading food and beverage packaging companies. Advertising directly placed the responsibility on the shoulders of people making selfish choices. It wasn’t the governments that didn’t provide trash cans, and it wasn’t the manufacturers that started producing non-recyclable products. No, you were the problem.
Chater and Loewenstein argue that behavioral scientists naturally tend to look at problems in the same way. Why do people not have enough pension savings? Because they are impatient and find it difficult to save rather than spend. Why are so many greenhouse gases emitted? Because switching to a green tariff for electricity is difficult and tiring. If your problem is primarily that fallible people make the wrong choices, behavioral science is a great solution. If the real problem is not individual, but systemic, then nudges will be limited at best and harmful diversion at worst.
Historians such as Finis Dunaway now argue that the Crying Indian campaign was a deliberate attempt by corporate interests to change the subject. Is behavioral public policy, by accident or design, such a distraction?
If you look at climate change policy, you might think that this is possible. Behavioral scientists themselves are quite clear that nudge is not a real substitute for the price of carbon – Thaler and Sunstein talk about it in push. Politicians, by contrast, have opted to bypass the price of carbon and go straight to a painless nudge. Nudge enthusiast David Cameron, in a speech given shortly before he became prime minister, stated that “the best way to get someone to cut their electricity bill” is to artfully reformat the bill itself.
It is politics as the art of avoiding difficult decisions. No behavioral scientist would suggest that this is nearly enough. However, they must be careful not to become proponents of the One Strange Trick approach to policymaking.
Behavioral science is commendably focused on hard evidence, but even that can backfire. It is much easier to conduct a rapid randomized trial of account reformatting than to evaluate anything systemic. These small quick wins only make sense if they lead us to harder wins, not away from them.
Another problem is that an empirically proven bad policy based on strict behavior can still be bad policy. For example, it has become fashionable to argue that people should be included in the organ donor registry by default because this dramatically increases the number of people registered as donors. But, as Thaler and Sunstein themselves constantly have to explain, this is a bad idea. In most cases, organ donation occurs only after consultation with the grieving family, and bloated donor registries by default do not help families understand what their loved one wanted.
Behavioral science is a great way to find small changes that can make a big difference in behavior. Such tweaks help if changing the behavior itself solves the problem, but that shouldn’t be taken for granted. It is easy to take a perfectly sound behavioral analysis and turn it into an unfortunate policy.
The most successful behavioral public policy has been automatic enrollment in pension savings plans, which has dramatically increased workplace pension participation in the UK. In the hotel and restaurant business, participation has grown from 5% in 2012 to over 50% last year. This is a triumph. However, huge problems remain in the pension system as a whole. Pension participation by the self-employed has collapsed over the past quarter century. Pensions are a clear demonstration of the strengths of behavioral policy as well as its weaknesses.
“We were unwitting accomplices,” write Chater and Loewenstein, “of the forces opposed to the creation of a better society.” This is too harsh on themselves and other behavioral scientists. We would have really great universal pensions, a healthy and healthy population, a low-carbon economy, if we were not distracted by push? Of course not. But behavioral science is too good at making the perfect icing for the political pie; practitioners should never forget the cake itself.
Written and first published in Financial Times May 6, 2022