‘Nudges’ Can’t Solve Poverty — But No One’s Saying They Can

By
Cass Sunstein, behavioral-econ "czar" for the Obama administration

Nudges have been big in recent years. The basic idea, popularized in Cass Sunstein and Richard Thaler’s best seller Nudge: Improving Decisions About Health, Wealth, and Happiness, is that taking human behavioral tendencies into account — particularly those pertaining to the biases and other cognitive shortcomings examined by behavioral economics — can help people make smarter decisions. Both the U.S. and U.K. governments have set up research offices devoted to behavioral economics, albeit with rather mixed results, and “behavioral economics” is itself now a popular buzz-phrase wherever people who want to sound smart gather.

Underneath all this excitement, there’s long been a political debate about the limits of nudges. Some have argued that a focus on nudges, which are usually presented as cheap and apolitical, distracts from broader, more structural problems. Poor people aren’t poor because they haven’t been nudged enough, this argument goes: They’re poor because we have a system that diverts available resources away from them, hanging them out to dry.

Eduardo Porter made a version of this argument in his Economic Scene column for the New York Times earlier this week, and I think he takes it just a bit too far. I’ll explain why.

First, the argument itself:

Yet, though lauded by policy makers as a powerful new tool in the policy kit, the approach poses a risk, too. It fosters a belief that tweaks based on an understanding of people’s psychology could lead to a vastly improved society at little or no cost to taxpayers.

“The insights from behavioral economics are beautiful from a research perspective,” said Eldar Shafir, a professor of psychology at Princeton who is an expert on decision-making and a leading proponent of the behavioral approach to economics. “But its popularity no doubt comes from a combination of lack of funds and political helplessness.”



America’s shortcomings are serious. The United States has the deepest poverty rates among rich countries. It is at the bottom of the pile in terms of infant mortality, obesity and diabetes, and has the highest teenage pregnancy rate among rich members of the Organization for Economic Cooperation and Development, the club of advanced industrial nations. The death rate from drug overdoses among young white adults is now about as high as the death rate from AIDS was at its peak in the mid-1990s.

It’s great to know that there are promising ways to improve society by developing a smarter email or changing the default choice on an application form. But if the question is whether policy makers can cheaply nudge Americans out of destitution onto a path to prosperity, the answer must be no.

But there just aren’t that many people arguing that we can nudge ourselves out of serious poverty. If you polled 100 researchers whose work lies at the intersection of behavioral economics and inequality, I bet 100 would say that nudging isn’t enough. There are much bigger, broader problems with the way things are structured, and everyone acknowledges this.

Plus, certain small tweaks can have big real-world consequences if they’re widely adopted. Take Shafir, the Harvard researcher. In the (wonderful) book he co-authored with Sendhil Mullainathan, Scarcity: Why Having Too Little Means So Much, the two write about a shortcoming of certain job-training programs for welfare recipients that simply wouldn’t have occurred to many people: They often happen at the same time every week, and attendance is mandatory.

There are two problems with this: One is that poor people have unpredictable schedules and are often juggling responsibilities other people can either outsource to paid help or ask friends or family to help with — something might come up that interferes with the job training, in other words. The other is that when humans are stressed out, overworked, and juggling responsibility, we have a tendency to forget stuff. This particular cognitive task falls hardest on the poor.

So it seems like a small tweak to simply structure job-training programs in a more flexible or redundant way — in many cases, it would cost little or nothing. But the idea is that in the long run these insights into human behavior will spread far and wide, and eventually everyone will be aware of them, and the old, behavioral-econ-naïve way of doing things will die out. In this manner, researchers and program administrators can chip away at “small” problems that, in the aggregate, sometimes have massive impacts. For a single poor person, being tossed off of welfare because of a scheduling conflict is a very big deal. For tens of thousands of them, the societal impact can be massive. In the aggregate, these tweaks have a lot of potential.

At the same time, of course, all sorts of other poverty-reduction programs — many of them on a bigger scale — should be attempted, because nudging can only get you so far, and in the long run it would be better if fewer people were forced to seek out welfare benefits in the first place. But still: If nudging became a habit, everything would become a bit easier, and there’s no good reason to think it will distract people from the full scope of the problems we face.