qualms about Behavioral Economics

In Sunday’s New York Times, Katrin Benhold describes how the current UK Government has embraced “behavioral economics.” The Cameron Government has been influenced by Richard H. Thaler’s and Cass R. Sunstein’s book Nudge to adopt policies like telling people who are late with their taxes how many of their neighbors have already paid. A government can improve your behavior (citizen) by showing that it knows what you are doing, by sharing that knowledge with your fellow citizens, and by demonstrating your similarity or divergence from the norm.

Another example is Mayor Bloomberg’s ban on huge sodas. You can still buy as much soda as you want in New York, but limiting the individual portions to 16-ounces confronts you with an explicit choice if you decide you want to drink more than that. Proponents say the government can get better results with less force by using such techniques, and they call it “libertarian paternalism.”

Real libertarians are not happy. Sean Collins writes in the libertarian magazine Spiked:

This paternalistic approach changes the relationship between government and citizens. Instead of government representing us, working for us, government now works on us, trying to change our interests. It would be one thing if government sought to convince the public in open debate, but those who would nudge or ban do not want to have open debate or discussions. As the term ‘paternalism’ implies, citizens are essentially treated like children who do not speak; they are only spoken to.

Followers of Michel Foucault are not libertarians, but they should be equally concerned. They should recall Foucault’s discussion of the “panopticon,” Jeremy Bentham’s scheme for a prison designed so that each prisoner can be observed at all times but cannot tell whether he is being watched:

“So it is not necessary to use force to constrain the convict to good behaviour, the madman to calm, the worker to work, the schoolboy to application, the patient to the observation of the regulations. Bentham was surprised that panoptic institutions could be so light: there were no more bars, no more chains, no more heavy locks ….

The Panopticon was also a laboratory; it could be used as a machine to carry out experiments, to alter behaviour, to train or correct individuals. To experiment with medicines and monitor their effects. To try out different punishments on prisoners, according to their crimes and character, and to seek the most effective ones. To teach different techniques simultaneously to the workers, to decide which is the best. ….

Although it arranges power, although it is intended to make it more economic and more effective, it does so not for power itself, nor for the immediate salvation of a threatened society: its aim is to strengthen the social forces – to increase production, to develop the economy, spread education, raise the level of public morality; to increase and multiply.” (Foucault, Discipline & Punish)

In defense of the “nudge” idea, I would say that governments have always influenced how choices are presented. It makes sense to be deliberate about the design of choices. I am fine with making you decide to buy soda 16-oz at a time.

Yet there are good reasons to be skeptical about behavioral economics as a tool of governance. Your overall reaction will depend on what most deeply concerns you. You may think that our main problem is unhealthy or immoral personal behavior–people failing to pay their taxes, for example, or drinking 32 ounces of sugary soda at a time. You may, furthermore, believe that to change their behavior by banning or taxing it is often too costly in terms of individual freedom, burdens on the state, or sheer cash. Then it will be appealing to use behavioral economics to influence citizens’ choices, just as it was very tempting to build state prisons according to the principles of Bentham’s panopticon. One guard, very few beatings and executions, yet everyone behaves.

A different stance begins with the idea that modernity poses a threat to the human being as an end-in-herself. Modern rationality is means/ends rationality: we constantly develop and refine tools for getting other people to do what we want, whether those tools are laws and surveillance, bureaucratic files, surveys, advertisements of all kinds, payments and rewards, or taxes and penalties. Each of those devices whittles away at people’s capacity to decide for themselves how to live. From that perspective, manipulation is a fundamental problem, worse than obesity or tax-evasion, and behavioral economics is just the latest and most sophisticated version of it.

Benhold uses the verb “manipulate” in her basic description of the British behavioral economic policies:

Manipulating behavior is old hat in the private sector, where advertisers and companies have been nudging consumers for decades. Just think of strategically placed chocolate bars at the checkout counter. But in public policy, nudge proponents study human behavior to try to figure out why people sometimes make choices that they themselves would consider poor. Then they test small changes in how those choices are presented, to see whether people can be steered toward better decisions — like putting apples, not chocolate bars, at eye level in school cafeterias.

It is better to eat apples than chocolate bars. And it is appropriate for a school to shape students’ behavior. But the classical republican ideal is that no one may influence your thoughts and actions without giving you an explicit justification, and you must have the right to respond if you don’t agree. No one can say, “Do this because I say so.” Your response to being coerced may be as modest as voting against the people who are trying to regulate you, but the exchange of reasons (on their part) and actions (on yours) respects your dignity. Moreover, to the greatest extent possible, the citizens of a republican regime must decide how to constrain and improve themselves and create their own norms. Each is accountable to the others, and nobody manipulates us.

In We Are the Ones We Have Been Waiting For (pp. 60-2), I discuss more democratic ways to change our behavior:

Human beings are distinctive because we can have ordinary desires plus desires concerning our desires. For example, I may want to put down a difficult scholarly book that I am reading so that I can watch a trivial television show, and at the same time want not to have that desire. … If I turn myself into someone who enjoys scholarly books more and trivial TV shows less, I am not only entitled to believe that I have done the right thing with my time, but that I have also improved myself. In that way, the self (personal identity) is connected to second-order volitions.

I introduce this concept here because we are capable of assessing and altering our own second-order volitions in ways that produce conscious development, not just random change. In the words of the Port Huron Statement that inaugurated the New Left in America, we “have unrealized potential for self-cultivation, self-direction, self-understanding, and creativity.” The Statement proceeds to note that this process of self-cultivation is not individualistic, on the model of a Romantic artist developing his or her own genius. “This kind of independence does not mean egoistic individualism—the object is not to have one’s way so much as it is to have a way that is one’s own. . . . Human interdependence is contemporary fact.”

Indeed, most of the effective techniques for improving our second-order volitions are “relational” and collaborative. Religious congregations, Study Circles and other deliberative forums, internally democratic associations, and participatory social movements convene people to decide on what they should want and then to hold one another accountable for changing their identities by changing what they want. …

Meanwhile, as a whole country, we have both desires and second-order volitions. We want to drive our big SUVs to work, and we want to be the kind of country that does not want to do that. Whereas individual consumer choices elicit our ordinary desires, civic acts such as making arguments in public and voting make us think about our second-order volitions. A good law is not a reflection of what we want but of what we think we should want.

Again, I do not rule out the possibility that a democratically elected government might put apples on the lower shelves of school cafeterias, ban 32-oz sodas, or even inform tax scofflaws how many of their neighbors have paid on time. But each of these acts is a potential threat to the dignity of the persons being regulated, and so it requires explicit public discussion and careful review.

I realize, by the way, that I have combined allusions to libertarianism, civic republicanism, and Foucault in this post, even though they represent very different perspectives. But their differences emerge mostly in what they say about how we should govern. Presented with “libertarian paternalism,” I think they would converge on the same hostile response.

(See also “the new manipulative politics: behavioral economics, microtargeting, and the choice confronting Organizing for Action” and “qualms about a bond market for philanthropy” for similar concerns about another popular idea, social investing.)

This entry was posted in civic theory on by .

About Peter

Associate Dean for Research and the Lincoln Filene Professor of Citizenship and Public Affairs at Tufts University's Tisch College of Civic Life. Concerned about civic education, civic engagement, and democratic reform in the United States and elsewhere.