Mark D. White
In today's New York Times, David Brooks comments on libertarian paternalism in "The Nudge Debate." There is not a lot in his article that is surprising or unreasonable, but it does suffer from some vagueness and misunderstandings. For instance, Mr. Brooks conflates interventions of a paternalistic nature (such as nudging people into retirement plans) and those of a nonpaternalistic nature (such as nudging people into registering for organ donation). While the mechanisms in both cases are similar—and raise the same issues of unconscious manipulation and subversion of rational decision-making processes—the purposes and motivations are very different, with only the former involving the policymakers substituting their interests for those of the decision-makers themselves.
Of more concern is Mr. Brooks' contention that libertarian paternalism does not involve value substitution. He writes,
Do we want government stepping in to protect us from our own mistakes? Many people argue no. This kind of soft paternalism will inevitably slide into a hard paternalism, with government elites manipulating us into doing the sorts of things they want us to do.
As I explain in The Manipulation of Choice, there is no way for the government to know what we value well enough to help us make decisions in our own interests. Because they lack this information, policymakers necessarily impose their idea of people's interests on them when they design nudges. Policymakers think that it's in our interests to save more; policymakers think that it's in our interests to drink less soda. These are not unreasonable assumptions, of course, but they are assumptions nonetheless, and it is pure hubris on the part of policymakers to presume that they bear any necessary relationship to people's actual interests.
Because Mr. Brooks apparently doesn't recognize this, he concedes the "theoretical" point but dismisses any real-world concerns:
I’d say the anti-paternalists win the debate in theory but the libertarian paternalists win it empirically. In theory, it is possible that gentle nudges will turn into intrusive diktats and the nanny state will drain individual responsibility.
But, in practice, it is hard to feel that my decision-making powers have been weakened because when I got my driver’s license enrolling in organ donation was the default option. It’s hard to feel that a cafeteria is insulting my liberty if it puts the healthy fruit in a prominent place and the unhealthy junk food in some faraway corner. It’s hard to feel manipulated if I sign up for a program in which I can make commitments today that automatically increase my charitable giving next year.
This last paragraph is illuminating, because it conflates three different types of nudges. The first, organ donation, is a social issue; such a nudge is not paternalistic and therefore does not raise any issues of value substitution (though, as I said above, the mechanism still subverts rational processes). The third, self-commitment, is vague; there is nothing manipulative in the concept of commitment, but if such commitment is elicited using a nudge that bypasses a person's rational decision-making faculties, then it's a problem. Only the cafeteria example is by definition a paternalistic intervention; Mr. Brooks may not be insulted by the management of the cafeteria putting their idea of his interests above his own and manipulating his actions in those imposed interests, but that does not justify an action which would insult many others.
Finally, I do not see the issue of libertarian paternalism as one of theory versus empirics—in the case of paternalistic interventions, the theory iself discounts any attempts to measure its success. Mr. Brooks finishes the paragraph above with this sentence: "The concrete benefits of these programs, which are empirically verifiable, should trump abstract theoretical objections." In the case of paternalistic interventions, the "theoretical objections" render any "concrete benefits" questionable and inherently unverifiable. How do you measure the "concrete benefits" of an action meant to improve people's choices according to their own interests if you have no way to ascertain those interests? Such knowledge is necessary in order to "verify" any benefits from such a program. With socially-motivated nudges, like automatic enrollment in organ donation programs, this makes some sense, but with measures explicitly intended to "help" people better make decisions in their own interests, the idea of verifying "concrete benefits" makes no sense whatsoever, given the inherent subjectivity of those interests.
Rather than an issue of theory versus evidence, the nudge debate is a matter of autonomy. Each person's right to further his or her own interests, in a way consistent with all others doing the same, is violated by policymakers who impose their own conception of people's interests on them and then design policy tools that subvert people's rational decision-making processes to steer them towards those imposed interests. Given Mr. Brooks' antipathy towards individualism, I am not surprised that he disregards concerns about autonomy as an "abstract theoretical objection." To some, however, the right to pursue their own interests without the government questioning them is a very "concrete benefit" to living in a free society.
Then again, if policymakers really knew our true interests, they'd know that already, wouldn't they?
—in particular the last point in #2 about information—but Zywicki and Smith delve much more deeply and broadly into problems with the CFPB itself, contributing a much needed public choice perspective to the issue and concluding with recommendations to improve the operation of the CFPB. This is an essential read for anyone interested in behavioral law-and-economics or "nudges," regulation, or paternalism in general, as well as the CFPB in particular.