How Oppressive Can A Nudge Be?
Alan Wolfe Argues that Behavioral Economics is the Real Threat to Liberalism
December 17, 2009
Probably the most distinctive innovation in the Obama Administration’s brand of liberalism is its interest in behavioral economics and the power of modest incentives–the “nudge,” as administration official Cass Sunstein calls them–as an instrument of policy. The theory, sometimes called “soft Chicago school” economics because of its association with Sunstein, formerly of the University of Chicago Law School, and Freakonomics co-author Steven Levitt of that university, is that government can guide society gently toward certain results by understanding our sometimes less-than-rational response to choices.
Mark Schmitt, Senior Fellow at the New America
Foundation and Executive Editor of The American
Prospect, offers a response to Alan Wolfe’s
recent New American Contract Policy Paper
critiquing libertarian paternalism. See also:
Wolfe responds to Schmitt.
Others call it “libertarian paternalism,” and the eminent political scientist and scholar of liberalism Alan Wolfe recently published a paper through the New America Foundation’s Next Social Contract Initiative (which I helped to launch, before coming to the Prospect), that calls it something much worse–incipient totalitarianism, if the Bolshevik allusion in his title, “All Power to the Choice Architects,” is to be taken seriously. Wolfe warns that while “conservatism may not offer much of a challenge to liberalism at the moment,” there are other threats, first among them the ideas about policy derived from behavioral economics, which “deny the key liberal idea that human beings are free to live their lives in ways they collectively decide for themselves.”
I generally agree that using subtle nudges to encourage worthy behavior is not the ideal way for a government to operate. From both a liberal and a mildly libertarian perspective, it would be preferable to have big, decisive, well-defined programs that fully guarantee key public goods–such as Social Security, defense, national health insurance, or anti-trust regulation–on one side, and a fairly open field for human activity on the other, with the line between public and private, regulated and unregulated domains, fairly obvious and well-guarded.
Further, while academics like Sunstein and Obama economic advisor Austan Goolsbee may be attracted to behavioral incentives as a scholarly matter, their analysis matches all too conveniently with cautious Democratic politicians’ preference for safe, soft compromises between real public action and doing nothing. Democratic politicians have long been far too tempted to try to build progressive government out of an array of modest tax credits to encourage good parenting, voluntarism, economic development, entrepreneurship, and so forth–nudges that rarely have an effect worthy of their cost. At the end of a column in 2007 noting the dozens of tax incentives that littered the campaign platforms of the Democratic presidential candidates, I suggested that it was time for liberals to “put down Freakonomics and read some Keynes.” And Bob Kuttner earlier this year wrote a skeptical article about Sunstein’s “radical minimalism.”
But Wolfe’s critique goes much further, in ways that are thought-provoking but ultimately left me more sympathetic to the use of behavioral nudges than I started. In the most bracing section of his essay, Wolfe attempts to retain some value from the “neoclassical paradigm” challenged by behavioral economics. While acknowledging that neoclassical economics, with its assumptions of rational, fully-informed behavior, “is anything but problem-free,” he notes “one major advantage: it subscribes to the conviction, long identified with the liberal tradition, that individuals are possessed with reason, that they have both the confidence and capacity to figure out many of life’s key questions for themselves.” This is a useful insight–whatever else one might say about neoclassical economics, it reflects a fundamental respect for individuals that is valid in its own right. For that reason, many people are attached to it for ethical or ideological reasons, regardless of its results.
And that’s a reasonable ethical position to take. I certainly have a similar view of politics and democracy. As a political liberal, for example, I might think it’s irrational for a voter to prefer Sarah Palin over Joe Biden to be a heartbeat away from the presidency, but I’m unalterably committed to the assumption that individual voters given good information have to be able to make their own judgments. I’ll assume them to be rational even if they aren’t, just as some economists and economic libertarians do. As Wolfe implies, political liberalism and neoclassical economic liberalism have a shared origin in this ethical assumption about human reason.
But, that said, there are plenty of public choices that even the most committed democrat wouldn’t leave to the assumed-rationality of voters. And the assumptions about rationality of neoclassical economics are indeed “anything but problem-free”–they quite nearly led to the collapse of the world economy barely a year ago! Even purely rational choices can lead to disastrous results at times, and as Keynes understood long before there was anything called “behavioral economics,” our consistent inability to think clearly about risk and especially about uncertainty almost always leads to cycles of enthusiasm and pessimism, prosperity and crash. Wolfe says that the behavioral economists view people as wholly irrational–“rats in a maze”–but in fact they understand that people are irrational in fairly predictable ways: We put more weight on the risk of a loss than on a possible gain, for example, or we tend toward inaction rather than an active choice.
Wolfe adapted his title from the Bolsheviks, but the essay brought to mind a very different bit of Marxian doggerel: The line attributed to Trotsky, “You may not be interested in the dialectic, but the dialectic is interested in you.” Wolfe may not be interested in behavioral economics, but that doesn’t mean that behavioral economics doesn’t exist or that the phenomena its researchers have identified have no effect. We could have a world without what Wolfe calls “choice architects,” but it would still be a world in which there are choices and incentives, and someone, somewhere, has made decisions that design those choices.
Countless policies that were created long before behavioral economics came along, as well as countless accidents in the world, have created circumstances that nudge individuals’ behavior strongly in one direction or another. The home mortgage tax credit, for example, was established in the 1950s as a way to support the housing industry and help build a middle-class society; today it is a powerful nudge toward one choice–buying a home–and against both other housing options and other forms of savings. A wise “choice architect” might suggest some modernization to that structure. Or, to take the classic example often cited by Sunstein, Goolsbee and others: Most current voluntary retirement savings plans require people to actively choose to participate. Changing them so that people have to opt-out if they don’t wish to save retirement preserves the choice but has the effect that far more people will elect to save, and thus will be better off in retirement.
Does such a change deny human autonomy? It’s hard to see it. If people retain their choice to save or not to save, but those who don’t give it much thought or don’t feel strongly about it are more likely to do what’s in their own economic interest and that of society as a whole, what harm is done?
Ultimately the question about behavioral incentives has to be what they’re replacing. If government uses behavioral incentives to nudge people in purely private realms of activity, then they certainly can be a threat to liberty. I wouldn’t want government to nudge people toward marriage or childbearing any more than I would want government to mandate those things–even if I thought they were social goods. What about food? It’s a tougher choice–obesity is a major public health problem, and incentives such as a tax on soda might be able to reduce it, but what to eat seems like a choice deeply related to individual dignity. On the other hand, there are already plenty of government incentives that nudge people toward overconsumption–corn subsidies high among them–so a bit of “choice architecture” that would reverse or offset those choices might be wise.
On the other side, there are problems for which we need big government, with its firm mandates, generous subsidies, and structured regulations. We can’t nudge our way to universal health care or economic recovery. There are more dangers in that direction–that we will continue to try to govern through tax incentives and a complex array of inadequate gestures instead of initiatives of a scale adequate to the problems–than that “libertarian paternalism” will encroach on our basic freedoms. But as a complement to other policies, in the space between public and private, understanding just how people make choices–and gently encouraging the obviously better ones–seems like a valuable addition to the arsenal of public policy.