All Power to the Choice Architects

If we go by election results, and no better guide to our national mood exists, the years of conservative hegemony in the United States are over, at least for the foreseeable future. Not only has the country elected a Democratic president and Congress, but conservative ideas, which once had seemed so innovative, are increasingly stale or non-existent. Meanwhile, the Republican Party seems both directionless and leaderless.

Under such circumstances, a revival of liberalism is very much on the "ideas agenda" for the United States. As that revival unfolds, it is important to keep in mind that, while conservatism may not offer much of a challenge to liberalism at the moment, other threats to it do exist. One of the most important comes from the popularity of ideas associated with evolutionary psychology and behavioral economics, both of which deny the key liberal idea that human beings are free to live their lives in ways they collectively decide for themselves.

Liberalism is a political philosophy based on the premise that autonomy and equality can best be achieved when each goal reinforces the other. Liberals believe that people ought to be in control of their lives. But they are also convinced that no person can lead a life under her own control if others cannot; if my ability to be safe or healthy is blocked because poverty leads other people into crime or creates susceptibility to an infectious disease, I have no real autonomy.

Autonomy and equality emerged as ideals in late eighteenth century Europe in reaction to forces that denied the legitimacy of both: religious orthodoxy and political absolutism. Liberals were not opposed to religion; indeed, one of England' greatest liberal philosophers, John Locke, was a devout Christian. Liberals instead objected to the use of a powerful state apparatus to impose upon people only one right way to live. If you are told that God has planned your life and that disobedience can result in damnation or death, you are neither free nor equal.

Because it separated church and state from the start, the United States never developed an alliance between priest and prince. Even though this country has seen the rise of various forms of conservative religion in recent years, those movements appear to have peaked with the election of Barack Obama in 2008. Liberalism is safe from religious tyranny in a way it never was two hundred years ago in Europe.

At the same time, the liberal goals of autonomy and equality are threatened from another powerful intellectual force. This contemporary threat can best be called scientism. In the long struggle against religious and political oppression, liberal thinking and scientific discovery were frequently linked; both required a commitment to reasoned discourse, open methods of inquiry, conclusions based on empirical data, and skepticism toward dogma. Understood in its broadest sense, science and liberalism remain powerful allies in the contemporary world, especially when conservative administrations shut their eyes to evidence of global warming or approach questions of disease transmission through ideological presuppositions.

But scientism is not the same as science. Scientism refers to the use of scientific language or techniques in order to promote conclusions about human happiness or welfare based on pure speculation. Two forms of scientistic discourse have become especially popular in the United States; as I have already mentioned, they are associated with evolutionary psychology and behavioral economics.

Evolutionary psychology should not be confused with the Darwinian theory of evolution. Darwin's explanation of how species evolve is as accepted a scientific theory as it is possible to have; those calling themselves adherents to intelligent design who spend so much time challenging the teaching of evolution in school are little more than ideologues. It is one thing to argue that species evolve and adapt over hugely long periods of time, however, and another to claim that choices made by human beings in the here-and-now--whether or not to worship, when and whom to marry, whether to break the law, where to live--are the product of our biological or genetic make-up. Human choices are shaped by culture and culture in turn is shaped by human choices. To deny the importance of human culture, or to view culture as just another form of genetic evolution, is to deny us our autonomy.

None of this would matter if adherents to evolutionary psychology just wrote scientific articles for peer reviewed journals. But because the presumed findings of evolutionary theory to everyday life are counterintuitive--who is not surprised to learn that love is a purely biological adaptation?--this way of thinking both influences best-selling writers such as Malcolm Gladwell and produces books from academics that attract widespread readership, such as Daniel Gilbert's Stumbling Toward Happiness or Helen Fisher's Anatomy of Love. Gilbert's book in particular captures the evolutionary zeitgeist. "Each of us," he writes, "is trapped in a place, a time, and a circumstance, and our attempts to use our minds to transcend those boundaries are, more often than not, ineffective." When it comes to thinking on our own, there really is no such thing.

The irony of all this is that while conservative Christians attack the Darwinian theory of evolution, sociobiology's popularizers share with the most right-wing religious believers an identical skepticism toward human purpose. Both believe in predestination. For the one we are stained with sin. For the other we are products of our genetic makeup. In neither case can we change our destiny. If conservative Christians and Darwinian popularizers would stop accusing each other of bad faith, they might discover how much they have in common.

Behavioral economics has also been storming the best-seller lists of late and many of its leading practitioners turn to evolutionary psychology to support their findings. But the scientistic nature of behavioral economics is also illustrated by its reliance on experiments frequently conducted under contrived circumstances using unrepresentative subjects, which are then generalized with abandon toward the human race as a whole. Despite the fact that its findings are wildly speculative, however, behavioral economics considers itself in possession of a revolutionary new way of understanding human behavior. Its conclusions are so important, its practitioners insist, that they offer us bold new approaches to public policy. Because the administration of Barack Obama includes a number of policymakers influenced by this way of thinking, behavioral economics may well emerge in the near future as a major force in shaping decisions over how health care, pensions, and other important social goods are distributed.

To understand the importance of behavioral economics, it is first necessary to say a world about the revolution in economic understanding it seeks to displace. For well over a century, economists have come to accept what is generally called the "neoclassical paradigm." According to this approach, human beings are fundamentally rational creatures who seek to maximize their utility, generally by making choices that offer them the greatest rewards with the least amount of pain. We can, the theory continues, view the price people are willing to pay for something as revealing what their preferences are; if they will pay more for a car, they must really want it, but if they shift to mass transit when cars become too expensive, they have a new preference. The point is that their behavior is predictable and can be modeled using sophisticated mathematical techniques.

The neoclassical paradigm is anything but problem-free. Its vision of human choice reduces all behavior to matters of calculation, producing a vision of human rationality lacking complexity and mystery. Politically, the neoclassical model, especially in its most extreme form, is associated with forms of libertarianism that prefer market-based policies and are skeptical toward any kind of liberal reforms that rely on government. In its most conservative versions, neoclassical economics elevates human rationality but downgrades social rationality: we are free to plan our lives, but the moment we try to use our collective intelligence to plan or guide our society toward ends chosen by ourselves, we are doomed to fail.

Behavioral economists reject the neoclassical assumption that we act out of rational self-interest. The only thing predictable about us, they conclude from the experiments they conduct, is that we are fundamentally irrational. We cannot judge the intrinsic value of the things to which we are attracted. We are miserable failures at understanding the risks posed by hurricanes or illness. We are myopic, choosing for pleasure things in the short run against our long-term interests. We think that medicines or wines that cost more must work or taste better. Given choices, we inevitably make the wrong ones. If we are therefore left free to decide the best course of life for ourselves, our choices may easily prove disastrous ones.

To compensate for our inherent flaws, behavioral economists argue that we must be prevented from acting on our instincts. Their preferred way of doing so is through what economist Richard Thaler and legal theorist (and head of the White House Office of Information and Regulatory Affairs) Cass Sunstein call "libertarian paternalism." As both scholars explained in their best-selling book Nudge, people should neither be left completely free to do what they want nor should they be coerced by government. Instead their choices should be constrained by organizing them according to design principles that will produce the optimum outcome for themselves and their society. Thaler and Sunstein call such design a form of "choice architecture." During the Russian Revolution, the Bolsheviks shouted, "All Power to the Soviets." Under the rules of libertarian paternalism, all power goes to the choice architects. The goal of the behavioral economists is to give people choices over how much to save for their retirement or to recycle their trash, but to design the choices in ways that will overcome the irrationality and ignorance that shape individual decisions.

"Libertarian paternalism," Thaler and Sunstein write, "is neither left nor right, neither Democratic nor Republican." They say this because they are convinced that their approach combines the best of both the free market, long identified with conservatives and libertarians, and government, upon which liberals and Democrats tend to rely. But it many ways it does the exact opposite: behavioral economics offers neither the commitment to reason associated with the free market nor the transparency and accountability that states demand.

Although the neoclassical paradigm reduces us all to utilitarian calculators, it does contain 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 the capacity to figure out many of life's key questions for themselves and to plan their lives accordingly. If we argue instead, as behavioral economists go out of their way to do, that we act like rats in mazes unable to find a sensible path through life's complexities, we are best off avoiding bold ambition and new insight. The underlying vision of human purpose contained in behavioral economics is an impoverished one, far more depressing than the utilitarian calculator associated with the neoclassical paradigm. Instead of recognizing how complex human behavior really is, behavioral economists replace the scientistic idea that we always act rationally with the equally scientistic one that we always act irrationally.

At the very same time, the theory of government associated with behavioral economics is more problematic than the one associated with the regulatory state that developed during the New Deal. Liberals have always understood that government can be a force for good. Since government has coercive powers, liberals have also long accepted that a person cannot be free to do anything he wants at any time. I might have a strong preference to own slaves, but a liberal society will not allow me to, for such a freedom for me would deny freedom to so many others. Less dramatically but still importantly, I might desire to opt out of the social security program and pay for my own old age assistance, but a liberal will recognize that, if I am permitted to take such an action, the principle that no one should face old age without public assistance would be abrogated with consequences too severe for a liberal society to accept. All of this is what distinguishes liberalism from libertarians; the latter believe that all actions by government rely on coercion and therefore diminish our freedom.

Reliance on government, like any social practice, is never perfect; we all know that regulations can become cumbersome, taxes high, and bureaucracies inefficient. Yet government acts in the name of all and is subject to rules of transparency that makes its business, the public's business, open to all. When government regulates, the rules are published, appeals are possible, and changes can be sought. Liberals believe in what has been called "positive freedom," or reliance on the power of government to help people lead fuller lives. At the same time, government must be viewed as the servant rather than the master of public purpose, and this can only be accomplished through what nineteenth century liberals called "publicity" and what we today call "transparency."

If the "libertarian" side of behavioral economics leans toward the market, the "paternalism" side leans toward government. Yet unlike New Deal liberalism, which devoted considerable attention to the form and function of government intervention, behavioral economists such as Thaler and Sunstein lack an account of where, how, and by whom key decisions will be made. Choices, in their view, have an architecture but the policymaking institutions that frame those choices do not. In contrast to the well-worked-out science of public administration that emerged from the New Deal, government in their scheme of things becomes a big black box whose internal workings are as mysterious as the irrational minds all human beings are cursed with possessing.

The fact of the matter is that policy proposals based on the assumptions of behavioral economics have become a growth industry all their own in Washington. The concrete measures designed by the Obama administration to respond to the financial meltdown it inherited from the Bush years, for example, possessed a distinct behavioral twist: the payroll tax cuts in the stimulus plan were spread out in small chunks rather than delivered in one big dose, because when we get a large sum, behavioral economists tell us, we are more tempted to save it rather than spend it, a propensity that would prove counterproductive when the economy needs money in circulation rather than in our pockets. Along similar lines, the financial regulatory plan introduced by the administration in June 2009, widely regarded as far too cautious and solicitous of the very banks that caused the panic in the first place, may have avoided stronger calls for regulation based on the suspicion behavioral economics has toward firm governmental action.

Given such a start, we can expect to see proposals for public policies based on the assumptions of behavioral economics both in Congress and throughout the executive branch. Raising taxes on gasoline will help us conserve energy by nudging us in the direction of mass transit. Computerizing health care records will enable patients to better question the recommendations of their physicians. Mobilizing peer pressure can help teenagers lose weight or smokers to give up their habit. Finding the right incentives can lead mortgage providers and home owners to avoid defaults that would harm both of them. Even Michelle Obama's highly publicized vegetable garden can be interpreted as a hint to vast numbers of Americans that there are alternatives to prepared foods. The idea of trying to change our behavior without raising our taxes is so great that few policy wonks can resist it.

Economists attracted to this way of thinking insist that, because behavioral economists put more information in the hands of ordinary people, it furthers the goal of transparency. "If instead of the 30 pages of unintelligible crap that comes with a mortgage you can upload it with one click to a website that will explain it and help you shop for alternatives," Richard Thaler explained to Time's Michael Grunwald, "you make it as easy as shopping for a hotel." Either Thaler has not shopped for a hotel lately--the web's hotel offerings are bewildering in their complexity--or he is being disingenuous. In fact, approaches based on behavioral economics, true to their conviction that people are fundamentally irrational, are impossible to square with the idea of citizens exercising effective control over those who govern them.

Given how policy is made in Washington, there is every reason to believe that nudging will prove to be opaque to those being nudged. Manipulating tax incentives is always good business for accountants and cumbersome for everyone else; given everything we know about how politics in the United States works, knowledgeable upper-middle class Americans will take advantage of incentives while poorer and time-strapped ones will not. No one can safely predict that the incentives policymakers introduce will turn out to work as they are expected to do; to be sure, more people will invest in old age pensions if they have to take the active step of opting out of them, but in more complex areas of human behavior, the same peer pressure aimed to help young people eat better could also help them become better consumers of dangerous drugs. Most harmful of all, policies along these lines are likely to reinforce invidious distinctions between people; "good" citizens will be defined as those who respond the way the theory predicts they should, while those who stubbornly resist will be viewed as harmful to the health care or retirement system that is being redesigned.

As if these were not problems enough, public policies based on the assumptions of behavioral economists may well play havoc with the American Constitution. According to that document, Congress ought to be the source of legislation. Yet if choice architecture is to be designed to overcome human mistakes, ways must be found to prevent human beings from influencing its design, for if the process were open to pressure from ordinary people or lobbyists, the mistakes that characterize their decision-making would be built directly into the system. As the most democratic branch of government, Congress is most exposed to people's irrationality. If government is to be used to control our myopia, Congress cannot be in charge.

Nor can courts, which are given to insuring that rights are respected and procedures followed, play much of a role if choice architecture is to be designed to achieve its purposes. A good court decision clarifies for the contending parties the nature of any disputes between them. But the choice architecture envisioned by behavioral economists works better under conditions of ambiguity than under conditions of clarity. Its aim is not to present us with options so that we will deliberate among them and choose the best one. Choice architecture works best when, in the terms used by Thaler and Sunstein, our thinking is automatic, unconscious, fast, and effortless rather than deliberative and deductive. Based upon principles of legal reasoning, courts require a reflective cognitive style quite at odds with the automatic system behavioral economists associate with our limited capacity for reason. The ability of courts to correct the abuses made by policymakers would therefore have to be limited under a well functioning system of choice architecture.

This leaves the third branch of government, the executive, as the proper home for choice architecture. Within that branch, moreover, it would surely not be the president himself who would be spending time designed the best choice frameworks; the choice of choice architecture will be delegated out to administrators within the executive branch who would have considerable discretion in designing the systems they think will work best. For eight years liberals tried to prevent the accumulation of power within a "unitary executive" that conservatives in the Bush administration viewed as the best way to keep decisions secret from the people influenced by them. Putting into effect a workable choice architecture would require a similar concentration of power among unelected officials in the executive branch.

In their book Nudge, Thaler and Sunstein never spell out details such as these. This lack of attention to how public policy is actually made, it seems fair to say, is not due to the fact that they forgot to include it. The specifics of who the choice architects are and what functions they perform are not spelled out, because if they were, they could not operate as choice architects. Choice architecture works best when people think they are acting as libertarians even as their behavior is guided by paternalists. If they had more knowledge about who was guiding them, people would be more resistant to being guided. What Thaler and Sunstein call "libertarian paternalism" can also be characterized as "hidden paternalism." People should on the one hand have more choices and on the other less information about how their choices were determined for them.

For all these reasons, neither goal for which liberalism stands--autonomy and liberty--would be promoted by greater reliance on behavioral economics. People's capacity to lead lives of their own choosing would be undermined, because decisions they believed to be their own would actually be the products of behind-the-scenes manipulation. And those who create the architecture responsible for that manipulation, because unelected and uncountable, would have considerably more power than those being manipulated. If it works, libertarian paternalism offers us a third path between the market and the state. If it fails, it leaves us simultaneously as slaves to our impulses and at the mercy of others.

Liberalism can do better than this. Having successfully won their campaigns for autonomy and equality by defeating the forces of religious orthodoxy in the past, liberals ought to be able to defend and advance their ideas in an age of scientism today. Politics and governance are serious matters. When it comes to providing health care or improving the environment or dealing with the effects of devastating recessions and natural disasters, we need to develop public policies based upon a careful understanding of human nature and human purpose. The last thing we should do is to allow ourselves to be attracted by the latest academic fashions, especially when there is little actual empirical support behind them.

Like the Progressives of the early twentieth century, scientism today claims to know what is best for us in spite of what we think might be best for ourselves. The elitism in Progressivism, however, was never well-hidden: reformers in that era were proud to minister to people's souls; uplift was their purpose and improvement their means. The scientistic speculation associated with both evolutionary psychology and behavioral economics, by contrast, in reducing who we are to one thing only, lowers our vision and distorts our purpose; it does not define our choices for us, it dumbs our choices down. Liberals in the age of absolutism showed courage and foresight in fighting the combined forces of church and state. Liberals in the age of Obama need to jettison the faddishness of unproven ideas and the manipulative policies based upon them.