Paternalism and Psychology: If Individuals' Ability to Make Rational Decisions Is Limited, Wouldn't Their Ability to Make Political Decisions Also Be Limited?
Glaeser, Edward L., Regulation
An increasingly large body of evidence on bounded rationality has led many scholars to question economics' traditional hostility toward paternalism. After all, if individuals have so many cognitive difficulties, then it is surely possible that government intervention can improve welfare.
As Christine Jolls, Cass Sunstein, and Richard Thaler write in a 1998 Stanford Law Review article, "bounded rationality pushes toward a sort of anti-antipaternalism--a skepticism about antipaternalism, but not an affirmative defense of paternalism." Even if the authors stop short of endorsing traditional hard paternalism, such as sin taxes and prohibitions, Jolls, Sunstein, and Thaler are enthusiastic about "soft" or "libertarian" paternalism, where the government engages in "debiasing"--changing default rules and other policies that will change behavior without limiting choice.
But flaws in human cognition should make us more, not less, wary about trusting government decisionmaking. After all, if humans make mistakes in market transactions, then they will make at least as many mistakes in electing representatives, and those representatives will likely make mistakes when policymaking.
While I generally share Jolls, Sunstein, and Thaler's view that soft paternalism is less damaging than hard paternalism and that in many cases some form of paternalism is inevitable, I respectfully disagree with their view that this type of paternalism "should be acceptable to even the most ardent libertarian." Soft paternalism is neither innocuous nor obviously benign.
Hard paternalism in the form of tax rates or bans is easy to monitor and control; soft paternalism is not. Soft paternalism often relies on stigmatizing behavior like smoking, drinking, or homosexuality, and that can lead (and has led) to dislike or hatred of individuals who continue to engage in the disapproved activities. Moreover, soft paternalism will surely increase support for hard paternalism, as it seems to have done in the case of cigarettes.
Finally, persuasion lies at the heart of much of soft paternalism. It is not obvious that we want governments to become more adept at persuading voters or for governments to invest in infrastructure that will support persuasion. Governments have a strong incentive to abuse any persuasion-related infrastructure and use it for their own interests, mostly keeping themselves in power.
THE SUPPLY OF ERROR
Psychology and social science have an enormously rich tradition of showing that individuals are heavily subject to social influence and that errors easily result from external stimuli. The pioneering social psychologist Solomon Asch demonstrated that individuals who have been shown illustrations of two different-sized lines are more likely to report that the shorter line is longer if they are exposed to planted confederates who declare that the shorter line is longer. Asch's basic result has been reproduced hundreds of times throughout the globe with many different types of questions. More generally, there is widespread agreement in the experimental literature that even modest changes in framing can create wildly different results.
Outside of the laboratory, there is also substantial evidence suggesting that suppliers are able to manipulate beliefs. In the legal sphere, competent attorneys are paid well to shape the beliefs of juries. Firms spend large amounts of money on advertising and other forms of belief manipulation. While some of this manipulation can be seen as correcting errors (that is, informing the consumer), not all advertising is strictly informative. In the pre-modern era, false advertising was common (touting the miraculous advantages of patent medicine for example) and presumably firms would not have spent on this unless it was having an effect.
Is there strong evidence that attempts at belief manipulation are successful on a large scale outside of the laboratory? …