I’ve written often about Thaler and Sunstein’s Nudge, which highlights the realm where decision framing can “nudge” individuals toward making better decisions. Since a nudge requires some design – there must be someone to do the nudging – the approach advocated in Nudge is described as a form of libertarian paternalism. The goal is to frame decisions by respecting individuals’ freedom (the libertarian) while encouraging a choice (the paternalism) which has some benefit for the chooser.
Here’s a link to a more in-depth analysis of the notion of paternalism from Cass Sunstein. Sunstein discusses different types of paternalism, nudges, and when paternalism should play a role in choice architecture. A preview of Sunstein’s 5 primary conclusions:
- Behavioral market failures are an important supplement to the standard account of market failures, and in principle, they do justify (ideal) responses, even if those responses are paternalistic. As in the case of standard failures, however, the argument for a government response must be qualified by a recognition that the cure may be worse than the disease, and that all relevant benefits and costs must be taken into account.
- Choice architecture is inevitable, and hence certain influences on choices are also inevitable, whether or not they are intentional or a product of any kind of conscious design.
- Some of the most intuitively appealing objections to paternalism rely on autonomy, but as applied to most efforts to remedy behavioral market failures, those objections lack force, because those efforts do not interfere with autonomy, rightly understood. There is also a risk that some of these autonomy-based objections are rooted in a heuristic for what really matters, which is welfare.
- The most powerful objections to paternalism are welfarist in character. In many contexts, those objections are a good place to start and possibly to end, especially insofar as they emphasize the importance of private learning and the risk of government error. But they depend on normative claims that are complex and highly contested, and on empirical claims that are often false. There is no sufficient abstract or a priori argument against paternalism, whether hard or soft.27
- The welfarist arguments against paternalism, new or old, are irrelevant insofar as choice architecture, and nudges, are inevitable. But insofar as paternalism is optional (and it often is), there is an intelligible rule-consequentialist objection to paternalism — though the strength of the argument depends on the form of paternalism. There are plausible rule-consequentialist arguments against (optional) paternalism, but those arguments depend on strong empirical assumptions, involving extreme optimism about markets and extreme pessimism about public officials, that are unlikely to hold in our world. The objections to paternalism are weakest when it is soft and limited to means; especially in such cases, there are many opportunities for improving welfare without intruding on freedom of choice.
(HT: Miles Kimball)