Richard Thaler’s very recent Nobel prize in economics will likely not be celebrated much in libertarian circles. We’re not big fans of paternalism, even of the “soft” variety. I briefly explain why in my Libertarian Philosophy in the Real World (chapter 4). A near-final draft of this section is reproduced below.
* * *
Recently, certain scholars have proposed and defended the use of what is known as “soft” or “libertarian” paternalism, most famously by Richard H. Thaler and Cass R. Sunstein in their book Nudge: Improving Decisions About Health, Wealth and Happiness. It is supposedly distinguished from the traditional kind by its requirement that the choice made by the superior authority be ratified by those affected, meaning that they are informed of it and offered an easy way to reverse it. It turns out, however, that most everything that its proponents claim as soft paternalism is entirely consistent with libertarian principles, and thus offers us nothing new.
One well-known suggestion offered by Thaler and Sunstein involves the default option with respect to worker enrollment in an employer’s 401(k) program. In most firms, if employees do not affirmatively request enrollment are they left out of the plan. Alternatively, under soft paternalism, the employer could, in the absence of a response, automatically enroll the employees up to the point where their contributions take maximum advantage of the company’s matching payments. This policy would be coupled with the employees’ right to withdraw from the plan or modify one’s contribution with a couple of mouse clicks.
But, this policy is not paternalism, soft or otherwise, but simply the exercise of a right held by the entity sponsoring the plan. Because the employer is using its resources to fund this benefit, the selection of a default option is rightfully its alone. At most, workers have an enforceable expectation interest of participating in a 401(k) plan, if this was offered as an inducement to accept employment.
It is analogous to the choice facing a person who gratuitously decides to knit sweaters for a few of her co-workers as Christmas gifts. She would be acting benevolently if she makes them in colors she believes the recipients will like, and maliciously if she selects those they will hate, but she is not acting paternalistically either way. Neither choice makes the recipients worse off than if she never knit the garments, so it cannot constitute an objectionable “interference…with another person,” as required by Dworkin’s definition.
Similarly, other purported examples of the soft variety turn out not to be revisionary. It would not be paternalistic under libertarian principles for regulators to employ cost-benefit analysis, assuming that the government has the right to regulate the activity in question, an unlikely concession (see Chapter 5). However, once we move beyond the uncontroversial cases, we are back in the world of old-school, objectionable paternalism, pursuant to which the government requires restaurant owners to post dietary information on their walls, sell soft drinks of less than a certain size, and so on. Accordingly, I agree wholeheartedly with libertarian pundit Will Wilkerson that, “The whiff of paradox in “libertarian paternalism” may have set up hopes for a category-defying revolution, but Nudge is the book where those hopes, and the tiny monster of an idea, prove flightless.”
 Richard H. Thaler et al., Nudge: Improving Decisions About Health, Wealth and Happiness (New Haven: Yale University Press, 2008)
 Will Wilkerson, “Why Opting Out Is No ‘Third Way’: The Perplexing Banality of ‘Libertarian Paternalism,'” Reason.com, October 2008, http://reason.com/archives/2008/09/29/why-opting-out-is-no-third-way.