A Third Theory of Paternalism

• Author: Cornell, Nicolas

Introduction

Paternalism is widely understood to have a negative connotation. Proposed laws - on topics ranging from seat belts1 to sodas2 to health insurance3 -are frequently criticized on the grounds that they are paternalistic. Those defending criticized laws or policies often respond by developing nonpaternalistic justifications for the proposed policies.4 Both sides of the political aisle routinely characterize the other side as paternalistic.5 In short, paternalism carries a very negative connotation in legal and public policy discourse.

This Article attempts to answer the question: What's wrong with paternalism? This isn't a rhetorical question meant to prompt reconsideration of whether paternalism is wrong. This Article takes as a starting point the assumption that paternalism is, at least generally, something to be avoided. This is not to say that paternalism is always wrong. But the fact that an action, law, or policy would count as paternalistic is at least a prima facie reason against adopting it- if a paternalistic action, law, or policy is permissible, it is permissible only in spite of its paternalism. The aim of this Article is to explore why paternalism is, in general, a bad thing. The answer to this question is important because it will inform when and to what extent a law or policy's paternalism should count against its adoption.

This Article considers three answers to this question -that is, three theories about what makes something paternalistic and what gives paternalism its normative character. The central thesis is that the two most common explanations for paternalism's character are mistaken. While philosophers and legal scholars have devoted a great deal of thought to the topic of paternalism, they have almost always focused on two elements: coercion and intent. What is wrong with paternalism, the common wisdom says, is either its use of coercion or the impermissible intentions behind it. According to these theories, paternalism is wrong because it prevents us from making our own choices, or because it intends to prevent us from making our own choices.

This Article sketches a third conception of paternalism- one that locates the normative significance of paternalism in neither coercion nor intention. The third - and I believe correct -theory holds that instances of paternalism are objectionable because of their expressive content. Paternalism is suspect because it implies that the other party is not capable of making good judgments for herself. While it is sometimes noted that paternalism is insulting, some other characteristic is usually taken to explain paternalism's impermissibility. The expressive account rejects this move. According to the expressive view, paternalism is objectionable because it constitutes an insult, and that is all that needs to be said. Focusing on the expressive content allows us to make sense of certain charges of paternalism and also to see where those charges are inappropriate. In particular, it highlights the way that paternalism's permissibility depends on various contextual factors.

This new understanding of paternalism's impermissibility has a significant bearing on policy. Recent high-profile policy proposals - most notably the Affordable Care Act- raised important questions about paternalism. But the ensuing debate can feel like ships passing in the night, with one side fixating on a visceral objection6 and the other side emphasizing more tangible consequences.7 Because both sides lack a coherent conceptual framework for thinking about paternalism, there is little meaningful engagement. My hope is that by clarifying the concepts underlying these arguments, we can begin to bridge this gap. This paper seeks to provide some framework for determining whether something like the Affordable Care Act or any other proposal is paternalistic in a problematic way. It attempts to point the critical lens at the appropriate set of considerations, many of which are often not appreciated.

Methodologically, this Article employs the tool of normative ethics and builds on and responds to recent philosophical work on paternalism.8 The goal of this philosophical project - a proper understanding of the normative significance of paternalism- is essential to clarifying our legal and policy discourse, in which paternalism serves as a routine bugaboo. Thus, I describe ways throughout that the abstract question - what's wrong with paternalism -has substantive practical implications.

The Article proceeds in the following way. Part I examines the view that paternalism is negatively charged because it involves coercion. In particular, I focus on the recent position of libertarian paternalism, which suggests that paternalism that is not coercive is less objectionable. Part II considers the view that paternalism is bad because of the intentions that it involves. I offer reasons to doubt that intentions can explain what is normatively significant, both in general and in the particular context of paternalism. Part III describes a third alternative - namely, that paternalism's normative character is based on its expressive content. A brief conclusion follows.

1. Coercion

   The simplest theory of paternalism holds that paternalism is bad because it involves coercion. What is wrong with paternalism, according to this view, is that it denies individuals the ability to make choices about their own lives. This view underwrites recent calls for so-called "libertarian paternalism." In the next Section, I examine this line of argument. After reviewing a common objection, I suggest that libertarian paternalism's real difficulty is that paternalism isn't necessarily wrong because it is coercive. Coercion, I argue, is not the touchstone for objectionable paternalism.

   1. Libertarian Paternalism

      The view that paternalism is bad because it involves coercion provides the theoretical framework for Professors Thaler and Sunstein's widely acclaimed book, Nudge,9 and for subsequent behavioral economics policy proposals.10 For Thaler and Sunstein, what makes paternalism wrong is that it violates the libertarian principle that one should respect individual choices.11 This assumption opens the possibility that noncoercive paternalism is not wrong. Thaler and Sunstein offer a panoply of suggestions as to how noncoercive paternalism can be realized.12 By altering the conditions under which choices are made, we can encourage people to make better choices-thereby improving their lives without coercing them at all. As long as the paternalism is libertarian, it is permissible.13 Thaler and Sunstein's admirably succinct idea and its audaciously wide-reaching applications have received praise and public attention from commentators.14 Legal scholars and public policy experts have begun to internalize the main idea as a versatile tool to support various positions.15 Even Barack Obama and David Cameron have openly embraced Thaler and Sunstein's proposals.16

      The basic idea is elegant: because all sorts of small factors affect the choices that people make, policymakers can greatly improve people's welfare by improving people's choices. In other words, Thaler and Sunstein explain an empirical proposition- that seemingly trivial features of our situation when we make a choice can alter what choice we make -and then deploy this empirical finding as a policy tool.

      Consider the empirical point first. As Thaler and Sunstein put it, "[S]mall and apparently insignificant details can have major impacts on people's behavior. A good rule of thumb is to assume that 'everything matters.' "17 The idea here isn't anything new. One might say that Thaler and Sunstein are just summarizing the main insight of behavioral psychologists and economists. But the collection of various examples, presented all at once as an armada of human decisional failings, is impressive. Seemingly irrelevant suggestions will alter our assessments - writing down your phone number will affect your guess about what year Attila the Hun sacked Europe.18 We are heavily influenced by peer pressure - given simple factual questions that they would invariably get right on their own, people will give the wrong answer surprisingly often if surrounded by others giving the wrong answer.19 We unreflectively choose available options -people will eat vastly more of something they acknowledge isn't very good, like stale popcorn, merely because it is placed in front of them.20 The list goes on and on. The point is that human choices are influenced by all sorts of small, seemingly irrelevant details.

      If one believes that people could make better choices in many circumstances, then creating more favorable conditions can induce people to make choices that will ultimately make them better off. The big idea is that by slightly changing the structures within which people make choices, we can alter what choices they make. Thaler and Sunstein offer the example of a cafeteria, which can encourage diners to eat healthier foods based on where in the cafeteria different foods are displayed (first or last, eye-level or off to the side, etc.).21 They refer to such subtle influencing as making changes in the "choice architecture"22- or, for short, "nudging."23 Of course, not everyone's choices will be altered, but even small changes in the choice architecture can induce significant shifts in distribution of choices by a population. And, if leveraged properly, this means that whoever determines the choice architecture can induce more people to make good choices, without shoving anything down their throats.

      This last point is important to Thaler and Sunstein. Nudges aren't shoves- they don't require anything.24 That is, changing the choice architecture isn't coercive in the way that simply mandating or prohibiting particular behavior is. It is gentler. Although Thaler and Sunstein concede that government intervention to try to make people better off is paternalistic,25 this...