NUDGE-PROOF: DISTRIBUTIVE JUSTICE AND THE ETHICS OF NUDGING.

• Author: Roberts, Jessica L.
• Position: Book review

THE ETHICS OF INFLUENCE: GOVERNMENT IN THE AGE OF BEHAVIORAL SCIENCE. By Cass R. Sunstein. New York: Cambridge University Press. 2016. Pp. ix, 202. $29.99.

INTRODUCTION

Nudges--factors that influence behavior while maintaining choice--are everywhere. And they can apparently get us to do all kinds of things that we might not ordinarily do, (1) like eat chicken livers, (2) wake up early, (3) or join a Facebook group for scrapbooking enthusiasts. (4)

According to Cass R. Sunstein, (5) one of the world's foremost scholars on nudging, nudges are "choice-preserving, low-cost tools" (p. 5) to influence behavior positively. (6) Nudges harness the power of choice architecture to influence people's decisions. Choice architecture refers to "the background conditions for people's choices" (p. 5). Our decisions do not occur in a vacuum. Take the seemingly simple decision of what to eat for lunch at a restaurant. The ways in which the restaurant presents the available choices--whether printing them on individual menus, listing them on a chalkboard, or having a server describe them--is the choice architecture behind your lunch order.

Not only are nudges ubiquitous, they are also inescapable. Because all choices have context, choice architecture is unavoidable. (7) The restaurant has to put the items on the menu in some order, and that order will affect what people decide to eat (p. 35). Insofar as it influences our behavior, even the weather is arguably a type of choice architecture (p. 35). When crafted intentionally, nudges capitalize on the reality that the circumstances under which people make decisions can have a powerful impact on what they decide. (8)

Yet it was only in the past ten years that governments attempted to realize the power of choice architecture. Ever since, nudging has been hailed as a sort of policy panacea, able to address some of society's stickiest problems-from the obesity epidemic to the energy crisis to the lack of retirement savings. President Obama jumped on the nudge bandwagon, establishing the White House's Social and Behavioral Science Team (p. 8) and issuing an executive order (9) requiring federal agencies to design and implement policies based on insights from behavioral economics and psychology. (Sunstein himself served as administrator of the White House Office of Information and Regulatory Affairs from 2009 to 2012.) The United States was not the first to engage in these efforts. Nudges as policy interventions have made a splash worldwide. The governments of other affluent, industrialized nations--mainly the United Kingdom, Germany, Australia, Denmark, Sweden, Canada, Singapore, Israel, the Netherlands, South Korea, and Mexico--have also embraced nudges to improve the lives of their citizens. (10)

While the Trump Administration may not keep President Obama's team and infrastructure intact, a complete move away from nudges of all kinds seems unlikely and shortsighted. Research shows surprisingly strong bipartisan support for many nudges. (11) Thus, in this era of extreme political divisiveness, nudges may actually represent an opportunity for consensus across party lines, making them more important now than ever as tools for policymaking. Whether you are on the left or on the right, nudges can help you get what you want. Perhaps nudges could offer solutions on such contentious issues as healthcare (12) and climate change. (13)

When used as policy interventions, nudges are based on the core insight of behavioral economics--that human beings do not always act in our own long-term best interests. We are subject to several well-documented cognitive quirks that lead us down the path of less than perfect decisionmaking. (14) Nudges can help people avoid what behavioral economists call "behavioral market failures" and, in the process, advance prosocial goals. (15)

In his latest book on the subject, The Ethics of Influence: Government in the Age of Behavioral Science, Sunstein takes a deeper dive into government efforts to nudge its citizens. Sunstein points out that, while for-profit companies, charities, and even self-help gurus readily employ nudges to influence behavior, when the state adopts these measures it can raise unique ethical concerns (pp. 13-15). So how do we know when a government's attempts to influence its citizens are ethical? As his framework, Sunstein offers four guiding values: (1) welfare, (2) autonomy, (3) dignity, and (4) self-government. (16) He argues that these four values should constrain governments and inform the actions of the ethical state.

Despite its theoretical elegance and flexibility, Sunstein's framework, like nearly all the literature on nudging, relies on one crucial and often unspoken assumption: people are "nudge-able." (17) In other words, they are susceptible to influence. But being nudge-able requires access to meaningful choices and the freedom to make those choices. (18)

Scholars outside the realm of behavioral economics have long recognized that the control we exert over our lives varies from person to person and from context to context, leaving some individuals with far fewer options. Put simply, we are not all equally autonomous. Any number of factors may limit our range of choices. For example, a food label indicating a product contains GMOs does little to affect the decision of a person who cannot read. A grocery store that prominently features healthy foods will not encourage a person to buy more fruits and vegetables if she simply cannot afford them. A pension plan that automatically deducts from a lower-in-come worker's paycheck will funnel her money away from immediate necessities, such as food or rent, and could have little or no ultimate benefit because poverty shortens her life expectancy. While advocates of nudging might point out that she could opt out of the pension, the worker might not be aware of its detrimental impact, or her relative lack of free time and resources could make opting out significantly more challenging for her than an affluent person, a phenomenon known in economics as an "effort tax" (p. 175). Thus, some individuals may make decisions that do not further their long-term interests--not because of cognitive errors and biases but because of structural barriers and social constraints.

Because we are not all equally autonomous, we are not all equally nudge-able. In fact, some populations, particularly the most vulnerable among us, may even be "nudge-proof in certain contexts. For members of these populations, no amount of subtle prodding or thoughtful construction of defaults will get to the desired result. This reality raises an important question: Is it ethical to adopt social policies that threaten to leave behind our citizens who are in the greatest need?

In all of the rich, ongoing discourse surrounding nudges and their ethics, to date no one has taken a close look at their distributional implications. This lack of scholarly attention represents a significant gap in the literature surrounding choice architecture. Failing to explicitly acknowledge that some people are nudge-proof threatens to erase them from the conversation, possibly leading to even greater inequalities in our already economically polarized nation. (19)

To his credit, Sunstein alludes to these concerns at various points throughout the book. (20) He does so, however, as part of his discussion of the four guiding values, particularly autonomy, dignity, and welfare. While Sunstein advocates convincingly for his four guiding values for ethical nudging, this Review argues that a fifth value--distributive justice--is equally important. Thus, it accepts Sunstein's framework and builds off of it to address the distributional implications of nudging. In so doing, I assert that, regardless of their Pareto impacts, nudges that fail to help the neediest citizens or that reduce those citizens' relative welfare are potentially ethically problematic. Importantly, I do not reject nudges outright, nor do I argue that distributive justice should be dispositive. Rather, I seek to encourage would-be nudgers to engage more deeply with distributional concerns.

The argument for considering the role of distributive justice in nudges proceeds in three parts. Part I describes why nudges can be effective and desirable policy interventions. Part II turns to the ethics of nudging, outlining Sunstein's four-value framework and demonstrating how even seemingly ethical nudges can raise distributional concerns. Finally, Part III proposes distributive justice as a fifth value for ethical nudges and explores how consideration of this additional value can lead to nudges that improve the lives of all Americans.

1. WHAT NUDGES ARE AND WHY NUDGES WORK ... SOMETIMES

   Nudges can include any number of interventions, such as default rules, disclosures, and warnings, and they can span a wide range of topic areas, like fuel efficiency, health care, and retirement savings. (21) Although, at first blush, it might seem that almost anything could count as a potential nudge, the term--as used by Sunstein--actually applies only to initiatives that have a particular set of characteristics and that operate in a specific way. Thus, before considering the ethics of nudges and their potential distributive implications, we need a working definition. So what is a nudge and why do nudges work?

   1. Why Nudges are Effective Policy Interventions

      As noted, nudges are "choice-preserving, low-cost tools" (p. 5) that encourage people to make better decisions. Maintaining the freedom to choose is essential to the nudging process. Sunstein explains that "[n]udges steer people in particular directions but also allow them to go their own way" (p. 20). Importantly, nudges are not mandates or bans. (22) They do not offer subsidies, nor do they impose significant costs (p. 21). Nudges must then be both cheap and avoidable. They do not bribe, mandate, or coerce.

      Nudges attempt to steer people in the right...