As philosophers have long recognized, human beings are imperfect and often make bad choices that have a negative impact on themselves and other people. Sometimes we recognize our own mistakes, and sometimes other people bring them to our attention. In the best case scenario we learn from our mistakes, but sometimes we need institutions to guide us to better ones, whether those come in the form of mentors to give us advice on how to live better lives or the state enforcing laws to protect citizens from each other.
In recent years, however, the emphasis of these institutions has changed in two significant ways. First, the focus has shifted from choices and their consequences to how we make those choices in the first place. Psychologists and economists have subjected the process of decision-making to new scrutiny, and their innovative and fascinating research revealed quirks and anomalies in how we process information and weigh options. Idealized models of ârational choiceâ are gradually being replaced by new conceptions of decision-making that incorporate empirical evidence of psychological imperfections. Put simply, we seem to know more about why we make bad choices as well as how we might make better ones.
Second, policymakers in government absorbed this new research and began to incorporate it into regulation and policymaking itself. Some uses were oriented toward public policy ends, such as making recycling bins more attractive (to promote proper disposal of waste) and painting flies in urinals (to reduce clean-up costs in public restrooms). 1 But policymakers also began to use behavioral insights to help people make better decisions for themselves, which can be characterized as paternalistic and is therefore subject to disagreement from those who believe a governmentâs proper role is to protect citizens from others, not from themselves.
As Gerald Dworkin defines it, paternalism is âthe interference of a state or an individual with another person, against their will, and defended or motivated by a claim that the person interfered with will be better off or protected from harm.â 2 While this definition covers paternalistic acts by individuals, the focus of this chapter is on paternalistic interventions on the part of the state. The canonical statement against paternalism was provided by John Stuart Mill in what came to be known as the âharm principleâ:
However, the new psychological insights into decision-making have led academics and policymakers to doubt that people regularly make decisions reliably in their own interests, which in turn motivated a more active role of government in helping people pursue those interests. 4The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right.⊠The only part of the conduct of any one, for which he is amenable to society, is that which concerns others. In the part which merely concerns himself, his independence is, of right, absolute. Over himself, over his own body and mind, the individual is sovereign. 3
In this chapter, I will summarize the development, key results, and main applications of the new behavioral science to government paternalism. Then I will offer several criticisms of this new brand of paternalism, focusing on the lack of knowledge among policymakers regarding peopleâs true interests and the nature of the manipulation inherent in such behavioral interventions. Lastly, I will survey private uses of these psychological insights, including private alternatives to government interventions, the topic of many of the chapters in this book.
Behavioral Science and Choice
For decades, mainstream economics enjoyed a broad consensus regarding its model of individual decision-making, known as expected utility maximization or constrained preference-satisfaction. 5 This standard choice model allowed economists to explain and predict typical economic behavior such as consumption, production, and investment decisions, in addition to noneconomic topics such as marriage and childbearing decisions, criminal activity, and discriminatory hiring. 6 This model consists of several simple elements, which I will describe briefly before introducing recent behavioral critiques of them.
First, a person has a set of preferences, a ranking of states of the world over which he or she has some choice, such as which size of laundry detergent to buy, which company to invest in, and which job to take. These preferences are normally assumed to be âgivenâ to the individual and do not change (an assumption arising less from a principled conviction and more from a desire for parsimony, as well as humility regarding our knowledge of the nature of preference change). 7 Furthermore, they are substantively empty: rather than stemming from a psychological state such as desire or need, they are ideally taken to be ârevealedâ by behavior, whatever their ultimate source may be. As long as a personâs preferences meet basic formal conditions such as consistency and completeness, they can be represented by a âutility functionâ that represents to what degree the personâs preferences are satisfied. 8
Second, persons are faced with certain constraints that limit satisfaction of their preferences. These normally take the form of material resource constraints, such as scarcity of wealth or capital, or limitations on the time available to âspendâ on various activities (such as in labor supply decisions). Like preferences, constraints are often assumed to be exogenous, although some can be based on earlier choices; for instance, wealth is based both on initial endowments and later choices about work and investment.
Third, individuals are assumed to have sets of beliefs over the likelihood of various uncertain states of the world. These include beliefs about themselves, such as which options will satisfy their preferences, as well as beliefs about external events, such as whether it will rain on their cousinâs wedding day. Sometimes the two types of beliefs combine, such as whether a person will be able to adjust comfortably to living in a new city where she is considering taking a job. These probabilities, often based on subjective judgment about uncertainty rather than precise objective risk, are used in the decision-making process to discount the amount of utility a person will derive from each state of the world.
These three elements together define decision-makersâ âproblemâ: maximizing their expected utility, based on their preferences and beliefs, within their constraints. This model grounded nearly the entirety of microeconomic theory until theoretical and empirical research by decision theorists and psychologists questioned its real-world applicability. For example, in the 1950s, Herbert Simon introduced the concept of bounded rationality, which encapsulated the imperfections of and limitations on human rationality. As Simon described the standard model of choice, the agent
As an alternative, Simon introduced the concept of satisficing, in which an individual facing a difficult choice lacks either the time or the mental capacity to select the perfectly optimal decision, and instead optimizes these scarce resources by making a decision that is âgood enoughâ (or satisfactory). 10 In this way, the person meets some internal threshold of preference-satisfaction even though it may not be the choice he or she would make with infinite time and computing power. Now a significant topic of study by economists, psychologists, and philosophers alike, Simonâs concept of satisficing was the first major challenge to the standard economic model of choice, which was revealed to be an abstract mathematical ideal that neglected to consider or incorporate the limited decision-making capacities of real human beings. 11is assumed to have knowledge of the relevant aspects of his environment which, if not absolutely complete, is at least impressively clear and voluminous. He is assumed also to have a well-organized and stable system of preferences, and a skill in computation that enables him to calculate, for the alternative courses of action that are available to him, which of these will permit him to reach the highest attainable point on his preference scale. 9
Simonâs bounded rationality emphasizes the limits of our deliberative and executive faculties: that our rationality and willpower are limited and do not always operate as predicted by textbook models of decision-making. Later, psychologists led by Daniel Kahneman and Amos Tversky added details and nuance to Simonâs ideas, supported by years of psychological experiments. 12 These researchers identified a number of common cognitive biases, dysfunctions, and heuristics that cause decision-making processes to differ from the ideal textbook depictions. For example, the endowment effect describes the phenomenon by which people place more value on an item when it is in their possession than when it is not. This is also related to loss aversion, by which people feel more harmed by a loss of given magnitude than they are helped by an identical gain. So we would rather give up $100 that we never received than lose $100 we have, and we would demand more money to part with an item we owe than we would have paid to get it in the first place. Also, hyperbolic discounting explains why our preferences seem to change over time, and why an unpleasant event (such as a root canal) that a person has no problem scheduling six months from now seems much worse the day before it is set to happen; this is very useful to modeling weakness of will, especially with regard to procrastination. 13
These cognitive biases affect our preferences, from which, in the traditional model of choice, values are derived. Other findings from behavioral research address how we incorporate new information into our set of beliefs. Instead of simply updating our beliefs with any new information we acquire, we do so in a way that privileges some information over other. For example, confirmation bias implies that we give more attention and weight to new information that confirms our current beliefs (and less attention and weight to information that challenges them); this can be seen in choices of news sources, in which people often choose cable news stations that confirm their preexisting political positions. Optimism bias, as the name suggests, describes our propensity to maintain exaggerated beliefs about ourselves and our futures; this can be illustrated by the âit canât happen to meâ effect when hearing mortality rates about smoking, and the statistic that well over 50 % of people think they are above average drivers (which is unlikely if we assume a fairly symmetrical distribution of driving skill).
Most relevant for the discussion of nudges, behavioral studies also demonstrate the power of framing in influencing how we make choices, which implies quirks in how both our preferences and information are affected. For example, patients are more optimistic when told that a procedure has a survival rate of 90 % than when they are told that it has a mortality rate of 10 %, a result that reflects the emotional impact of how a choice is framed. We see this also when there is a widely publicized plane crash and more people take road trips, despite the much higher mortality rate on the highway, because of the visceral impact of the news. Finally, marketers are testaments to framing effects, expert at promoting products in order to increase sales. 14
Behavioral Economics and Nudge Policy
In recent years, behavioral economists have been incorporating the numerous insights of Kahneman, Tversky, and their colleagues into standard models of choice and experiments of th...