Nudge – Improving Decisions about Health, Wealth and Happiness

Citation

Kendall, K. (2010), "Nudge – Improving Decisions about Health, Wealth and Happiness", Sustainability Accounting, Management and Policy Journal, Vol. 1 No. 1, pp. 106-112. https://doi.org/10.1108/20408021011059296

Regulation as a nudge in the right direction: insights from behavioural law and economics Introduction

Introduction

This is a review and critique Nudge, a recent US publication that uses insights from behavioural law and economics to make recommendations regarding regulation design. Such regulations considered are very varied and include environmental considerations. The approach taken is to position Nudge in the broader literature and then critique its fundamental arguments in light of related literature.

This review finds that Nudge is a useful volume in as much as it uses insights into human behaviour to suggest improved methods of regulation to produce desired behavioural outcomes. The fundamental argument, that classical economic thinking regards individuals as making decisions in their own best interests (making it an inappropriate framework for regulation design), is flawed on two grounds: classical economics does not make this claim, but rather claims that individual are better placed to make decisions in their own best interests than a regulator; and regulators, as individual decision‐makers, suffer from the same decision‐making limitations as private citizens. The insights highlighted are valuable in their own right, but the book also serves the useful purpose of forcing adherents to classical methodology to respond to such a frontal attack, thereby refining their own framework and making it more useful in the long run.

Behavioural law and economics is an embryonic discipline. This review, as with the book itself, highlights the insights that this discipline provides, which has had limited attention given its recent development.

Imagine that you are the manager of a school cafeteria. One of your tasks is to determine the specific presentation of the food that is on offer to students (and staff). If you keep the menu items the same, and in the same quantities, will the order and placement of the items on offer make any difference to their sales level?

This is a modified version of the hypothetical with which Nudge (Thaler and Sunstein, 2008) opens. The answer presented is yes, presentation does matter, and often dramatically so. By rearranging items, sales can be increased by as much as 25 per cent. This observation has clear implications for your job as manager, regardless of your ultimate motive. For example, observing that whatever item you place at customer eye level invariably sells the most, your choice of which item to place in that position will depend on the goals you are pursuing. If your objective is to maximise the cafeteria's profits, then the item with the highest profit margin would be the logical choice. Alternatively, if you are more concerned with the nutritional content of your students' diets, then you would place healthier foods at eye level.

This example introduces the notion of choice architecture. Choice architecture is the context in which people are required to make decisions. In this case, the food display represents the choice architecture under consideration. This is an especially appropriate example for the purpose of the overall thesis proposed, as the cafeteria illustrates a situation in which the manager is forced to influence the customer's choice, as the food must be displayed in some way.

Choice architecture is one part of the broader concept dealt with in Nudge, which is libertarian paternalism. The two elements forming this notion, libertarianism (simply put) being overall freedom from interference and paternalism being external influence over decisions in order to promote whatever is deemed to be in an individual's best interests, may seem like an oxymoron. Indeed, some critics have said exactly this (Klein, 2006; Becker and Posner, 2009).

The authors, though, defend their concept on the basis that people do not always act in their own best interests. For example, people continue to smoke and eat fatty foods, despite being aware of the health risks. People speed on public roads, despite being aware of the dangers. People do not invest in retirement funds, despite the knowledge that government‐funded retirement benefits are hopelessly inadequate. Considered more broadly, people consume more energy from fossil fuels than necessary, drive when walking is a viable alternative, purchase products from companies known to employ child labour or who employ environmentally unfriendly production processes (such as emitting too much carbon, disposing effluent in waterways, catching other species of marine life when fishing).

It is in this respect that paternalism plays its part. In cases where there is an objective preferable alternative (healthier food, lower driving speeds, more saving, more socially and environmentally responsible production processes), regulations can be designed to promote these particular choices.

The libertarian aspect is incorporated by the authors' requirement that free choice be left unfettered. In other words, while a particular decision is being promoted, individuals remain free to make a different choice. So, in the cafeteria example, if promotion of healthy eating habits is the desired outcome, the libertarian paternalist would place fresh fruit at eye level and fried fatty foods in a less conspicuous place. Individuals who want fatty foods are still able to select these choices unhindered. One would expect, though, that more customers would end up choosing fresh fruit than the less healthy options due to the placement described. This may be contrasted with what may be termed a pure paternalistic approach, in which only fresh fruit is offered for sale. In this scenario, free choice is impeded, as the customer no longer has the ability to choose unhealthy food. Under the libertarian paternalism position, though, free choice is maintained since all options remain available; it is just that the specific choice architecture employed promotes a particular choice. Or, to use the title of the book, individuals are given a “nudge” in the right direction.

While the labels may contain political connotations, libertarian paternalism sits within the developing interdisciplinary field of behavioural law and economics. This discipline represents the intersection of three distinct areas: psychology; economics; and law. Nudge's authorship reflects this; Richard Thaler is a Professor in the University of Chicago's Booth School of Business and Cass Sunstein a Professor at Harvard Law School (although was at the University of Chicago Law School at the time of writing Nudge). Thaler's particular specialisation is the more established, albeit still developing, area of behavioural economics, which essentially uses insights from the psychology literature to challenge classical economic assumptions about individual behaviour[1].

Behavioural law and economics may best be thought of as the application of behavioural economics to legal or regulatory problems[2]. As noted, behavioural economics is a development from classical economics, which models individual behaviour in order to predict probable outcomes in specific situations. Importantly, classical economics does not purport to predict what a specific individual would do in a given situation; rather, it predicts the likely behaviour that would be exhibited in the aggregate. Allowances are made for an individual to behave differently from the model, but such behaviour is cancelled out by opposite deviations from the model (that is, for example, someone not saving enough for retirement is offset by someone else saving too much).

The classical economic model predicts behaviour based on a set of assumptions, the most notable of which is rationality. In essence, all forms of economics is an attempt to forecast the likely behavioural response to incentives. A modern Australian example in the environmental field is the Rudd Government's proposed emissions trading scheme[3], which seeks to impose a price on carbon‐based pollution. Under classical economics, as the price of pollution increases, economic actors will respond by polluting less.

Behavioural economics uses the insights from psychology to forecast that aggregate behaviour will deviate from the classical economic model in certain predictable ways. The cafeteria example at the beginning of this review demonstrates this; classical economics predicts that the specific placement of food items will not have an effect on their popularity (all else being equal). Behavioural economics, though, recognises a phenomenon known as framing; that the response elicited can be affected by the manner in which the choice is presented, which can explain the effects on sales of changing the placement of food items in the cafeteria.

Libertarian paternalism, as presented by Thaler and Sunstein, seeks to exploit these insights into human behaviour through the design (or introduction) of regulation in particular fields. Part I is dedicated to explaining the concept of libertarian paternalism and how nudges can be used to guide people's behaviour, so that society can reap particular benefits (becoming healthier, wealthier and happier). Parts II‐IV apply libertarian paternalism to specific fields, such as the credit markets, retirement savings, marriage, organ donations, health insurance and school choice. Part V ends with some specific regulatory recommendations and responding to some criticisms.

Of particular interest is Chapter 12, which sets out some considerations in the area of environmental regulation. Emissions trading schemes (labelled “cap‐and‐trade” systems, as is the custom in the USA) are promoted as consistent with libertarian paternalism. Given that most environmental regulation has been of the command‐and‐control nature, such as mandatory car emission limits or technological mandates, Thaler and Sunstein advocate for the more incentive‐based regulation that an emissions trading scheme represents. Such schemes effectively take the position that people can choose to continue to pollute, so long as they pay for the damage caused. Individual choice is preserved. This is compared with a mandate requiring a certain device be implemented or emissions reduced to a specific level (however achieved), options which limit if not eliminating choice.

Thaler and Sunstein then present other opportunities for nudges in environmental regulation, largely based around the demonstrated effects that feedback has on human behaviour. Such suggestions include additional disclosure requirements, with such regulations introduced requiring disclosure of hazardous materials stored or released into the environment in the wake of the Chernobyl nuclear plant accident in 1986. While not designed to influence environmental behaviour (the regulations are presented as “essentially a bookkeeping measure” (Thaler and Sunstein, 2008, p. 190), the disclosure requirements enabled environmentally active groups to name and shame the worst polluters. In an effort not to appear on any such blacklist, companies tended to reduce their pollution. The information made available through the disclosure requirements provided a form of feedback that allowed behaviour to be monitored more accurately, leading to a modification of that behaviour.

An important element of such feedback and information disclosure for the desired effects to be brought about is that the information provided needs to be meaningful. Fuel economy stickers, ubiquitous amongst car dealerships in the USA, provide a case in point. Initially, these stickers were required to show the fuel mileage that the vehicle would typically attain. Thaler and Sunstein, though, argue that these measures were largely ineffective, since fuel mileage figures in isolation contain very little information value to the typical car purchaser. New stickers, in which the estimated annual fuel cost (expressed in dollars) and a graphical comparison of that car's fuel economy against other similar models cars is anticipated to be much more effective.

A more ambitious example provided is at the household level, in which feedback is provided on energy use through the use of an “Ambient Orb”, which changes colour to indicate the level of energy consumption; green for modest use, red for heavy use. This form of feedback is expected to have benefits (there is no present formal move for Ambient Orb to be used widely) as the problem normally with household energy use is that it is not readily observable. The effects of leaving the air conditioner on overnight, for example, are not immediately apparent. The Ambient Orb provides immediate feedback, with the anticipated effect that people who wish to reduce their energy use will be motivated to do so more effectively.

The common thread with all these examples is the maintenance of freedom of choice, hence consistency with libertarian paternalism. Companies can continue to pollute, so long as they are prepared to pay for that right. Companies can release toxic chemicals into the environment, if they are prepared to bear the brunt of criticism from environmental activists. People can still choose to purchase a car with high fuel consumption, or to use large amounts of energy in the house. Choice has not been impeded.

Nudge is not the first time that these authors have espoused the virtues of libertarian paternalism. The first instance was in 2003 (Thaler and Sunstein, 2003), which provoked quite strong criticism from some quarters (Klein, 2006; Klain, 2006; Mitchell, 2005), provoking an equally spirited defence (Sunstein, 2006b; Sunstein and Thaler, 2003). Sunstein has also used libertarian paternalism in his book, Laws of Fear: Beyond the Precautionary Principle (Sunstein, 2005; Mandel and Gathii, 2006). Nudge, though, is the first attempt to bring libertarian paternalism into the world beyond academia, written in a manner much more accessible to non‐scholars through not only the choice of language and tone, but also the everyday examples to which the theory is applied and illustrated[4]. In doing so, Thaler and Sunstein maintain a University of Chicago tradition of taking heavy scholarly theories and concepts and applying them to everyday problems in a manner accessible to a mainstream audience[5].

There is an intuitive appeal to Thaler and Sunstein's thesis. After all, if an individual can still costlessly (or at negligible cost)[6] choose an alternative that is not that deemed to be “best”, then what harm can come from designing regulations in a particular manner? This is particularly attractive in situations (such as the cafeteria) where the choice architect has no option but to exert influence in some way (that is, food has to be displayed in some order). While there may be scope to pursue a particular agenda, including political ends, in scenarios where this influence is unavoidable, the notion of libertarian paternalism does seem innocuous in the potential harm and very powerful in terms of promoting public policy goals.

The real danger, though, comes in when no choice architecture needs to be provided, yet the principles of libertarian paternalism are applied in any event. For example, there is no necessity to disclose fuel mileage on cars, or for people to save for their retirement at all. These two examples, taken from Nudge, may appear spurious; in that, it is difficult to argue that disclosing fuel economy and encouraging people to save for their retirement are bad things per se. The problem, though, is that regulating these areas represents a significant slippery slope, raising the question where is the line to be drawn? This is a simple exercise where choice architecture must be provided; if there is no intrinsic requirement to regulate an area of society, then do not apply these principles. Once libertarian paternalism is applied in areas in which regulation is avoidable, the door has been opened for political ends to be pursued and individual behavioural biases manipulated to suit the regulator's ambitions[7].

Notwithstanding these concerns, there are two significant contributions made by this exposition and application of libertarian paternalism. The first is as a design mechanism. Knowledge of predictable actual human behaviour can only be invaluable when designing regulations, necessary or not. In the environmental field, while command‐and‐control regulation has tended to be the norm (as noted in Nudge), regulation through nudges shows a good deal of promise. If regulation designed in accordance with the precepts of libertarian paternalism is more effective (however measured) in achieving the desired environmental ends, then a regulator would be foolhardy not to take such understanding into account.

The second arises through the recurring criticism of libertarian paternalism. An early premise put forward in Nudge is that classical economics assumes people make choices in their own best interests. This, however, is easily debunked, and Thaler and Sunstein do so, through simple observation. People eat too much, smoke too much, pollute too much, drive too fast and save too little despite being cognisant of all the associated dangers. Consequently, the fundamental assumption that individuals can make their own best judgments does seem flawed.

This is a common criticism of classical economics, one of the many base‐level assumptions that critics argue proves that classical economics has no relevance to the real world[8]. The criticism, though, is slightly off the mark. Classical economics does not posit that individuals make decisions in their own best interests, even only some of the time. The position that classical economics takes, which may be more correctly associated with the so‐called Chicago School of Economics (which, amongst other precepts, is highly sceptical of government intervention) that dominates classical economic thinking, is that the individual is better placed to make the decision that is in their own best interest than a third party regulator (Becker and Posner, 2009, pp. 148, 150). Indeed, as Becker and Posner notes (Becker and Posner, 2009, pp. 148‐50), libertarian paternalism's essential foundation is that individual decision‐making ability is limited[9]. This provides the justification for a regulator to step in and influence the decision‐making process; except that regulator is also an individual, who is also subject to the same limitations in making decisions. There is no evidence to expect that government officials (or any other group in society) are exempt from these inherent human limitations.

Through responding to the critique that libertarian paternalism provides, though, the classical economic position is made stronger. Flaws in the manner in which arguments have been put forth (framed even) are rectified through the necessity to respond to critique. In this way, it is perhaps appropriate that the central protagonists and both (apparently) competing paradigms all are associated with the University of Chicago. The Chicago School has earned a reputation over the last century as a bastion for fearless inquiry and critique, through which theory is only strengthened (van Overtveldt, 2007). Through the constant critical feedback that each paradigm provides the other, both will become stronger over time in the quest to describe an indescribable world, thereby carrying on another Chicago tradition.

Notes

Another example in this vein is at the end of Chapter 1, in which a nudge used on Lake Shore Drive, the main road in Chicago bordering Lake Michigan, is described. At particularly dangerous turns that have a high rate of accidents, lines have been marked at regular intervals leading into the turns. Once the bend has been reached, the lines are painted closer together. This creates the sensation in the driver that the vehicle is traveling faster than it actually is, due to the conditioned visual cues that is manipulated at the critical time, triggering an automatic response in the driver to slow down. This nudge has been more effective than lowering the speed limit (mandatory regulation) in reducing the rate of accidents at these places (Thaler and Sunstein, 2008, pp. 37‐9).

Contained in the Carbon Pollution Reduction Scheme Bill 2010 and associated bills.

For a more detailed overview of behavioural law and economics, see Jolls et al. (1998).

For a recent critique in this vein, see Boyle and Simms (2009).

One of the early examples of this type of work was Nobel Laureate Milton Friedman with his co‐publication of Free to Choose (Friedman and Friedman, 1980). Other more recent examples include Steven Levitt's Freakonomics (Levitt and Dubner, 2005) and many of Posner's (2006) recent publications, for example, Not a Suicide Pact: The Constitution in a Time of National Emergency.

Thaler has been at the forefront of the behavioural economics movement since its recognition as a separate discipline in the early 1980s. For example, see Thaler (1980) and Kahneman et al. (1986, 1990, 1991).

This concern applies to all political agendas, for example, promoting union membership, or encouraging investment in business as opposed to real estate. Particularly with respect to framing, noted earlier, individuals may be manipulated, such as in the particular phrasing of a referendum question to promote a specific outcome.

This is a critical criterion for a regulatory alternative to satisfy Thaler and Sunstein's notion of libertarian paternalism. Anything more than a negligible cost (however measured, including time and effort) impermissibly impedes on individual freedom and, therefore, does not qualify as an example of libertarian paternalism. An example of such negligible cost may be filling out a simple form or ticking an additional box on a form that already needed to be filled out (such as opting out of organ donation when renewing a driver's licence).

This is a phenomenon referred to as “bounded rationality”. For a recent application of bounded rationality in a regulatory context, see Sunstein (2006a)