Friday, June 16, 2017

An Introduction to “Nudge Science” and the Ethics of Influence

Ronald F. White, Ph.D.
Mount St. Joseph University
Department of Philosophy
Mount St. Joseph University

Let’s begin by addressing the most obvious question. Given the vast number of books published on political science every year, why would the Association for Politics and the Life Science (APLS) and it’s journal Politics and the Life Sciences (PLS) expend time, energy, and resources publishing a multiple-author analysis of a book that contains little (if anything) about the life sciences, Darwin, or evolution? The answer is that The Ethics of Influence provides an excellent opportunity for APLS to further expand its commitment to interdisciplinarity; especially, behavioral economics.
Cass R. Sunstein, a prolific author, has written several books and scholarly articles defending “libertarian paternalism.” (Sunstein 2008, 2014, 2016, 2017) Libertarian critics have long-argued that the conjunction of “libertarian” and “paternalism,” is oxymoronic; and that the “liberty principle” or the “principle of autonomy” excludes paternalistic intervention on behalf of rational, competent adults. Over the years, with varying degrees of success, Sunstein has addressed many, if not most lines of criticism emanating from both the political left and right. And, like many scholars, his views have evolved, over time, based on that criticism. This roundtable review will focus on the arguments presented in most recent book and how those arguments may (or may not) apply to recent public policy issues. By way of introduction, I will introduce some of the more enduring elements of the conceptual framework that underlies Nudge Science; most notably, the concepts of: choice architecture; political bans, mandates, and influence, political ethics, and paternalistic intervention. 

Choice Architecture

 In recent years, the philosophical debate over “free-will v. determinism” in the context of human decision-making has been usurped by the social and biological sciences. Sunstein, a pioneer in the field of behavioral economics, has consistently argued that human decision-making is “framed” by “choice architecture;” a set of causal variables that influence (if not determine) our ultimate decisions.  We are not always aware of this framework and therefore we often confuse “freedom to choose” with “freedom to choose within architectural constraints.”
Given the vast number of architectural variables that shape our ultimate decisions, one might still question how much room is left for freedom of choice? But, for now, let’s leave these metaphysical arguments for those lingering pre-scientific philosophers.
Today, philosophical analysis initially focuses on clarifying the meanings of key concepts. In ordinary language we use the term “influence” in many different contexts. In physics, scientists use it to indicate natural causal relationships between non-living things; that is, the sun “influences” the orbit of the earth. Biologists employ the same term to describe a wide variety of relationships between living things and their environment. And finally, in the human sciences we use the term influence to designate a wide array of relationships between humans; but not without risking confusion between causality with influence.
At least some of the choice architecture that underlies our decision-making is contextual, and therefore shaped by our own individual or collective life history and/or unique genetic makeup. Interpersonal influence can be exercised between individuals (family, friends, and/or strangers) or between collective organizations (businesses and/or governments). However, “Nudge Science,” seeks to identify the (more or less) universal psychological forces that constitute human choice architecture. Sunstein and others argue that, scientific knowledge of these universal mechanisms will improve our own individual decision-making by exposing our own natural biases. But it will also advance our individual and collective ability to influence the behavior of others (hopefully for the better); and perhaps even increase our own ability to resist unwanted interpersonal influence.
One way to explore the relationship between choice architecture and our ultimate choices is to analyze a few relatively simple, clear-cut case studies. For example, we all agree that our decision whether or not to go on a picnic on any given day is influenced (or nudged) by the external physical environment, especially the weather. In short: “nature nudges.” (Sunstein, 2016: 35) However, there may be conflicting opinions of what constitutes good and bad picnic weather. My personal rule of thumb would be: “the hotter the better.” In the case of conflicting weather predictions, our would-be picnickers might “trust” one local forecaster more than another. If it turns out to be unexpectedly cold and/or raining on the day of the picnic, the availability of a picnic shelter may be turn out to be highly influential, if not decisive. 
Although nature and the external physical environment, obviously, influence many of our decisions, it does not, necessarily, determine those decisions. That’s because our decisions may also be influenced by external social and/or political variables. Our decision to picnic at any given park might also be externally influenced by legal bans and legal mandates. Many public parks legally ban the consumption of alcohol, or legally mandate dogs to be on leashes. Hence, consumers of alcohol and/or dog owners might choose one park over another based on legality. But all legal bans and mandates require monitoring and enforcement, which require a costly, ever-vigilant police force, and judiciary. Depending on whether a local government is willing and/or able to monitor and enforce these legal bans and mandates, those laws might be undermined by a “black market effect,” whereby would-be picnickers might deliberately choose to violate those bans and mandates. In fact, many utilitarians follow Sunstein and argue that it’s often more cost effective for political leaders to exercise influence over citizens, rather to monitor and enforce legal bans and mandates.
While our choices are obviously “framed” by external physical and sociopolitical architecture, those choices are also influenced by internal, psychological forces. Psychologists now agree that human decisions are ultimately shaped by cognitive operations that have evolved over millions of years, which are located in specific regions of the human brain. System 1 cognitive operations are “fast, automatic, and intuitive;” which include perceptual and emotive operations. They tend to be associated with the parts of the brain responsible for perception (back) and part responsible for emotive responses (inner). System 2 cognitive operations are “slow, calculative, and deliberative.” (Sunstein, 2016: 28) “Rational” operations emanate from the frontal lobes or cerebral cortex. As scientific knowledge of System 1 and System 2 operations advance, so will our individual and collective ability to influence and/or manipulate the behavior of others. However, that same knowledge might also advance our ability to resist unwanted political influences. The long-term challenge for Nudge Science is to distinguish between universal choice architecture that underlies all human decision-making, from  architectural determinants that are contextual, and/or relative to specific individuals and cultures. 
Historically, the most obvious applications for Nudge Science have been in the domain of business marketing; as the most successful business leaders have long employed knowledge of various cognitive operators in order to more effectively market their products and services to consumers. Sunstein and others seek to expand the exercise of influence into politics.   

Political Bans, Mandates, and Influence 

Worldwide, political regimes tend toward either authoritarianism or democracy. For better or worse, all regimes exercise both coercive force and influence over citizens. Authoritarian regimes rely almost entirely on coercive force, by monitoring and enforcing legal bans and mandates; which often dictate, not only what is good, but also how to pursue it. So what are the moral limits to the exercise of political influence within liberal democracies?
Although Democratic Republics respect individual autonomy, they also employ combinations of political coercion (bans, mandates) and political influence (System 1 and System 2 Nudges). However, unlike Authoritarian regimes, Republican regimes must morally justify bans, mandates, and nudges. For example, most Western democracies employ System 2 nudges that mandate that private corporations conduct scientific research on the costs and benefits of the products and services they offer; and clearly and accurately “label” those products so consumers can make informed decisions whether to purchase those products or services. In the U.S. recent labeling proposals include mandating labels that identify Genetically Modified Foods (GMOs) and labels that disclose the caloric and sodium content of foods. As a general rule, most of us do not object to labeling nudges. There are also Public Education nudges, whereby governments mandate that corporations provide information concerning the risks associated with dangerous or unhealthy activities such as smoking, drinking, distracted driving, and childhood obesity. Again, most of us do not object to Public Education Nudges based on System 2.  
However, sometimes governments supplement System 2 nudges with System 1 nudges, by requiring graphic labels that instill fear of dangerous products. Fear is a powerful motivator. Other System 1 nudges are based on universal perceptual defaults, including our natural preferences for products presented at eye-level, and our inclination to choose the first alternative. Therefore, many democracies mandate that cafeterias present unhealthy, high caloric deserts at the end of the line, below eyelevel, in less conspicuous places. Although most of us oppose health-related bans and mandates, we often appreciate, or at least tolerate, health-related nudges. 
Most of us are also naturally wired by choice architecture (individually and collectively), to maintain the status quo, as evidenced by the formation of personal habits and cultural traditions. Both habits and traditions are notoriously difficult to change. That’s why we also have a propensity to avoid making life-changing decisions. Many liberal democracies, therefore, mandate that we make choices, such as deciding whether to be an organ donor before getting a license. Other democracies employ Automatic Enrollment via default decisions, which offer passive “opt-out” benefits rather than active “opt-in” benefits, such as enrollment in Pension Plans or the use of Green Energy options. We are also naturally programmed more to avoid harm, than to pursue benefits. This explains why governments, worldwide, invoke fear rather than merely promise a better life.
Finally, we are also naturally predisposed to act in pursuit of short-term benefits, and avoidance short-term harms; less inclined to pursue long-term benefits or avoid-long term harms. In part, that’s why System 2 scientific arguments that warn of long-term harms associated with obesity, poverty in old age, breast cancer, and global warming, tend to fall on deaf ears. All of this suggests that System 2 nudges must occasionally be supplemented by System 1 nudges, and even bans and mandates.   

Political Ethics 

Much human behavior is teleological (or goal-directed) and therefore, requires the analysis of both means and ends. Empirical psychologists describe what human beings, in fact, pursue (ends) and how they pursue those ends (means). Ethicists, however, prescribe not only the ends that we all ought to pursue, but also how we ought to pursue those ends. As a libertarian, Sunstein embraces the idea that liberty consists in the freedom to choose which ends are worth pursuing, as long as that pursuit does not harm others. Therefore, the Ethics of Influence limits the application of Nudge Science to helping us achieve the universal ends that all humans value. Sunstein argues that unlike “ends nudges” (which dictate what we ought to pursue) “means nudges” are (by definition) “freedom preserving.” (Sunstein, 2016:
Ethics, like psychology, political science, and economics, is a scholarly discipline. For centuries, theologians and philosophers dominated that discipline. The Western Legal and Moral Tradition focuses on both “knowing” (what to do or not to do) and being willing or able (to do or not do it). Thus the determination of moral responsibility for one’s actions involves both rationality (the ability to know what’s right and what’s wrong) and free will (the ability to do what’s right and not do what’s wrong. The ability to “know” is widely regarded as a System 2 brain function and the ability to “do” is usually considered to be a product of System 1.  For centuries moral philosophers argued that ethical behavior is determined by either System 1 “feelings” or System 2 “knowledge.” Historically ethicists argued over whether ethics is anchored in System 1 emotions or feelings or System 2 rationality?  Today we know it’s both.
Philosophers have also long-debated the role that moral rules and/or principles play in moral decision-making. Political philosophers have also questioned the degree to which liberal democracies ought to rely on legality (laws) and morality (moral rules). Sunstein identifies four values, or “foundational commitments” that constitute the ethics of political influence in a liberal democracy: welfare, autonomy, dignity, and self-government. (Sunstein. 2016: 25) However, the simultaneous advancement of all four values is often problematic. In a liberal democracy (Republic), self-government often trumps welfare, autonomy, and dignity. Therefore, unpopular bans, mandates, and nudges carry with them a political cost.
Sunstein’s four commitments are non-controversial and deeply embedded in the Western liberal tradition. However, many philosophers would amend that list to include utility, justice, or non-maleficence. Others might question whether his four commitments logically independent.  For this review we’ll, assume that these four principles capture the essence of morality in a modern liberal democracy. Given these four moral commitments, how might the U.S. government go about justifying the deployment of specific bans, mandates, and nudges?
Recall that political scientists have long observed that authoritarian political regimes rely almost exclusively on coercive force; in the form of “bans” (Don’t do X or you’ll be punished by the state!) and “mandates” (Do Y or you’ll be punished by the state!)  Democratic regimes, in contrast, set moral and/or legal limits on the use of coercive force. Left-leaning democratic regimes tend to emphasize human welfare, often at the expense of autonomy, and self-government. Sometimes welfare-liberals are even willing to employ coercive bans and mandates in pursuit of welfare. Right-leaning political philosophers, in the libertarian tradition, value autonomy and dignity over both welfare and liberal democracy. For Sunstein, the battleground lies in the ethics of state paternalism.  

The Ethics of State Paternalism

 Given the complexities presented by choice architecture and the potential for unintended self-destructive decisions, paternalistic intervention by liberal democracies has become increasingly common. Paternalism, by definition, involves treating an adult as a child, thus violating his/her autonomy and dignity in order to either remove a harm or provide a benefit. State paternalism, involves government treating rationally competent adults as if they are irrational, incompetent children. The conceptual puzzle stems from the apparent contradiction between the Western concept of “human agency” (autonomy and dignity) and rapidly growing body of research associated with “behavioral economics.”(Sunstein 2017) For Sunstein, the central question of political ethics is whether autonomy, necessarily, trumps welfare, dignity, and democracy. If not, under what conditions might paternalistic intervention be justifiable?
In his classic work book, On Liberty, John Stuart Mill, laid the conceptual foundation for anti-paternalism in liberal democracies. He argued that the only justification for the use of political force (legal bans and mandates) is to prevent or remove “harm to others.” However, “harm to self” by rational competent adults is protected by the liberty principle. According to Mill, paternalistic intervention by government officials must be limited by the above “very simple principle.” The first step is to determine whether that potential beneficiary is a rational, competent adult. If so, then that official may also inquire whether that person, knows that this action will (in fact) result in harm to self. If so, that official may present rational argument in order to change his mind. But ultimately, physical coercion must be avoided. Thus, according to Mill, paternalistic intervention on behalf of rational competent adults must be limited to the determination of System 2 competence. If that person is a rational competent adult, then paternalistic intervention must be limited to providing information and issuing a warning. Sunstein seeks to soften Mill’s anti-paternalistic stance based on state-of-the-art behavioral science.
There are three longstanding philosophical problems are associated with anti-paternalism: an epistemic problem, a moral agency problem, and the impure paternalism problem. The first two problems are deeply embedded in the Western liberal tradition, which says that rational, competent adults must be treated as “moral agents,” who can be held legally and morally responsible for their actions toward themselves or others. The third problem is the direct result of the corporatization of welfare.
The epistemic problem arises from the fact that “human agents” (rational, competent adults) may not “know” (for lack of information) what’s good to do in order to promote their own well-being (do exercise), or what not to do (don’t smoke). Libertarians, in the John Stuart Mill tradition, argue that once a rationally competent adult is informed (warned) of the self-regarding risks associated with any activity, government cannot forcefully interfere with that informed choice. But how much information does a rational, competent, adult need in order to make a truly informed self-regarding decision? What should that paternalistic governmental official do, if there is conflicting public information concerning the degree and/or probability of a specific activity? How much information does a rational, agent need, before he/she engages in a self-harming activity, such as smoking? If a genetic test were available that might “inform” that would-be smoker of his/her own cancer risk, and that would-be smoker refuses to take the test, is that person really informed? Should government sometimes mandate those tests?
            Moral agency is usually ascribed to rationally competent adult humans. Thus, human agency hinges on the standards of knowing and doing. Even if a rational person knows” what’s good for him/her (do exercise, don’t smoke), that person may not be willing or able to do or not do it. Informed moral agents might freely choose to smoke because “by their own lights” the immediate pleasure of smoking outweighs the long-term health risks. In short, a person may know what to do or not do, but will not or cannot do it or not do it. Some humans obviously lack moral agency, including young children and/or adults; who, by their very nature, lack rationality or free will. Today, given a widespread explosion of conflicting scientific information that is readily available; rational competent adults may be regarded as de facto incompetent.  Many substances and activities are now regarded as addictive, including: tobacco, alcohol, heroin and even gambling. Addiction is widely regarded a mental and/or physical disease, that justifies paternalistic medical intervention. Thus the battleground lies in the question of whether person P knows what to do, and if so, can person P do it? As psychologists continue to plumb the depths of choice architecture, the more difficult it will become to defend absolute standards of moral agency. The third problem is impure paternalism, whereby paternalistic intervention by the state (in the form of bans, mandates, and nudges) benefits not only the intended beneficiary, but other third parties, who benefit by providing that benefit or removing that harm. Sometimes these third-parties are governmental officials, or their friends or relatives; and sometimes they are for-profit private businesses, non-profit charitable organizations. Americans often object to impure paternalistic interventions that provide minimal benefit to the intended beneficiary, while providing enormous benefits to third parties. In recent years, the Affordable Care Act included both an employer and employee mandate to purchase costly health insurance. That mandate has proven to be increasingly unpopular among small businesses and young healthy workers.    

Summary and Conclusion

In sum, many (if not most) humans lack the intellectual capacity to acquire and/or process the vast amount of information that is now available. And many lack the will to act upon that information. Many instances of state paternalism benefit third-parties. This sets us up for the basic issue that underlies the Ethics of Influence: Is Mill’s anti-paternalistic stance still justified? Is paternalistic intervention exercised by the leaders of left-leaning liberal democracies ever morally justified? If paternalistic intervention is sometimes justified, under what circumstances might liberal democracies violate autonomy and/or, dignity in pursuit of welfare? The answer lies in self-government. Liberal democracies must refrain from the deployment of political bans, mandates, and influence that are enormously unpopular. In Appendix A (204-208) Sunstein lists 34 nudges and their corresponding approval ratings, which range from 86% approval for Mandatory Labels for GMOs to 21% approval rating for “Default assumption of Christianity for census data.” Liberal Democracies must also be wary of paternalistic interventions that line the pockets of third parties; most notably corporations that generously contribute to political campaigns.
In the final analysis, The Ethics of Influence is NOT about freedom and determinism. The fact is that our all of our decisions are influenced by choice architecture. Sunstein’s goal is to determine whether liberal democracies ought to employ freedom-diminishing bans and mandates or freedom-preserving nudges. Those decisions, he argues, ought to be made in conjunction with moral values: welfare, autonomy, dignity, and self-government. When the deployment of influence is morally/legally justified, then leaders must, then, decide which kinds of “nudges” (System 1 or System 2) are morally justified and/or efficient?   As nudge science advances, the exercise of political influence by liberal democraciesmay become increasingly more efficient. The Ethics of Influence is all about when to mandate, ban, or influence. If nudges are morally required or morally permitted, what kinds of nudges are best employed in various contexts, and for how long?
So how might recent advances in Nudge Science be applied in addressing some of todays’ most vexing public policy issues in the United States, including: obesity and dietary nudges, post-employment poverty and retirement savings nudges, breast cancer and mammogram screening nudges, and environmental degradation and environmental nudges?        



Sunstein, Cass R.; Thaler, Richard (2008). Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven, Connecticut: Yale University Press.


Sunstein, Cass R. (2013). Simpler: The Future of Government. New York: Simon & Schuster.


Sunstein, Cass R. (2014). Why Nudge?: The Politics of Libertarian Paternalism (The Storrs Lectures Series). Yale University Press.


Sunstein, Cass R. (2016) The Ethics of Influence: Government in the Age of Behavioral Science. New York, Cambridge University Press,


Sunstein, Cass R. (2017) Human Agency and Behavioral Economics: Nudging Fast and Slow (Palgrave Advances in Behavioral Economics) Switzerland: Palgrave McMillan.


David V. Johnson (2017) “Twilight of the Nudges: The Quest to Keep Behavioral Economic in Policy After Obama’s Presidency.” New Republic


Sunstein, Cass R. and Richard Thaler “Libertarian Paternalism” The American Economic Review. Vol 7 Number 2 ( May 2003) pp. 175-179






No comments: