Homo Probabilis, Behavioral Economics, and the Emotional Life of Neoliberalism

Michael Millner (bio)

Abstract

Neoliberalism often operates by privatizing what was once public and by turning questions of moral value into questions of market finance. This essay expands our understanding of these operations by examining the way neoliberalism takes hold at the most intimate level—the level of feeling. It argues that the field known as behavioral economics has helped to produce the neoliberal subject as a person of feeling. While it shows the alignment of behavioral economics with neoliberalism, the essay also suggests how the field might put its observations to work in very different ways.

“The hot hand is bunk.”

LeBron James’s Miami Heat was playing a tight NBA championship series with Tim Duncan’s San Antonio Spurs in June 2013. I had just offered a rather inept analysis of the Heat-Spurs matchup, simply as a way of continuing a cocktail conversation on a subject about which I knew next to nothing. “The Heat will win if LeBron gets the hot hand” was all I could muster. To which my friend countered, “The hot hand is bunk.” In retrospect, I see this exchange as my introduction to one of the hegemonic principles of our moment.

One thing I thought I knew about basketball was that in the final few seconds of a close game you look to get the ball to the player with the so-called hot hand—the player with the flow, who is completely in the zone, hitting everything from everywhere. In one way or another, the idea of the hot hand seems to apply to almost any athletic contest, but especially to basketball, a game characterized by a fluid stream of infinitesimal, instantaneous decisions and movements of great complexity by a group of players essentially untethered from playmaking coaches (who have only so much control over the back-and-forth stream of action). It makes a kind of folk sense that talented players might suddenly be able to put all this complexity together, if only for a few minutes, through a rare synchronicity of mind and body—and shoot the lights out. You better get them the ball while it lasts.

But my acquaintance in this conversation—not incidentally, it seems to me now, a law professor of the economics and law persuasion—proceeded to blow up that folk wisdom. He set out for me Thomas Gilovich, Robert Vallone, and Amos Tversky’s debunking of the hot hand, which originally appeared in 1985 in the journal Cognitive Psychology. It is worth noting that Tversky was the Israeli psychologist who, with his close friend and later Nobel laureate Daniel Kahneman, laid the foundation for the field known as behavioral economics through a series of extraordinary experiments and observations in the 1970s and 1980s. As Gilovich, Vallone, and Tversky explain (writing in this case without Kahneman), the sense that a basketball player has the hot hand derives from the common tendency to misperceive a particular experience involving probabilities. People tend to see significant coordination in small numbers of sequential events where more careful statistical reasoning would show much less coordination. For instance, in a paradigmatic example, naïve weekend gamblers may believe they are on a hot streak when they’ve tossed three or four good rolls of the dice, but, of course, their odds of getting a good roll haven’t changed at all (each roll is distinct with exactly the same odds). The dice rollers are simply experiencing a statistically expected repetition of good rolls, which exists in any significant sequence of rolls. The problem is even easier to see when discussing coin flips. We know that coin flips are essentially random and if we flip a coin a very large number of times it will land nearly 50 percent of the time “heads” and 50 percent of the time “tails.” But within this large number of flips, there may be sequences of several “heads” in a row (or several “tails”). In the case of coin flips, we might be intrigued by such repetitions but we tend not to significantly misperceive the probability of these sequences (we don’t seriously talk of hot streaks in coin flipping) because the coin flip is so clearly understood as random. But in the case of more complex activities, like a series of basketball shots, we do often fall into such misperceptions. The hot-hand essay relies on empirical evidence to argue that a series of successful shots by an individual player during a couple of minutes in a single game is best understood as a statistically expected sequence of successful shots—not evidence of a hot hand—within the context of the player’s season-long shooting average. Put slightly differently, but to the same end: the next jumper that the player takes doesn’t have any significantly greater chance of going in than his or her average from the field determined by a large number of shots across a number of games. As Tversky, Vallone, and Gilovich found by running various experiments and by closely observing real-game shooting statistics, the laws of probability do a better job of predicting the next shot than does a sense of the hot hand. So the best thing for the coach to tell the team in the final seconds of a nail-biter is to get the ball to the player with the best shooting percentage across the season.¹

In my experience, debunking the hot hand tends to conjure up disbelief in just about everyone, and in die-hard sports junkies this disbelief can turn quickly to outright hostility. Indeed, a part of me simply couldn’t believe it either when I watched the Heat and the Spurs in 2013—and perhaps still doesn’t want to believe it today. For both the player and audience, the experience of the hot hand is so intense, feels so accurate and real, and is associated so deeply with an altered state of mind and body, that it’s nearly impossible to relinquish it simply on account of a statistical analysis. The apparent naturalism of the hot hand, as well as the emotional investment it conjures, is at the heart of the argument put forth by Gilovich, Vallone, and Tversky. Their point is not simply that we typically misunderstand statistical reasoning and probabilities—a fact unsurprising to anyone who has taken a stats course—but that those misunderstandings actually feel just right, are again and again our regular initial response, and are almost impossible to shake, even after we fully understand the logical probabilities. We feel in our bones a truth that is an illusion.

In this essay, I’m interested in this feeling of certainty. And not just this feeling, but other closely associated feelings that arise in response to gambles, calculations, or predictions we make under pressure. One of the striking observations of behavioral economics (the hot-hand essay from 1985 is foundational in the field) is that we often feel our way through probabilistic decisions. Whom to throw the ball to for the final shot? What retirement fund manager to choose? How much money to borrow for an education? What insurance contract to purchase? Such questions are only the starting points in our nearly daily encounters with probabilities. In a neoliberal world ever more characterized by the privatization of risk and the financialization of everyday life, we contend with probabilities relentlessly. We have become what the political theorist Ivan Ascher calls “homo probabilis” (85 and passim). And I will add to Ascher’s helpful conceptualization that homo probabilis is, counterintuitively in many respects, a man of feeling (or more generally, a person of feeling). Behavioral economics has played a considerable role in the theorization of this new person of feeling, a role that has had mass appeal and influence. Behavioral economics has popularized particular ideas of the subject, reason, and agency, which call for interrogation. I will argue that behavioral economics has helped produce homo probabilis as a person of feeling, and that the field’s theorization and instrumentalization of affect around probabilities is in league with neoliberalism. But, in the end, I will also argue that ideas about what may be called “probabilistic affect” shouldn’t be completely critiqued away, but rather redirected towards ends very different from those valued by neoliberals.

This argument will unfold by first outlining the contours of homo probabilis—that is, the neoliberal subject—and by emphasizing the importance of understanding subjects who feel their way through probabilities. This subject isn’t natural—it had to be made—and the two middle sections of the essay examine how that making began with the Chicago School economists in the early 1960s and continued through the rise of behavioral economics in the 1970s through 1990s, culminating with its recent popularity. The account offered runs against behavioral economics’s history of itself as a radical break from the Chicago School. My description of behavioral economics also presents the field as developing a theory of affect that it takes to describe natural experience. “Affect” is not a term behavioral economics itself uses, but application of the term—highly developed in the humanities and in sectors of the social sciences—helps provide a critical perspective on behavioral economics. The penultimate section of the essay presents this theory of affect as ideological and aligned with neoliberalism, while the final section suggests different directions we might move in, putting the observations of behavioral economics to work.²

Who Is Homo Probabilis?

Ivan Ascher uses the term homo probabilis to distinguish the older idea of economic man—homo œconomicus—from the newer, neoliberal subject—homo probabilis, or speculative man. The older idea of homo œconomicus, so important to classical economics from Adam Smith onward, sees the individual as a partner in exchange. This individual partners with others in exchange so as to maximize his utility—in the language of economics, the term “utility” refers to his needs and wants. In doing so, this classically liberal subject is thought of as free, agential, and rational. But for the neoliberal subject, homo probabilis, the central activity is not exchange but investment, and that investment is often an investment in himself, that is, in his own human capital. Using a memorable phrase, Michel Foucault calls this new neoliberal subject an “entrepreneur of himself, being for himself his own capital” (226). Like an entrepreneur, the neoliberal subject invests capital in an enterprise, and that enterprise is himself. These investments are risky, and like all investments require consideration of probabilities. However, Foucualt writing in 1979 could not foresee the rise of financialization and the intensification of privatization that have made playing the probabilities ever more complex, becoming one of the primary characteristics of homo probabilis.

In Portfolio Society, Ascher explains that homo probabilis is a fairly recent creation—or, perhaps better put, a fairly recent necessity. In the last forty years, homo probabilis has been cultivated in relation to the rise of a society backed by financialization (what Ascher calls a “portfolio society,” or a society that is viewed as an investment portfolio that balances risk with potential returns). With the dismantling of the security provided by more socialized forms of education, retirement, health care, and living assistance, the neoliberal homo probabilis has been compelled to securitize itself in financial markets. This subject seeks its security by turning itself into a kind of security that can be bought and sold, aggregated and spun off, complete with futures contracts and other kinds of derivatives. Homo probabilis has little choice but to take risks in the market: student debt for education and 529 plans for the education of future generations, 401K investments in retirement, a 30-year loan in the mortgage market, 3-to-6-year financing in the auto loan market, various insurance contracts, and often significant credit card debt (necessitated by slow wage growth). This subject is also managed in terms of risk in the market. Among other things, homo probabilis is a credit score with associated hazards to banks and credit card issuers, an age and gender with associated risks for health insurers, a collection of points with associated probabilities to auto insurers, and, if this person is an immigrant, a collection of data associated with dangers to national security and a panoply of additional concerns.³ This subject speculates in all these markets when it makes decisions about its future—its human capital—while simultaneously being the object of speculation in these markets. Ascher notes that homo probabilis is what some political theorists have recently begun to refer to as a “dividual” (89).⁴ A “dividual” is no longer a subject that is “not divisible,” but one who is divided into parts, separated into data about this and that, and aggregated into pools of other subjects with similar data. A dividual is a means, not an end. This dividual subject, so entwined with speculation and probabilities, invests in itself as the means of collecting dividends and in the hope of having, say, a retirement and medical care in the future—or simply a future in the future.

There is much more to Ascher’s important book, but for now it will suffice to say that it helpfully expands the baseline definition and history of neoliberalism through an examination of financialization and the subject that emerges, homo probabilis. But this subject didn’t simply emerge—it was made, as Ascher well knows: “this neoliberal Homo probabilis, this ‘entrepreneur of the self’ or this investor on which today’s capitalist mode of prediction depends, does not exist in the wild: it has to be bred” (106). And the central question may not be how this subject was made, but why it has had such an astonishing grip—so strong that it has not been broken by economic meltdowns, wide disparities between the 1% and the 99%, a deep sense of precarity in health care and education, and other tribulations fairly easily traceable to the new political economy that began to develop in the early 1970s. As William E. Connolly notes, “most [accounts of neoliberalism] may not come to terms sharply enough with the subjective grip the state, media, and neoliberal combine exerts on the populace even after it has been rocked” (23). The strength of the grip is in part explained by the making of homo probabilis in very intimate, affective ways. Homo probabilis, as we will see, is constructed by behavioral economics as a very particular kind of feeling subject.

History: Toward “A Colossal Definition”

Behavioral economics outlines an array of theories about the everyday, intimate, affective operations of homo probabilis, and through both mainstream media popularity and the advancement of multiple policy initiatives, the field has put those theories into practice to shape homo probabilis. Perhaps no other body of research has thought so thoroughly about the routine, nearly automatic processes we use to make decisions about the future, to calculate risks, to understand, forecast, and analyze data—all the skills that will make one a good homo probabilis and all things that we do affectively as often as we do rationally, according to behavioral economists. But before turning to behavioral economics per se, I want to spend a moment examining its prehistory, where we can first see neoliberalism’s interest in affect and emotion. This history will be nothing like the history that behavioral economics presents of itself. The economic behaviorists often represent their findings as a break from the rational choice theory articulated by the Chicago School economists of the 1950s and 1960s—Gary Becker, Milton Friedman, and others. For instance, in the bestselling Nudge: Improving Decisions about Health, Wealth, and Happiness, perhaps the most influential book of policy suggestions derived from work on cognitive misperceptions, economist Richard Thaler and legal policy scholar Cass Sunstein mock the idea of a “species” of “Econs” who are “Mr. Spock”-like and believe in an old-fashioned rational agent (7, 22). In The Undoing Project: A Friendship That Changed Our Minds, the account of Kahneman and Tversky’s extraordinary and tumultuous friendship and the work it produced, Michael Lewis paints the two psychologists as outsider hero-geniuses who deconstruct the idea of rational choice. The rational actor is what gets undone in The Undoing Project.⁵

From one perspective, this notion of a radical break rings true. The work in behavioral economics deeply complicates the standard economic model of an individual who maximizes utility through decisions made about exchanges with others. That model understands the individual as guided only by payoffs determined rationally. That rationality is unaffected by emotion, the framing/context of decisions, or the difficulty individuals have predicting the future—all areas that behavioral economics has focused upon. As postwar economists “constructed their new approach to economics upon the foundation of utility,” economist George Lowenstein explains, “they rapidly became disillusioned with utility’s psychological underpinnings and sought to expunge the utility construct of its emotional content [developed by philosophers like Jeremy Bentham and John Stuart Mill]” (426). This is certainly true in the main, if we think of emotions as passions that might upset reasoning. But even at the University of Chicago, ground zero for rational choice theorizing, and especially in the work of Gary Becker (“the most radical of the American neo-liberals,” as Foucault remarks [269]), there was interest in something like feeling or emotion from the beginning of the reconstruction of the economics discipline in the post-World War II era.⁶ Regarding this interest, there are two central points to make—one about human capital and utility and another about rationality.

Becker’s idea of “human capital”—a transformative idea in twentieth-century economics—was essential to expanding the notion of what might be considered a utility so as to include aspects of life that are hard to pin down as rational, like satisfaction and wellbeing. A bank account, a stock portfolio, and an assembly line traditionally count as capital because they might produce income or other useful yields. Forms of human capital, including “[s]chooling, a computer training course, expenditures on medical care, and lectures on the virtues of punctuality and honesty” are “capital too,” Becker explains, “in the sense that they improve health, raise earnings, or add to a person’s appreciation of literature over much of his or her lifetime” (15-16).

That rational agents might trade in “appreciation of literature” or the “virtues” of certain values (or the satisfactions of motherhood, another area of interest to Becker and the Chicago School) ranks as a significant expansion of the traditional idea of utility. Foucault in his remarkably prescient 1979 Collège de France lectures on “American neo-liberalism” calls such emotional payoff “psychical income” (244). In the case of the mother-child relationship, the child benefits from the mother’s investment in obvious ways that maximize utility—better care and education lead, potentially, to a better income—but the mother also benefits: “She will have the satisfaction a mother gets from giving the child care and attention in seeing that she has in fact been successful” (244), as Foucault explains, summarizing the Chicago School position. When such “satisfaction” can be incorporated into economic analysis, an economic rationality is implemented all the way down—the human at a very intimate level is a homo œconomicus. “[T]he generalization of the economic form of the market beyond monetary exchanges,” Foucault remarks, “functions in American neo-liberalism as a principle of intelligibility and a principle of decipherment of social relationships and individual behavior” (243). Economics can be applied to the most private of realms.

The rationalization of “psychical income”—like a sense of virtuousness, appreciation, or satisfaction that may become an object of economic analysis—is only part of Becker’s proto-neoliberal project. What Foucault calls Becker’s “colossal definition” (269) involves significantly expanding the notion of rationality to incorporate affective response. Foucault observes that Becker’s idea is in essence a new conception of labor. The Chicago School replaces older classical understandings of labor, as well as the Marxist understanding of labor, with a new interpretation of it as essentially decision making (or “choice,” as the Chicago School proponents like to say)—and decision making, ultimately, involves affective response. As Foucault puts it, the Chicago School inspires economists to take up

the task of analyzing a form of human behavior and the internal rationality of this human behavior. Analysis must try to bring to light the calculation—which, moreover, may be unreasonable, blind, or inadequate—through which one or more individuals decided to allot given scarce resources to this end rather than another. (Italics added, 223)

Labor takes the form of “calculation,” and the object of study for the new Chicago School economics is essentially the study of how decisions are made using “internal rationality” and how calculations lead deciders to “allot given scarce resources” in a market.

But what exactly is “internal rationality” for Becker? In the traditional economic view, individuals, firms, and markets are deemed rational because they consistently maximize the utility function. A “calculation” that is “unreasonable, blind, or inadequate” doesn’t sound like a form of rationality under any definition—economic, philosophical, or commonsensical. But Becker develops just this position in his 1962 essay “Irrational Behavior and Economic Theory” (Foucault’s point of reference). Here Becker enters the fray over what might count as rationality in economic theory and how economic theory might approach the “impulsive,” irrational behavior of individuals and firms (“Irrational Behavior” 5, 6, 7, 9, 11, 12). He says that his purpose in this essay is “not to contribute yet another defense of economic rationality” but rather to argue that “economic theory is much more compatible with irrational behavior than had been previously suspected” (1, 2). Specifically, Becker reasons in his technical analysis that economic theory is not best understood in terms of rationality but in terms of “systematic” (4, 5, 11, 12), “consistent” (1, 3, 7, 8, 13) responses to the environment. Becker concludes that “[s]ystematic responses might be expected, therefore, with a wide variety of decision rules, including much irrational behavior” (12). “Rational conduct,” Foucault writes, explaining Becker’s view, “is any conduct which is sensitive to modifications in the variables of the environment and which responds to this in a non-random way, in a systematic way, and economics can therefore be defined as the science of the systematic nature of responses to environmental variables” (269). “Homo œconomicus,” Foucault summarizes, “is someone who accepts reality”—that is, accepts those “environmental variables”—in a regular, predictable way. Foucault calls Becker’s take on rationality “a colossal definition, which obviously economists are far from endorsing” (269). It is “colossal” in the sense that so much calculation and decision-making fall under the definition of “internal rationality,” which isn’t logical rationality, or even the more narrowly defined rationality of neoclassical economics, but rather the systematic, predictable, and nonrandom response to an environment. Foucault also suggests, in terms that again seem accurately predictive, that this notion of rationality will require a new kind of “environmental psychology” (259). Becker’s “colossal definition,” Foucault explains,

has a practical interest, if you like, inasmuch as if you define the object of economic analysis as the set of systematic responses to the variables of the environment, then you can see the possibility of integrating within economics a set of techniques, those called behavioral techniques. (270)

That was then—1979. Foucault doesn’t have much more to say about these “behavioral techniques” beyond quickly gesturing toward B. F. Skinner’s behaviorism and remarking on the techniques’ impersonal application (they are “brought to bear on the rules of the game rather than on the players” [201]). But since 1979, many economists, psychologists, and even the general reading public, it seems, have been willing to accept Becker’s “colossal definition.” I am suggesting here that Foucault is describing in Becker, avant la lettre, the form of rationality developed more fully by behavioral economists. From this perspective, behavioral economics looks more like a reformist project, not a radical break from the Chicago School economists. In the next section, I will turn more fully to this new rationality and its emphasis on affect.

Theory and Policy: The Project in Probabilistic Affect

Foucault foresaw in 1979 the development of what would eventually become a mainstream project, as evidenced by the popular consumption of books and other media that investigate our “systematic responses to the variables of the environment” and suggest in turn various “behavioral techniques.” A list of such mainstream work would include Daniel Kahneman’s overview of his and Tversky’s research, Thinking, Fast and Slow, which has remained in the top ten of the New York Times Business Best Sellers List since its publication in 2011. Lewis’s The Undoing Project climbed to number four on the New York Times Best Seller List a few weeks after its publication. Sunstein and Thaler’s Nudge has been a best seller since its publication in 2006, and its policy influence has been considerable. Sunstein served in Barack Obama’s administration as head of the Office of Information and Regulatory Affairs, where, by his own account, he implemented dozens of nudge policies.⁷ All of these books call upon a great deal of specialized research and make it available to a general audience. I will refer to this threading together of technical research and popular outreach as “the project in probabilistic affect.” It deserves to be called a project because it has a set of policy aims. I also want to give it a name because I want to describe it in terms different from those that behavioral economics would use to describe itself. In doing so, I emphasize its affectual dimensions and its particular constructed-quality (rather than its universal naturalness) as a project that serves specific ends and that could possibly take a different form.

First let me describe the project in probabilistic affect without focusing on the criticisms I want to make (to which I’ll turn in the next section). The project stems from two different but related fields of psychological research: the psychology of decision-making and the psychology of cognitive dual processing. Both have been important to the field of behavioral economics. As we have already begun to see, the former finds that in many different instances, but especially when involving probabilities, our decisions are guided by feeling rather than what would traditionally be called rationality. The latter—the psychology of cognitive dual processing—provides a theory of affect that matches up well with and underwrites the research on probabilistic decision making. Both of these areas of research are applied through a number of policy suggestions, especially those put forth by Thaler and Sunstein in Nudge. Let me explain each of these areas of work in a little more detail, focusing on how together they produce and instrumentalize the project in probabilistic affect.

The psychology of decision-making is vast, but the findings most often repeated in the popular literature of behavioral economics involve bets, gambles, and other games of probability and risk. Experimental subjects aren’t very good at these calculations, at least from the perspective of maximizing their utility. They make irrational decisions (“nonstandard” decisions, in the parlance of economics), which often seem based on a feeling or impulse rather than on rational analysis. Importantly, they make the same mistakes consistently. Some examples of the research:

• • The gambler’s fallacy, where after rolling several good rolls of the dice, you expect a bad roll (or vice versa), on the theory that the odds in a game of chance must even out. In reality, you are operating more on inclination than on rational analysis. The probability of each roll is exactly the same (as almost everyone knows) and, as with the hot hand, any large number of rolls can exhibit strings of good/bad rolls. However, the intuitive-feeling decision that events are connected and your luck will change is hard to shake. That feeling affects experts who should know better and who are not playing games of chance. Researchers have shown that even experienced immigration judges are susceptible to the gambler’s fallacy when deciding asylum cases, as are bank loan officers when making decisions about mortgages, and professional baseball umpires when calling balls and strikes.⁸

• • Loss aversion, where you feel a loss much more intensely than you feel a gain. In lab experiments, subjects will characteristically risk losing $100 on a 50/50 coin flip only if they might win $200. From the perspective of the standard model of utility maximization, it is irrational to require so much surplus upside before you are willing to risk a downside. But because you dread a loss so much more than you appreciate a gain (about twice as much, it seems from the laboratory experiments), you consistently respond to risks in skewed ways, following your initial felt response more than your rational analysis. Loss aversion shapes many everyday decisions facing homo probabilis: goods bought for discounted trial periods are hard to give up when the trial ends; stocks are sold in panic at relatively minor losses; more resources are sunk into lost causes in the hope of avoiding further losses.⁹

• • The endowment effect, closely related to loss aversion, where you irrationally endow something you own with more value than you place on the same object on the open market. In the well-known experiments on this effect, participants who are given coffee mugs or pen packets (to keep as their own) typically require twice as much payment to sell their mugs or pens than they would spend to buy the mug or pens in the first place. Standard economic thinking finds it illogical to attach yourself to a mug (or a house, job, or plan for the future) so much more intensely simply because you’ve owned it for a few minutes.¹⁰

• • Predictions/projections of the future, where you are terribly and consistently bad at thinking logically about the future, and instead follow your nose and trust your feelings. There are many different varieties of this problematic relationship with the future. Overconfidence in the ability to predict the future has been shown in surveys about health, where individuals on average underestimate the possibilities of future hospitalizations. Just about everyone knows that it will take longer to finish a complex project than predicted, even as this knowledge doesn’t make one less susceptible to bad predictions about the completion date (so powerful is the intuitive sense of how the future will unfold). In field observations, CEOs and employees alike overestimate the prospects of their companies, irrationally staying highly invested in company stock, and stock traders overestimate the precision of their information about companies and make bad trades. The standard theory of economic utility doesn’t take account of such systematic overconfidence. And akin to overconfident predictions are many other kinds of prediction problems. For example, people tend to project current states into the future in ways that again and again prove wrong (yet they continue to make such projections). One of the key findings in studies of wellbeing is that individuals often believe a certain event will bring happiness or unhappiness in the future, but then it doesn’t. For example, in studies assistant professors report that they will be very unhappy in the future if they do not receive tenure and very happy if they do, but when the event occurs, there isn’t much change in reported levels of happiness. This misperception of future states plays mischief with the standard model of economic utility, which assumes that one can predict what will be satisfying in the future.¹¹• Framing, where your decisions are shaped by the way they are presented, even when you know the decisions are presented from a particular perspective. It is perhaps not surprising that the framing of a choice will play some role in the decision. For instance, when subjects are presented the same gamble in somewhat different ways (with different framing), they make very different decisions in a predictable manner. What is more surprising is just how much individuals underestimate the influence that the framing of a decision has on the decision. Research has shown the ways experimental subjects’ decisions are shaped by framing even when they know that the presentation of their choices is being manipulated. Often, however, it is very difficult to recognize the framing, as is the case in our world of media and information saturation, which taxes our limited capacity for attention. Instead of responding logically to all the available data in complex, information-saturated choices, laboratory and field subjects tend to attach themselves to the most recent and the most familiar information. The standard model of utility does take account of incentives and the interestedness of buyers and sellers, but not how much we allow ourselves to be pulled and coaxed hither and thither, even when we recognize the manipulation.¹²

All of these examples touch on probabilities—how we evaluate risk and make decisions about the future—and all of these examples suggest that we often act irrationally when it comes to probabilities—that we make such decisions by feeling rather than by calculative deliberation in relation to maximizing utility. We feel something in the ownership of the mug, we sense the hot hand, and we just know that the way we feel now will be the way we feel in the future. It is also important to recognize that these responses are not random one-offs, but are generally consistent and systematic. The findings indicate that humans are not simply irrational in these respects, but systematically irrational (and thus rational in Becker’s sense of economic rationality, as a systematic response to one’s environment). From a distance and with adequate time for deliberation, we can all recognize the irrationality, but in the moment such decisions feel just right.

There is in these findings a notion of affect, if not a full-blown and fully articulated theory. In outlining this point more fully, it is worth taking a moment to distinguish affect from emotion. As Brian Massumi says, “affect” frequently indexes a force that is “irreducibly bodily and autonomic,” rather than stemming from our beliefs, desires, or mental relations with the world (like emotions) (28).¹³ Affects don’t so much mediate, frame, or shape decisions (the way our emotions might), as much as they are the basic response upon which the decision is made: get the ball to the player with the hot hand; a series of good outcomes must be followed by a bad result; avoid losses more than appreciate gains; if it’s not working, sink more resources into it; if it looks like it will bring unhappiness in the future, it will bring unhappiness in the future. These are affective responses rather than emotional responses. It is thus salient that it’s hard to put a name to the intense response that many have to the hot hand, for instance. We have names for emotions, which derive from our somewhat discernible beliefs, desires, and mental representations of the world, but we don’t often have names for affects. Possibly a “gut reaction” or “reflex,” a “sensation” or “perception,” an “impulse”? These possibilities don’t name feelings so much as different kinds of responses to a choice or situation, responses that are feeling-related in that they reference a bodily reaction, rather than one based in reason. In other words, they index, without precisely naming, affect.

We now come to the second field of psychological research, dual processing. If the research on decision-making only gestures toward a theory of affect, the work on dual processing, important to a great deal of the key ideas of behavioral economics, provides a much more elaborate account. The dual-processing view splits cognitive processing into two very different systems or types often called System 1 and System 2 (or Type 1 and Type 2). Decisions made with System 1 are involuntary, intuitive, and uncontrolled; they are made without deliberation or rationalization. They are also made very quickly. System 1 isn’t exactly natural and nonconscious because in it one does learn rules through practiced nurture—like 2 + 2 = 4—but it often feels natural and nonconscious. It also is a fait accompli: it is hard to change or control System 1—it is what it is, and it accepts the environment it encounters, reacting to it automatically. We use System 1 when we respond to something with disgust, turn to a loud and sudden sound, or do highly practiced activities like drive a car on an open, straight road. We also frequently use System 1 when we say “get the ball to the player with the hot hand,” or when we rashly hold onto a Bitcoin investment after it loses half its value. System 1 has systematic reactions to probabilistic situations, but it’s just not very good at probabilities. Because it is automatic, and because its decisions feel nonconscious and natural, System 1 can be closely associated with the way we think of affect.

Decisions made with System 2 are the opposite in many respects from decisions made with System 1: they rely on logic, deliberation, and working memory, and thus require effort. We use System 2 when we reason out complex math problems, or identify hard to see patterns, or evaluate probabilities consciously and deliberately. System 2 isn’t affective, but rationalist. Because System 2 requires effort and the use of working memory (the kind of memory needed to do long division in one’s head), it is a limited resource requiring considerable energy and focus. The two systems have different time signatures: if System 1 decisions are fast, then System 2 decisions are slow (hence, the title of Kahneman’s book, Thinking, Fast and Slow).

The second key point to make about the dual-processing structures concerns the relationship between System 1 and System 2. They are in many instances understood as being in conflict with each other. System 1 says “get the ball to the player with the hot hand,” and System 2 says that the hot hand is bunk if you carefully examine the probabilities. However, the relationship is more complex than a pure, diametrically opposed conflict, in that the two systems are understood as working together while still in opposition. Jonathan Evans and Keith Stanovich, two preeminent researchers of dual processing, refer to the relationship as “default-interventionist”: “Default interventionismallows that most of our behavior is controlled by Type 1 [or System 1] processes running in the background. Thus, most behavior will accord with defaults, and intervention [from System 2] will occur only when difficulty, novelty, and motivation combine to command the resources of working memory” (236-37). Many of the policy suggestions developed out of the findings of behavioral economics employ this default-interventionist conceptualization of System 1 and System 2.

A nudge policy typically overrides or corrects System 1 decision making by figuring out how to allow System 2 interventions and make them easier. From the perspective of nudge theorists like Thaler and Sunstein, when System 1 encounters certain areas of experience—like probabilities—it needs to be restrained and guided by the slower, more effortful system in order to achieve rational outcomes. A “nudge” is a policy that provides a gentle prod or a soft push in a particular direction, not by directing or explicitly forcing, but by shaping the decision-making environment, by jiggering the “rules of the game” (to use Foucault’s previously quoted phrase describing the new “behavioral techniques”). A nudge policy isn’t a mandate or a demand, and it doesn’t provide a direct incentive like a financial fine or a tax refund. Importantly, in Thaler and Sunstein’s conceptualization it doesn’t significantly limit choices. In their view, a nudge seeks to leave open potential choices while using frameworks and structured environments to nudge an individual’s behavior toward specific decisions. (Some critics of nudges, as we will see, have found this delimiting of possible choices in turn limiting of agency and autonomy, but Thaler and Sunstein emphasize that nudge policies try to limit choices as little as possible.) Some nudges use “defaults,” like the popular nudge that automatically enrolls individuals in retirement accounts. Policy makers frequently combine this nudge with another one in which the default retirement account is low-cost and routinely rebalances according to the owner’s proximity to retirement. A similar default strategy is used when automatic enrollment plans nudge utility consumers into slightly more expensive but more environmentally beneficial “green” electricity plans. Users can always opt out of a default nudge—they can choose to invest their retirement in their own way or to continue with the established mix of coal, oil, and gas energy—otherwise it wouldn’t be a nudge under Thaler and Sunstein’s definition, but a mandate. Another category of nudge doesn’t work with defaults but shapes the decision environment when a “choice architect” (3) arranges choices in specific ways. For example, if a cafeteria is organized so that the French fries are slightly more difficult to reach and the salad is up front, it might lead to healthier decisions by molding availability while allowing all options to remain available. As Thaler and Sunstein explain, “A choice architect has the responsibility for organizing the context in which people make decisions” (3). Most (perhaps all) of the nudges that Thaler and Sunstein discuss involve managing and correcting for our difficulties with probabilities. Default enrollment, automatic investment allocation, and choice architectures of various sorts find their rationale in the difficulty that humans have with calculating the present’s probable relationship to the future. But whether a default nudge or a choice-architecture nudge, the System 1 affectual response to probabilities is guided towards System 2 ends. Nudges turn non-rational, affective responses into responses that better match up with the rational, standard model of utility.

I don’t want to underestimate either the technical expertise (and brilliant creativity) or the pragmatic value of this three-prong project in probabilistic affect (the psychology of nonstandard decision making, the psychology of cognitive dual processing, and the development of nudge policies). We are better off recognizing the hot-hand fallacy in our evaluation of a potential retirement fund manager, and we are well served by understanding our propensity to irrationally endow with value, say, our current job or marriage, and it’s also certainly a good idea to appreciate the tendency to irrationally risk more and more on lost causes. But we should also ask, at what costs the pragmatic gains of the project are purchased.

Critiques: “Someone Who Accepts Reality”

There are a number of established criticisms of the project I’ve just laid out. I will not belabor the more technical critiques internal to psychological research, which have questioned the differences between naïve experimental subjects and experienced decision makers, the extrapolation of individual behavior to behaviors of markets, and the reproducibility of various laboratory studies. More aligned with my discussion here are broader criticisms of nudge policies like the critiques by the legal theorist and philosopher Jeremy Waldron. Waldron speculates that nudges affront the autonomy and even the dignity of individuals because they rely on a “choice architect”—”an elite who kn[ow] the moral truth and could put out simple rules for the natives (or ordinary people) to use” (21). “There’s a sense underlying such thinking,” Waldron goes on to say, “that my capacities for thought and for figuring things out are not really being taken seriously for what they are: a part of my self” (21). Waldron is particularly interested in the claim that nudge policies are not interested in people learning from their mistakes and thus practicing a kind of agency and autonomy. Sunstein for his part (writing without Thaler in this case) has directly responded to Waldron’s argument by offering survey data that suggest just how much people approve of and even appreciate many kinds of low-cost, unobtrusive, pragmatic nudges that they feel improve their lives—hardly an affront to dignity in Sunstein’s view if so many people find nudging useful (7).

If questions of morality and of the autonomy of the self trouble Waldron, the political theorist John McMahon is concerned with the ideological work of behavioral economics. In an argument akin, in part, to the one I’m offering, McMahon sees behavioral economics as a “technology of neoliberal governance” (146) and outlines several ways that it furthers neoliberalism’s valuing of the market and the market’s diffusion into all aspects of life and policy. “Behavioral economics problematizes not the market itself, only the assumptions made about the actors on the market, whose behavior is then measured against the truth of the market,” McMahon explains. Behavioral economics “seeks not to challenge or defy the market but to provide tweaks so as to better assimilate all to the market” (146). We have seen this dynamic in nudge interventions in and corrections of System 1 “mistakes” that aim to produce more systematic (if not quite rational) actors who will better invest in retirement funds and other parts of the “portfolio society” that Ivan Ascher describes. As I have suggested here, behavioral economics isn’t so much a break from the initial American theorists of market-based social policy—like Gary Becker and the Chicago School—but should be understood as an effort to reform and thus continue fundamental aspects of their thinking.

But if behavioral economics is a “technology of neoliberal governance,” why is it so popular? This question returns us to William Connolly’s challenge to consider the “grip” that neoliberalism seems to hold on the contemporary moment, especially in light of its manifold failures. Part of the answer is that behavioral economics offers specific tools with which to address the neoliberal order’s demand that each of us becomes a homo probabilis. To some degree humans cannot escape thinking probabilistically, but the necessity of such thinking has surely intensified for people in the economically developed world over the last forty years. Behavioral economics provides specific tools for living in this kind of new world—for being a homo probabilis. Those tools are also grounded in experimental psychological science and seem to be closely connected to natural ways of being. This sense of naturalness is attractive and reassuring, but it can obscure questions about the forces behind the intensification of probabilistic thinking in everyday life.

The attachment to the project of behavioral economics also stems from the familiarity of its mode of operating. It functions along the lines of a century of self-help and therapeutic mass culture. In this well-established model, the self is understood as fractured and conflicted; it can only be put back together with the proper therapy, with various self-care products and regimes, with an idealized love. The project in probabilistic affect follows the same design: as we’ve seen, the deficiencies of the System 1 part of the subject (it is often wrong and conflicted, especially when it comes to probabilities) mean that it’s in need of an intervention in order to achieve its better System 2 self. It seems appropriate, even wise, to put guardrails in place in order to achieve a more ideal rationality. In this schema, the typical indicators of fractured selfhood—anxieties, addictions, and destructive desires—are replaced with System 1’s systematic irrationality and wayward affects in relation to probabilities. Note that the fracturing is still located at a very intimate level—the level of affect. The typical solutions to fracturing and conflict—good love or a sense of wholeness and wellbeing—are replaced with the achievement of System 2 rationality. Part of the popularity of the recent decision-making books is that they repeat the established form of therapy culture.

But there is reason to wonder if this neat scheme of conflict and repair isn’t more of a construction than it often appears to be in work on behavioral economics. The concept of System 1 and System 2 processing has produced significant and complex debates among cognitive psychologists, which are often ironed out in the writings by behavioral economists. Some of the complexities of these debates are outlined in a series of highly technical articles in Perspectives in Psychological Science from 2013. The current debates are wide-ranging and fundamental, and include questions like the following: Is dual processing actually better conceived as one process? Are the two processes associated with two different parts of the brain or with multiple different parts of the brain (thus perhaps suggesting multiprocessing rather than dual processing)? And is it accurate to understand the relationship between System 1 and System 2 as “default-interventionist” (where System 1 defaults continually run in the background with necessary interventions from System 2 as needed), or is the relationship better described as “parallel-competitive” (where the two systems operate simultaneously in competition with each other)?¹⁴ These criticisms frequently extend beyond calling for alteration in, amendments to, or further research into the theory of dual processing: instead, they often advocate abandoning it. Gideon Keren concludes his criticisms: “two-system theories offer a good story—one that ‘pleases the mind,'” but “[a]fter two decades in which two system models have blossomed yet added little if any new insights, the time is probably ripe to divert our scientific efforts into new and more promising avenues” (260-61).¹⁵

It is impossible to adjudicate these debates here, but my point is that the dual-processing understanding of cognition is far from settled science. Significantly, the further one gets from the psychological research literature in the technical journals—and the closer one gets to the mass-cultural, best-selling presentation of these ideas—the more simplified and metaphoric the dual-processing structure becomes. In their 2008 Nudge, Thaler and Sunstein present dual processing as a given, but in Thinking, Fast and Slow Kahneman speaks of dual processing as a “metaphor” (13), as does Sunstein in more recent writing. While Sunstein and Kahneman seem to recognize the debates over dual processing, this recognition doesn’t dissuade them from using the idea to structure their arguments. Kahneman writes in the introduction of Thinking, Fast and Slow: “the labels of System 1 and System 2 are widely used in psychology, but I go further than most in this book, which you can read as a psychodrama with two characters” (21, see also 48). Similarly, Sunstein references the debates but says he thinks of the identification of System 1 and System 2 “as a helpful metaphor, not a reference to something concrete in the human brain” (5). It is difficult to know what to think of these descriptions of dual processing—as a metaphor and a psychodrama. These authors seem to use the term “metaphor” or the genre of “psychodrama” as synonymous with “generalization,” “idealization,” or “helpful fiction” in that System 1 and System 2 are generalizations or idealizations that are helpful to think with—”a good story,” in Gideon Keren’s words. Ironically, rational choice theory, from which behavioral economics is attempting to escape, is itself an idealization/generalization/fiction used as a thinking tool. As a metaphor dual processing may still be useful, but we may want to ask the kind of question humanities professors like me typically ask: “What is this particular metaphor in the service of?”

The answer I’ve been suggesting is that the dual processing metaphor underwrites a popular therapy-like framework so important to the instrumentalization of behavioral economics. Homo probabilis might need something more, but in light of the constantly risk-assessing life of homo probabilis, this therapy has its attractions. We find ourselves drawn to it because it takes account of the dividual nature of neoliberal experience, in that it sees the subject not as rational, agential, and free, but as reacting to an environment and making the best of it. Not only does this therapy framework jettison liberal principles that are difficult to maintain in the contemporary world (like rationality, autonomy, and freedom), it also provides a vision of the acting and feeling subject that makes sense of that jettisoning. This new subject is better off without a false confidence in its ability to be rational, agential, and free, and it is better off with the recognition that it needs help, guidelines, and nudges. In a world where economic crises have become ever more frequent,¹⁶ this idea of the subject serves as a kind of cushioning device. It posits a subject who can ride out these crises by following pragmatic rules rather than by battling for its agency and rationality against the current of neoliberal risk-taking and dividualism. The neoliberal, to recall Foucault’s description, “is someone who accepts reality” (269).

Redirecting Probabilistic Affect

Reading through the policy theorists who ground their work in behavioral economics and enthusiastic reports from the field of choice psychology in popular media, I am frequently reminded of Herman Melville’s “Bartleby, the Scrivener: A Story of Wall-Street.” The unnamed narrator of Melville’s story—a lawyer on Wall Street in antebellum New York—is a great nudger, if only his scrivener, the strange Bartleby, would allow himself to be nudged. Bartleby, however, famously refuses. Initially he refuses his mind-numbing job as a scrivener who repetitively copies documents and reads contracts, and eventually he declines to speak, eat, or even move. The generally kind and well-intentioned boss-narrator tries to nudge Bartleby to cooperate and participate in the law firm’s work by suggesting that Bartleby will be healthier, wealthier, and happier if he does so, if he would simply be reasonable. Variations on the word “reason” appear a dozen times in the story. But in agreeing to be nudged, Bartleby knows, on some level, he will be offering consent to a larger system or supposition—a system of reason and labor that is also a system requiring a particular kind of subject and sense of agency—and he prefers not to do that. Eventually, he prefers not to live at all.

Bartleby isn’t a neoliberal subject. He doesn’t live by risk, probabilities, and investments in his own human capital, as much as he lives by an older economy of factory-like regimentation that restricts his autonomy and freedom. He doesn’t live by the logics of twentieth-century therapy culture (although the lawyer-narrator does call upon the logic of sentimentalism, a nineteenth-century version of self-help). But I do think we can imagine a current-day Bartleby. He would say no to the policies derived from behavioral economics for a number of different reasons. They limit agency and prevent us from learning from our decisions because they so often operate as defaults and through difficult-to-discern choice architecture (as Waldron suggests). They seem like individualized solutions to large-scale, public problems—how to take care of an aging population or impending environmental catastrophe. They have not been able to confront the structural inequalities associated with race and gender. But perhaps the current-day Bartleby (who, like the nineteenth-century Bartleby, resists interpellation) would also say “no” to nudges because they underwrite and produce a particular kind of subject, one who is seen as in need of rules and nudges in order to be a better calculator of probabilities and thus a better inhabitant of neoliberalism.

Alternatively, it is difficult to see what our contemporary Bartleby might respond to with a “yes.” It’s an important question because we do need projects that go beyond critique. In imagining such a project, we might well want to pay close attention to the same kind of everyday reasoning, choice, and error that cognitive scientists and behavioral economists detect, if only we could take those lay responses in a different direction. What I’ve called the project in probabilistic affect has explained everyday experiences, but it has often been bent on correcting that experience toward a standard of rationality that will allow people to mesh better with markets. How might we use this knowledge to a different end?

A redirection might begin by exploring what is learned if we don’t correct what has been called System 1 responses. The scholarly investigation of affect in the humanities and the social sciences is diverse, but it takes as a central interest the breakdown of the long-held dichotomy between feeling and thinking, affect and reason. Affect is understood across an array of fields—by neuroscientists like Antonio Demasio and by philosophers like Martha Nussbaum (who have very different ideas of what affect is)—as inseparable from decision making and discerning interactions with our world. The choice psychologists and the behavioral economists inadvertently and unwittingly overlap with this work in the sense that they take seriously what was previously cast off as mere impulse or irrationality, and, importantly, they develop an understanding of that affect as nonrandom and systematic. But when it comes to the question of what to do with this affectual knowledge, the answer often given by behavioral economists has been to correct it into a different kind of knowledge. If instead we ask what these affective responses might teach us, we might find ourselves in a different place. For instance, what is the idea of the hot hand good for? Most of the limited work on this question focuses on its evolutionary importance. The hot hand requires few resources and limited information, and it is good at identifying short-term patterns quickly. For these reasons, it may have been evolutionarily selected. Rhesus monkeys, a recent study by the psychologist Tommy C. Blanchard and others suggests, use the hot hand to find food, and the strategy is particularly successful when encountering “clumpy resources” (280) in the environment (often the case when foraging for food) rather than randomly distributed resources. In other words, food tends to be lumped together in the wild because of environmental conditions that are hard to discern rationally, but the affectual draw to the hot hand experience might help identify these resources. This way of understanding the hot hand shifts the focus from the subject who is mistaken in his thinking to the environment, where something is different or changed and a new pattern has emerged. To put this shift in terms of basketball, the sense of the hot-hand might be used to identify a “clumpy resource” in, say, the form of an unrested, slightly injured, or hungover defensive player. It’s not that the offensive player is “hot,” but that the environment has changed. Such changes in the environment can be extremely difficult to identify rationally on the fly; the affective hot-hand sense might be better at noticing them. As the psychologist Gorka Navarrete notes, “From an ecological standpoint it is rare for one to observe sequential events that are completely independent of each other” (1), as in situations characterized by chance. How we might incorporate the knowledge available in hot-hand experiences (or the experience of loss aversion or the endowment effect) into our larger social structures remains an open question.

One of the most surprising findings in the behavioral-economy laboratory is that individuals are not as purely self-interested as the standard economic model of the rational agent has assumed. In one particularly simple lab experiment, called the Dictator Game, subjects will give away part of their holdings to anonymous others without social pressure. In addition, where the standard economic model suggests that self-interested people only work with as much effort as required by the market, in field observations people have worked harder for higher pay and gifts (and also have been less productive in response to lower pay and labor disputes). In these experiments and observations, individuals seem to have a social preference—a feeling for others—rather than a preference for pure self-interest. The literature on nudges, dedicated as it is to individual choice making, has thus far not been particularly interested in how to maximize this preference or affectual orientation to any great degree, when in fact it might well be one of the most important orientations for any society to maximize.¹⁷

A modern-day Bartleby might well say “yes” to further thinking about the knowledge available through so-called System 1 responses, finding them rich terrain for developing the social compact rather than responses to be corrected. Because these responses are nonrandom and predictable, they may serve as a foundation for policy, and because there seems to be at least a tendency toward social preferences, such policies might also tend toward the nonindividualistic and intersect with an ethics or might help provide a structure for public values. As Amanda Anderson has made us aware, recent attention to affect in the humanities has often ignored the need for both foundational structure and ethics (9–10). Our imagined Bartleby may also helpfully note that what we have been taught to think of as System 2 (and value as our goal) comes with its own deficiencies. Daniel Kahneman, often less sanguine about the policy uses of his findings than others who use his work, observes that System 2 misses things as it focuses its working memory and other deliberative energies on calculation and reasoning. Christopher Chabris and Daniel Simons in The Invisible Gorilla discuss what is perhaps the most famous example of the shortcomings of System 2 dependence: about half of the viewers of a short video miss the appearance of a women in gorilla suit (clearly on screen for nine seconds) because they have been asked to count the number of times a ball is passed between players on one team while ignoring the other team. Kahneman notes about this experiment that “we can be blind to the obvious, and we are also blind to our blindness” (24). In other words, appealing to System 2 as an interventionist corrective to System 1 may be little better than relying on the standard model of economic utility to describe or normalize human behavior. We may not want to discard System 2, but we may want to think carefully about its overvaluing.

The picture of the subject that begins to emerge from placing more emphasis on System 1 responses—more attention to social preferences, and less faith in System 2 deliberations—is very different from the picture of the subject that emerges in either liberal theory (agential, autonomous, rational) or neoliberal theories and policies (someone who accepts reality, a dividual who needs guidance). Within behavioral economics there are findings that might be put to work in very different ways than have thus far been used. What this essay’s observations recommend is not the dismissal of behavioral economics as a technology of neoliberalism, but a much-expanded project in probabilistic affect that redirects some of its findings, hopefully toward the unmaking of homo probabilis and toward a kind of new subject, a new “homo___________.”

Footnotes

1. The original hot hand essay has produced a considerable body of follow-up research and meta-analysis across a number of fields (and a number of sports and other activities). Most of this work supports the debunking of the hot hand, but some of this work debunks that debunking. For a helpful overview of at least a sample of the research, see Michael Bar-Eli, et al., “Twenty Years of ‘Hot Hand’ Research.”

2. For important accounts of neoliberalism and affect that serve as touchstones for this essay, see Lauren Berlant, Cruel Optimism and Brian Massumi, The Power at the End of the Economy. My essay also engages with the historiography and critique of neoliberal subject making, and it contributes to recent but still underdeveloped criticisms of behavioral economics as a technology of neoliberalism (this work is discussed in detail in the section “Critique: ‘Someone Who Accepts Reality'”). Because this essay offers a critical perspective on behavioral economics as popular psychology, I count as kin—if at times distant kin—works like Eli Zaretsky, Secrets of the Soul; Eve Kosofsky Sedgwick, “Epidemics of the Will”; and Daniel Horowitz, Happier?

3. Ascher allows us to see the connections to an early form of capitalism when he explains that what we have seen in the last forty years is a new kind of “enclosure of the commons”—an “enclosure of the market” (85-107). If for Marx one of the foundational moments of capitalism’s development was the enclosure of land held in common by communities, resulting in the forced migration of people to the cities where they had little choice but to enter the labor force as supposedly “free labor,” then for Ascher one of the foundational moments of neoliberalism is the “enclosure” of welfare state programs (pension funds, state-sponsored medical care, public schooling), forcing people into markets (401Ks, private insurance, privatized public universities). Individuals, rather than communities as a whole, face risk. The switch to such limited and individualistic focused thinking is a continuation of the shift made by the Chicago School toward human capital and away from questions of large-framed constituencies of labor and capital. For complementary accounts of this recent history and the structuring of the neoliberal subject, see Michel Feher, “Self-Appreciation,” and Wendy Brown’s reading of Foucault’s neoliberalism lectures in Undoing the Demos, 47-78.

4. The concept of the “dividual” is important to Brian Massumi, The Power at the End of the Economy, as well as to others who often trace it back to Gilles Deleuze’s “Postscript on the Societies of Control,” but a fuller account of its history and origins in anthropological research is provided by Arjun Appadurai, Banking on Words, 101-124.

5. The break between traditional economic theory and behavioral economics is repeated in just about all the books about decision making and behavioral economics written for a popular audience. See for instance Dan Ariely, Predictably Irrational. Ariely notes that “the very basic idea, called rationality, provides the foundation for economic theories, predictions, and recommendation, … [but] “wouldn’t it make sense to modify standard economics, to move it away from naïve psychology” (xx) that understands humans as acting rationally. For similar understandings of a break with traditional economics, see Ori Brafman and Rom Brafman, Sway, 17-19, and David Halpern, Inside the Nudge Unit, 6-7.

6. Michel Feher quips: “Becker largely remains a neoliberal theorist trapped in a utilitarian imagination. Thus his relationship to the neoliberal condition may one day be described as that of G. W. F. Hegel to Marxism—or, for that matter, as that of Moses to the Promised Land” (27).

7. I have already mentioned books written for a popular audience like Ariely, Predictably Irrational; Brafman and Brafman, Sway; and Halpern, Inside the Nudge Unit. Other significant books in this niche of mainstream publishing include Jonah Lehrer, How We Decide; Sheena Iyengar, The Art of Choosing; and Barry Schwartz, The Paradox of Choice. Center and center-left news outlets like National Public Radio and the New York Times energetically update their audiences on the findings and arguments of the science of decision making and its economic ramifications through columns like “Economic View” and shows like “Freakonomics Radio.” TED Talks on the subject abound.

8. In these bullet points I have, of course, greatly simplified the discussion of these various findings, and I have concentrated on areas of research that are often repeated in the popular literature. One of the best technical overviews of the findings in behavioral economics is Stefano DellaVigna, “Psychology and Economics: Evidence from the Field,” and I have used DellaVigna’s work as a guide to the key academic papers (on the gambler’s fallacy, see 344-345). For field research on the gambler’s fallacy in immigration court cases, loan applications, and major-league umpiring, see Daniel L. Chen, et al., “Decision Making Under the Gambler’s Fallacy.” On the similarities and difference between the gambler’s fallacy and the hot-hand fallacy, see Peter Ayton and Ilan Fischer, “The Hot Hand Fallacy and the Gambler’s Fallacy.” Kahneman focuses on these problems with large and small numbers (109-118). Thaler and Sunstein discuss the gambler’s and the hot-hand fallacies in Nudge (27-31).

9. On the $100/$200 gamble, see Kahneman 284; the surrounding pages discuss loss aversion and associated issues in great detail (278-288). DellaVigna outlines the research on loss aversion (324-326), and the examples of some of its real-world consequences are taken from those pages. Thaler and Sunstein address loss aversion at several points (33-34, 122-123) and propose several investment and personal finance policies to correct for its mistakes.

10. For the foundational accounts of the endowment effect in the technical literature, see Richard Thaler, “Toward a Positive Theory of Consumer Choice,” and Daniel Kahneman, et al., “Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias.” DellaVigna discusses application of the endowment effect to understand behavior in housing, financial, insurance, and labor markets (328-335). Kahneman dedicates a chapter to the endowment effect (290-299), and Thaler and Sunstein present a number of policies to mitigate the endowment effect in retirement investing, such as default plans which automatically rebalance retirement funds by proximity to retirement (132).

11. DellaVigna offers an overview of the academic research on prediction and projection problems (341-344, 346-347), and the examples of the real-world consequences are taken from these pages. The work on future emotional states, like happiness, has been investigated by the psychologist Daniel Gilbert and made popular in his bestselling Stumbling on Happiness; see especially 113-117.

12. Kahneman frequently discusses framing; see especially 363-374. DellaVigna discusses decision-making when the framing is well-known as well as decision-making under conditions of attention overload (347-360). In Thaler and Sunstein, the academic work on framing is essential to the conception of “choice architecture” (11-13, and passim) explained in this essay.

13. “Autonomic” refers to the autonomic nervous system. Massumi is one of the central proponents of affect as bodily and pre-personal, and, as such, a category wholly separate from emotion and feelings, which are personal and social in his view. But this particular understanding of affect, and what follows from it, are by no means accepted by all in the voluminous writings on the subject, and I don’t wholeheartedly accept Massumi’s position. Rather, I find that the distinction between affect and emotion/feeling is helpful in describing the experience of probabilistic decision making as it is discussed by Tversky, Kahneman, Thaler, Sunstein, and other behavioral economists. In these pages I will think of the behavioral economists as theorizing affect—even though they don’t use this term—in ways that overlap in some respects with the affect theorists proper, like Massumi. At various points throughout, I attempt to navigate some of the central debates in affect studies, although adjudicating these debates is not my purpose here. For a critical overview of affect studies (helpful for mapping the field as well as providing a specific argument about its shortcomings), see Ruth Leys, “The Turn to Affect.”

14. On the single-system position, see Arie W. Kruglanski, “Only One? The Default Interventionist Perspective as a Unimodel”; on the multi-system position, see Elizabeth A. Phelps, et al., “Emotion and Decision Making”; on “default-interventionist” and “parallel-competitive” theories, see Evans and Stanovich, “Duel Processing Theories” (227).

15. It may seem strange that the unconscious—so central to the understanding of the self and decision making in the humanities—plays no role in this account of System 1 and System 2, but that is certainly the case. Behavioral economics is little interested in the Freudian tradition. I’ve tried in this essay to focus on some of the problems with what behavioral economics does endorse rather than consider traditions that it ignores, but the absence of the unconscious calls for further inquiry.

16. There is little doubt that financial crises are more frequent than ever before. A 2017 Deutsche Bank report (a missive from the heart of the beast) found that “prior to the post WWII Bretton Woods system, financial crises existed, but the frequency was not as intense as the post Bretton Woods world [early 1970s to present]. Interestingly this period between the mid-1940s and early 1970s was the longest stretch without an observable financial crisis for 200-300 years” (Reid et al. 10).

17. DellaVigna offers an overview of research in social preferences (and the Dictator Game) (336-341). Thaler and Sunstein have little to say about social preferences but suggest two nudges focused on charitable giving that involve an automatic giving plan and a special credit card/account for charitable gifts (231-232).

Works Cited

• Anderson, Amanda. The Way We Argue Now: A Study in the Cultures of Theory. Princeton UP, 2006.
• Appadurai, Arjun. Banking on Words: The Failure of Language in the Age of Derivative Finance. U of Chicago P, 2016.
• Ariely, Dan. Predictably Irrational: The Hidden Forces That Shape Our Decisions. Rev. and expanded ed., HarperCollins, 2009.
• Ascher, Ivan. Portfolio Society: On the Capitalist Mode of Prediction. Zone Books, 2016.
• Ayton, Peter, and Ilan Fischer. “The Hot Hand Fallacy and the Gambler’s Fallacy: Two Faces of Subjective Randomness?” Memory & Cognition, vol. 32, no. 8, Dec. 2004, pp. 1369-78. Crossref, doi:10.3758/BF03206327.
• Bar-Eli, Michael, et al. “Twenty Years of ‘Hot Hand’ Research: Review and Critique.” Psychology of Sport and Exercise, vol. 7, no. 6, Nov. 2006, pp. 525-53. ScienceDirect, doi:10.1016/j.psychsport.2006.03.001.
• Becker, Gary Stanley. Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. 1964. 3rd ed., U of Chicago P, 1993.
• Berlant, Lauren. Cruel Optimism. Duke UP, 2011.
• Blanchard, Tommy C., et al. “Hot-Hand Bias in Rhesus Monkeys.” Journal of Experimental Psychology: Animal Learning and Cognition, vol. 40, no. 3, 2014, pp. 280-86. APA PsycNET, doi:10.1037/xan0000033.
• Brafman, Ori, and Rom Brafman. Sway: The Irresistible Pull of Irrational Behavior. Doubleday, 2008.
• Brown, Wendy. Undoing the Demos: Neoliberalism’s Stealth Revolution. Zone Books, 2015.
• Chabris, Christopher F. The Invisible Gorilla: And Other Ways Our Intuitions Deceive Us. Crown, 2010.
• Chen, Daniel L., et al. “Decision Making Under the Gambler’s Fallacy: Evidence from Asylum Judges, Loan Officers, and Baseball Umpires.” The Quarterly Journal of Economics, vol. 131, no. 3, Aug. 2016, pp. 1181-242. Oxford Academic, doi:10.1093/qje/qjw017.
• Connolly, William E. The Fragility of Things: Self-Organizing Processes, Neoliberal Fantasies, and Democratic Activism. Duke UP, 2013.
• Damasio, Antonio R. Descartes’ Error: Emotion, Reason, and the Human Brain. G. P. Putnam, Avon, 1994.
• Deleuze, Gilles. “Postscript on the Societies of Control.” October, vol. 59, 1992, pp. 3-7. JSTOR, jstor.org/stable/778828.
• DellaVigna, Stefano. “Psychology and Economics: Evidence from the Field.” Journal of Economic Literature, vol. 47, no. 2, 2009, pp. 315-72. JSTOR, doi:10.1257/jel.47.2.315.
• Evans, Jonathan St B. T., and Keith E. Stanovich. “Dual-Process Theories of Higher Cognition: Advancing the Debate.” Perspectives on Psychological Science, vol. 8, no. 3, 2013, pp. 223-241. JSTOR, doi:10.1177/1745691612460685.
• Feher, Michel. “Self-Appreciation; or, The Aspirations of Human Capital.” Public Culture, vol. 21, no. 1, Jan. 2009, pp. 21-41. Duke University Press, doi:10.1215/08992363-2008-019.
• Foucault, Michel. The Birth of Biopolitics: Lectures at the Collège de France, 1978–1979. Edited by Michel Senellart, translated by Graham Burchell, Palgrave Macmillan, 2008.
• Gilbert, Daniel Todd. Stumbling on Happiness. Alfred A. Knopf, 2006.
• Gilovich, Thomas, et al. “The Hot Hand in Basketball: On the Misperception of Random Sequences.” Cognitive Psychology, vol. 17, no. 3, July 1985, pp. 295-314. ScienceDirect, doi:10.1016/0010-0285(85)90010-6.
• Halpern, David. Inside the Nudge Unit: How Small Changes Can Make a Big Difference. W. H. Allen, 2015.
• Horowitz, Daniel. Happier?: The History of a Cultural Movement That Aspired to Transform America. Oxford UP, 2018.
• Iyengar, Sheena. The Art of Choosing. Twelve, 2010.
• Kahneman, Daniel. Thinking, Fast and Slow. Farrar, Straus and Giroux, 2011.
• Kahneman, Daniel, et al. “Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias.” Journal of Economic Perspectives, vol. 5, no. 1, Mar. 1991, pp. 193-206. American Economic Association, doi:10.1257/jep.5.1.193.
• Keren, Gideon. “A Tale of Two Systems: A Scientific Advance or a Theoretical Stone Soup? Commentary on Evans & Stanovich (2013).” Perspectives on Psychological Science, vol. 8, no. 3, 2013, pp. 257-262. SAGE Journals, doi:10.1177/1745691613483474.
• Kruglanski, Arie W. “Only One? The Default Interventionist Perspective as a Unimodel—Commentary on Evans & Stanovich (2013).” Perspectives on Psychological Science, vol. 8, no. 3, May 2013, pp. 242-47. SAGE Journals, doi:10.1177/1745691613483477.
• Lehrer, Jonah. How We Decide. Houghton Mifflin Harcourt, 2009.
• Lewis, Michael. The Undoing Project: A Friendship That Changed Our Minds. W. W. Norton, 2017.
• Leys, Ruth. “The Turn to Affect: A Critique.” Critical Inquiry, vol. 37, no. 3, 2011, pp. 434-472. JSTOR, doi:10.1086/659353.
• Massumi, Brian. Parables for the Virtual: Movement, Affect, Sensation. Duke UP, 2002.
• —. The Power at the End of the Economy. Duke UP, 2015.
• Melville, Herman. “Bartleby, the Scrivener: A Story of Wall-Street.” Melville’s Short Novels: Authoritative Texts, Contexts, Criticism, 1st ed., Norton, 2002, pp. 3-34.
• Navarrete, Gorka, et al. “Small Samples and Evolution: Did the Law of Small Numbers Arise as an Adaptation to Environmental Challenges?” Frontiers in Psychology, vol. 6, Feb. 2015. PubMed Central, doi:10.3389/fpsyg.2015.00029.
• Nussbaum, Martha C. Upheavals of Thought: The Intelligence of Emotions. Cambridge UP, 2003.
• Phelps, Elizabeth A., et al. “Emotion and Decision Making: Multiple Modulatory Neural Circuits.” Annual Review of Neuroscience, vol. 37, no. 1, 2014, pp. 263-287. Annual Reviews, doi:10.1146/annurev-neuro-071013-014119.
• Reid, Jim, et al. Long-Term Asset Return Study: The Next Financial Crisis. 18 Sept. 2017, financedocbox.com/Investing/67145295-Long-term-asset-return-study.html.
• Schwartz, Barry. The Paradox of Choice: Why More Is Less. Revised ed., HarperCollins, 2016.
• Sedgwick, Eve Kosofsky. “Epidemics of the Will.” Tendencies. Duke UP, 1993, pp. 130-42.
• Sunstein, Cass R. Human Agency and Behavioral Economics: Nudging Fast and Slow. Palgrave Macmillan, 2017.
• Thaler, Richard. “Toward a Positive Theory of Consumer Choice.” Journal of Economic Behavior & Organization, vol. 1, no. 1, Mar. 1980, pp. 39-60. ScienceDirect, doi:10.1016/0167-2681(80)90051-7.
• Thaler, Richard H., and Cass R. Sunstein. Nudge: Improving Decisions about Health, Wealth, and Happiness. Yale UP, 2008.
• Waldron, Jeremy. “It’s All for Your Own Good.” The New York Review of Books, vol. 61, no. 15, Oct. 2014, p. 21.
• Zaretsky, Eli. Secrets of the Soul: A Social and Cultural History of Psychoanalysis. Alfred A. Knopf, 2004.