When might behavioral-economics accounts of misfiring rationality fail to recognize our status as “sporting people”? When might we will by telling ourselves: “Let’s make this more interesting”? When I want to ask such questions, I pose them to Richard Robb. This present conversation focuses on Robb’s book Willful: How We Choose What We Do. Robb is Professor of Professional Practice at Columbia University’s School of International and Public Affairs, where he teaches courses in microeconomics, economic foundations of capital markets, and finance. He is CEO of Christofferson, Robb & Company, an investment management firm in New York and London, which he co-founded in 2001.
¤
ANDY FITCH: Many contemporary brain scientists might agree with Willful’s premise that we often arrive at rationales for our actions after the fact. But could you describe some of the broader philosophical underpinnings for this book’s conception of human motivation? Whose preceding models of human willing stand out as most meaningful for your own work?
RICHARD ROBB: Ten years ago, when I started writing this book, I expected I’d end up attributing a lot to Nietzsche. In fact, I titled the 2009 paper that attracted the interest of my editor, and ultimately resulted in the book, “Nietzsche and the Economics of Becoming.” But over time, it became apparent that much of what I cared about in Nietzsche’s work can also be found in Schopenhauer, Hume, and Kierkegaard. I do want to stress at the outset that not all action results from pure acts of will. I’m trained in traditional neoclassical economics, which I teach enthusiastically to my students. I still consider that model (of agents ranking options and picking the one that leads to the best possible expected outcome) our most powerful tool for understanding human behavior. But certain actions don’t fit that paradigm. We may fail to see this because our tendency to rationalize after the fact only makes the two-fold nature of our experience more obscure.
Then on a more autobiographical timeline, Willful describes you coming to this book’s present understanding of human volition in a pretty roundabout way, through introspective efforts to account for certain life decisions by applying some airtight University of Chicago conception of rational-choice theory. Willful speaks in quite personal terms of your own happily workaholic engagements, your blissful immersion in collective endeavors and competitions — all occurring independently of any comparable external reward. So here especially for peers confronting their own disorienting personal (and sometimes quite purposeful-feeling) departures from theories of rational choice, how might you start to graft your conception of homo ludens onto more familiar models of homo economicus?
Well, for the past 40 years, my field has offered behavioral economics as an alternative for those who feel uncomfortable seeing themselves purely in terms of the homo economicus, the rational agent governed by predictable calculations. Behavioral economics provides a catalog of around two hundred cognitive biases identified in laboratory (or sometimes natural) experiments. It conceives of our actions as purposeful (trying our best to get what we want) but prone to mistakes, falling prey to biases that may have served early humans but hinder modern ones. The core assumption from rational-choice models still carries through: we compare and rank possible outcomes, consider the trade-offs, and then try to get what’s best in light of our resources.
As you mentioned, I left Chicago with the feeling that rational choice explains a lot but not the whole story. Since it seemed like no better explanation was available, I acknowledged that behavioral biases may come into play from time to time, shrugged, and moved on. But the doubt I’d experienced as a student carried forward into my career in finance and teaching. I felt increasingly convinced that rational choice, whether or not it’s augmented by behavioral bias, still misses a lot. That sensation really crystallized for me during a spell of unemployment. This was, to be clear, comfortable unemployment. Yet all this leisure time, which, from a neoclassical economic perspective, should have increased my utility, wasn’t satisfying at all. I would wake up around 6:30 AM, help make my children’s breakfast, send them off to school, run around the reservoir, read The New York Times, and then it would be nine in the morning. Now what? The day is long. All these activities that I’d assumed would fill my days with meaning did not. And when you’re missing a goal worth pursuing, you can’t just go to the striving store and buy more striving. You can only fully embrace challenges that feel in some way authentic.
Once this became apparent to me, I gradually started to explore, develop, and test in my own way the theory that some actions, as a matter of logic, do not map onto well-ordered preferences. They are neither rational nor irrational, but rather one-time acts of will that no one, not even the individual who undertakes them, can predict. Each of these actions stands for itself — so that’s how I began to refer to this entire realm of human activity. The for-itself centers on the homo ludens, the sporting person, who chooses challenges to overcome. Rather than continuously maximizing some weighted function of well-being on behalf of present and potential future selves, the homo ludens gets lost in the chase.
Could you sketch a few basic sociological, and/or economic insights likewise gained by clarifying this distinction among the rational, the purposeful, and the for-itself — perhaps specifically in terms of human labor, entrepreneurial activity, financial investment?
Well, in my schema for human action, “purposeful choice” refers to the combination of rational choice and behavioral bias. Purposeful choice is contrasted to the for-itself. “For-itself” describes choice without the intermediation of preferences. It does not allow for optimization, because in this realm we cannot meaningfully compare one outcome to another.
This framework tries to clarify the way we think about our behavior through time. Rational choice imagines that we choose at each moment on behalf of the present and potential. Economists formalized this idea in a model called geometric discounting. If we assume perfect certainty about the future (to keep things as simple as possible), then the math works. With geometric discounting, as time unfolds, you should willingly stick to the path you had initially judged to be optimal. But this formula doesn’t square with the real world. It’s the product of mathematical convenience rather than experiential or psychological realism.
It turns out to be impossible to reach an effective preference ordering about potential paths for long-term well-being. To put it another way, if a demon forces you to commit to the future path that seems best from today’s point of view, your future self will still want to stray from this path. Any trade-off of well-being in the distant future, for well-being in a slightly more distant future, will look quite different once that distant future becomes the present. When future me changes his mind and no longer wants what I chose long ago, someone might feel tempted to say: “This person has a problem. He can’t stick to the plan he made.” But is this true? Does changing my mind indicate a pathological relationship with time?
Behavioral economics would say yes. Behavioral economics pathologizes this behavior, calling it “hyperbolic discounting” or “time-inconsistent preferences.” But I argue that this only appears to be a defect if we apply the homo economicus model where it does not belong. There is no demon who forces us to commit once and for all to future actions. The more we can make peace with real-life dynamics, and accept that behavior over time is better framed in terms of the homo ludens, the more clearly we’ll understand how and why humans seek challenges to overcome. Even in the absence of new information, straying from the path that once seemed best doesn’t have to mean we’ve lost our way.
Some people get worked up about commitment devices. These are tricks to force yourself to stick with your decisions, like when Odysseus binds himself to the mast and instructs his sailors to stick wax in their ears so they can sail past the sirens without leaping overboard. To take a more up-to-date example, maybe you deliberately place your alarm clock across the room because you want to commit to waking up early and not hitting the snooze button. Or you might give your car keys to a friend the second you arrive at a party, because you know you’ll drink too much to then drive home. These are all cases of rational planning for anticipated periods of temporarily impaired judgment. But to live one’s whole life this way (always calculating in advance, always binding oneself, always pledging to work harder, always saving more) needlessly ties our planning into knots.
Rather than taking commitment devices and failures of geometric discounting to mean that we have rational preferences but fail to follow through on them, we should accept that some decisions about how we arrange our action through time are not dictated by preferences. In these cases, another kind of experiential process plays out. That doesn’t mean we should abandon the neoclassical economic model entirely. We should simply recognize that some of our actions are self-justifying and for-itself.
The purposeful model does not explain, nor can it be adjusted to explain, the importance of work in our lives, or why we are sloppy with personal finances, or start quixotic businesses, or stick with projects over the long-term without evaluating whether to quit. All of these are natural activities for the homo ludens that cannot be fit into the narrow world of the homo economicus and his preferences.
For example, the homo ludens won’t beat himself up over procrastination. Procrastination may or may not improve results by harnessing “good stress” (I have no idea). But either way, it allows us to feel the bite of time. Postponing dull tasks right up to the deadline can make the game more exciting. To still feel as rational as possible, we may tell ourselves a story: “Other priorities forced me to put this task off until the last minute,” or “I suffer from a minor psychological defect that makes me a procrastinator but am fine otherwise.” The true explanation may be even simpler: let’s make this more interesting.
Here again in terms of why our own lived experience may not feel at ease with strict rational-choice models, could you outline, perhaps with Dostoyevsky in mind, Willful’s emphasis upon our desire to will — regardless of whether such willing comports with the supposedly objective, predictable dictates of one’s self-interest? Why might the inclination to will push us in fact (and almost by definition) in some alternate direction?
Have you read Notes from Underground?
Multiple times.
Yeah, I still remember my exhilaration on first reading it. In the passage that has stuck with me the most, the underground man says, “that’s the whole bane of it, that this tricky profit doesn’t fall into any classification, doesn’t fit into any list.” This seems to me a takedown of Jeremy Bentham’s “felicific calculus,” a formula for achieving utmost well-being by maximizing 14 different pleasures and minimizing 12 pains. I see that particular kind of utilitarianism as a burlesque version of rational choice — where you’re just born into this straitjacket. Rational choice at least gives you some rein to form your own idiosyncratic preferences.
While I suspect Dostoyevsky’s “list” refers to Bentham’s pleasures and pains, in modern lingo this list consists of the inputs to an economist’s “utility function.” Some of my economics colleagues have reacted to my theory by asking: “Why do you need a whole new frame of reference? Why not just toss acts of will into the utility function?” But it doesn’t work. Plugging acts of will into the list would reduce us back to passive agents simply tabulating pre-existing calculations, and so our actions would no longer be willed.
Still, while this for-itself realm can’t be subsumed within neoclassical economics, the idea that some actions are self-justifying is not new or revelatory. You can find it in the works of early existentialist writers like Kierkegaard, Dostoyevsky, and Nietzsche. Also, depictions of individuals rationalizing their actions post-hoc have circulated for a long time. My job in this book is to revive the argument that there are two realms of action, and to demonstrate that not all activity fits in the domain of purposeful choice. Recognizing the boundary between the purposeful and the for-itself can help us make better decisions, and be more at ease with the decisions we already made.
So here could we step back to your friends’ recommendations to just add a few more complicating factors to the overall equation? Let’s say that, in Willful’s account, behavioral-economics approaches still posit a driving force of personal choice, but now including factors such as habit, inertia, misperception, bias, and illogical association within a vectored calculus. Could one argue that your for-itself model does something relatively similar — adding the previously underrecognized value of expressing one’s will into the mix? Or how else might you position various subtle gradations between acting purposefully and acting for-itself?
It was absolutely my aim with this book to go beyond adding new entries to the catalogue of behavioral biases. I try to treat behavioral economics with a certain amount of respect, but am reluctant to infer too much from low-stakes lab experiments. Often, if we work hard enough, we find that behavior appearing to result from biases turns out to be rational.
Your question, why can’t we just add some complicating factors, is the question. The burden of proof is on me to show that, even allowing for behavioral bias, the purposeful paradigm is insufficient. If I’m making the picture more complicated, I’d better have good reason. My argument hinges on three aspects of human action that transcend this paradigm. First, as we’ve discussed, intertemporal planning. Rational choice attempts to optimize on behalf of a sequence of future selves. These attempts fail unless we make assumptions that, on close scrutiny, are strange and implausible and bear little resemblance to our real experience of the flow of time. Second, our stubborn commitment to certain beliefs even when adopting others would help us get ahead. Our beliefs make up our identity, and so may not be up for sale. Whenever possible, we stick with who we are, rather than optimizing on behalf of the new person we might become. Third, a subset of unpredictable, intrinsically human altruistic gestures that aren’t “better” than other gestures in any coherent way.
As you say, the purposeful and the for-itself are often bound together in the same activity. For example, we might do a favor for an acquaintance in hopes of future favors, or to signal our cooperative nature. That’s as purposeful as can be. But if we spontaneously and inexplicably choose to help more than is necessary to advance any clear goal, then this is partially for-itself. The most important facet of life where these two realms comingle may be on the job. We choose what work to do and how much effort to exert partly to gratify preferences for money, leisure, and other factors like status or camaraderie with coworkers. At the same time, for many people, work is the ultimate team sport. We want to pit ourselves against rivals within our firm, or help the firm beat out the competition. Or we might simply want to achieve goals at work that we have deemed (perhaps without clear reason) to be important. How else to explain why some people work long hours after they have enough money, sometimes until the day they die?
Your formulation of human agents as often deliberative and rational, sometimes biased, sometimes “none of the above,” also makes me wonder if/when this book conceives of the will as unitary, and when as pluralized. Here I think, for instance, of Nietzsche’s aesthetic priest willing “against” himself, if only so that he can will anything at all. Or I think of Freud’s civilizing superego paradoxically deriving its strength from the id’s primal projective identification with some authority figure. So of course any willing on our part takes place amid a teeming array of social/environmental forces. Of course our will faces a host of external obstacles, deterrents, challenges, rivals. But do you see such divisive struggles, or such partial and capricious and yet essential-feeling “victories,” also taking place at times within ourselves, “against” ourselves?
Freud supposedly said that Nietzsche knew himself better than anyone who had ever lived before or was likely to live in the future. The psychological insights of the thinkers this book draws on are profound beyond my ability to weigh in. And the internalizing questions you raise are enormously important, but I don’t have much to say about them. I mostly want to understand our actions, rather than the underlying sensations. I want to differentiate the conditions under which we try to optimize from those where we do not.
In purposeful choice, a single motive (optimization) drives all action. The for-itself lacks an analogous unifying motive, although it is generally characterized by the exercise of will. By this I mean something like Keynes’s concept of “animal spirits” (which he actually borrowed from the ancient Greeks), though not in the vulgar, dumbed-down sense it’s now used to refer to a kind of irrationality that grips markets and market participants. Keynes deserves better than that. Tom Nagel gets at the truer meaning of “animal spirts” when he describes the blind spot that hides something we cannot take into account while acting, because it is what acts. This is what I’m referring to as “will.” I’m interested in exploring the many different manifestations of the will rather than its source or the innermost feelings that drive it. I happily leave these hard questions to others to sort out.
Rational choice forces us to contemplate questions like: what will the version of me who exists x units of time from now want? How would a homunculus optimizing for one particular metric guide me in this decision? If I adopt new beliefs that turn me into a somewhat different person, what will that person want? As a rational agent, how should I act in order to optimize in any given context? Willful action is a counterpoint to this unwinnable fight of, as you call it “within ourselves, against ourselves” — somehow truer and more organic, more a remedy to the fragmentation that arises from an obsessive commitment to interpreting our actions in terms of rational choice.
Sure, and in terms of blind spots we might be better off just living with, Willful notes the persistent need, in most rational-choice models, to find some commensurate rate of exchange among seemingly unlike goods, qualities, experiences, principles. Here could you again flesh out Willful’s lived introspective musings (“Does removing everything from its context to determine the rate of exchange at which we would trade this for that…impoverish our experience?”), while also retaining a practical, Trolley Problem (“Whom should be saved when resources are limited? How much should we spend to save them?”) recognition that we do need to rationalize our way to the best possible calculus among incommensurate entities all the time?
Right, I don’t want to take the extreme approach of calling everything incommensurable and saying we shouldn’t ever bother trying to determine the best possible course of action. I’d consider that a big mistake. We know, of course, how to weigh calculations, such as: which works better for a grilled cheese, gruyere or cheddar? It might depend on the circumstances, but…
No I was with you on the grilled cheese.
Great. Agreed, gruyere is better. Maybe you’d also agree that a roller coaster is better than a Ferris wheel. But which is better: cheese or a roller coaster? Even a child recognizes that you can’t compare the two. Yet, in our lives, we might have to give up the grilled cheese to make it to the amusement park on time. We have to compare completely unlike things to live an effective life in the world.
This book explores the seemingly straightforward example of choosing where in Manhattan to live: downtown, midtown, or uptown. I have to weigh cost as well as the joint preferences of my wife and me. To me, the most important concern is the commute to midtown (where my office is located) and to uptown (where I teach). This commute, of course, involves many factors, not just distance.
Express train versus local.
Right, and every time you take a taxi those same kinds of questions come up [Laughter]. Rational-choice models would tell us we need to abstract certain concerns (such as whether to spend money on apartment size, or quality) and to squeeze competing desires down into one single dimension, so we can maximize our preferences. But choosing on a single reason would waste a lot of valuable information and would ultimately be inefficient.
Here again, Nietzschean and Humean conceptions of temporality also came to mind as I progressed through Willful’s account of time as an ongoing flow, forever spilling beyond the bounds of economic models fixated on the supposed single moment in which decisions or transactions happen. Nietzsche in fact declares that no “doer” exists at all, no “deed” except for an infinite string of overlapping befores and afters. How might that more abstracted perspective on causality and temporality relate to your specific concerns with rational-choice models failing to account for aspirational aspects of human motivation — for our anticipated obligations and reward, themselves always getting reshaped by a horizon of new future prospects (at least until, as Kierkegaard might have it, we die)?
Your description of Nietzsche’s account of action is exactly what I have in mind. The for-itself doesn’t have a “doer” or a “deed.” Rational choice, though, conceives of a model where time stops, we evaluate alternative bundles, pick what’s best, and then experience the consequences once time restarts.
Here, as part of reconceiving the future-tending stakes of our present choices, Willful also parses in fascinating ways how altruistic behavior can play out somewhere between an instrumentalized social vision (in which deliberate actions directed at certain human objects produce certain predictable consequences), and a much more authentic, personally identified (perhaps overly so) vision in which we basically see others as extensions of ourselves. Willful offers clarifying explanations of how, say, effective altruism might forgo the generosity of the spontaneous act, prioritizing the meticulously strategized outcome. What else can you say here about how even the for-itself will might find itself acting generously?
People act altruistically for many different reasons. Some altruism is selfish in clear-cut ways — you want to cultivate a particular image, or predispose others toward helping you in the future. Altruism sometimes results from a desire to adhere to good manners or social norms. Or maybe you’re usually well-mannered, but you do a quick cost-benefit analysis and decide not to hold the door for someone when you’re in a hurry [Laughter].
In the strongest forms of caring, the welfare of another person becomes part of your own welfare. To use Gary Becker’s example, consider the deep care a mother might feel for her child. Becker argues that, in this type of bond, even a “rotten kid” might act altruistically. The selfish child recognizes that when something hurts the mother more than it benefits the child, the whole household’s well-being decreases. The mother consumes less of everything that matters to her, including dedicating fewer resources to the child’s well-being. The child loses out when the household loses out. So the child optimizes on behalf of the household, exemplifying a very peculiar kind of selfish altruism. And not all care is as easily observed as is the mother’s in this example. It doesn’t necessarily make you a hypocrite to say you care about something without doing anything about it. You care, just not enough to act.
But while many different forms of altruism fit in the purposeful realm, some do not. In my favorite example, the Good Samaritan starts out by checking on a beaten man on the side of the road. Given the Samaritan’s values, that might be perfectly rational — it’s what he’d always do. But then the Samaritan takes the man to the inn and pays the innkeeper to look after him. If the Samaritan did this for everybody in need, he’d soon run out of money. If the Good Samaritan wanted to devote himself to helping humankind in some coherent way, he would have helped this beaten man get back on his feet, and then rushed off to help the next most needy person. We might respect and admire the Samaritan, but we can’t call it rational to lavish all this attention on one random man. The Samaritan just did as he liked in an unpredictable act of will.
If you asked the Samaritan “Why’d you do it?” he might provide a rationale. But the real answer is: just because. To claim instead that the Samaritan fell victim to the availability bias or something like that and failed at fulfilling his broader commitments seems a pretty big stretch.
Can even seemingly impulsive, irrational, for-themselves acts of generosity like this sometimes reinforce an under-recognized social value — that of dignity? Can such spontaneous gestures provide public recognition that this person, precisely because of his/her own intrinsic worth, deserves assistance (perhaps from anybody) in this particular circumstance?
I like that question. Another biblical act of mercy comes to mind: Abraham, fresh off massacring thousands, pleading with God to save the people of Sodom and Gomorrah, effectively Abraham’s enemies. He risks his political capital in an attempt to save exactly the wrong people — it’s not rational, but who would say he’s caught up in a behavioral bias? I discuss an episode in 1972 when an American general “shut down” the Vietnam War for several days, sending out over a hundred aircraft to rescue one lieutenant the enemy had captured. This flies in the face of any rational cost-benefit calculation. The general had to decide by himself as an act of will. It’s a dramatic recognition of the dignity of a single person.
Questions of dignity return us to your incisive description of economic models often failing to account for why stubbornly holding onto one’s beliefs might not indicate an “irrational,” or “delusional” disposition, so much as an understandable inclination to think for-oneself. And here, to take just one glaring present-day example, I wonder what it might mean for us to conceive of, say, at least some recalcitrant protectionist voters (particularly those in displaced postindustrial communities) as truly protecting, truly defending, not just their nostalgic or bad-faith policy contradictions, but their identity, their relevance, their dignity. Here could you talk through the very real concerns that such Americans face when summarily told to “reinvent” themselves? Could you sketch how a political rhetoric calling for slower, steadier economic change might prove much more persuasive, fair, reachable, even reasonable to such audiences?
David Ricardo tells the story of Portuguese winemakers and English clothmakers both finding themselves better off when the countries agree to trade with each other. Everybody knows this story of comparative advantage. By transferring money to the losing English vintners, it’s theoretically possible to make everyone better off. So we can criticize protectionists for trying to preserve their own jobs. We can ask: “Don’t these people understand the potential efficiency gains and economic benefits we’d achieve by moving textile manufacturing to Vietnam, while workers in South Carolina get re-trained in consulting services?” And they very well might. They have almost certainly heard the argument and have the intellectual capacity to follow it. Instead, they glom onto demonstrably false “job stealing” narratives to rationalize their belief.
Traditional neoclassical economics lacks the language to talk about how individuals embrace challenges and seek obstacles to overcome. Even if displaced textile workers can theoretically end up with more money and more leisure than they had before, they are not clamoring for monetary transfers. They want their old jobs back. We all have our own ways of doing the work we consider dignified, and of providing for our families, for ourselves, and for society. And society may value those efforts even when it could procure the same goods and services more efficiently from a developing country. I’m certain, in the long run, efficiency will win out. But as a society, we could drag our heels.
Finally then, returning to questions of lived experience, what has your own diverse range of professional, entrepreneurial, intellectual, pedagogical pursuits taught you most broadly about how the field of economics as we know it today might emphasize the empirical tracking of consumption, the logics of rational choice, in part because such points of focus can give economists themselves authority as information gatherers and interpreters and evaluators? How/when might this discipline benefit by incorporating more first-person accounts of semi-impulsive, semi-irrational living — never fully accounted for by the mechanisms of theoretical purposeful choice? How might you steer readers less towards seeking some “correct” explanation for their actions, and more towards exploring the “art of crafting meaningful challenges”? And/or how else might economics better integrate doers among its teachers?
Well, one early peer reviewer of my manuscript wanted to see an all-out attack on economics. He wanted this book framed as: “Here’s a guy from the inside. He went to Chicago. He teaches at Columbia. And now he’s gone rogue.” And I’m sorry if I end up disappointing some people, but I’m no heretic. That’s just not the way I see it — neoclassical economics is a powerful tool for understanding human behavior. I try to show my students the primacy of rational choice for analyzing decision making and public policy.
In my graduate seminar, “Foundations of Individual Choice,” I always ask the students: “What interested you in taking this class?” One after the other they say: “I don’t see myself as a fully rational creature, so I want to know more about behavioral economics.” But is that all we have to offer — the idea of life as a series of passive calculations maximizing fixed preferences that we sometimes mess up?
In the class, we explore why it’s impossible to shoehorn everything into the rational framework. But for me, adding to the list of biases doesn’t deliver a very satisfying alternative. I want my students to see that it can be acceptable, even therapeutic, to go beyond explanations for all action in terms of the purposeful.
Since most of my students have never encountered Aristotle, I introduce them to the Nicomachean Ethics. We start by asking: “What is happiness? What is human flourishing?” They assume that we won’t find an answer to these big questions. But Aristotle’s answer is precise and even practical: happiness is a process of the soul acting in accordance with virtue, which is a mean between two extremes. By engaging in virtuous acts, you become virtuous. I like to present this theory as an antidote to rational-choice economics. Conceiving of ourselves and others as nothing more than passive preference-maximizers can lead to what Keynes called “a thinness, a superficiality, not only of judgment, but also of feeling.”
I’m not going to take the bait though, and suggest that economists need to get out more. For sure, giants in the field like Ricardo and Keynes led active business lives that shaped their theories. But others, like Adam Smith, Paul Samuelson, and Edmund Phelps, spent equally productive careers as scholars, learning about the world from data and keen observation.
