I’ll begin with some confessions. When the facts change, I usually don’t change my opinions unless I’m backed into a corner, and then I’ll change them by as little as possible.

But it doesn’t feel that way. We may choose badly or make the same mistakes again and again, but at some level we feel as if we are trying to get what we want. When we do act without a purpose, we invent a reason after the fact, like a sleeping person who hears a barking dog and weaves it into the narrative of her dream. Inventing reasons in this way preserves our self-image as rational. It might sound like I’m rejecting the backbone of economic theory, rational choice, but to do so would be a mistake. Rational choice has illuminated huge swaths of behavior by emphasizing that we do our best to satisfy our desires with the information and resources at our disposal; we compare all available options and choose the one we prefer over the others.

I’m not launching an attack on this theory; far from it. I start out each semester defending rational choice against two objections that students usually raise: they don’t feel like calculating machines and they are not materialistic.

The first concern is unwarranted because your actions may adhere to rational choice whether you know it or not. Arthur Schopenhauer tells the story of an elephant traveling through Europe, crossing many bridges. The elephant stops dead at one rickety bridge even after seeing men and horses cross, having sensed that the bridge cannot bear its weight. The defiant elephant illustrates the intuition behind much of economics: when a decision really matters, people and even animals are pretty smart.

As for the second objection, economics does not assume that people care only (and unattractively) about themselves and their material well-being. The satisfaction, or utility, that an individual chooses to maximize might depend on inputs like altruism, the well-being of others, or adherence to ethical standards.

Even after allowing for altruism and accepting that calculations can be intuitive, the idea of yourself as a strictly rational actor may leave you a bit queasy. Conventional thinking offers a palliative: behavioral economics. Behavioral economics has extended rational choice to account for biases and heuristics. A person acting with a behavioral bias also tries to satisfy her desires, but routinely misses the mark. Behavioral economists hope that identifying biases will help people mend their ways and act in conformity with economic models. If rational choice theory conceives of people as robots whose behavior is determined by their preferences, then behavioral economists believe that those robots are badly programmed.

Both rational choice and behavioral economics assume that action is purposeful, that people seek the outcomes that best gratify their preexisting desires. People either know their preferences and can describe them out loud, or sense them and act as if they understood what they wanted. The purposeful choice model can explain many things, but not everything. Certain actions are undertaken not for any tangible benefit but for their own sake. They cannot be ranked against, or traded for, other actions. These actions belong to a second realm of behavior that is neither rational nor irrational, but for-itself.

Suppose a woman is about to jump into a river to save her drowning husband. We would not expect her to behave rationally — that is, to calculate the “present value” of the future benefits that she might derive from keeping her husband alive multiplied by the probability she will be able to save him (net of the probability he would save himself without her help) and then deduct the probability that she would drown multiplied by the value she attributes to her own life. It’s good enough that the drowning person is her husband, whom she loves. Any justification, any model or calculation, any attempt to validate her action as a realization of some general principle, would be weaker than that fact. Any additional reason for her decision would be, in the words of the philosopher Bernard Williams, “one thought too many.”

The distinction here is not in the magnitude of the decision. A great deal of everyday non-husband-rescuing behavior belongs to the for-itself realm. In the 1942 Preston Sturges screwball comedy The Palm Beach Story, an elderly Texas sausage magnate, the “Wienie King,” decides to lend a hand to penniless Claudette Colbert. She reminds him of himself when he was young and poor, so in a spontaneous, one-time act of mercy, he peels off $700 from his money roll, gives it to her, and says, “so long.” The Wienie King can’t help everyone he meets even though other potential recipients may be more worthy of aid. His for-itself gesture to Colbert was not predictable; he just did as he liked.

While neither the husband-rescuer nor the Wienie King acts on the basis of any calculation in these instances, they surely do in other contexts. I’m not asking you to jettison purposeful choice altogether, only to recognize that there’s more to the story.

Perhaps most of your behavior fits into the purposeful model — sometimes you’re a rational agent, confident of the best course of action and able to explain your reasoning. Sometimes you’re a super-smart elephant who knows intuitively what action is optimal, and sometimes you’re the victim of behavioral biases. But then, at other times, you’re none of the above.

Admittedly, I’m an unlikely advocate for the idea that motives don’t have to be purposeful and behavior doesn’t have to be maximizing — that we’re not always trying to pick the best available option given the information at hand. My stance is incongruous not only because I am an economist but also because I was trained at the University of Chicago, the high temple of rational choice economic theory, and still teach it enthusiastically to my students.

Drunk on Theory

I came to the realization that not all our actions have a purpose in a long and roundabout way. I began the 1980s drunk on neoclassical economics, the theory that assumes people choose rationally and that supply and demand are in equilibrium, and then tries to explain as much of the world as it possibly can. As a PhD candidate at the University of Chicago, I saw people acting rationally everywhere I looked. Economic theory applied not just to money and markets, but to everything.

Why did the A&P package fresh green beans in little cartons? Simple: if the store placed loose beans in large bins, consumers would hunt for the best ones up to the point where the extra benefit equaled their wage. The store eliminated wasteful search by selling randomly selected beans to everyone. Consumers would pay more to avoid wasting time competing for quality.

Should the A&P put the best beans on top of each package where they’d be visible to consumers? No, because the store would have to pay workers to hide the lower-quality ones, and rational consumers would learn to discount appearances. My classmates and I told stories like this all day long. Gradually, we thought, the world was revealing its unseen order.

Not that doubt didn’t creep in around the edges. We wondered why we’d chosen to live at a lower standard of comfort than if we had tried some pursuit other than graduate school at Chicago. We had little money. It was freezing cold. My apartment was so infested with roaches, I’d given up trying to kill them. Approximately 80 percent of the entering class would be tossed out before receiving a PhD.

A few of us had been accepted to equally prestigious programs but chose to tough it out at Chicago with its notoriously difficult qualifying exams. We told ourselves that attending the University of Chicago was the best way to build human capital — capital that would lead to reasonably high earnings in stimulating academic careers. But deep down, we knew that wasn’t the real reason. Somehow, we liked that it was hard. Our attraction to struggle seemed perverse as we tried to reconcile our actions with a cherished theory that felt not quite right.

Early on in graduate school, my classmates and I stumbled on behavioral economics, which was then emerging as an alternative to rational choice orthodoxy. Cognitive biases were documented in all sorts of lab experiments. In one famous experiment, subjects were indifferent between receiving $10 immediately and receiving $21 in one year. They were also indifferent between paying $10 immediately and paying $15 in one year. Since a rational person ought to be willing to trade small amounts of cash now for cash in one year at a single discount rate, whether paying or receiving, this discrepancy was interpreted as evidence of “gain-loss asymmetry” — meaning that people need more compensation to delay gains than they are willing to pay to delay losses.

Maybe it was that easy. If this were the case, all we had to do was document biases through experiments, like the one on gain-loss asymmetry, and adjust our models accordingly. Rational choice, with all its insights into markets and many other aspects of human behavior, could largely be preserved. But in the end, behavioral economics did not seem to be the solution to what we thought neoclassical theory was lacking.

In most circumstances, when behavioral economics offered a psychological solution for some ostensible puzzle, we could explain the data with rational choice if we worked hard enough. With the gain-loss asymmetry experiment, what about the cost of collecting the debt from the professor running the experiment? Subjects receiving a payment should be inclined to take the money now rather than have to track down the professor in a year and convince him to pay. Compensation of $11 for credit risk and inconvenience of collecting seems reasonable. Likewise, subjects who have to pay would be smart to gamble $5 in hopes that the professor would forget all about collecting and they’d never hear from her again.

Considering these factors, the experimental results made sense. Ultimately, my classmates and I likened the behavioral economists’ experiments to optical illusions: entertaining and sometimes instructive, but hardly central to everyday life. In the absence of any better ideas, I made an uneasy peace with economic theory. I accepted that behavior is purposeful and choice is mostly rational with a bit of cognitive bias tossed into the mix.

After graduating in 1985, I took a job in the bond business in Chicago. As time passed, I remained convinced of the power of neoclassical economics and wary of the popular alternatives. Yet I also grew increasingly uncomfortable with the extent to which the traditional model failed to square with my own life.

First, I found plenty of truth in the saying that the journey is more important than the destination, even though the journey has little place in a worldview predicated on purposeful choice. On the job, I became blissfully lost in challenges that took on their own meaning. Sport often seemed like an apt analogy for how I competed to outsmart the markets and how my firm competed as a team against other firms.

Second, I was troubled by how I clung to my beliefs, more or less, even when they came in conflict with new data or the views of experts. I was amazed by the wide range of opinions I encountered outside the graduate school bubble. Why didn’t all these supposedly rational actors converge on the common view that was best supported by the evidence?

Third, while some of my dealings with other people could be understood in terms of rational choice as I had been taught, many could not. They were more complicated, or perhaps less complicated, than I could explain. Why, for example, would I give this person a break today but not tomorrow, and why not someone else equally close to me or equally worthy? A cost-benefit calculation didn’t always apply.

Finally, I began to wonder what effect seeing the world in terms of rational choice has on our inner lives. Does removing everything from its context to determine the rate of exchange at which we would trade this for that — even if subconsciously — impoverish our experience? I wondered whether John Maynard Keynes might have been right when he warned, the “pseudo-rational view of human nature [leads] to a thinness, a superficiality, not only of judgment, but also of feeling.”

Theory Collides with Evidence

In 1992, I moved to New York City to become the head options trader for the derivatives subsidiary of the Dai-Ichi Kangyo Bank (DKB), Japan’s largest bank at the time. I came to love DKB. My job was a sport that I could play every day.

The work felt important — we were solving problems that mattered to the bank and its clients — and the challenges we faced were stimulating and continually shifting. Eventually I was promoted to global head of DKB’s derivatives and securities subsidiaries in New York, London and Hong Kong.

The job left me with little leisure time, but that was OK. There was nothing I’d rather do. My most memorable experience at DKB came in November 1998, during the Asian crisis. Vaunted Japanese financial institutions like Nippon Credit, Long-Term Credit Bank of Japan and Yamaichi Securities had gone bust, and DKB was teetering on the brink. We were set to underwrite our third Japanese auto-loan-backed security for the giant consumer finance company Orico, but the managers in Tokyo told me to cancel the deal. They were worried we would fail to sell Orico’s securities to investors and embarrass the bank.

I was enraged. I had committed to raising this money for Orico and was looking forward to demonstrating that DKB could proceed with business as usual even when others had lost their nerve. I told my bosses in Tokyo that I would quit, and probably everyone else on the team would, too, unless we were allowed to complete the deal. In response to this threat (a bluff), we were allowed to proceed. We agreed to cut the size of the offering and promised to sell every last bond. If we failed, we would not have the chance to quit; we would be fired. The DKB bond sales force rallied to the challenge, and the day the deal closed was one of collective joy.

Why should I have cared whether DKB canceled a deal? Why am I still talking about it 20 years later? An economist might argue that I was concerned that a failure would cause my future earnings prospects to drop, either within the bank or if I looked for a new job. But that wasn’t it. The grown-ups were going to take away our ball in the middle of the game. It was no more than that.

It was getting harder and harder for me to cling to the view of work as a sacrifice of leisure to get money for consumption. I could not have bought my experiences at DKB — they were not for sale, and even if they were, it would have made no sense to pay for them. Nor could I determine a price at which I’d be indifferent to trading them for other things. It gnawed at me that I could not squeeze the attributes of my job into the framework of rational choice. Another sort of analysis was needed.

The game truly did end in 2000 when DKB merged with Fuji Bank and Industrial Bank of Japan. There was no role in the new bank for me. My boss told me, in a masterpiece of Japanese tact, “Frankly speaking, you are free to work anywhere you like.” I gestured to our trading room. He looked down and mumbled, “Anywhere but here.” I decided to enjoy some leisure time, which after 10 years with the bank I could afford. But it was not at all what I expected.

Each morning in my early retirement, I read The New York Times, helped make breakfast for my 9-year-old daughter and 14-year-old son, sent them off to school and jogged around the reservoir in Central Park. At that point, it was 9:00 a.m. I discovered that the day is long. Weekends blended in with weekdays. Sure, I was free from the annoyances that had cropped up at my job, but I didn’t like it at all.

My dissatisfaction didn’t arise from low income, as rational choice theory would suggest. My time at DKB had left me with ample savings, and I spent a few hours each week trading for myself. Shuffling around the apartment in my slippers was a comfortable way to make enough money to cover my family’s expenses.

I puzzled over this uneasiness until, one day during my abundant leisure time, I took my daughter to the Bronx Zoo. We watched the zookeeper feed the tigers a lunch of fish encased in blocks of ice. The tigers had to strip away the ice before they could eat; they seemed to relish the struggle. Suddenly, it became clear: I was like the tiger, except the zookeeper was feeding me all the ice-free fish I wanted, whenever I wanted. I knew my situation was fortunate and, to many, enviable, yet I craved a challenge. Eventually, I found two.

First, I started a hedge fund with partners in New York and London. Our fund aimed to raise money to invest in European structured credit — debt securities that package the risk of loans to individuals or companies. I could try to rationalize this undertaking in terms of wealth maximization, but in reality, starting a business put my nest egg in jeopardy. A nail-biting adventure was the way to conquer the doldrums.

I had invested in two entrepreneurial ventures before this fund, and both were disasters. They dragged on and on, sucking up more money than anyone had intended because we always believed that a change in our fortunes was just around the corner. I hoped that things would be different this time, even though the past is supposed to be the best predictor of the future and all that.

Our fund got off to a rocky start. Although we were able to swiftly raise $11 million, that’s not enough for a viable hedge fund. We met with around 100 potential investors and kept hearing the same message: “You’ll be a safer investment at $50 million than at $11 million. You’ll be more diversified and have better access to financing. Come back when you’re bigger.”

We needed to reach a critical mass quickly. If no one invested, no one else would invest. If many invested, many more would join in. We were stuck in the bad equilibrium. Because we had chosen to register with the SEC, the law required us to raise $25 million in the first 90 days. We fell short of the mark and faced the embarrassing prospect of having to de-register. Our original investors began to make noises about getting their money back.

All at once, our luck turned. The SEC inexplicably gave us a one-month extension, and then a new investor swooped in with $40 million. Other investors fell into place, and within five years we had $2 billion.

My miserable spell of unemployment and sense of rejuvenation prompted by starting the fund led me right back to the same old dilemma. I could no longer accept the traditional economic model that saw a trade-off between work and leisure and emphasized the importance of consumption. Yet I wasn’t ready to turn my back on the orthodoxy altogether.

In a way, my experience trading bonds in Chicago, working at DKB and starting a hedge fund actually reinforced my faith in rational choice theory. Practically every day, useful insight came from assuming that people optimized and markets were in equilibrium. Economic theory often felt like my secret weapon.

The second challenge I took on was teaching microeconomics at Columbia’s School of International and Public Affairs, first as an adjunct and then in a full-time position. In 2002, a year after I started teaching, I had the great fortune of meeting Edmund Phelps. With his project at Columbia’s Center on Capitalism and Society, he intended to reformulate economics for the modern world. The center sought a theory to describe “real human beings who are not only acquisitive and risk-averse but also inquisitive and adventurous and who sometimes feel the need to take a plunge, to leap into the unknown.” This was new. I especially liked the phrase “not only.” There had to be room for both.

The realization that a second realm sits alongside purposeful choice was the turning point for me. Why should one grand system explain it all? People not only seek to gratify desires, but also choose obstacles to overcome. If they succeed, or tire of their project and abandon it, new challenges will arise. It’s natural that this is hard to see, since action can feel more intentional than it really is. We often act first, then invent a cause for that action that we usually describe in terms of seeking pleasure or avoiding pain.

This Tricky Profit

Of course, I didn’t discover this realm. Fyodor Dostoevsky describes acts that cannot fit into lists of preferences (in economic lingo, inputs into utility functions). He asks:

How does it happen that all these statisticians, sages and lovers of mankind, in calculating human profits, constantly omit one profit? … It’s no great trouble just to take it, this profit, and include it in the list. But that’s the whole bane of it, that this tricky profit doesn’t fall into any classification, doesn’t fit into any list.

This “tricky profit” that eludes all systems and models is the exercise of will. Dostoevsky continues:

Man, whoever he might be, has always and everywhere liked to act as he wants, and not at all as reason and profit dictate; and one can want even against one’s own profit, and one sometimes even positively must. [A]ll this is that same most profitable profit, the omitted one, which does not fit into any classification, and because of which all systems and theories are constantly blown to the devil.

Dostoevsky’s “most profitable profit” cannot fit into a list alongside all other desires in the rational choice framework. In that model, an individual is controlled by her preferences. If the exercise of will were one of those preferences, then the resulting actions would be predictable and automatic, and therefore no longer “willed.” So acts that result from the exercise of will rather than the pursuit of preferences must belong to a different realm: what I call “for-itself.”

Each for-itself action stands for itself without regard to whether it is better than some alternative. It is undertaken “just because.” It can be a flow or a process, a self-justifying game, or a struggle to overcome a challenge that is not important in any objective way. For-itself behavior includes acting confidently based on beliefs we hold and that matter to us (whether or not those beliefs are accurate) because that’s who we are. Our beliefs constitute our identity and so are not up for sale; “I yam what I yam,” declares Popeye the Sailor Man in a kind of for-itself anthem. The apparent opposite of entrenching ourselves in our beliefs — an impromptu leap out of character — is for-itself, too.

Above all, the for-itself realm emphasizes agency — an individual acting on the world. It considers time as a flow rather than as a single moment, which is all purposeful choice can truly accommodate. The purposeful model cannot explain (or be adjusted to explain) how we decide to consume now or later, to retire or continue working. Nor can it explain why we procrastinate, are inattentive to personal finances, start quixotic businesses and persevere with projects for long stretches without evaluating whether to quit. In contrast, the more fluid for-itself model captures behavior as it unfolds over time in response to the challenges we encounter. This behavior includes spontaneous altruistic acts that no one could have predicted and heroism of the one-thought-too-many variety (e.g., the woman rescuing her husband) that transcends any rational calculus.

Just as rational choice, and perhaps behavioral economics, has shed light on the sphere of purposeful choice, a methodical approach can be developed to analyze for-itself action — although not with the same degree of rigor. My intent is to understand real-world phenomena that matter to individuals and firms, and that impact public policy. My method involves drawing the boundary between these two modes of action — the purposeful and the for-itself — and then explaining for-itself conduct with as much precision as I can. Most of my examples are drawn from business and investing because that’s my area of expertise, but a similar analysis might be fruitful in fields like politics, education and history.

Although my thinking has evolved out of personal experience, I believe my thesis is nearly universal. For most of us, who try to see our behavior as rational, embracing one realm to the exclusion of the other leaves us with a picture of the world that is incomplete. We strain to fit for-itself conduct into a realm where it doesn’t belong and, despite all the ingenuity we may deploy, end up with unsatisfying results. Of course, I recognize that some people, when choosing how to live their lives, discount purposeful action and instead cultivate spontaneity. These people will be unmoved by my claim that there’s more to life than rational choice. But these happy few may have an opposite quandary: how to reconcile the rational side of their actions with their natural grace. They, too, may wish to explore the boundary between the for-itself — their home base — and the purposeful.

On the one hand, I don’t want to claim too much for my thesis. My approach will not lead to a completely unified theory of human action because such a theory is impossible. Nor do I want to deny the supremacy of the purposeful approach to behavior. While my early exhilaration with rational choice economics has faded, my conviction in its power remains.

On the other hand, I am equally convinced that the purposeful choice model has serious limitations. Only by venturing beyond purposeful choice — with or without behavioral biases — can we improve our theories of many practical matters. Important aspects of our behavior (even in financial markets) are best understood not by neoclassical economics or mathematics or adding to the catalog of behavioral biases, but rather by reassessing basic motives along with assumptions about how those motives relate to action. Some of our actions arise from for-itself impulses rather than the purposeful intent to satisfy desires.

Recognizing the twofold nature of our behavior will generate a truer understanding of ourselves — one that is more at ease with our experience. As Nietzsche observed, “Honey, says Heraclitus, is at the same time bitter and sweet; the world itself is a mixed drink which must constantly be stirred.”

Two Realms of Human Behavior

Before diving into the details, let’s take a closer look at the categories of purposeful and for-itself behavior, define our terms, and introduce the key themes that will emerge in our investigation of this second realm.

The diagram summarizes my schema for human action.

Purposeful Choice

While much of this book will focus on the bottom fork of the diagram, for-itself action, we’ll start with a quick tour of the top fork: purposeful choice. But first, let’s clarify some terms. I’ll use “purposeful” as an umbrella term when contrasted to “for-itself” or when I want to allow for the possibility of behavioral bias. I’ll use “rational choice” to refer to economists’ traditional neoclassical models or to emphasize that I’m considering choice that is free of behavioral bias. “Cognitive bias” and “behavioral bias” will be treated as synonyms. “Preferences” will refer to desires in the purposeful realm — we know whether we would prefer to satisfy one desire over another and can rank them. The terms “wants,” “needs” and “desires” will emphasize purposefulness. But, at some high level of abstraction, for-itself conduct also concerns desire and its synonyms: someone who doesn’t desire an action wouldn’t undertake it. The key difference is that purposeful choice deals with desires that can be compared to and traded for each other.

When we understand and evaluate the strength of our desires, we enter the purposeful sphere. Purposeful choice assumes people mostly know what they want, and when they talk about desires that they don’t intend to act on, they are merely confused.

As the old joke goes, two economists are walking down the street. The first one says, “I’d like to buy a car like that.” The second one answers, “No, you wouldn’t.” That’s the whole joke. The first economist doesn’t want the car enough to make the required sacrifices or else he would already have bought it. He reveals his true preferences through his conduct.

Behavioral economics assumes that people understand their preferences, but that defects in their mental apparatus impair decision-making. At least 150 behavioral biases have been identified, mostly through laboratory experiments, from the “ambiguity effect” (ruling out options when we can’t assign probabilities to possible outcomes) to the “zero-risk bias” (spending unwarranted amounts to reduce small risks to zero while ignoring bigger ones). Presumably, once people are made aware of their biases, they will try to correct them, choose more wisely, and become better off. Until then, the field seeks to build more accurate models of behavior.

Both rational choice and behavioral economics are effective in explaining our actions to the extent that we are, indeed, seeking to gratify our desires and can foresee what the consequences of our actions will be. This does not mean that we can predict what will happen with absolute certainty, but that we at least know enough about the structure of the world to assess the possible outcomes and the probability that each will occur.

In practice, we’re often only vaguely aware of what might happen. Rather than facing crisp, easily modeled decisions — if we take action x, then y will happen with probability p and so forth — we may feel awash in a sea of ambiguity. This reality doesn’t imply that we need to toss out the apparatus of purposeful choice altogether, but does suggest that under such conditions, economists should be modest in their claims. In purposeful choice, people do their best with the information they have and spend resources to gather more information when that’s the optimal course.

Nor does purposeful choice require that we have perfect knowledge of ourselves. We might get exactly what we want and then feel disappointed. We can learn through experience and, in recent decades, through positive psychology. Also called happiness research, this second branch of purposeful choice can be regarded as an aid to learning about our nature and preferences, helping us make more satisfying decisions.

Assuming that at least some people know whether or not they are happy, discovery of the cause of that happiness should be an empirical task: ask a lot of people how happy they are, find out about their lives, and analyze the data. Do happy people have high incomes? High income relative to others around them? Rising income? Or does happiness correlate with some other factor?

If people are similar, we can learn about ourselves via the systematic study of others. Here, self-realization becomes a group project: having conducted surveys or brain studies and assumed people are homogeneous, positive psychologists advise us to spend more time with friends, find shorter commutes to work and worry less about making money after reaching the upper-middle-class. Done in this way, happiness research effectively generates a collection of self-help tips. It may also be used as a guide for public policy — if we can objectively measure happiness, then governments can try to promote it.

Whether or not we have an accurate understanding of our desires, the steps for applying purposeful choice are the same: determine, as best we can, what those desires are; rank them, consciously or unconsciously; and choose how to satisfy them, given our resources.

For-Itself

Turn now to the south fork of the diagram: for-itself.

The first branch, beliefs, explores how we come to our beliefs and how they shape our actions. Much of the time, we stick with our beliefs regardless of the evidence. Does this make us irrationally stubborn? I don’t think so. We hold onto our beliefs because they’re part of our identity. Stick-to-it-ness belongs to the for-itself realm, as does an occasional leap out of character.

This is neither rational nor irrational; it’s just the sort of creature we are.

A for-itself commitment to our beliefs can explain rigidities in companies and markets, and why we sometimes fail to make profitable investments. An entrepreneur, for instance, may be lit up with conviction for her project and who knows? She may turn out to be right. But the venture capitalist whom the entrepreneur approaches simply can’t share her enthusiasm. The venture capitalist may be fully incentivized to unearth promising startups; she may think clearly and be open-minded, courageous, honest and intelligent. Yet she rejects the pitch. Nothing the entrepreneur says can bridge the gap in their beliefs.

That’s because the entrepreneur has been galvanized by a one-time opportunity. Its essential appeal is unique so it can’t be compared to opportunities that have come before, at least as far as the entrepreneur is concerned. It stands for itself. This murky realm is where the money is made, so limits to the transmission of belief is a key neglected topic in finance.

The for-itself dimension of belief also explains why people go to so much trouble to form their own opinions. Why not simply adopt the views of an expert? The expert might be incompetent, but at least she’s gone through a vetting process and, odds are, knows more than you. Still, we resist experts who tell us what we’re supposed to believe.

Defying them is a for-itself act of will. When expert opinion conflicts with our core beliefs, the for-itself response is to discredit the expert, shop for maverick experts, and either hang on to preexisting beliefs for as long as possible or adjust them as little as possible.

The middle for-itself branch, people, analyzes social relations. Without a doubt, many of our social dealings belong to the realm of purposeful choice. Performing favors in anticipation of receiving reciprocal favors down the road can help us get ahead. Or we may care so deeply about other people that their well-being enters into our own preferences and becomes part of what we maximize.

Most Sundays before my wife wakes up, I go to a bakery around the corner to buy her a blueberry scone. She values the scone. I don’t mind a short errand that gets me out of the house and I care about her well-being, so buying the scone optimally improves my well-being. If it’s cold or rainy or I’m particularly busy, the cost to me of buying the scone is greater than the benefit, so I rationally skip it.

Yet there are situations that draw us into the unpredictable realm of spur-of-the-moment good deeds. I often see riders on the New York City subway hold the doors so strangers can scramble on. Not only does this break the Metropolitan Transportation Authority’s rules, it strikes me as antisocial. The calculus just doesn’t add up: holding the door saves one stranger the five-minute wait for the next train, but delays 300 other passengers by 10 seconds, increasing the total waiting time by 45 minutes. If the door-holder’s aim is to benefit humankind, he or she should let the doors close.

That’s what I’d always think during my 10 seconds of extra waiting. For years, I considered this behavior irrational. Perhaps the door-holder wanted to maximize social welfare but suffered from a cognitive bias that attributed too much importance to salient information that had captured his attention — the one stranger about to miss the train. Or perhaps he or she was too confused by the calculations to reason properly. But maybe the confusion was really mine: I was trying to fit a for-itself act into the purposeful model. My preference for letting the doors close is the result of a calculation, while holding them open is an act of mercy that transcends calculations.

But even with this more nuanced understanding, I still let the doors slam shut. I adhere to an ethical principle — promotion of the common good — unless I have a personal connection to whomever is about to miss the train. If the cost and benefit were approximately equal, I’d be inclined to favor the single person over the abstract calculation, but a 45-minute social cost for one person’s five-minute benefit is just too lopsided. Still, I can’t really object to the door-holder’s defense: in a wholly spontaneous gesture, he reacted to a need as it arose.

The way we act on the world as we move through it will be treated more broadly with the final branch, time. Often activity that unspools over time is best understood in terms of flow, as a for-itself process of continuous choosing.

In order to explain our actions, we infer a beginning, an end and a causal link between them. This abstraction makes it easy to freeze each moment in conscious thought, but is it an accurate characterization of life? Kierke-gaard observes that we cannot continually keep ourselves “on the spear tip of the moment.” Rather, he asks us to

Imagine a captain of a ship the moment a shift of direction must be made; then he may be able to say: I can do either this or that. But if he is not a mediocre captain he will also be aware that during all this the ship is ploughing ahead with its ordinary velocity.

Simply put, time never stops so that we may evaluate alternative bundles, pick what’s best, and then experience the consequences once time restarts. Purposeful choice requires that we ignore the flux and interpret action in terms of static categories. The for-itself perspective will provide us with an alternative picture of time and the way action unfolds.

From the for-itself perspective, rather than maximizing some measure of present and future well-being, we choose the obstacles that we then struggle to overcome. The chase takes on a life of its own. The importance of the chase explains the impulse to become a professional basketball player or a rock star, and to persevere even as evidence mounts that the chance of success is approaching zero. Likewise, a student might complete a degree even after deciding to quit the field; she wants to finish what she started. (A rational choice economist could argue that finishing the degree raises her income because it signals grit. But the benefit of that signal is unlikely to outweigh the cost of a year of her life plus tuition.)

These projects make sense only if we accept that an activity can matter beyond its ostensible purpose. A quest is about fighting over time and, when necessary, against the odds.

These sorts of battles don’t have to be major undertakings. They can instead be relatively minor, or even pointless. A cook prepares a meal according to a difficult recipe even though the subtleties are likely to go unnoticed, and a tenured professor diligently improves a class after it’s oversubscribed.

Some of my students pull all-nighters before exams in hopes of improving their grade from an A− to an A. Even after they’ve secured the job they want and the baseline GPA to graduate, these students lose none of their zeal for obtaining the highest grade. They are like athletes, and this is their sport. Wearing themselves out studying, the students feel that the challenge of the exam and celebration of a victory if they win are all part of their process.

In the case of important challenges, success is often followed by a different kind of test: finding a new, equally meaningful quest. For example, what to do after you have enough money and your one and only skill is making more? Perhaps you turn to conspicuous consumption and start accumulating the status symbols that you mocked not long ago. This new pursuit might feel a bit hollow — after all, you chose it because you didn’t know what else to do. Then you’ll have to try something that feels more natural to you. Maybe philanthropy or resetting your sights on overtaking your enemies who are more successful still.

The Significance of Authenticity

In the morning, I generally desire a cup of coffee. The quality can differ in all sorts of ways, satisfying my desires to a greater or lesser degree. I could assign a price to the small differences if I thought about it. But there is no sense in which I desire this particular cup of coffee; I just care about how hot it is, what it tastes like, what it costs, and so on. All potential cups are substitutes for other cups.

In purposeful choice, our aims are commensurable — one can be traded for another at some rate of exchange. This is not the case when acting for-itself. I might take on this challenge or commit to that eccentric belief for reasons I can’t fully explain. I’m not optimizing when I spend 90 minutes on a Sunday emailing with a student to clear up a misunderstanding on a minor point in the lecture notes — I have many students and they have many misunderstandings. I’m not going to pretend that I break my back attending to all of them. I could spend that time more profitably and help more students by organizing a group review session. Why bother with this one?

While I can’t give a precise answer, it matters that this exchange arises naturally: A current student is genuinely grappling with this week’s material. It feels like he wants to understand, not to cozy up to me for a letter of recommendation or a better grade. Before the email arrived, I did not think to myself, “I hope a student emails me seeking assistance.” But when it does arrive, there’s a chance I’ll take the bait.

A for-itself challenge can be relatively trivial or objectively important. No matter the magnitude of the challenge, it must come about organically to be experienced as authentic. A government program that paid people to dig holes and then fill them up again would be unpopular not only with taxpayers but also with diggers, since they’d know it was contrived to keep them busy.

Context is critical to for-itself action because this realm deals with unique events. This is the polar opposite of purposeful choice, which values options according to their various attributes and how well those attributes satisfy a person’s fixed desires. In fact, the Nobel Prize-winning economist Gary Becker counted stable preferences among the three key axioms of neoclassical economics (along with optimizing behavior and markets in equilibrium). If our tastes bounced around randomly, rational choice theory would lack its predictive power. Of course, we may prefer a cold drink in the summer and a hot one in the winter. We may prefer variety: after an orange is chosen over an apple five times, oranges yield diminishing marginal utility and we switch to apples. Viewed at a sufficiently general level, these choices reflect coherent, stable preferences.

An ostensible paradox formulated by behavioral economist Richard Thaler (another Nobelist) has a simple resolution if we consider the importance of authenticity. Thaler points out the inconsistency of a man refusing to hire somebody to mow his lawn for
$8 but then turning down the opportunity to mow a neighbor’s similarly sized lawn for $20. Thaler attributes this paradox to cognitive bias.

Rational choice economists might point to different reasons: if the man hired himself out as a mower, he could suffer diminished status among his neighbors or incur transaction costs in negotiating the terms of the job, as well as additional income tax. If he hired a mower, coordinating and monitoring would impose additional costs. But this is unconvincing: People would mow their own grass but not their neighbor’s for pay, even if they could work in disguise to uphold their status, income taxes were eliminated and quality was easy to monitor.

This paradox arises in a strict purposeful choice framework, with agents evaluating work according to factors like pay and pleasantness of the task, while ignoring authenticity. For-itself theory provides a simple explanation: some people enjoy the physical challenge of yard work. Every day the grass grows and the homeowner likes keeping it under control. Hiring oneself out as a paid laborer does not arise naturally in this context, and no practical amount of money could convert it into a challenge worth overcoming.

If a behavioral economist were to explain to the person who mowed his own lawn but not his neighbor’s for pay that he has been duped by cognitive bias, it’s unlikely he would set up shop as a gardener. He doesn’t experience this behavior as problematic, and neither should we.

The Lottery Winner’s Curse illustrates the meaning we derive from engaging in authentic challenges and how lost we feel when they disappear. Imagine you are a 35-year-old vice president for a big bank working in New York City. Your compensation is $200,000 annually, but after taxes you have just enough to pay basic living expenses and rent the two-bedroom apartment where you live with your stay-at-home spouse and two children.

Altogether, it’s a fulfilling life. Your job involves the usual worries, which occupy much of your attention. You come home from work tired but satisfied that you’re putting bread on the table. You have a plan to claw your way to the top and look forward to the luxuries your family will someday be able to afford.

Your lucky day comes sooner than you expected. From out of the blue, a long-lost relative dies and leaves you a fortune, say $10 million tax-free. Now what? Your job at the bank, which now adds only a pittance to your material circumstances, will start to feel pointless, yet you don’t have quite enough money (or the inclination) to devote yourself to philanthropy full-time.

If you complained about your dilemma, you’d find no sympathy. You’re rich! And rational choice theory would offer no support: you have all the same options as before, plus the new ones that your wealth opens up. According to for-itself theory, however, maybe your good fortune wasn’t so good after all, since it sabotaged your engagement with authentic challenges.

You can’t go back to your old life, not really, and the difficulties of striving to get ahead have been hollowed out. So far, no new challenge has presented itself in place of the old one. Deep down, you may even regret the windfall. But sadly, there’s nothing to be done about it.

If a demon suddenly appeared, offering a winning lottery ticket, I imagine nearly everyone would take it. I would. But a windfall might make you miserable, as it did our banker, and, evidently, many real-life lottery winners. That result doesn’t fit with purposeful choice; it belongs to the realm of for-itself.

In purposeful behavior, we optimally satisfy our preferences within the constraints of our limited resources.

If those constraints are relaxed (typically through more money), we’ll do better. But this is not so in the for-itself realm. Powerful intervening figures like the demon with the winning lottery ticket will yank individuals from their lives with offers of free gifts, control over the future, or favorable trades. But the deals will backfire, exposing the importance of context and the limits of purposeful choice.

Authenticity may not play such a crucial role in the purposeful-choice model, but it’s intrinsic to our engagement with everyday life. Adventures, like the time the car got a flat tire and we had to walk five miles to a gas station at night, may offer meaning richer than the satisfaction we could derive from consumption. But we do not choose such experiences — they choose us. If we were to slash the tire intentionally, the resulting experience would feel like an exercise in cynicism. Still there’s no reason to characterize this adventure as for-itself: once the car breaks down, rational choice ought to kick in.

Neither Realm Can Eclipse the Other

Suppose you’re setting out to rent an apartment. You’ll consider the features of each available option, how much you value those features, and whether they justify the price. If your income goes up, other things being equal, you’ll rent a bigger and better place. Maybe you are seduced by a cunning broker or by marketing gimmicks that exploit your behavioral biases and so you choose one that’s too expensive. Whether you choose well or poorly, it’s purposeful: you seek to satisfy your desires for a convenient location, ample size, aesthetic charm, length of lease, quality of construction and a dozen other factors, while keeping in mind the alternative uses of the money you might save by renting someplace cheaper.

If we tried to analyze this decision through any lens other than rational choice (or purposeful choice, if biases are in play), we would miss out on all its many insights. No reasonable person chooses an apartment purely on impulse, as a one-time plunge into the chaotic unknown. If you behaved that way, you’d be in big trouble. Consciously or unconsciously, we weigh the consequences of various alternatives and pick the one that seems best, knowing what we know. Rational choice makes this comparison possible by converting disparate options into a common currency.

This mechanism goes beyond ranking and aggregating easy comparisons. Suppose I’m considering three apartments, one downtown (D), one in midtown (M) and one uptown (U), and assume that my wife has delegated the decision to me. Midtown would provide the easiest commute because I could walk to my office and take a relatively short subway ride uptown on the one or two days each week I go to Columbia. My commute from uptown would be second best since I could walk to Columbia and ride the subway to my midtown office. Living downtown, I would have to take a medium subway ride to reach the office and a long one to reach Columbia. Comparing the commutes one at a time, I can arrive at a ranking:

Commute: M > U > D

The largest of the three apartments is uptown, the second largest is downtown, and the smallest is in midtown.

Size: U > D > M

My wife, Ianthe, prefers downtown to midtown and midtown to uptown. As discussed later in the book, care altruism makes my utility a function of hers, so her preferences matter to me.

Ianthe: D > M > U

There are many other factors to evaluate in choosing an apartment. But to simplify the decision, consider just these three. I might envision the internal debate as three homunculi residing in my mind, each advocating for a preference. Homunculus 1 focuses on the commute, homunculus 2 on apartment size and homunculus 3 on pleasing my wife. Suppose the homunculi cast one vote each. In elections considering two apartments at a time, M would beat U, since homunculi 1 and 3 would vote for M, while U would beat D, since homunculi 1 and 2 would vote for U. Yet D would beat M, since D would get the votes of homunculi 2 and 3.

Election: M > U > D > M

The homunculi would arrive at a stalemate, leaving my wife and me without a place to live. I would go round and round until I fired the homunculi and switched to a different approach.

I could simplify the decision by choosing to act on a single reason, focusing on the feature that is most important. For me, that’s the commute, and the shortest commute is from midtown. But this simple-minded approach, which throws out troves of valuable information, is no formula for getting ahead. The smarter tack would be to quantify each variable so I can weigh factors such as “how much better is the commute from midtown than the one from downtown?” and “how much do I care about the commute versus pleasing my wife?”

Because these disparate factors become comparable with a common currency, purposeful choice can handle this decision. I might wish the decision were less complicated, but I’ve got to think it through.

Not all decisions are driven by such clarity of purpose. Consider the following thought experiment: after watching a documentary on TV, I decide to donate to a charity devoted to refugees. On my way to the post office to mail a $100 check, a gust of wind catches a $100 bill that happened to be tucked into my shirt pocket and blows it away. At just that moment, I receive an email from a friend who saw the same TV show and was stirred to send $100 to the same charity.

If I’m motivated solely by concern for the refugees’ welfare, I should now tear up the check I was about to mail. I wanted a world in which I had $100 less and the charity had $100 more. That world is now a reality by virtue of the lost $100 bill and my friend’s generosity. If I still send a $100 check, then either I truly preferred a $200 donation in the first place or the act of donation is about more than the money.

Sending the check is a one-time, unique gesture. I don’t donate up to the “optimal point” where the marginal benefit of the last dollar to the refugees equals its next best use in my life. Nor do I investigate every other charity to ensure that this one is the most worthy. This act of giving stands for itself.

Commingling of the Two Realms

The purposeful and for-itself are often bound together in the same activity. Work is perhaps the ultimate example of an activity that is at once purposeful and a high-stakes for-itself game. Many aspects of the labor market that are inadequately accounted for by purposeful choice can be explained neatly once we recognize that money is in part a byproduct of work, rather than its exclusive driver.

The thrill of winning and engagement with colleagues in a team sport are intangibles generated by work that many people cannot access in any other way. These intangibles help explain why some work long hours and retire late in life or even deliberately work until the day they die.

They account for a housecleaner who hires cleaners for his own house and a cobbler whose children have no shoes. Both are so caught up in their jobs that they neglect duties at home. This kind of work effort is inconsistent with the rational choice trade-off between labor and leisure, but consistent with the for-itself theory of overcoming obstacles.

Many members of the middle- and upper-class would end up with more money if they devoted one hour less a week to work and one hour more to optimizing investments, organizing tax records, and reducing interest and fees paid to banks or credit card issuers. By focusing on personal finances, a homeowner might, for example, realize that redeeming shares in mutual funds that invest in mortgage-backed securities and using the proceeds to repay her own mortgage would cut several layers of fees. But she doesn’t need to be told any of this: if all she cared about was maximizing steady-state consumption, she would have done so already.

The commingling of purposeful and for itself also explains the phenomenon that economist Herbert Simon called “satisficing.” Satisficing describes decision-making that falls short of optimization. A satisficing person evaluates courses of action only until she arrives at one that is “good enough.” In Simon’s theory, the satisficer doesn’t try for the optimum because the calculations are too hard.

A rational person with all the brainpower she needed, however, might want a reasonably good outcome in terms of satisfying preexisting desires while leaving room for choice. In this case, she would serve up a few acceptable options and pick one as if at random. The one she picked would come as a surprise, both to herself and to an observer. Although this might look like an irrational failure to optimize, it can be a deliberate attempt to preserve her ability to exercise her will.

Satisficing thus enables us to go through our day, choosing (somewhat) freely while still doing pretty well in terms of satisfying our preferences.

The two explanations for not-quite optimizing — that is, cognitive failures and for-itself acts of will — can be distinguished empirically: unlike satisficing, for-itself choosing applies to complicated decisions as well as simple ones that require little effort or attention.

Defaulting to Purposeful Choice

Purposeful choice is easy to understand, think about and explain. We are so steeped in this way of thinking that we take the question “why did you do this?” to mean “why is this optimal?” When the true nature of our behavior is obscure, we’ll default to an explanation that lets us maintain a rational self-image.

Self-image matters. In Thus Spoke Zarathustra, Nietzsche describes a bloodthirsty murderer who, ashamed, robs his victim to provide a motive retroactively. Having established one, he can see himself as a rational actor rather than an incomprehensible monster.

Though we may have little else in common with such a criminal, every one of us fabricates explanations for our behavior at least some of the time. We pride ourselves in our rationality, and when prompted to explain our own actions, may feel sheepish answering “just because.” As natural storytellers, we invent motives after the fact to explain actions that were without clear purpose, whether to some external audience or simply for our own benefit. Conditioned as we are to thinking in terms of causal relationships, we reflexively assume that our choices are motivated by a reason and buy into our own justifications.

On rare and spooky occasions, we catch ourselves splicing an event into our memory before it could have occurred. But while we only infrequently feel a sense of déjà vu, rationalizing a motive after we take action and experience the resulting pleasure or pain is woven into our experience.

This compulsion to identify rational motives for our actions can make it difficult to accept that some actions are simply play. Someone who earns more than she can consume in the short term is rational if she enjoys the work or wants to save for the future.

Behavioral scientists examined what happen in the absence of both conditions with an experiment in which the work was designed to be unpleasant and the subjects couldn’t save. During the first five minutes, subjects could choose to listen to obnoxious white noise or to music. They earned small chocolate bars in proportion to the amount of time they listened to the obnoxious noise. They then had five minutes to consume their earnings, after which they forfeited any chocolate they did not eat.

Although they had been warned of this rule in advance, subjects earned significantly more chocolate than they consumed. Additionally, subjects who earned higher wages —that is, more chocolates per unit of noise — worked about as long as subjects who received a lower wage.

The researchers decided that these results pointed to a pathological tendency to “over-earn,” comparable to overeating. A Harvard Business School professor told The New York Times that the study’s implications were “enormous.” But were they really? Instead of suffering from a newly diagnosed mental disorder, perhaps the subjects simply treated the experiment as a game.

Whether the wage was high or low, subjects were experiencing a novel challenge and wanted to see how well they could perform. Under this interpretation, over-earning is no more irrational than playing a video game. Why pound your fingers on a controller and stare at a screen for no pay?

Although we may feel uncomfortable admitting it, play is pervasive in life, even for adults. For instance, motorists sometimes drive around looking for cheaper gas when the cost of gas used in the search outweighs the likely savings. They might defend this behavior by claiming that they don’t want to be gouged or that they want to punish greedy gas station owners. But unless they spend resources punishing profiteers in general, this does not seem to be their real concern.

Couldn’t consumers simply see gas station owners as adversaries in a game? Why should we question the judgment of consumers sacrificing efficient consumption so they can win? While it may be tempting to suggest more efficient solutions to someone hunting for cheaper gas, fixing her own car or knitting his own sweater, doing so would be inconsiderate. If they wanted to think harder about the obvious inefficiencies, they would already have done so.

The same logic applies when, rather than fighting an adversary, we’re helping someone in need. Some people give money to panhandlers occasionally. No one can argue that this is the most efficient way to reduce poverty. This capricious gesture cannot be cast in any cogent way as a purposeful optimization of fixed preferences.

We may tell ourselves that charity will lead to rewards in this life (if we buy into the Western take on karma) or the afterlife (if we believe in a God who actively judges human affairs). Yet given that other types of charity would be more effective, these justifications amount to nothing more than efforts to defend our self-image as rational. Whatever we may tell ourselves, this spontaneous act of mercy is for-itself.