Misbehaving: The Making of Behavioral Economics

by Richard H. Thaler

  That turned out to be a poor forecast. Although there are a handful of psychologists who have formed successful collaborations with economists over the years, Drazen Prelec and Eldar Shafir being notable examples, behavioral economics has turned out to be primarily a field in which economists read the work of psychologists and then go about their business of doing research independently.† One of our early participants, Stanley Schachter, is a case in point. He tried his hand at doing some research on the psychology of the stock market, but grew frustrated with the reactions he got from the referees at mainstream finance and economics journals and eventually abandoned the research program.

  There are several possible reasons why psychologists might have failed to engage. First, since few have any attachment to the rational choice model, studying departures from it is not considered inherently interesting. A typical reaction would be: “Of course people pay attention to sunk costs! Who would have thought otherwise?” Second, the psychology that behavioral economists have ended up using is not considered cutting-edge to psychologists. If psychologists started using supply and demand curves in their research papers, economists would not find the idea very exciting. Finally, for some reason the study of “applied” problems in psychology has traditionally been considered a low-status activity. Studying the reasons why people fall into debt or drop out of school has just not been the type of research that leads academic psychologists to fame and glory, with the notable exception of Robert Cialdini.

  Furthermore, we behavioral economists have not been particularly successful in generating new psychology of our own, which might breed the kind of cross-fertilization that we originally expected. Most of the advances in the field have been to figure out how best to modify the tools of economics to accommodate Humans as well as Econs, rather than discovering new insights about behavior. Of the emerging group of economists that have become the leaders of the field, only George Loewenstein has really created much new psychology. Although trained as an economist, George is really a talented psychologist as well, a fact that might be partially attributed to good genes. His middle initial F stands for Freud; Sigmund was his great-grandfather.

  Although this effort to get economists and psychologists working together did not succeed, Eric Wanner remained committed to helping foster the field, even if it consisted almost entirely of economists. The Russell Sage Foundation’s small size meant that it could not be the primary source of research funding if the field were to grow beyond a few hard-core members, so Eric convinced the board to continue to support the field in a limited and highly unusual way. And unlike the initial effort, it has been a huge success.

  Here is the plan Eric devised. In 1992, the foundation formed a group of researchers called the Behavioral Economics Roundtable, gave them a modest budget, and tasked them with the goal of fostering growth in the field. The initial members of the Roundtable were George Akerlof, Alan Blinder, Colin Camerer, Jon Elster, Danny Kahneman, George Loewenstein, Tom Schelling, Bob Shiller, Amos Tversky, and me, and within reason, we could spend the money we were given any way we wanted.

  The Roundtable members decided that the most useful way to spend our limited budget (which began at $100,000 per year) was to foster and encourage the entry of young scholars into the field. To do this, we organized two-week intensive training programs for graduate students to be held during the summer. No university was then teaching a graduate course in behavioral economics, so this program would be a way for students from all over the world to learn about the field. These two-week programs were officially called the Russell Sage Foundation Summer Institutes in Behavioral Economics, but from the beginning everyone referred to them as the Russell Sage summer camps.

  The first summer camp was held in Berkeley in the summer of 1994. Colin, Danny, and I were the organizers, with several other Roundtable members joining for a few days as faculty members. We also had some guest stars, such as Ken Arrow, Lee Ross (a social psychologist), and Charlie Plott. In the spirit of encouraging young scholars to join the field, we also invited two economists who had received their degrees quite recently to participate: Ernst Fehr and Matthew Rabin. Both had independently decided to take up careers in behavioral economics.

  Ernst Fehr is the most aptly named economist I know. If you had to think of a single adjective to describe him it would be “earnest,” and the topic that has interested him most is fairness. An Austrian by birth, Ernst has become a central figure in the behavioral economics movement in Europe, with a base at the University of Zürich in Switzerland. Like Colin, he has also become a prominent practitioner of neuro-economics.

  The first paper by Fehr that captured our attention was experimental. He and his coauthors showed that in a laboratory setting, “firms” that elected to pay more than the minimum wage were rewarded with higher effort levels by their “workers.” This result supported the idea, initially proposed by George Akerlof, that employment contracts could be viewed partially as a gift exchange. The theory is that if the employer treats the worker well, in terms of pay and working conditions, that gift will be reciprocated with higher effort levels and lower turnover, thus making the payment of above-market wages economically profitable.

  In contrast, Matthew Rabin’s first behavioral paper was theoretical, and was at that time the most important theory paper in behavioral economics since “Prospect Theory.” His paper was the first serious attempt to develop a theory that could explain the apparently contradictory behavior observed in situations like the Ultimatum and Dictator Games. The contradiction is that people appear altruistic in the Dictator Game, giving away money to an anonymous stranger, but also seem to be mean to others who treat them unfairly in the Ultimatum Game. So, does increasing the happiness of someone else make us happier too, or does it make us less happy, perhaps because of envy? The answer, Rabin suggested, hinges on reciprocity. We are nice to people who treat us nicely and mean to people who treat us badly. The finding discussed earlier, that people act as “conditional cooperators,” is consistent with Rabin’s model.

  Matthew is also a character. His normal attire is a tie-dyed T-shirt, of which he seems to have an infinite supply. He is also very funny. I was one of the referees who were asked to review his fairness paper when he submitted it for publication in the American Economic Review. I wrote an enthusiastic review supporting publication, but added, without providing any details, that I was disturbed that he had left out an important footnote that had appeared in an earlier draft. The footnote referred to the game economists refer to as “chicken,” in which the first person to concede to the other loses. Here was his footnote, which was restored in the published version: “While I will stick to the conventional name for this game, I note that it is extremely speciesist—there is little evidence that chickens are less brave than humans and other animals.”

  So we had an all-star faculty lined up for our summer camp, plus the up-and-coming young guys, Fehr and Rabin. But having never done this before, we did not know whether anyone would apply. We sent an announcement to the chairs of the leading economics departments around the world and hoped someone would want to come. Fortunately, over 100 students applied, and the group of thirty that we picked was packed with the future stars of the field.

  These summer camps have continued in alternate years ever since. After Danny and I grew too busy/tired/old/lazy to organize and participate in the entire two-week program, it was taken over by younger generations. For a while Colin and George organized it, and David Laibson and Matthew Rabin have run the last several camps.

  One indicator of the success of these summer camps is that David was a student at the first one, so the group is becoming self-generating. Many of the other faculty members who participate now are also camp graduates. I should be clear that we make no claims about turning these young scholars into stars. For example, David Laibson had already graduated from MIT and taken a job at Harvard before he arrived at our summer camp. Others were also clearly stars in the making. Instead,
the primary accomplishment of the summer camps was to increase the likelihood that some of the best young graduate students in the world would seriously consider the idea of becoming behavioral economists, and then to provide them with a network of like-minded economists they could talk to.

  The talent level of the campers that first year is evidenced by the number who have gone on to fame. One was Sendhil Mullainathan, who had just completed his first year of graduate work at Harvard. I had gotten to know Sendhil when he was an undergraduate at Cornell, completing degrees in economics, mathematics, and computer science in three years. It was not hard to see that he had the talent to do almost anything, and I tried my best to interest him in psychology and economics. Luckily for the field, my pitch worked, and it was his budding interest in behavioral economics that tipped him from computer science to economics for his graduate training. Among his other accomplishments, Sendhil founded the first behavioral economics nonprofit think tank, called ideas42. He, Matthew, and Colin have received a so-called “genius” award from the MacArthur Foundation.

  Other notable first-year campers were Terry Odean, who essentially invented the field of individual investor behavior, Chip Heath, who with his brother Dan has published three successful management books, and two of my future coauthors, who will soon make their appearances in this book: Linda Babcock and Christine Jolls.

  In the summer of 2014 we held our tenth summer camp, and I have yet to miss one. There are now about 300 graduates, many holding positions at top universities around the world. It is largely the research produced by those summer camp graduates that has turned behavioral economics from a quirky cult activity to a vibrant part of mainstream economics. They all can thank Eric Wanner for helping them get started. He is the behavioral economics’ founding funder.

  ________________

  * There are some exceptions to this generalization, such as neuroscience, where scientists from many different fields have productively worked together, but in that case they coalesced around specific tools like brain scans. I don’t want to say that all interdisciplinary meetings are a waste of time. I am just saying that in my experience, they have been disappointing.

  † To be clear, the field of judgment and decision-making that was kick-started by Kahneman and Tversky in the 1970s continues to thrive. Their annual meeting, sponsored by the Society for Judgment and Decision Making, attracts over 500 scholars whose work often intersects with behavioral economics. There are also a number of notable behavioral scholars in marketing, including my old friend Eric Johnson, several of my former students, and many others who do research on topics such as mental accounting and self-control. My point is that a typical meeting of behavioral economists does not include any psychologists, and I am one of the few economists who regularly attends the SJDM meeting.

  20

  Narrow Framing on the Upper East Side

  The contributions of the Russell Sage Foundation to behavioral economics were not limited to the creation of the Roundtable. The foundation also has a wonderful program for visiting scholars, who spend a year in New York living in a subsidized apartment near the foundation’s office on the Upper East Side. A scholar’s only responsibility is to show up for a tasty—and dare I say it, free—lunch. The rest of your time is available to spend thinking and writing. For the academic year 1991–92 Colin Camerer, Danny, and I had applied as a team, and Danny’s wife, Anne Treisman, also a psychologist, also joined as a visiting scholar. As a bonus, Amos made periodic visits, so we were poised to have a great year. Danny and I hoped to somehow recreate the magic we had experienced earlier at Stanford and Vancouver. But the stars were not aligned.

  It did not help that I was going through a divorce, and an enormous fire burned Anne and Danny’s Berkeley home to the ground. But those were only two of the distractions we had to overcome. Over the six years since we had been in Vancouver, we had both become too busy to ignore everything else and work intensively on a joint project. We had PhD students who needed attention, Danny and Anne had a lab full of graduate students back in Berkeley, and we all had colleagues at our home universities who wanted us to weigh in on various departmental dramas. Our days of being able to work on one thing, seven days a week, for months at a time, had ended.

  But there was an idea in the air that we were both thinking about independently, and this idea also played a role in project I worked on with Colin. The idea is called “narrow framing,” and it is related to a more general mental accounting question: when are economic events or transactions combined, and when are they treated separately? If you go on a vacation, is each component of the cost of the trip (travel, hotel, meals, outings, gifts) considered a separate transaction, or are they pooled into the vacation category and evaluated together, as they would be in an all-inclusive cruise trip? The specific question that Danny and I were each pondering is: when do people get themselves into trouble by treating events one at a time, rather than as a portfolio?

  Danny’s work on this problem arose in a project with Dan Lovallo, a graduate student at Berkeley who joined us that year as our research assistant. Their idea was that managerial decision-making was driven by two countervailing, but not necessarily offsetting, biases: bold forecasts and timid choices. The bold forecasts come from Danny’s distinction between the “inside view” and the “outside view.”

  To convey the distinction, Danny tells the story of a book project. The full story is described in detail in Thinking, Fast and Slow, but for those who have shamefully failed to memorize that book, here is the short version. A team of scholars with different backgrounds was tasked with the job of devising a curriculum on decision-making for middle school students. After working on the project for several months, Danny started to wonder how long it might take to complete. He took a poll of the various team members, having each write down their guess separately to get a set of independent guesses. The estimates for time to completion ranged from eighteen to thirty months. Then Danny realized that one member of the team was an expert in curriculum development, and had observed many such teams in action over the years. So Danny asked this expert to evaluate their team compared to the others he had seen, and based on his experience to say how much longer the project would take. The expert, whose own guess had been among those in the range between eighteen and thirty months, became a bit sheepish. He reluctantly told the group that in his experience, no group had finished a similar task in less than seven years, and worse, 40% of the teams never finished!

  The difference between the expert’s two estimates illustrates Danny’s distinction between the inside and outside views. When the expert was thinking about the problem as a member of project team, he was locked in the inside view—caught up in the optimism that comes with group endeavors—and did not bother thinking about what psychologists call “base rates,” that is, the average time for similar projects. When he put on his expert hat, thereby taking the outside view, he naturally thought of all the other projects he’d known and made a more accurate guess. If the outside view is fleshed out carefully and informed with appropriate baseline data, it will be far more reliable than the inside view.

  The problem is that the inside view is so natural and accessible that it can influence the judgments even of people who understand the concept—indeed, even of the person who coined the term. After learning of Amos’s illness and short life expectancy, Amos and Danny decided to edit a book that contained a collection of papers on decision-making, but Amos passed away before the book was completed. Danny had the daunting task of writing an introduction that they had intended to write together. Amos died in June 1996, and I remember talking to Danny that fall about the book and when he thought it would be done. He said it shouldn’t take more than six months. I started laughing. Danny got the joke and said sheepishly, “Oh, you are thinking of that book [meaning the one featured in his story about the inside view]. This book is completely different. It is just a collection of papers, most of them already published. I just have to
get a few stragglers to finish their new papers and complete the introduction.” The book came out, shortly after the last paper arrived and the introduction was finished, in 2000, almost four years later.

  The “timid choices” part of the Kahneman and Lovallo story is based on loss aversion. Each manager is loss averse regarding any outcomes that will be attributed to him. In an organizational setting, the natural feeling of loss aversion can be exacerbated by the system of rewards and punishment. In many companies, creating a large gain will lead to modest rewards, while creating an equal-sized loss will get you fired. Under those terms, even a manager who starts out risk neutral, willing to take any bet that will make money on average, will become highly risk averse. Rather than solving a problem, the organizational structure is making things worse.

  Here’s an example to show how this works. Sometime shortly after our year in New York, I was teaching a class on decision-making to a group of executives from a company in the print media industry. The company owned a bunch of publications, primarily magazines, and each executive in the audience was the head of one of the publications, which were run pretty much independently. The CEO of the firm was also in attendance, sitting in the back of the room, watching and listening. I put to the executives this scenario: Suppose you were offered an investment opportunity for your division that will yield one of two payoffs. After the investment is made, there is a 50% chance it will make a profit of $2 million, and a 50% chance it will lose $1 million. (Notice that the expected payoff of this investment is $500,000, since half the time they gain $2 million—an expected gain of $1 million—and half the time they lose a million—an expected loss of half a million. The company was large enough that a million-dollar loss, or even several of them, would not threaten its solvency.) I then asked by a show of hands who would take on this project. Of the twenty-three executives, only three said they would do it.