Misbehaving: The Making of Behavioral Economics (7 page)

In the published version of prospect theory, Amos and Danny included the following defense of their methods: “By default, the method of hypothetical choices emerges as the simplest procedure by which a large number of theoretical questions can be investigated. The use of the method relies on the assumption that people often know how they would behave in actual situations of choice, and on the further assumption that the subjects have no special reason to disguise their true preferences.” Essentially, they were saying that if their subjects were reasonably accurate in predicting the choices they would actually make in such cases, and their indicated choices were inconsistent with expected utility theory, then that should at least create a presumption of doubt about whether the theory is a good description of behavior.

This defense apparently satisfied the journal editor but remained a bugaboo among economists for years. Prospect theory gradually gained acceptance because it proved useful in explaining behavior in a variety of high-stakes settings where it was possible to observe actual choices, from individual investors to game show contestants. But I don’t think any economist would have come up with this theory, even granting them Kahneman and Tversky’s psychological insights. An unwillingness to rely on hypothetical questions would have kept them from learning the nuances of behavior that Kahneman and Tversky were able to discern.

I found the idea that you could just ask people questions and take their answers seriously to be quite liberating. Up to then, the items on the List were merely thought experiments. It seemed obvious to me that if readers were confronted with one of my hypothetical examples, they would check their intuition and then agree that the behavior existed. (This was, of course, naïve.) And, although the survey method was not considered authoritative, it was surely better than a survey of my own intuitions.

A few years later I got a nice lesson on how to do this from the masters themselves. They took my clock radio and television shopping example from the List and turned it into shopping for a jacket and a calculator, and then asked people what they would do. Here it is, with two different versions indicated by the numbers in parentheses or brackets:

Imagine that you are about to purchase a jacket for ($125)[$15] and a calculator for ($15)[$125]. The calculator salesman informs you that the calculator you wish to buy is on sale for ($10)[$120] at the other branch of the store, located a twenty-minute drive away. Would you make the trip to the other store?

Sure enough, real subjects said they would be more willing to take the drive to save $10 on the cheaper item, as I had conjectured, and now there was data to support it. I soon started using this method as well, though sparingly. But Danny and I would rely almost exclusively on the answers to hypothetical questions seven years later in a project about perceptions of fairness, discussed in
chapter 14
.

When I was not wandering the hills with Danny, I was hunkered down at NBER with nothing to do but think. Victor Fuchs played the role of guilt-inducing Jewish mother, periodically asking me about my progress. A paradox confronted me. I had what I thought was a big idea, but research proceeds through a series of small steps. And I did not know which small steps would advance the big idea. Big ideas are fine, but I needed to publish papers to stay employed. Looking back, I had what science writer Steven Johnson calls a “slow hunch.” A slow hunch is not one of those “aha” insights when everything becomes clear. Instead, it is more of a vague impression that there is something interesting going on, and an intuition that there could be something important lurking not far away. The problem with a slow hunch is you have no way to know whether it will lead to a dead end. I felt like I had arrived on the shores of a new world with no map, no idea where I should be looking, and no idea whether I would find anything of value.

Kahneman and Tversky ran experiments, so it was natural to think that I should be running experiments, too. I reached out to the two founders of the then nascent field called experimental economics, Charlie Plott at Caltech and Vernon Smith, then at the University of Arizona. Economists traditionally have used historical data to test hypotheses. Smith and Plott were practitioners of and proselytizers for the idea that one could test economic ideas in the laboratory. I first took a trip down to Tucson to visit Smith.

Smith’s research agenda was, at least at that time, different from the one I was imagining for myself. When he and Danny shared the Nobel Prize in economics many years later, I told a reporter that the difference between their respective research agendas that won them the prize was that Smith was trying to show how well economic theory worked and Kahneman was doing the opposite.
‡

At the time I visited him, Smith advocated using something he called the
induced value
methodology. Instead of trading actual goods or gambles, markets were created for tokens, in which each subject was given their own private value for a token. My token might be worth $8 while yours would be worth $4, meaning that these were the amounts we would receive from the experimenter if we ended up holding a token at the end of the study. Using this method, Smith was able to test economic principles such as supply and demand analysis. But I had some worries about this methodology. When you go to the store and decide whether to buy a jacket for $49, no one is telling you how much you are willing to pay for it. You have to decide that for yourself, and that value might depend on all sorts of issues such as what the retail price of the product is, how much you have already spent on clothing this month, and whether you happened to have just gotten your tax refund. Many years later I finally got around to testing my concern about this method by replacing tokens with coffee mugs, as we will see in
chapter 16
.

I then combined a family trip to Disneyland with a pilgrimage to Caltech to meet Charlie Plott, who was also pioneering this field (and could easily have shared the Nobel Prize with Smith). Perhaps because of the Caltech setting, Plott liked to use a wind tunnel analogy to describe what he was doing. Rather than showing that the basic principles of economics worked in the lab, he was more interested in testing what happened when the rules of the market were changed. Charlie, for whom the word garrulous seems to have been invented, was also warm and friendly.

As kind and impressive as Smith and Plott were, I was not ready to declare myself to be exclusively, or even primarily, an experimental economist. I wanted to study “behavior” and remain open-minded about the techniques I would use. I planned to run experiments when that method seemed to be the best way of observing behavior, or sometimes to just ask people questions, but I also wanted to study the behavior of people in their natural habitats . . . if I could just figure out how to do it.

A
t some point during my year in Stanford I decided I was going “all in” on this new venture. The University of Rochester was not an ideal venue given the intellectual proclivities of the senior faculty, who were deeply wedded to traditional economic methodology, so I looked elsewhere.
§

When you interview for a job in academia you present a paper in a faculty workshop, and that presentation, along with the papers you have written, determines whether you will get the job. My “Value of a Life” paper with Rosen was already pretty widely known, and I could have played it safe by presenting some additional work on that topic, but I wanted an environment that would tolerate a little heresy, so I presented a paper about the economics of self-control, cashews and all. Any place that would hire me after hearing that paper was likely to be at least moderately open to what came next. Fortunately, offers arrived from Cornell and Duke, and I settled on Cornell. My next move would be 90 miles down the road from Rochester.

________________

*
   Answer: Drive the same speed the whole way. The chance of getting a ticket is proportional to the time you are driving, holding everything else constant.

†
   There are, of course, exceptions to this generalization. In that era, George Stigler and Tom Schelling come to mind as great writers.

‡
   I was referring to Smith’s early work, cited by the Nobel committee. Later he delved into other more radical areas, including a series of experiments in which he could reliably produce an asset pricing bubble (Smith, Suchanek, and Gerry, 1998).

§
   Academic insiders might wonder how I landed a job in the Rochester business school after being a student in the economics department. Universities usually do not hire their own graduates. The answer is a long story, the short version of which is that I had been teaching at the business school while a graduate student, and when my first job fell through at the last minute, Bill Meckling, the school’s dean, offered me a one-year position as a stopgap measure, and I ended up sticking around for a few more years.

6

The Gauntlet

I
accepted the job at Cornell about halfway through my time at Stanford, and would start there in August 1978. I had work to do on two fronts. First, I had to produce research that showed what we could learn from the new approach I was suggesting. Second, and just as important, I had to be able to offer convincing replies to a series of one-line putdowns I would hear almost any time I presented my research. Economists had their way of doing things and would resist change, if for no other reason than that they had invested years building their own particular corner of this edifice.

This fact was brought home to me at one early conference where I gave a talk on my recent work. During the question and answer period that followed, a well-known economist asked me a question: “If I take what you are saying seriously, what am I supposed to do? My skill is knowing how to solve optimization problems.” His point was that if I were right, and optimization models were poor descriptions of actual behavior, his toolkit would be obsolete.

His reaction was unusually candid. The more common response, for those who engaged at all, was to explain what I was doing wrong, and what obvious factors I had ignored. I soon had another list: reasons why economists could safely ignore behaviors such as those on the List. Among friends I would call this series of questions the Gauntlet, since any time I gave a talk about my work it felt like running a medieval gauntlet. Here are a few of the most important ones, along with the preliminary responses I had worked up at the time. To some extent people are still arguing about these points; you will see them reappear throughout the book.

As if

One of the most prominent of the putdowns had only two words: “as if.” Briefly stated, the argument is that even if people are not capable of actually solving the complex problems that economists assume they can handle, they behave “as if” they can.

To understand the “as if” critique, it is helpful to look back a bit into the history of economics. The discipline underwent something of a revolution after World War II. Economists led by Kenneth Arrow, John Hicks, and Paul Samuelson accelerated an ongoing trend of making economic theory more mathematically formal. The two central concepts of economics remained the same—namely, that agents optimize and markets reach a stable equilibrium—but economists became more sophisticated in their ability to characterize the optimal solutions to problems as well as to determine the conditions under which a market will reach an equilibrium.

One example is the so-called theory of the firm, which comes down to saying that firms maximize profits (or share price). As modern theorists started to spell out precisely what this meant, some economists objected on the grounds that real managers were not able to solve such problems.

One simple example was called “marginal analysis.” Recall from
chapter 4
that a firm striving to maximize profits will set price and output at the point where marginal cost equals marginal revenue. The same analysis applies to hiring workers. Keep hiring workers until the cost of the last worker equals the increase in revenue that the worker produces. These results may seem innocuous enough, but in the late 1940s a debate raged in the
American Economic Review
about whether
real
managers actually behaved this way.

The debate was kicked off by Richard Lester, a plucky associate professor of economics at Princeton. He had the temerity to write to the owners of manufacturing companies and ask them to explain their processes for deciding how many workers to hire and how much output to produce. None of the executives reported doing anything that appeared to resemble “equating at the margin.” First, they did not seem to think about the effect of changes in the prices of their products or the possibility of changing what they paid to workers. Counter to the theory, they did not appear to think that changes in wages would affect either their hiring or output decisions much. Instead, they reported trying to sell as much of their product as they could, and increasing or decreasing the workforce to meet that level of demand. Lester ends his paper boldly: “This paper raises grave doubts as to the validity of conventional marginal theory and the assumptions on which it rests.”