Issue 35, Fall 2008

William A. McEachern, Editor

PREDICTABLY IRRATIONAL?

Three years ago The Teaching Economist discussed the publishing hit Freakonomics by Chicago economist Steven D. Levitt and journalist Stephen J. Dubner. The book, which shows the power of some simple economic propositions, has now been on the New York Times best-seller list for more than three years. It has yet to appear in paperback (that's like a movie playing in theaters for more than a year before migrating to cable or DVD). This wild success has launched other economics books aimed at a general audience. New entries that distill the power of economic reasoning include Discover Your Inner Economist: Use Incentives to Fall in Love, Survive Your Next Meeting, and Motivate Your Dentist (Dutton, 2007) by Tyler Cowens of George Mason University and The Logic of Life: The Rational Economics of an Irrational World (Random House, 2008) by Tim Harford, a Financial Times columnist.

But Freakonomics has also given rise to another strain of books that, rather than sharpen the economic toolkit, challenge the assumptions of economic reasoning. One getting a lot of attention is Predictably Irrational (Harper, 2008) by Dan Ariely of MIT. The book, which debuted number five on the NYT best-seller list, tries to show through dozens of experiments that people are irrational, but in predictable ways. Ariely argues that because economic theories assume rationality, they are unrealistic: "Wouldn't economics make a lot more sense if it were based on how people actually behave, instead of how they should behave?" (p. 239).

As in Freakonomics, the subject matter in Predictably Irrational is all over the map. Much of it is not new or surprising (e.g., people procrastinate and cheat), but Ariely's experiments are always interesting. I'll offer a quick chapter-by-chapter summary, comment on what he views as his major challenge to economics, then, in the next two sections, I'll discuss findings of classroom relevance.

Here are one-sentence summaries of each of the 13 chapters: (1) As consumers, we don't know what we want or are willing to pay until we see things in context (e.g., the highest priced item on the menu increases a restaurant's revenue even if nobody buys it—that high price offers context for the prices below it, making them appear more reasonable). (2) What consumers are willing to pay can be manipulated by suppliers (more on this below). (3) People get irrationally excited about a zero price (e.g., they are willing to wait in long lines for something "free"). (4) Social exchange and market exchange each has its own dynamic, but problems arise if the two markets intersect (e.g., never leave the price tag on a gift). (5) Young men are poor predictors of how they will behave when sexually aroused. (6) People procrastinate (see a classroom application below). (7) People overvalue what they already own (but, I might ask, what about buyer's remorse?). (8) People pay too much to keep their options open. (9) Expectations play an important role in shaping how much we will value something. (10) The placebo effect is powerful and works even through prices (e.g., aspirin believed to be pricier works better). (11) If given the opportunity, most people cheat a little, but cheating does not increase much even if it becomes easier, and cheating is sharply reduced when a moral dimension is introduced (again, see a classroom application below). (12) People are more honest when cash is involved. And (13) restaurant patrons show their independence by ordering something different from what others at the table ordered even if that item is not what they prefer most.

Many of his experiments reflect Ariely's psychology and marketing interests (he has Ph.D.s in each field but not in economics). The finding most challenging to economists arises in Chapter 2, where Ariely questions the validity of supply and demand curves: "[W]hat consumers are willing to pay can easily be manipulated, and this means that consumers don't in fact have a good handle on their own preferences and the prices they are willing to pay for different goods and experiences" (p. 45). Consumers, he argues, are highly suggestible when it comes to price, especially where they have little experience with the product. For example, he says that Starbucks could charge $4 for a cup of coffee because the company positioned the product as a completely new experience. Maybe so, but sooner or later doesn't market competition put downward pressure on the price? For example, Starbucks is now closing stores, while lower priced rival Dunkin' Donuts is expanding. And VCR's originally sold for about $30,000; you can find them now for as little as $30. Besides, economists never said they had much to contribute when it comes to the origin of tastes and preferences. Ariely's argument that tastes come from experiencing the goods in context doesn't pose the threat to economics that he claims.

Procrastination

Ariely's chapter on procrastination discusses an experiment that should be of interest to instructors. Certainly it's no surprise that we all procrastinate to some extent; we also see the results of student procrastination every term. To test some issues of procrastination, Ariely assigned three papers that counted for most of the grade in the course. He wanted to uncover the effects of different deadlines on student performance, so he varied the deadline structure across each of his three consumer behavior classes.

In the first class, he had each student set his or her own deadlines. By the end of the first week, each student had to specify a deadline for each of the papers. Any date before the end of the term was acceptable. Once deadlines were set, late papers would be penalized. Although the most flexible option was to pick the final day for all papers, most students recognized their tendency to procrastinate and chose to bind themselves with deadlines spaced throughout the term. Ariely argues that it's irrational to procrastinate, but students chose deadlines to offset their irrationality. What's irrational about that?

In a second class of the same course, there were no deadlines at all. Students had to submit papers by the end of the last class period. In a third class, Ariely dictated deadlines spaced throughout the term. Here's the question: which class did the best and which the worst? The group with deadlines set by Ariely did the best. Penalizing procrastination was the best cure for it. In the class where students set their own deadlines, those who spaced them out did as well as those with the instructor-imposed deadlines. On the other end, Ariely found that papers submitted by those with no intermediate deadlines or poorly spaced deadlines were rushed and poorly written (pp. 111-117).

Cheating

Ariely and colleagues conducted an experiment that was set up to determine whether and how much a group of participants would cheat on a math test (a test independent of any course). Test takers were paid for correct answers, so they had some incentive to cheat. A control group was tested under conditions that prevented cheating. Two other groups were given the same test under conditions that made cheating by a particular participant virtually undetectable (without getting into the mechanics of the experiment, the cheating involved how participants reported the number of correct answers and collected their winnings). Before they took the test, participants in one group were asked to write down the titles of up to ten books they read in high school. Participants in the other group were asked to write down as many of the Ten Commandments as they could recall. The researchers were curious whether introducing a moral dimension like the Ten Commandments would reduce cheating.

The group asked to recall book titles appeared to cheat, as they scored on average 33% higher than the control group. The Ten-Commandment group apparently did not cheat at all, as their test average was about the same as for the control group. The Ten Commandment exercise appeared to evoke honesty. What's more, students who could remember only one or two Commandments were as honest as those who remembered all ten. Ariely concludes that "it was not the Commandments themselves that encourage honesty, but the mere contemplation of a moral benchmark of some kind" (pp. 208-209). (I might add, what's irrational about that?) In another experiment, test takers were asked to sign a statement to the effect that "I understand that this study falls under the MIT honor system." This group didn't cheat either (even though MIT has no formal honor system).

Top                                                                                                                                                                                          Next