The same basic principle would also apply if the government one day decided to impose a tax that doubled the price of gasoline. Under conventional economic theory, this should cut demand. But would it? Certainly, people would initially compare the new prices with their anchor, would be flabbergasted by the new prices, and so might pull back on their gasoline consumption and maybe even get a hybrid car. But over the long run, and once consumers readjusted to the new price and the new anchors (just as we adjust to the price of Nike sneakers, bottled water, and everything else), our gasoline consumption, at the new price, might in fact get close to the pretax level. Moreover, much as in the example of Starbucks, this process of readjustment could be accelerated if the price change were to also be accompanied by other changes, such as a new grade of gas, or a new type of fuel (such as corn-based ethanol fuel).
I am not suggesting that doubling the price of gasoline would have no effect on consumers’ demand. But I do believe that in the long term, it would have a much smaller influence on demand than would be assumed from just observing the short-term market reactions to price increases.
ANOTHER IMPLICATION OF arbitrary coherence has to do with the claimed benefits of the free market and free trade. The basic idea of the free market is that if I have something that you value more than I do—let’s say a sofa—trading this item will benefit both of us. This means that the mutual benefit of trading rests on the assumption that all the players in the market know the value of what they have and the value of the things they are considering getting from the trade.
But if our choices are often affected by random initial anchors, as we observed in our experiments, the choices and trades we make are not necessarily going to be an accurate reflection of the real pleasure or utility we derive from those products. In other words, in many cases we make decisions in the marketplace that may not reflect how much pleasure we can get from different items. Now, if we can’t accurately compute these pleasure values, but frequently follow arbitrary anchors instead, then it is not clear that the opportunity to trade is necessarily going to make us better off. For example, because of some unfortunate initial anchors we might mistakenly trade something that truly gives us a lot of pleasure (but regrettably had a low initial anchor) for something that gives us less pleasure (but owing to some random circumstances had a high initial anchor). If anchors and memories of these anchors—but not preferences—determine our behavior, why would trading be hailed as the key to maximizing personal happiness (utility)?
SO, WHERE DOES this leave us? If we can’t rely on the market forces of supply and demand to set optimal market prices, and we can’t count on free-market mechanisms to help us maximize our utility, then we may need to look elsewhere. This is especially the case with society’s essentials, such as health care, medicine, water, electricity, education, and other critical resources. If you accept the premise that market forces and free markets will not always regulate the market for the best, then you may find yourself among those who believe that the government (we hope a reasonable and thoughtful government) must play a larger role in regulating some market activities, even if this limits free enterprise. Yes, a free market based on supply, demand, and no friction would be the ideal if we were truly rational. Yet when we are not rational but irrational, policies should take this important factor into account.
Reflections on the Existence of Well-Defined Preferences
One of the lessons from Chapter 2 was that we generally believe we have precise and well-articulated preferences, but in reality, we only think that we know what we want. Here’s an example of an experience where I went into a situation with one set of ideas about what I wanted and emerged with a very different understanding.
When I turned 30, I decided it was time to trade in my motorcycle for a car, but I could not decide which car was right for me. The Web was just taking off, and to my delight, I found a site that provided advice on purchasing cars. The Web site, which is now defunct, asked a series of questions ranging from my preferred safety rating to my desired braking distance, my ideal turning radius, the number of passengers I’d like to be able to bring along, and, of course, my price range.
I spent fifteen minutes answering these questions. At the top of each page, I watched the progress bar inch closer to my result. It was exciting—I was really interested in seeing what kind of recommendation the site would come up with. The final screen displayed all the answers I had provided in the last fifteen minutes; all I had to do was click on “Submit” to receive my tailored recommendation. The second I did, I learned that my perfect car was (drum roll, please) a Ford Taurus.
What?
Now, I might not know much about cars (in truth, I know very little about them), but I knew that I did not want a Ford Taurus (and I don’t mean any disrespect to what I am sure is generally a fine automobile). The problem was that, having just surrendered my motorcycle, I couldn’t see myself driving such a sedate sedan. I was now facing a dilemma: I had tried a deliberative and thoughtful process for my car selection, and I didn’t like the answer I got. So, I did what I think anyone in my position would do. I hit the back button a few times, backtracked to earlier stages of the interview process, and changed many of my original answers to what I convinced myself were more accurate and appropriate responses. I lowered my interest in safety and the number of passengers I wanted to take with me, and changed many of my answers to fit what I deemed a more appropriate motorcycle replacement. From time to time, I checked to see how the different responses translated into different recommendations.
I kept this up until the car-advising Web site suggested a Mazda Miata. The moment the program was kind enough to recommend a small convertible, I felt grateful for the fantastic software and decided to follow its advice. A few weeks later, I became the proud owner of a Miata, which served me loyally for many years.
WHAT HAPPENED HERE? On one hand, I knew that buying a car was no trivial matter, and I wanted to approach such a large decision by carefully weighing the cost and benefits in a cold, calculated, and sensible way. At the same time, I knew I was making an important and symbolic move into adulthood, and I understood that kids and the inevitable minivan (which I drive these days) were awaiting me. Nevertheless, my brain and my heart were engaged in a practical tug-of-war. Deep down, what I really wanted was a car that felt closer to a motorcycle—something that was fun to drive.
Taking the systematic and calculated approach to solving this problem did not yield the “correct” answer, so I went back and fudged around with my responses, letting the computerized method rationalize my choice for me. This way, I ended up with a decision that made me happy, and at the same time, it was a decision that I could easily explain to myself. With a neat and programmed computerized process, it was now obvious why the small convertible was, in fact, the right choice for me.
This elaborate computerized justification process might seem artificial and extreme, but I suspect that the same basic elements end up playing out in many of our important decisions. This experience taught me that sometimes we want our decisions to have a rational veneer when, in fact, they stem from a gut feeling—what we crave deep down. I suspect that in our attempts to make sure that we end up with decisions that seem well-reasoned and thoughtful, we commonly undergo a lot of unnecessary mental gymnastics and justifications, particularly when the choices are large and significant. Sometimes these rationalizations are complex and time-consuming, and sometimes we have the benefit of a software program to help us with more efficient rationalization. Perhaps this was the real function of the Web site I used—it was not necessarily designed to help me make a better decision, but to help me justify my choice and feel confident about it.
In the end, following our gut feelings and rationalizing them after the fact is not always bad. It can sometimes lead us to pick a satisfactory outcome or, at the very least, prevent us from ending up with a car we really don’t want.
CHAPTER 3
The Cost of Zero Cost
Why We Often Pay Too Much When
We Pay Nothing
Have you ever grabbed for a coupon offering a FREE! package of coffee beans—even though you don’t drink coffee and don’t even have a machine with which to brew it? What about all those FREE! extra helpings