Dawson has not only observed this phenomenon but has quantified it in a series of experiments. In one she enlisted 227 undergraduate volunteers to log on to a Web page and check out a photograph of a hip-looking stainless steel study lamp. The test subjects agreed that the lamp was pretty cool and well worth the price of $99.95 plus $3.95 for shipping and handling. They were then asked to imagine that another well-known Web site wholesaler was offering the lamp at a discount. The subjects were split into six groups, and each group was offered a different level of discount: 8 percent, 25 percent, 45 percent, 65 percent, 85 percent, and 100 percent (that is, free). In the 8 percent case, participants were told the lamp was being offered for $91.95 plus shipping and handling. In the free case, participants were told they need pay nothing except shipping and handling. The test subjects were then told to use three different 100-point scales to rate the lamp’s aesthetic value, quality, and usefulness, as well as the likelihood that they would buy it.
Dawson theorized that participants offered a large discount would find this desired object even more desirable, and as it turned out, she was correct. The students judged the usefulness and aesthetic value of the lamp in precise proportion to its price: the cheaper the lamp, the better their opinion of it. Meanwhile, no level of discount—not even 100 percent—diminished the students’ assessment of the lamp’s quality. “Positive prior attitudes combined with a large discount will fuel a process of motivated perception resulting in a more favorable product assessment,” Dawson and her coauthor, Yale colleague On Amir, concluded in their assessment.
Since this study tested subjects who had an intrinsic attraction to the product, it says nothing about the allure of discounts for consumers who are neutral or negative about a product. So Dawson and Amir ran another experiment in which 537 men and women subjects were asked to give their impressions of a pricy eye cream. Not surprisingly, the male subjects were far less enthused about the eye cream than were the female subjects, and said so in a pretest. The volunteers were again divided into six groups, and discounts of the eye cream assigned at 7 percent, 20 percent, 40 percent, 60 percent, 80 percent, and, yes, 100 percent. As in the previous experiment, the subjects were asked to rate the products on a scale of 1 to 100 for aesthetic value, quality, and usefulness. Once again, the larger discounts prompted an overall greater desire to purchase the eye cream. But when the results were broken down by gender, it became clear that the women were more likely to purchase the cream at a greater discount, and the men were somewhat less likely to purchase the cream at a greater discount. As predicted, participants were motivated by the higher discounts to purchase the eye cream only when it held intrinsic appeal for them. When the product lacked intrinsic appeal, the discount seemed only to confirm what the male subjects already believed: that the product was not for them.
“When products have great appeal, retailers can get by with moderate or even very small reductions in price,” Dawson told me. “And when consumers have no idea as to the quality of a product, or feel neutral or negative about it, deep discounts can actually discourage them from making a buy.”
It is not very surprising that preconceived ideas skew our response to price. A Rodeo Drive or Madison Avenue address predisposes us to anticipate that a shop will be expensive and to be grateful for any price break it offers. Deep discounts at Big Box stores may prompt the opposite reaction: When something is cheap, making it cheaper still might push us to question its value. But behavioral scientists are learning that consumer reactions go beyond this predictable pattern; they are rooted in something deeper than simple preconceptions. It is hard to know what customers are willing to pay for any given item at any given time, and even harder to balance that knowledge with considerations of cost and profit margins. Price is burdened by memory and guilt. Our reaction to it is personal, particular, and almost impossible to predict.
Consider for a moment that you have a family heirloom, perhaps a quilt lovingly stitched by a beloved ancestor. You cherish the quilt but no longer have a need for it, and have decided, albeit reluctantly, to let it go. You’ve seen what looked like similar quilts selling for $150 at flea markets, and you’ve also seen what looked like similar quilts appraised at thousands or even tens of thousands of dollars on
Antiques Roadshow.
You’re no fool, but you’re no expert, either. Which quilts do you decide will appropriately “anchor” the price of your quilt? How you “frame” it—whether as a precious heirloom or as a funky castoff—has everything to do with what price you and your potential customers will consider fair. In this case, as in many cases, price is not about the object per se. It’s about context.
Ariely conducted an ingenious experiment that demonstrates the remarkable power of “anchoring” to prime the mind for price. He gathered a group of students and asked each one to recite his or her Social Security number. Then he asked each to guess the price of a bottle of 1998 Côtes du Rhône. (He used Côtes du Rhõne rather than beer because he assumed college students were unlikely to know its price.) Students whose Social Security numbers contained low numbers in the last two digits estimated the bottle at an average of $8.64 (not far from the actual price).Students whose Social Security numbers contained high numbers in the last two digits priced the same bottle at an average of $27.91. What did their Social Security numbers have to do with the price of a French wine? Not a thing. How much influence did the recitation of their Social Security numbers have on what they thought was the price of a bottle of French wine? Enormous.
Findings like these offer clues as to why pricing is so difficult. We all rely on external cues when we estimate the value of things, and it’s almost impossible to know in advance what those external clues might be. This is especially true of things with which we’ve had little or no experience. It is for this reason that works of art and antique cars are so often sold at auction. No one really knows in advance how much these items are worth to buyers. The price is quite literally what the highest bidder is willing to pay. The price of auctioned goods starts low and then rises to the precise point where supply meets demand. The amazing success of eBay is testament to the psychological appeal of auctions, which seem to satisfy buyer and seller alike.
But perhaps this satisfaction is built on illusion. For decades economists agreed that auctions offered the perfect compromise, that they were immune to the passions inflamed by factors such as rabid discounting. But recently Ulrike Malmendier, an economist at the University of California, Berkeley, put that assumption to the test. She tracked 166 eBay auctions of CashFlow 101, a personal-finance-themed board game retailing for $195. eBay sellers generally offered an opening price of $45 and set a one-click “buy-it-now” price of roughly $125. That is, if a buyer was willing to bid $125, he or she got the CashFlow 101 immediately, without further bidding. This was a terrific deal, allowing bidders to end the auction and pay less than the going price; or, if they preferred to take a gamble, they could bid an even lower price. Astonishingly, 43 percent of CashFlow 101 buyers blew right past the “buy-it-now” price and paid more. Some even exceeded the $195 retail price. Why didn’t they just click in at $125? Malmendier and her team wondered if there might be something special about CashFlow 101 buyers and decided to test other goods. They observed thousands of auctions for iPods, and, again, 45 to 50 percent of eBay users paid more than the “buy-it-now” price. Malmendier’s team then expanded their investigation, to include auctions of men’s cologne, perfume, and autobiographies of both Barack Obama and Bill O’Reilly (presumably to catch both Democratic and Republican eBayers.) Incredibly, they found that, again, between 40 and 50 percent of buyers paid more than the “buy-it-now” price. Buyers seem to derive significant pleasure from bidding—even more pleasure, it seemed, than they did from getting the desired item at the lower price. The Romans had a term for this weird phenomenon: They called it
calor licitantis,
bidder’s heat or auction fever. In Rome, bidders afflicted with
calor licitantis
were sometimes forgiven their lapse and had their money returned. Unfortunately, eBay makes no such provision.
LEGIONS OF PH.D.’S
have for decades studied the intricacies of why people are willing to pay what they do, yet pricing remains a most imperfect “science.” In the words of one economist, “For most businesses, pricing is a profit-leaking paradox.” Blame this on Wanamaker’s scrappy invention, the price tag. Virtually the opposite of auctions, price tags fix the price and discourage bargaining, thereby de-skilling the salesclerk’s job. Lower-skilled jobs are lower-paying jobs, so this saves retailers money. But fixed prices also carry a significant cost: By reducing or even eliminating the possibility of price negotiation, they create gaps between supply and demand, leaving stores holding too much of what customers aren’t willing to pay for and selling too cheaply merchandise for which customers would be willing to pay more.
At the Fordham pricing conference, I met Bidisha Burman, a young professor of marketing from Appalachian State University in North Carolina. Burman studies the relation of price to a response with which most of us are familiar: regret. Regret is a much maligned emotion, but it is not necessarily irrational. According to University of Michigan psychologist Janet Landsman, regret is “a defensible response to a world that inevitably presents decision makers with irreducible conflicts that at times require acting in opposition to one of their various judgments.” Having regret implies that we believe we have done something bad or foolish, or neglected to do something good or wise. We regret because no matter how logical and carefully made our decisions, we wish we had not made them. People who pride themselves on their decisiveness often claim to never feel regret, but they are likely lying. Were such a “decider” presented with making the choice of submitting a beloved pet to surgery for a broken leg, scheduling the operation would be a perfectly logical choice. But if the pet dies on the operating table, whatever satisfaction this decider might get from having made this logical choice would probably be overwhelmed by regret. Yes, the decision was rational, but it is also regrettable. Humans regret because we have the ability to compare real outcomes with possible alternative outcomes. We can extrapolate beyond what is and imagine what could have been.
Regret cannot change the past, but it can influence the future. It can teach us not to repeat mistakes or overlook opportunities. And like all emotions it influences us to act or not. Burman told me that regret comes in two flavors: regret of decision and regret of outcome. The first form, regret of decision, is rather surprising. Who would regret a decision if its outcome was neutral or favorable? If I cheat on a test, get an A, and no one is the wiser, do I regret it? Well, yes, I do. But why? Burman responded to this question with an example. Imagine waking up early on a Saturday morning to the realization that you had driven rip-roaring drunk the previous night. The outcome of this decision is not the problem. You found your way home and safely to bed. But thinking back on this reckless action, you may well be horror-stricken. You may regret the
decision
to drive drunk despite the fortunate outcome. Then again, you may not. Regret, Burman points out, is a cognitive response; you have to think to experience it. Thinking about driving while under the influence may flood you with remorse and fill you with determination not to repeat the folly. But not all of us are so introspective, and not all of us experience the same level of decision regret. In fact, some of us experience almost no decision regret at all. That is why some people feel triumphant when they wake up alive and well the morning after a drinking spree. They don’t regret their bad judgment. They believe their decision to drive drunk was rational (“I know my limits; I wasn’t that drunk”) and convince themselves that the fortunate outcome—no fatal car accident—is proof of their own good sense.
Those of us with what psychologists call “high cognition” tend to experience high levels of decision regret. These folks take a long time to make decisions and are more likely to respond negatively if they later perceive a decision to be a poor one. People classified as “low cognition” are more impulsive, less thoughtful, and less likely to suffer decision regret. They tend to worry less about making a bad decision and therefore seem more decisive and less likely to blame themselves if the decision turns out to be disappointing. Burman told me that high-cognition types are not necessarily more intelligent, just more introspective. Highly introspective people are not always the most successful, particularly when their introspection leads to indecision. “Deciders” such as Donald Trump and President George W. Bush are not known for their introspection, and there is no evidence that they suffer greatly from decision regret. They don’t seem to suffer from much outcome regret, either, although this is mere inference.
Burman is fascinated by the concept of regret and, in particular, by the impact of this noisome emotion on our everyday lives. She told me that the power of regret to shape consumer decisions is profound, particularly in the case of discounting. In an elegantly simple experiment, she offered three groups of research subjects a chance to buy a digital camera “on sale” for $249. One group was told the retail value of the camera was $799, the next group that it was $499, and the third group that it was $299. Burman asked each group how they felt about their deals. Interestingly, while few subjects in the first group actually believed that the camera was worth $799, they reported the highest overall satisfaction with their hypothetical “purchase.” For some reason just being told that the price was more than three times what they paid for it gave them great pleasure. Members of the other two groups believed what they were told about the real value of the cameras and were pleased with their purchase, but not to the same degree.