Priceless: The Myth of Fair Value (and How to Take Advantage of It) (20 page)

One of the items auctioned was a bottle of 1998 Côtes du Rhône. My social security number ends in 23, so the first question I would have had to answer is “Would you pay more or less than $23 for this bottle of wine?” The second question is “How much would you be willing to pay?”

As expected, the results showed impressive anchoring. Bidders with “low” social security numbers (defined as those ending in the digits 00 through 19) were willing to pay an average of $8.64 for the bottle of Côtes du Rhône. Those with “high” numbers (ending 80 through 99) were willing to pay an average of $27.91. It wasn’t just the wine mystique. There were similar differences in bids for the chocolates, a cordless keyboard and trackball, and a design book—all because of the social
security numbers. For the most part, the students who happened to have higher-ending social security numbers walked away with the merchandise. The lower-number people missed out. I will leave it to you to decide who were the real winners—and who were the suckers.

Ariely was serving in the Israeli military when a magnesium flare, used to illuminate battlefields at night, exploded near him. He received third-degree burns over 70 percent of his body. For the next three years, Ariely was a veritable “English patient,” covered in bandages and largely immobile. His treatment required regular replacement of the bandages. There was no way of making this anything but torture. Ariely’s nurses were compassionate souls with experience in this unpleasant task. They believed in ripping the bandages off quickly. It produced a burst of agony, fading to mere pain. Ariely had a lot of time to ponder the psychophysics of pain. He concluded it was better to pull gradually, for the pain to be less intense yet of longer duration. One can become adapted to slow, steady pain. There was no way to become adapted to the nurses’ high-contrast technique. He had little luck in convincing his nurses of this. They had a different perspective. It hurt them, too, to see their patients suffer, and they preferred that this distasteful part of their work be accomplished quickly.

Once he was able to leave the hospital, Ariely studied psychology at Tel Aviv University. (He met Amos Tversky, who gave a lecture there.) Ariely acquired a grounding in psychophysics, reading the work of S. S. Stevens and others. He did experiments on pain, sometimes with himself as a subject, using heat, cold water, pressure, and loud noises. As his interests turned to economic decisions, it was natural to see money as a stimulus, and price as a magnitude scale.

Ariely pioneered a widely influential thesis: that remembered prices obscure how inexact the human price sense is. Were a shopper required to guess the price of an elliptical trainer, he would try to remember what he’d paid for exercise equipment in the past, or what he’d seen elliptical trainers advertised for. He’d make adjustments for quality and features and come up with a figure that might not be too far off. Yet in some sense, he’d be like Oscar Wilde’s definition of cynic, knowing the price of everything and the value of nothing.

The MIT auction was designed to strip away some of the effects of memorized prices. They chose items that MIT students were unlikely to have purchased and items known to have a large range of prices. (Wine and fancy chocolates are popular gifts in part because it’s hard for the recipient to guess how much the giver paid.) The question was not
Do you remember the price?
but
How much is this worth to you?

The auction’s results looked much like those of S. S. Stevens’s experiments with magnitude scales. There was consensus on ratios, though little on absolute values.

The chart above shows the average bids, broken down into five ranges of social security number endings. Each line indicates the bids for a different item. For the purposes of this experiment, social security numbers can be considered random. You’d normally expect that five random groups would have the same average valuations. Instead, all the lines trend upward. The people with the low social security numbers (left) bid much less than the ones with the high numbers (right). This indicates anchoring.

Within any social security number range, the relative valuations of
different items were approximately the same as in every other group. All groups concurred that the keyboard was the most valuable item, and the chocolates were at or near the bottom. The rare wine was consistently valued higher than the average wine, and by approximately the same ratio (1.5 times more) in all groups.

Ariely, Loewenstein, and Prelec theorized that their subjects were retroactively imposing self-consistency. They wrote,

Suppose that a subject with a social security number ending with 25 has an a priori WTP [willing to pay] range of $5 to $30 for wine described as “average,” and $10 to $50 for the “rare” wine. Both wines, therefore, might or might not be purchased for the $25 price. Suppose that the subject indicates, for whatever reason, that she would be willing to purchase the average bottle for $25. If we were to ask her a moment later whether she would be willing to purchase the “rare” bottle for $25, the answer would obviously be “yes” because from her perspective this particular “choice problem” has been solved and its solution is known: if an average wine is worth $25, then a rare wine must be worth more than $25! Moreover, when the subject is subsequently asked to provide WTP prices for the wines, then that problem, too, is now substantially constrained: the prices will have to be ordered so that both prices are above $25 and the rare wine is valued more.

Ariely’s group published these results in a 2003 paper in
The Quarterly Journal of Economics
, “ ‘Coherent Arbitrariness’: Stable Demand Curves Without Stable Preferences.” The same paper includes yet more impressive proof of the memory theory.

“We wanted something where people don’t have a strong reference price,” Ariely explained. They needed a brand-new product to price, and the product was pain. One hundred thirty-two MIT students listened to an annoyingly loud high-pitched tone (a 3,000-Hz triangular wave similar to that used for emergency warnings) through headphones. Onscreen instructions read,

In a few moments we are going to play you a new unpleasant tone over your headset. We are interested in how annoying you find it
to be. Immediately after you hear the tone, we are going to ask you whether you would be willing to repeat the same experience in exchange for a payment of 10 cents [50 cents for another group].

Participants were asked to name their prices for listening to 10 seconds, 30 seconds, and 60 seconds of the annoying sound. As expected, those who received the low anchor (10 cents) quoted consistently lower prices than those who got the high value (50 cents). Everyone’s prices were scaled appropriately to the length of exposure. Furthermore, repeated trials with the same subjects did not erase the effect of the initial anchor. Most stuck with their original pricing, unaware that it was prompted by a meaningless anchor.

Some volunteers were asked to name their price for the sound and also to rank it on a list of minor annoyances. The peeves included “discovering you purchased a spoiled carton of milk,” “forgetting to return a video and having to pay a fine,” “having your ice cream fall on the floor,” and seven other items. Overall, the annoying sound came in #2 on the list, behind “missing your bus by a few seconds.” The telling thing is this. The 10- and 50-cent price anchors had no effect on the
ranking
of the annoying noise. Everyone approximately agreed on how bad the noise was, relative to life’s other little annoyances.

Another group of volunteers consented to put a finger in a vise. The experimenter tightened the vise until the subject said he was beginning to experience pain (the “pain threshold”). Then the vise was tightened a millimeter more. The subject was instructed to remember that level of pain. Their fingers released from the vise, the volunteers were then asked which torture they would prefer: 30 more seconds in the vise, or 30 more seconds of the annoying sound.

Most opted for the sound. Again, the anchoring had no statistical effect on whether people preferred the sound or the vise. The anchors affected only the prices.

Economists had long articulated an ideal of decisiveness and self-consistency in financial matters. Apparently, this was not just an abstraction for Ph.D.s; it is a widely shared ideal that average folks try to live up to. We all pretend to have the self-consistent reserve prices of theory and common sense. But the unspoken truth is, all we know are relative valuations. We are ratio wise and price foolish.

One of America’s longest-running Guy Grand pranks takes place every day just off 1-40 in Amarillo. A giant steer statue is blazoned with a sign advertising a
FREE
72
OZ STEAK
. It’s the signature dish of the Big Texan Steak Ranch, and it comes with salad, shrimp cocktail, baked potato, roll, butter, and a very big catch. The catch is that the customer has to consume everything within one hour. Otherwise, the price is $72.

In our litigious age, a deal like that merits some fine print. Customers must pay the $72 up front, to be refunded if and when they clean their plate. Rule #5: “You don’t have to eat the fat, but we will judge this.” No third party is allowed to touch the food (lest they palm a baked potato?). Diners must sign a waiver accepting responsibility for any and all health risks. Those ordering the 72-ouncer are the Big Texan’s de facto floor-show: they have to sit on a special platform, in view of all, and aren’t allowed to leave the table during the meal. And just in case you’re wondering, anyone who vomits is disqualified even if they want to continue. A bucket is supplied.

Whatever the 72-ounce steak does to the Big Texan’s ambience, it has earned the restaurant more publicity than it knows what to do with. It’s become a perennial favorite of TV food and travel shows. A
Simpsons
episode had Homer tackling a 256-ounce “free” steak.

Since 1960 (when the price was $9.95), about 60,000 trenchermen have taken the challenge. The restaurant reports that 8,500 have managed to eat everything, an overall success rate of about 14 percent. Not many women try, but about 50 percent have succeeded. Those who
order the 72-ounce steak probably feel it’s a good deal no matter what. It’s a dollar per ounce of beef, and unlike all-you-can-eat promotions, the customers are allowed to take home leftovers.

It’s a can’t-lose proposition—uh, until you realize you just paid $72 for dinner in Amarillo.

The “free” 72-ounce steak stands at the nexus of vernacular and professional pricing. Big Texan owner Bob Lee came up with the gimmick on his own in 1960, long before the age of menu consultants. His promotion anticipates several principles now espoused by academics and marketing professionals alike. Most important, the 72-ounce steak is an anchor. You can’t come anywhere near the Big Texan without being exposed repeatedly to the
idea
of eating a 72-ounce steak. Though the vast majority of Big Texan customers will never order it, the exposure subtly raises diners’ estimates of how much they can eat and what they’re willing to pay. One of Daniel Kahneman’s anchoring experiments is worth mentioning in this connection. He and Karen Jacowitz tried asking:

(a) Does the average American eat more or less than 50 pounds of meat a year?

(b) How much meat does the average American eat in a year?

The median answer was 100 pounds of meat. They asked another group whether the average American ate more or less than
1,000
pounds of meat a year. For this group, the median estimate was 500 pounds.

The Big Texan promotion is also a simple example of
nonlinear pricing
. “Nonlinear” means that the price (or price per ounce) is not a straight line—it varies with the amount consumed. The 72-ounce steak costs $72 until you finish everything, and then the price plummets to zero.

This type of pricing casts a hypnotic spell. It is one of the most common tricks of price consultants, used for everything from cell phone bills to airfares. A hungry customer at Big Texan does not know whether he’s going to be paying $72 or nothing. That uncertainty renders the $72 a bit less real. There is an alternate way of judging the deal: by price per
ounce. A chart of the price per ounce is a curve slanting steeply downward, then slowly approaching $1, and finally dropping to zero. A diner who ordered the 72-ounce steak, ate a 1-ounce mouthful, and put down the fork would be paying the outrageous price of $72 an ounce. But someone who ate a couple of pounds of meat would be paying $2.25 an ounce, and someone who ate nearly the whole thing would be paying just over $1 an ounce. That’s pretty reasonable. The customer is so concerned with getting a “deal” that he pays a price he wouldn’t otherwise stomach.