Priceless: The Myth of Fair Value (and How to Take Advantage of It) (27 page)

Charm prices inaugurated the study of psychological pricing. In 1936 Columbia University’s Eli Ginzberg published a one-page note on what he called “customary prices.” “For many years, retail prices in this country have been quoted at one or two cents below the decimal unit—$.49, $.79, $.98, $1.49, $1.98, tell the tale.” Ginzberg reported on the informal experiment of an unnamed large retailer. The firm was curious enough to print multiple versions of its catalog, some with the already customary 9-ending prices and others with the corresponding round amounts.

To Ginzberg the results were “as interesting as they were perplexing.” Some products sold better with charm prices, and some sold worse. His brief article did not supply statistical detail. “The vice-president in charge of merchandising ventured the guess that the losses were balanced by the gains. He realized full well that a repetition of the experiment might . . . permit more definite conclusions.” With money on the line, “the experimental zeal, even of a daring business man, was . . . held in check.”

For nearly half a century, much informed opinion held that charm prices were a harmless superstition. This didn’t keep retailers from using them. By the 1980s, the Kahneman-Tversky revolution had revived interest in psychological pricing. In eight studies published from 1987 to 2004, charm prices were reported to boost sales by an average of 24 percent relative to nearby prices.

Don’t take that quotable figure too seriously. The increase in sales varied from insignificant to over 80 percent. Take an experiment done by Eric Anderson of the University of Chicago and Duncan Simester of MIT. They found a mail order house willing to print up different versions of its catalog. The company sold moderately priced women’s clothing and normally used whole-dollar prices ending in 9. One of the items tested went for $39. In experimental versions of the catalog, the company offered the same item for $34 and $44. Each catalog was sent to an identically sized random sample of the company’s mailing list.

There were more sales at the charm price of $39 than at either of the other prices. The key finding was that more people bought at $39 than at $34. At the charm price there was greater volume
and
greater profit per sale.

This fits in with what’s known of the balance sheets. In 2002
Forbes
magazine concluded that the 99 Cents Only chain’s gross margin was an astonishing 40 percent, twice that of Wal-Mart. On average the chain was paying only about 60 cents for the items it sold for 99 cents. A typical coup: David Gold bought a Fruit of the Loom closeout lot of 700,000 packs of
Star Wars: The Phantom Menace
underwear and sold them when the next
Star Wars
movie came out. Discerning shoppers might have wondered why they were selling
Phantom Menace
and not
Attack of the Clones
underwear—but the 99-cent price answered any questions.

Why do charm prices work? You may feel the answer is obvious. Shoppers must round numbers down, or at any rate focus their attention on the first significant digit. A price like $29.99 registers mentally as twenty-something dollars, while a price of $30.00 or more gets pegged as thirty-something. Twenty-something seems so much less than thirty-something.

This explanation has been widely debated in the marketing and psychological literature. Charm prices actually raise some intriguing questions about how the mind works. Numbers are arbitrary mileposts on the endless highway of magnitude. Does the brain have a deep understanding of quantities signified, or does it manipulate numbers in only superficial ways?

There is a body of psychological research implying that people, even young children, have a decent grasp of magnitudes. They understand that 29 is only a little less than 30. Anchoring experiments have also shown that magnitudes (not just numbers per se) influence estimates and decisions.

Mental rounding alone can’t explain results like Anderson and
Simester’s. If shoppers paid attention to the first digit only, you’d expect that both $34 and $39 would be understood as thirty-something dollars. Sales at both price points would be about the same. Instead, buyers were more likely to buy at the higher price of $39. Nine truly is a magic number.

An alternate theory says that charm prices convey the message that the price has been discounted. Once upon a time, a small-town gas station charged 20 cents a gallon. A new station went up across the street, undercutting the price by a penny: 19 cents. The first station retaliated by charging 18 cents . . . The cultural memory of long-ago price wars has perhaps led us to associate numbers like 19 with competitive pricing and round numbers like 20 with monopolies and poorer values. Unquestionably, something like this is going on even now. Harlem’s 98-cent Plus store was named to undercut the 99¢ Plus store, and it briefly had competition from a 97 Cent store.

Charm prices are informative to any astute shopper. A good way of judging the ambitions of a restaurant or hotel, and sometimes the quality, is whether the prices are in whole dollars or end in .99 or .95. Nordstrom’s department store makes a point of not using charm prices. They mean to say, “We’re not Wal-Mart, come here for quality and expect to pay for it.” This may be why charm prices sometimes don’t work. Price consultant Frank Luby tells of an automaker that thought it wanted to sell a car for $19,999. His research showed the car would sell just as well at $20,000+. Possibly the car’s buyers didn’t want to feel they were buying a “cheap” car. Some retailers, such as Eddie Bauer and J. Crew, have adopted 99-cent endings only for reduced items. Costco uses 97-cent endings to signal that an item is discontinued or slow-moving. To someone who knows this code, the charm prices speak loudly. Of course, customers don’t have to be aware of any explicit rule to respond unconsciously.

In Anderson and Simester’s experiment, there was no significant difference in sales when a garment was priced at $44 or $34. This is further proof that buyers don’t have a strong innate sense of value. It was the $39 price that boosted sales. One hypothesis is that charm prices seem cheap in mental comparison to the round price.

The catalog company was in the habit of putting items on sale and marking them with old and new prices: “Reg $X
SALE
$Y.” The researchers had them print up some catalogs with the sale prices but without any indication that they were discounted. As you’d expect, they saw higher sales when the sale prices were highlighted as such. Buyers didn’t know that $Y was a bargain price unless the catalog told them it was.

Sale price markers were more powerful motivators than charm prices. Consumers were more likely to buy an item marked with the sale price on the left than with the charm price on the right.

Anderson and Simester tried both gimmicks together, using sale-marked charm prices like “Reg $48
SALE
$39.” This had the strongest effect of all. The effect was not additive, though. It boosted sales only a little more than the sale price alone did. This could mean that sale prices and charm prices exploit the same mental principle. Standing on its own, a charm price implies a discount that’s not there. It’s like a mime faking a glass wall. The price’s audience reacts to the virtual discount in much the way they react to an actual one.

Supporting this interpretation is the fact that the charm prices had a bigger effect on new items that the catalog had not carried before. Customers would have had the weakest notion of value with new items and depended more on price cues.

There’s nothing crazy about liking bargains (when the bargain
is
a bargain). A price of $19.99 means what—marked down from $20.00? Gee, thanks. Even that old standby, 99 cents, is only a 1 percent discount from a round dollar (see page 192). By reasonable standards, that should be too trivial to affect behavior much. Yet this fits in with studies of consumer choice and trade-off contrast. When there are many hard-to-evaluate options, attention wanders. It is drawn to easy comparisons, to options that are clearly superior to another, even if the difference is slight. The imagined round-number price becomes a foil for the 99-cent price, bathing it in an unaccountably alluring glow.

The ultimate discount is to FREE!—as in FREE GIFT!! Beloved by marketers, the “price” of zero triggers some unique psychology. In one experiment by Dan Ariely, Nina Mazar, and Kristina Shampanier, they set up a chocolate stand offering Hershey Kisses and Lindt truffles. You don’t have to be much of a chocophile to know that Hershey Kisses are about the lowest form of chocolate, and Lindt truffles are better. They offered the Hershey Kisses for 1 cent apiece, and the Lindt truffles for 15 cents. A prominent sign said
ONE CHOCOLATE PER CUSTOMER
.

Of the people who bought, 73 percent chose the Lindt truffle. With apologies to Hershey’s, no mystery there. Then they reduced both candies’ prices by 1 cent. They offered the Lindt truffles for 14 cents, and the Hershey Kisses for free (still with the one-to-a-customer rule). This reversed the preferences. Sixty-nine percent of customers took the free Hershey Kiss, and only 31 percent bought the Lindt truffle.

Ariely and company were selling the Lindt truffles for about half the wholesale price. Most customers were passing up a 14-cent discount in order to get a free candy they didn’t especially like that might be worth about a cent.

Ariely believes this is largely due to the certainty effect. Any purchase carries a risk of buyer’s regret. The chocolate I bought may not taste as good as I’d thought . . . I might find out I could have bought it cheaper somewhere else . . . what about my diet? . . . etc., etc. Free things are different. You can’t regret spending your money on something free because you didn’t spend any money. By overvaluing certainty, we overvalue anything that’s free.

•   •   •

The magnitude scales of psychophysics are said to have meaningful zeros. On a scale of loudness, the “sound” of silence should be a zero. In practice, things aren’t quite that simple. It can be a challenge to distinguish a barely audible sound from true silence. People tend to say they experienced something—or didn’t—based on power of suggestion, and they feign consistency. (This is why Paul Hoffman had to go to all the trouble of building a fake optometrist’s office in Oregon.)

Much the same applies to the price scale. Consumers do not recognize true worthlessness when they see it. Among items of low value, there is a largish zone of confusion in which it’s unclear what is worth paying for and what is not. This was demonstrated in an already famous 2006 paper by Dan Ariely, George Loewenstein, and Drazen Prelec, “Tom Sawyer and the Construction of Value.”

The title alludes to the classic preference reversal of American literature. Mark Twain’s trickster-hero Tom Sawyer is given the irksome chore of whitewashing a fence. Tom would much prefer to let someone else do it. To achieve that, he pretends to enjoy the job so much that his friends want some of the fun. They beg Tom to let them help, to paint a few strokes at least. Tom refuses, then finally gives in—on the condition that his friends pay him for the privilege of painting the fence. Mark Twain’s mordant point is that there are no absolutes in life, and those who say otherwise are as big a fraud as Tom was.