E pluribus tunum

What kind of stupid undergrad students willing to pay $1.5 to download a song? The maximum amount I am willing to pay to download a song is $0. Why pay a dime if you can get it for free? A song only worth the cost of electricity and bandwidth to download it.

Oct 22nd 2009 – From The Economist
Uniform prices for online music are no way to maximise profit

THE term “bargain basement” was supposedly coined to describe the first outlet of Filene’s Basement, an American discount shopping chain whose bags scream “I just got a bargain!” in large red letters. Those bags celebrate a feeling most shoppers will be familiar with: the pleasure of having bought something for less than they were willing to pay. This difference between the price paid and the most a buyer would have agreed to shell out is known as “consumer surplus” and is the part of the overall economic benefit from any trade that the buyer gets.

The rest of the surplus from a trade goes to the seller. To maximise his share of the economic surplus, a merchant wants to sell to each customer at a price as close as possible to the most that client would be willing to pay. Naturally this is impossible if everyone is charged the same price. Buyers who would have been willing to pay more get a big chunk of the surplus, leaving less for the seller. In addition those willing to pay less than the uniform price are simply left out, even though they may be willing to pay enough to allow costs to be covered. Lowering the price enough to attract them would reduce the amount that customers who value the product more could be charged, reducing overall profit.

Where possible, the solution is to tailor prices to people’s willingness to pay or price-sensitivity, much as vendors in street markets do when they offer naive tourists a higher price than seasoned locals. Most of the time it is hard to know how much people value things. But sellers can still encourage people to sort themselves into groups which can be charged different prices. Those who are particularly keen to get their hands on a copy of a new book, for instance, may be willing to pay a higher price for it, which may explain the persistence of expensive hardbacks. The logic behind cheaper student editions is similar.

These tactics and others, like selling things in bundles rather than individually, all exploit differences in people’s sensitivity to price. They ought to be better for sellers (and possibly for some customers too) than a uniform price. So many found Apple’s initial pricing strategy on iTunes, its popular online music-store, perplexing. Until April this year, when it switched to a multi-tiered pricing system, every song available on iTunes cost a uniform 99 cents in America, 79 pence in Britain and so forth. This had the benefit of simplicity. But it seemed likely that other ways to price music online would be more profitable.

Knowing how sellers and buyers would act under different pricing schemes requires information about how those who buy music online value different songs. New research* by Ben Shiller and Joel Waldfogel, two economists at the University of Pennsylvania’s Wharton business school, tries to answer this question. In January 2008 the researchers presented nearly 500 undergraduate students at Wharton with clips of the 50 most popular songs on iTunes earlier that month. Having listened to each clip, the students were then asked to write down the most they would be willing to pay to download the song in question. Data on more than 23,000 song valuations resulted, allowing the professors to get a sense of the actual demand curves for popular songs. Similar data were also collected in January this year, though this time some older and less popular tracks were also included.

The exercises showed that even a uniform price per song that maximised revenue among the students was quite high—$2.30 in 2008 and $1.46 in 2009. Wharton students may be particularly fond of music, but it is also possible that the market would sustain a higher uniform price than 99 cents. More important, knowing the uniform price that maximised revenue also allowed the authors to evaluate other ways to price online music.

One alternative is song-specific pricing, much favoured by record companies. (Apple has already moved a bit in this direction with its multi-tier system.) But the research suggested that this would increase profits by a mere 3%. Part of the problem was that people who valued one song highly also tended to place a high value on others. This implies that person-specific, rather than song-specific, pricing would be more efficient. But sellers’ data are not refined enough to set different prices for different people. People may resent such pricing anyway, so it could harm sellers’ brands. Crude profiling—by race or sex, say—would be illegal. In any case, the authors found that basic demographic information did not tell them much about musical tastes.
Rhythm ’n’ choose

Charging an “entry fee” for use of the service and then a small, fixed per-song cost for downloads turned out to benefit both the seller and the buyer. The most revenue, according to the 2009 survey data, would be generated by charging the students $21.19 for entry and 37 cents a song. This could raise the producer surplus by 30% compared with uniform pricing. Consumer surplus would also rise in this instance, because some people would buy songs they would have not have done at a higher uniform price. Spotify, a rival to iTunes, has a model somewhat like this for its premium service, where it charges a monthly fee for songs without limit.

Selling several songs together in a bundle (much like an album) had almost identical results. The authors reckon that there is a range of prices for bundles that would make both buyers and sellers better off than a single price per song. For example, charging $16.95 for all 50 songs studied (which would have cost $49.50 to download) would leave profit unchanged because enough people would buy. But it would increase consumer surplus by around 50% by attracting new buyers. The costs of implementing pricing schemes go up with their complexity. But it should be possible to set prices that both increase profits and leave at least some music-lovers with the thrill of a bargain.

Comments are closed.