When commerce began to move online, economists predicted two big benefits for consumers. Prices would become lower and more uniform. And the selection available to consumers would increase.
Identifying those benefits has been challenging. Online prices have proved to surprisingly diverse. And while the selection of products online is indeed vast, many are niche products such as self-published books for which demand is scant to non-existent. E-commerce is still a net plus. But papers presented at this year's meeting of the American Economic Association demonstrate that its value arises in ways that economists did not foresee, and that are not easily captured by measures such as GDP.
Consider used books. Glenn Ellison and Sara Ellison of the Massachusetts Institute of Technology (MIT) collected prices on 335 titles and found that on average, the typical title sold for $17.80 online, 50% more than in stores. Paying more for an identical product would normally leave you worse off, not better. But in this case, the Ellisons argue the opposite is true：higher prices are a sign that buyers are being better matched to books they want.
For example, only a few bookshops might carry an out of print title such as “The Reign of George III, 1760-1815”, published in 1960 by Oxford University Press. They may never be visited by the readers most keenly interested in that book, who are scattered around the country (or beyond). By posting its inventory online, all those readers are added to the potential demand for the store's copy. Higher demand translates into higher prices which clearly makes the booksellers better off. But so is the reader since without the internet he would not have found the book. One of the authors some years ago bought a 30-year-old academic tome on pharmaceuticals online that the MIT library didn't have. She paid $20 and upon arrival, saw that it had $.75 written in pencil inside the front cover, then erased. It had “evidently been languishing on the shelf of some used book store for years, and not a single customer...was willing to pay even $.75.”
These benefits are less likely to hold for easy-to-find, commoditised products；online prices of popular, usually in-print, books were less dispersed and closer to offline prices.
Rare used books are an example of a “long tail”：a vast expansion in variety. But since the newly available products are often of niche interest only, the aggregate benefit to consumers is small. For example, since a minority of account for the vast majority of music sales, the fact that songs available for digital down-load tripled between 2000 and 2010 might not be a big deal：most of those new songs max be by marginal artists of no great interest.
Joel Waldfogel of the University of Minnesota believes this understates the internet's contribution. The demand for cultural products is much harder to predict than for conventional products such as shoes or soda. Seasoned publishers have only a vague idea what book, film or song will be hit. A major record labels can sign only a fraction of the artists available, knowing full well it will unwittingly reject a future superstar.
Thanks to cheap digital recording technology, file sharing, YouTube, streaming music and social media, however, barriers to entry have been dismantled. Artists can now record and distribute a song without signing to a major label. Independent labels have proliferated, and they are taking on the artists passed over by major labels. Hit songs are still a lottery, but the public gets three times as many lottery tickets.
This seems at odds with the collapse in recorded music revenue since 2000 which suggests declining, not rising, music industry output. Mr Waldfogel says that is misleading：because of piracy, revenue understates how much music the public has really consumed. He calculates the quality of songs recorded since then has been either stable (based on the number that made it onto critics' “best-of” lists) or significantly improved (based on the pattern of sales and airplay). This is corroborated by the success of indie-affiliated bands such as Arcade Fire and Mumford & Sons. Indie labels' share of the Billboard to 200 selling albums grew from 13% in 2001 to 35% in 2010. Mr Waldfogel and a coauthor reckon that tripling the selection of songs available has produced 15 times as much benefit for consumers than tripling the selection of a more predictable product.
That prices do not properly capture the benefits of online commerce is reinforced by a paper examining the search behaviour of users on eBay. Such people are often assumed to know what they want and ruthlessly pursue the lowest possible price for it. But a paper by Tom Blake of eBay, Chris Nosko of the University of Chicago and Steve Tadelis of the University of California at Berkeley, paints a more nuanced picture.They followed 500,000 eBay users chosen randomly on one particular day, and found just 16% of their searches resulted in a purchase. Users seemed to find browsing almost as costless, since they were willing to conduct an additional search to achieve an average saving of just 25 cents. Moreover, many searchers were not trying to get a better price for one product but explore different products. A user who typed in “opera DVDS” examined offers of Puccini, Verdi, Wagner and Bizet before returning and purchasing the Puccini.
This, the authors reckon, suggests price savings are but a small part of the value users derive from search；some of the value comes from what they learn while searching, much as many people enjoy window-shopping at bricks-and-mortar outlets. This is clearly a problem of mainstream economics, which generally assumes search costs make people worse off, not better. It also underlines how far away measures like GDP are from capturing the benefits of the internet.
