Blockbusters and the long tail (part 3)

Adapting Hausman and Leonard’s model to our setting, we found that the economic value consumers gained from being able to access obscure books online was between $700 million and $1 billion a year in 2000 – nearly ten times the value consumers gained from lower online prices. The main determinant of value for online consumers, in other words, wasn’t saving a few dollars on products that they could already buy in physical stores.

Rather, it was the value generated from their new ability to discover, evaluate, and consume millions of products that didn’t fit within old brick-and-mortar business models. This value increased during the 2000s, as we showed in 2008, when we used new data to revisit our earlier analysis. In the 2008 study, we found that three changes had significantly increased the value that consumers gained from the increased product variety they found online.

First, internet book sales had increased from 6 percent of total book sales in 2000 to nearly 30 percent in 2008. Second, in 2008 consumers were even more likely to buy niche titles than they had been in 2000. Third, consumers had many more books to choose from in 2008. The number of new titles printed each year had increased steadily from around 122,000 in 2000 to around 560,000 in 2008.

Our study showed that together these changes caused the value consumers received from increased product variety to quintuple between 2000 and 2008, to between $4 billion and $5 billion per year. And recent work by Luis Aguiar and Joel Waldfogel suggests that these figures may underestimate the true value gained from increased product variety online. That’s because, as we noted in earlier articles, no one knows which products will go on to become big hits. Publishers, labels, and studios do their best to anticipate which titles will fly off the shelves, but the process is imperfect.

As a result, when technology gives previously undiscovered artists access to the market, some of these new artists – maybe even many of them – will surprise industry gatekeepers and land in the head of the sales distribution rather than the tail.

Aguiar and Waldfogel tested their theory by analyzing value creation from new music. They first observed that technological change has caused an explosion in new music, and that the number of new recorded-music products tripled from 2000 to 2010. Then, by applying their theory, they showed that the economic value created by these new products increases by a factor of 15 if one includes the possibility of blockbusters’ emerging from the long tail.

One might ask whether these results extend to the tail of the tail – for example, to truly obscure books that have languished on the shelves of used-book stores for years. In some ways, these are exactly the sorts of products that McPhee’s theory predicts will fail to deliver value. That’s certainly Anita Elberse’s view. In a 2008 article in the Harvard Business Review titled “Should You Invest in the Long Tail?” she writes (citing McPhee): “Although we might believe that ‘the out-of-the-way book is at least a delight to those who find it,’ in reality, the more obscure a title, the less likely it is to be appreciated.”

Does this effect show up in the data? A recent empirical paper by Glenn Ellison and Sara Fisher Ellison may shed some light on this question. Ellison and Ellison studies the economic value generated by online markets for used books, motivated in part by an experience one of the authors had had while searching for an obscure out-of-print title:

Several years ago, one of us wanted a thirty-year-old academic book on the pharmaceutical market which the MIT library did not have. The book had long been out of print, and looking for a used copy in brick and mortar stores would be like looking for a needle in a haystack. A quick search on Alibris, however, produced four or five copies for sale. A copy was ordered, for around $20, and it arrived shortly, with $0.75 written in pencil on the inside front cover and subsequently erased! The book had evidently been languishing on the shelf of some used bookstore for years, and not a single customer who noticed it was willing to pay even $0.75. A researcher needing the book happily paid $20 and would have paid significantly more.

To test the possibility that similar things happen to other niche books, Ellison and Ellison collected a detailed dataset of the prices of used books from both online and physical stores. Their analysis of the data shows that the ability to find just the right thing among the millions of obscure used books available online generates a great deal of economic value for both consumers and booksellers. In short, products which may not be appreciated by the vast majority of consumers, can still generate a great deal of delight – which economists equate to economic value – when discovered by the right consumer.