Only a “Longish” Tail

Prior to the Internet, many industries catering to mass audiences, particularly entertainment industries, tended to have very skewed sales distributions with most sales volume accruing to a small share of products (titles) generally known by terms such as “blockbusters.” The long tail hypothesis claims that increased Internet usage will weaken the dominance of blockbusters in these industries, so that the large number of obscure products, in aggregate, will become an important, perhaps primary, portion of the market. Such a change would have significant implications for firms’ operations. We test this hypothesis for the book industry with two analyses. First, using unusually rich data on book sales, we examine the temporal share of sales taken by niche titles as Internet retailing grew from obscurity to ubiquity. Changes in the distribution of sales by title popularity are weakly consistent with the direction of the hypothesis but not with the magnitude. Second, we exploit the 2011 bankruptcy of the Borders bookstore chain, and the resulting sudden shutdown of all its physical stores as a natural experiment. This analysis also finds a shift in sales away from the most successful titles. But the major beneficiaries are not the most obscure titles but instead titles that would regularly be found in most bookstores. Both analyses support a version of the long tail where the share of obscure titles changes from one very small value to a somewhat greater but still very small value. These results are inconsistent with proposed radical changes in management practices to accommodate a presumed large increase in product variety.