Do we need brick-and-mortar bookstores in the era of digitisation, where the next book is just “one click away”? This question is at the centre of an ongoing and international legal and policy debate. Proponents of regulations such as fixed book price laws point out that traditional bookstores are important to maintain a vivid book market. Opponents argue that such regulations hinder more efficient distribution channels such as e-commerce, which might be more convenient and cheaper for customers. The answer to the question depends on whether buyers consider the offline and the online channel as close substitutes, and therefore switch easily between the two channels without adversely affecting the pattern and volume of book sales.

This blog post distils the findings of an article in which we empirically analyse the substitutability between brick-and-mortar stores and e-commerce based on data for Germany. Our findings imply that a lower number of bookstores leads to decreasing book sales. If a local bookstore closes, a significant share of consumers does not switch to an e-commerce platform to order physical books online or to buy e-books but simply consumes fewer books. These former book buyers potentially choose other leisure activities instead of reading. This means that the offline and online retail channels are no perfect substitutes for consumers.

It is a policy goal in many European countries to get people to read more books. The finding that a reduction in the number of bookstores leads to people buying (and reading) fewer books indicates that traditional brick-and-mortar stores are crucial to achieving that goal. Given that a central element of fixed book price systems is to protect brick-and-mortar bookstores (as, for instance, fixed in paragraph one of the German law on fixed book prices), our research provides empirical support for a positive effect of these regulations on the promotion of books as a cultural good vital for society.

To analyse to what extent book buyers perceive brick-and-mortar stores and e-commerce as substitutes, we compiled an extensive data set. We received support by the German Publishers and Booksellers Association in collecting these data, which is gratefully acknowledged. The data set was provided by Media Control and covers almost all sales of physical books in Germany from 2011 to 2017. These data were merged with information on the number of brick-and-mortar stores in Germany. We also used consumer-panel data provided by GfK, Germany’s largest market research institute, to account for sales of e-books and audiobooks.

In Germany, the sales of physical books went down by approximately 11 per cent from 2011 to 2017. However, revenues remained almost constant, which indicates that book prices increased on average. Sales of e-books and audiobooks show an increasing trend. Notably, the sales of e-books account for at most five per cent of yearly sales over the period 2014-2017. In the same period, the number of brick-and-mortar stores went down by approximately 20 per cent.

The statistical analysis of the relationship between book sales and the number of bookshops sheds light on the degree of substitution between brick-and-mortar stores and e-commerce. A central challenge to this analysis is to solve a chicken-and-egg problem: do book sales decrease because the number of bookshops goes down or does the number of bookshops decline because the demand for physical books decreases? For further information on how to solve the chicken-and-egg problem methodologically see the full article.

Our results indicate that the drop in the number of brick-and-mortar bookstores explains about 33 per cent of the decrease in book sales. Taken literally, this means that without the drop in the number of bookstores the decrease in book sales would have been 11.3 million books smaller over the entire period 2011-2017. Even if we account for e-books, book sales decrease significantly and a share of 26 per cent of this decrease can be explained by the drop in bookstores. These findings imply that e-commerce and brick-and-mortar stores are not perfectly substitutable: When local bookshops close some consumers buy fewer books or stop purchasing books at all.

We observe that when an average bookstore closes, sales of fiction titles decrease by around 4,250 books, whereas the sales of non-fiction titles decrease by around 1,340 books. The extent to which consumers perceive brick-and-mortar stores and e-commerce as substitutes differs between genres. Our findings indicate consumers perceive e-commerce a closer substitute to brick-and-mortar stores for non-fiction-titles than for fiction-titles.

We find that around 23 per cent of the overall decline in fiction-sales can be explained by the closure of bookshops. This finding indicates that other factors such as, e.g., a change in leisure activities are more responsible for the drop of fiction sales than the reduction of the number of bookshops. These other factors explain the decrease in sales of, e.g., non-fiction titles to a smaller degree: almost 60 per cent of the overall decline in non-fiction sales are attributable to the decrease in the number of brick-and-mortar bookstores.

Do we still need bookshops in times when consumers can purchase books in a few seconds using their smartphones? Although some consumers clearly prefer the online channel, there is still an economically significant number of consumers who still favour their corner book shops. When their local book shop closes those consumers buy fewer books. This is an important finding that should not be overlooked in the debate on the regulation of book markets.



Georg Götz is a professor in the department of economics at Justus-Liebig-University Gießen.




Daniel Herold is a postdoc researcher in the department of economics at Justus-Liebig-University Gießen.



Jan Thomas Schäfer is a research associate in the department of economics at Justus-Liebig-University Gießen.




Phil-Adrian Klotz is a research associate in the department of economics at Justus-Liebig-University Gießen.