The text below was prepared for the Danish Ministry of Culture’s Book and Literature Panel. On 3 May 2016, I was invited to Copenhagen to give a presentation about the state of e-book lending through public libraries in the Netherlands. After the meeting I wrote the text with some updated information. The text was finalized on 26 August 2016 and has been translated into Danish for a publication of the Danish Ministry of Culture. The original English and Danish language versions of the publication in which it was included can be downloaded from this site.
1. The public library landscape in the Netherlands
Since the turn of the millennium, the number of public library organizations in the Netherlands has declined considerably, from 544 in 1999 to 156 in 2015 (Statistics Netherlands, 2016). A policy program called Library Renewal (‘Bibliotheekvernieuwing’, 2000-2008) succeeded in arriving at a limited number of larger organizations through mergers of local libraries. Library organizations are providing public library services in multiple municipalities. The 156 organizations are currently operating
- 782 branches with a minimum of 15 opening hours per week
- 215 service points, opened 4-15 hours per week
- 55 mini service points, opened less than 4 hours per week
- 67 delivery points
- 11 self-service libraries without staff
- 10 book mobiles with 141 stops
(Netherlands Public Library Association, 2016).
Public library membership in the Netherlands is a paid membership, at least for adults. Most library organizations offer free memberships for children and youngsters, but for those aged 16-17 it is common they have to pay a fee. From the age of 18 (19 in Amsterdam), a full membership fee is due, the height of which varies between library organizations and is also dependent on the service level. Typically, libraries offer a choice between three membership options: a regular membership, a cheaper reduced version for incidental borrowers, and a more expensive top version for library aficionados. As an example, in the public library of Amsterdam the corresponding rates are € 35 (€ 25 for 19-22 years and over 65s), € 20, and € 55 (€ 45 for 19-22 and over 65s). What is included in each of the membership versions is dependent on local conditions.
Only one in every eight euros of the total public library budget is paid by registered members.
It is important, however, to stress that the Dutch public library system, as elsewhere, is financed largely by the local, regional and national governments. In 2015, subsidies constituted over 82 per cent of the total budget of public libraries (519.4 million euros), whereas 12 per cent derived from membership fees (the remainder of the budget coming from provinces, other grants and contributions, and on-charged costs; the budget for the national e-book portal is not included in this sum). In other words, only one in every eight euros is paid by registered members.
National statistics reveal a considerable decline in the number of registered adult users, from 2,3 million (1999) to less than 1,5 million (2015). The number of children and youngsters (0-17 years) declined somewhat between 2000 and 2005 from 2,1 to 2,0 million. Since then, a steady increase has set in. Currently, 2,3 million children and youngsters are registered members, bringing the user total on (1,5+2,3=) 3,8 million on a population of 17,0 million.
Figure 1 Registered public library users in the Netherlands, 1999-2015 (in millions)
Likewise, the collection sizes have been in decline. In 1999, the total collection size numbered around 43 million items, of which 25 million remained in 2015. Most significantly, the number of books lent out was cut in half between 1999 and 2015, from 145 to 73 million.
Figure 2 Volumes lent out by public libraries, 1999-2015 (in millions)
2. E-book lending through public libraries in the Netherlands
The national e-book portal
The national e-book lending portal was launched in January 2014. Registered users of the 156 public library organizations can make an account on the portal with their membership number and start borrowing e-books. At the time of writing, some 11,500 e-book titles are available for borrowing, a number corresponding with 25 per cent of the commercially available titles. Recently published books are in short supply, yet titles that appeared 6 months ago or longer are common to be on offer. Which titles appear on the e-book shelf is subject to negotiations between the Public Library Association and the Royal Library on the one hand, and publishing companies on the other.
Since the launch, the number of registered users on the portal gradually increased to 314,000 accounts on 1 July, 2016 (out of the 3,8 million library members who could register to the service at no additional cost). This is not to say that all these accounts are used. In the first half of 2016, 143,000 accounts were actively used (at least one book borrowed). In the first quarter of 2016, an average of 213,000 borrowings took place monthly, as opposed to a monthly average of 232,000 borrowings in the second quarter.
Figure 3 Quarterly sales and loans of e-books in the Netherlands
The amount to be paid to publishers varies, depending on such factors as whether or not titles are licensed in bulk or individually, whether or not an up-front payment (fixed cost) is required in addition to the per-lending fee (variable cost), and particularly whether it is a recently published title or one which has almost exhausted its commercial potential. According to a 2014 research report, the pay-per-loan licensing model in the Netherlands had the lowest average cost per loan of all European models studied (Mount & Huysmans 2014). It is not sure if this is still the case. The Royal Library, which is now in charge of licensing negotiations, does not wish to disclose financial details for obvious reasons.
The Vakantiebieb app
In addition to the national lending portal, the Dutch public libraries have introduced the Vakantiebieb (‘holiday library’) in the summer of 2013. After downloading an app, everybody – both members and non-members of one of the public libraries – can register and borrow and read a limited number of titles during holiday periods, especially during the summer months. The service proved very popular. In the summer months of 2015, 578,000 users registered, who together borrowed 2.25 million e-books. The amount needed to remunerate rights holders is covered by subsidy from the national government (Ministry of education, culture, and science) through the Royal Library.
Both in terms of registered users and loans, the Vakantiebieb app can be called more successful than the regular e-book-portal. In comparing these figures, one should bear in mind that borrowing e-books in the holiday library is free for everyone (17 million inhabitants) whereas the e-books in the national portal are only available to the 3.8 million library members. Furthermore, the figures do not give any evidence of the extent in which the downloaded books are actually read. It is conceivable that holiday library ‘members’ simply download all available books (in the summer of 2016: 30 books for adults and 30 for children) with the intention of selecting titles to read when arrived at the holiday destination. Likewise, users of the regular e-book portal can download titles from the 11,500 on offer without ever finding the time to read. Without further information about the access/download ratio of both models, drawing conclusions about the relative ‘success’ is speculation at best.
Audiobooks: the LuisterBieb app
There is a separate app for borrowing audiobooks called LuisterBieb. It currently (August, 2016) offers 634 titles in the rubrics adult books, youth/children’s books, foreign/translated and informative books. Over half of the titles are available for use for both members and non-members of the public library. The remainder is available for library members only (‘premium’ titles).
3. Purchasing and borrowing e-books: evidence for ‘cannibalization’?
In developing e-lending models, the fear by rights holders of ‘cannibalization’ of sales by lending has been a factor of importance. The fear is based on a series of assumptions including at least these three:
- profit margins on a sold title are higher than on a title lent out
- persons borrowing a title will not buy the same title
- despite protection measures (‘digital rights management’, DRM), titles lent out might be stripped of that protection by tech-savvy users and shared digitally on peer-to-peer platforms, or physically on USB sticks and CDs/DVDs
Although the assumptions are logical at first sight, there are also counterarguments suggesting that e-lending might be beneficial to rights holders in the short and long term:
- through e-lending, a lot more readers may be reached by the same title, leading to an additional revenue stream
- persons borrowing a title may want to read more titles by the author, leading to borrowing and/or buying of other titles – which might otherwise not have been the case
- although copyright violations are a real threat to publishers’ and authors’ revenues, enhancing/promoting a culture of reading might in the long term result in the selling and lending of more titles, with additional revenues as a consequence.
To provide empirical evidence of the validity of the arguments is extremely hard in real-life situations, as only in rare cases (quasi-) experimental situations arise that could settle the dispute once and for all. In practice, therefore, more often than not publishers and libraries find themselves in a difficult negotiation situation. The 2014 study mentioned earlier, comparing 19 e-lending models in European countries, Canada and the United States, paid attention to the possible conflict of interest between rights holders and libraries affecting the negotiations over an ‘ideal’ e-lending model. Overall, it appears that providing access to titles via e-lending is easier to accomplish in the larger language communities (English, German) than in smaller ones (see Mount & Huysmans, 2014; see also Mount, 2016).
It appears that providing access to titles via e-lending is easier to reach in the larger language communities (English, German) than in smaller ones
For the Dutch e-lending portal, there is some evidence that the level of cannibalization is limited at best. The logistical service provider CB, the de facto monopolist in the Dutch market for selling, renting and lending e-books, stated in its 2016 Q1 e-book monitor: “loans have little or no negative influence on sales”. This is substantiated by three pieces of evidence:
- “A top-25 loaned book is on average 3 years old”, whereas a book from the sales Top-25 has been published only slightly over 1 year ago;
- “In numbers, the top-100 sales is twice as big as the top-100 loans”
- “For both sales and loans the large numbers come from the long tail”
What is more, over the last year CB consistently (in every quarter year) the top 25-loaned books to sell more copies in the month they become available for e-lending, not less. Evidently, this does not give any information on how many copies would have been sold more if titles hadn’t been available for e-lending. Looking at the lifecycle of top-selling e-books, however, CB shows that the sales numbers wear off quickly. Six months after publication, the spike in sales belongs to the past, and it is safe to say that 1,5 years the commercial value approaches zero (CB, 2016b).
All in all, it’s an indication that the cannibalization effect is limited and in any case not absolute (like that every copy that is lent would otherwise have been sold). Convincing as these observations may appear, they point to the established situation only. Although some recent and bestseller titles are on offer in the public libraries’ e-lending portal, it should be borne in mind that it has proven easier for libraries to license older titles than newer and bestselling ones. Under the copyright regime in the European Union, publishers have the power to effectively block the inclusion of titles in the e-lending portal.
This had led to two separate markets, one for sales (recent, bestsellers) and one for lending, which overlap only to a limited extent. (As was mentioned above, only 25% of titles available in e-form are available in the e-lending platform.) Although from an economical/commercial point of view this situation has some logic to it, from a cultural perspective it is far from perfect. Public libraries strive for inclusion of every available title in the e-lending platform as a matter of principle, like they have the right to do in the physical realm. A sound test of the cannibalization thesis, therefore, is not provided by the Dutch case. For that to materialize, it would be necessary to (temporarily) include a larger number of commercially viable titles in the e-lending platform and see whether this impacts on the sales figures.
Figure 4 Comparing the Dutch and Danish e-book portals: loans per 1000 inhabitants
4. Future developments
E-lending models, in the Netherlands, Denmark and elsewhere, are developing in a complex interplay of factors of a political, economical, cultural and educational nature. It is therefore not very likely that a one-size-fits-all e-lending model will emerge that can be applied to diverse socio-cultural contexts. A recent report of a working group established by the Directorate-General on Education and Culture (EAC) of the European Commission illustrates how local and global influences can affect reading promotion in general and e-lending in particular. Within the culturally and linguistically diverse context of the European Union, it might be advisable to create possibilities for a diversity in e-lending models to cater for, for example, linguistic and/or ethnic minorities (like for instance Italian speakers on the Istrian peninsula of Slovenia and Croatia) and global diasporas (like for instance the Estonian one (see European Union, 2016).
Public libraries strive for inclusion of every available title in the e-lending platform as a matter of principle, like they have the right to do in the physical realm.
A potentially very influential development is the opinion of one of the Court of Justice of the European Union‘s Advocates General on questions posed by the District Court of The Hague in a legal procedure between the Dutch Public Libraries Association (VOB) and the Dutch Public Lending Rights Foundation (Stichting Leenrecht). In his opinion, published on 16 June 2016, Advocate General Szpunar advised the CJEU to rule that the European Union’s Rental and Lending Rights Directive (Article 1(1)) must be interpreted in the sense of including the right to lend electronic books included in a library’s own collection (Szpunar, 2016). Whereas the communis opinio hitherto had been that e-lending should not be included under a public lending rights scheme (as has been the case in a lot of European countries for paper books), Szpunar holds that the directive, although it was not intended to include e-lending at the time of adoption, should co-evolve with the evolution of technology and could therefore include the electronic lending as well as physical lending.
Should the Court of Justice of the European Union follow the AG’s opinion (the judgment is scheduled for delivery on 10 November 2016), this would principally enable the inclusion of e-books in the existing public lending rights schemes – if the member states would decide accordingly. The AG names a one-copy-one-user model as a model that could serve that purpose, which in principle would enable public libraries across the EU to include every digitally available title in their collections. Despite its limited user-friendliness as compared to the current Dutch one-copy-multiple-users-model, from an accessibility point of view this would be a step forward. For this to materialize, however, a lot of political and legal discussions as well as negotiations will have to take place before the dust settles and a new era for e-lending begins.
CB (2016a, 21 April). E-book barometer Q1 2016. Retrieved from http://www.cb.nl/nieuws/ebookbarometer-2016-q1/, 8 August 2016.
CB (2016b, 12 July). E-book barometer Q2 2016. Retrieved from http://www.cb.nl/nieuws/e-bookbarometer-2016-q2/, 8 August 2016.
European Union (Directorate-General for Education and Culture)(2016). Promoting Reading in the Digital Environment. Luxembourg: Publications Office of the European Union. Retrieved from http://bookshop.europa.eu/en/promoting-reading-in-the-digital-environment-pbNC0116151/, 8 August 2016.
Mount, Dan (2016). E-lending: Challenges and Opportunities. In: Barbara Lison, Frank Huysmans & Dan Mount, Research for E-Cult Committee – Public Libraries: Their New Role (pp. 69-106). Brussels: European Parliament. Retrieved from http://www.europarl.europa.eu/RegData/etudes/STUD/2016/585882/IPOL_STU%282016%29585882_EN.pdf, 8 August 2016.
Mount, Dan & Huysmans, Frank (2014). A Review of Public Library E-Lending Models. Retrieved from http://tinyurl.com/nuh5gs7, 8 August 2016.
Netherlands Public Library Association (VOB)(2016, 3 August). Bibliotheekvestigingen in Nederland (Library Branches in the Netherlands). Retrieved from http://www.debibliotheken.nl/de-vob/publicaties/nieuws/bericht/bericht/bibliotheekvestigingen-in-nederland/, 8 August 2016.
Statistics Netherlands (2016, 22 July). Public libraries 1999-2015. Retrieved from http://tinyurl.com/NLPublicLibraries, 8 August 2016.
Szpunar, Maciej (2016, 16 June). Opinion of Advocate-General Szpunar, Case C-174/15, Vereniging Openbare Bibliotheken v Stichting Leenrecht (Request for a preliminary ruling from the Rechtbank Den Haag (District Court, The Hague, Netherlands)). Retrieved from http://curia.europa.eu/juris/document/document.jsf?docid=180332&pageIndex=0&doclang=EN, 8 August 2016.
E-book lending in the Netherlands in European perspective by Frank Huysmans is licensed under a Creative Commons Attribution 4.0 International License.
Interessant Frank, heel erg bedankt voor dit artikel. Het is inderdaad zo dat ‘E-lending models, in the Netherlands, Denmark and elsewhere, are developing in a complex interplay of factors of a political, economical, cultural and educational nature. It is therefore not very likely that a one-size-fits-all e-lending model will emerge that can be applied to diverse socio-cultural contexts’. Toch lijkt me belangrijk dat naar bepaalde niveau van harmonisatie wordt gezocht. Moeilijk te bepalen hoe dat geregeld kan worden. Bij het lenen van fysieke boeken is het vanzelfsprekend. Daar waren bibliotheken heel innovatief in. Wellicht was het toen ook niet makkelijk om dit voor elkaar te krijgen. Misschien hebben we tijd nodig om binnen de bibliotheekwetenschap standaard oplossingen te blijven vinden voor nieuwe vragen.