Agency or Distribution? The Struggle for Mastery in e-book Pricing

May 12, 2011

E-books have been the coming thing for some years now. Google’s scanning initiative and the Gutenberg project have been running for some years, with Google books in particular running into a storm of difficulties in the US and other countries. Gutenberg played safer by concentrating only on books well out of copyright. Some brave authors have even released self-published e-books online only.

Growing popularity of e-reading

Yet despite all this investment and activity, e-reading has until recently remained very much the province of early adopters and the material available for those readers has been limited in quantity. However, a combination of factors – including the introduction of the iPad and the surge of interest in alternative tablet devices, alongside the development and refinement of electronic reading devices such as Amazon’s Kindle – have all made consumption of electronic media a far better and more pleasant experience. This seems to have coincided with the amount of reading material available reaching a tipping point at which the less adventurous are prepared to give e-reading a try and, as a consequence, the e-book market appears to have been successfully unlocked, moving beyond the early adopter niche.

Indeed, whatever the tipping factor was, it seems to have tipped very abruptly, with Amazon announcing on 28 January 2011 that, across its entire US book business, it is ‘now selling more Kindle books than paperback books. Since the beginning of the year, for every 100 paperback books Amazon has sold, the company has sold 115 Kindle books’.

The announcement by the Office of Fair Trading on 1 February 2011 that it had opened an investigation ‘following a significant number of complaints’ into the pricing of e-books is therefore potentially of great significance to publishers and book retailers alike. In short, the OFT will be reviewing whether agreements between certain publishers and retailers concerning the sale of e-books (in particular, the pricing of e-books) are compatible with competition law.

Move to an agency model

Much of the debate in the UK, including significant consumer comment on a variety of blogs, followed Amazon’s publication of an open letter on its Kindle forum on October 19 in which it criticised the decision of a number of publishers to move to an agency model for the distribution of e-books. Under this model, unlike the normal reseller distribution model that exists for physical books, retail prices are set and controlled by the publisher, not the retailer. Following adoption of the model in the US, Amazon has reported retail price increases and a significant decline in demand for books distributed under this model, with growth rates on agency titles slowing to only half that of other Kindle titles.

Under a normal distribution model, any attempt by a publisher to fix the price (or even a minimum price) at which a retailer may sell its products is considered a hardcore infringement of competition law. Only in very exceptional circumstances will such resale price maintenance (RPM) be permissible. The default position is that any such agreement is not only automatically void but also exposes the parties to the agreement to the possibility of financial penalties and even follow-on damages actions by consumers. That contrasts with the agency model, where the publisher retains full control over all terms and conditions of sale, including price.

Agency models are not new. From a legal perspective, whether an agreement is a ‘genuine agency agreement’ will usually turn on whether title and risk in the contract goods remains with the principal until the point of sale, and the level of financial and commercial risk borne by the agent in fulfilling his duties. But even genuine agency agreements may fall within the general prohibition of competition law where they facilitate collusion. Hence the OFT will be looking closely at the factual matrix leading up to the roll-out of agency models by publishers over the course of the past few months, with a view to determining whether any collusion has taken place, or whether the collective outcome supports a finding of a concerted practice among the publishers.

Position in France

This aspect of the investigation was given a further twist on 2 March 2011, when the European Commission confirmed it had carried out dawn raids on publishers ‘in several member states’, citing ‘suspicions of collusion to keep prices high’. At such an early stage of their investigation, precise details are somewhat sketchy, but it appears that the inspection visits were largely focused on French publishers. This is particularly interesting as France maintains very stringent RPM pricing laws in respect of physical (but not, as yet, electronic) books; the law is aimed at protecting independent book sellers from heavy discounting by large resellers by prohibiting discounts of more than 5% from the (publisher-set) cover price.

In October last year, the French Senate introduced an extension to this law, seeking to ensure that there was one price imposed on all books, however distributed (ie covering both physical and electronic distribution). The proposed law also sought to apply not only to distributors established in France, but also to those established overseas (such as Amazon, located in Luxembourg). Following reservations expressed by the European Commission in January that such an extraterritoriality clause would breach Community law, the French National Assembly adopted at first reading a variation limited to distributors established in France, sending the Bill back for further review by the Senate. In France at least, there is a clear desire to extend restrictions on book retail prices to ensure continued protection of independent book resellers.

The one unifying strand for both the European Commission and the OFT is therefore the question of collusion, whether in terms of setting the cover prices, or in terms of collective agreements to move towards an industry-wide adoption of the agency model.

Which model is most appropriate?

Leaving aside the question of collusion, there remains the question of which model is most appropriate for the retail of e-books. It seems clear from the blogs and various press reports that Amazon and at least some consumers are suspicious of a model which places all pricing decisions in the hands of the publisher – not least because the outcome is similar to the position that previously existed under the Net Book Agreement. Yet not only is it questionable where the real consumer interest lies when looking at the bigger picture, it is also far from clear that a wholesale distribution model is appropriate for e-book distribution.

There is no title to electronic files and no physical stock. The normal questions of when title and risk pass from supplier to distributor therefore become moot. Whether set up under an agency or distribution model, delivery of the digital text to the retailer is likely to share the following characteristics.

Creation of a digital format by the publisher. This is often created using optical character recognition technology, which essentially scans the printed text and creates a digital image. This is not foolproof and is the source of many of the complaints concerning the poor grammatical quality of some e-books. While many formats are available, the epub open standard format is widely used as it can be read by many popular applications, including Apple’s iBooks, Barnes and Noble’s Nook and the Sony Reader.  

Delivery of the digital content (including the text itself, any associated artwork and metadata) by the publisher to the retailer. In simple terms, this would essentially result in the digital file being stored somewhere on the retailer’s IT network.

Creation of copies for download by ultimate consumers. For all practical purposes, therefore, the digital file that is ultimately downloaded by the end customer is created only at the point of purchase. Many e-book retailers will enable customers to access their e-book purchase from multiple devices and to download additional copies at no cost where the ‘original’ has been lost or otherwise corrupted. As a consequence, the product purchased by the end consumer is a form of intangible licence that comes into effect only at the point of sale and only for the purposes of that particular sale.

This ties in closely with the copyright analysis of the transaction. The copyright in the book belongs – or is exclusively licensed – to the publisher. A physical book embodies two divorced rights; the copyright in the work and the title to the chattel on which the copyright work is printed. When a retailer receives an e-book from a publisher, the bookseller gets a licence to keep the electronic file and to copy it over to consumers. The consumers get a licence from the bookseller because the publisher grants the bookseller permission to grant a licence to the consumer under the publisher’s title. The bookseller is a classic agent, being a person who can change the legal relationship between the principal and the third party. This is the case even if the existence of the principal is not disclosed.

No physical product

The absence of any physical product goes to the heart of the debate over which distribution model is most appropriate. Given the lack of any physical product, there is no title or risk to transfer from publisher to retailer. Furthermore, any costs of storage are de minimis for each title, amounting to no more than the cost of storing a few megabytes of data. The cost is also fixed at this low absolute amount per title, irrespective of how many sales the retailer makes of that particular e-book.

In many ways, an agency model therefore more accurately reflects the reality of the commercial agreement – the retailer’s obligation to make any payment to the publisher arises only at the point of sale. Not only does no physical title vest in the agent, but there is no financial or commercial risk – the agent merely passes the licence fee to the publisher, minus a commission for their services.  

The OFT’s position

All of this places the OFT in something of a dilemma. Wearing its consumer champion hat, the OFT’s announcement of the investigation made reference to a ‘significant number of complaints’. There is no doubt that adoption of the agency model has resulted in some perplexing outcomes for consumers, with e-book pricing even exceeding the price at which retailers such as Amazon and the supermarkets retail hardback editions of popular titles. Yet, from a publisher’s (and also potentially an author’s) perspective, such an outcome may be not only defensible, but desirable. Publishing books is an expensive undertaking, requiring significant investment. While the cost of publishing an e-book may be low in absolute terms, in reality it cannot be divorced from the publishing costs associated with a physical title (there are at present almost no titles published solely in electronic form). Volume and discount book retailers are understandably more focused on a narrow spread of the most popular titles, and this is where historically the biggest retail discounts have been. Lawful regulation of retail prices through one channel can therefore be seen as one string to a strategic bow aimed at protecting continued investment in potentially less popular, but no less valuable, titles.

While the battle lines on both sides of the argument are well-entrenched, the OFT are in an invidious position. On the one hand, in the absence of any demonstrable collusion between the publishers in how these new arrangements were formulated and implemented (a question of fact for the investigation), the evidence supporting a conclusion that the arrangements constitute normal agency arrangements appears compelling. Furthermore, given the importance of the issues, it can in any event be assumed that publishers took specific and detailed legal advice on the potential implications of adopting an agency model.

What is clear, however, is that this investigation will potentially have as dramatic an effect on the online publishing world as the abolition of the Net Book Agreement did on previous generations of publishers, booksellers, and consumers.

Stephen Smith and Andrew Hobson are Partners at Reynolds Porter Chamberlain: