Editor’s Note: This post is the text of a presentation Joe Esposito delivered April 14, 2011, at the Allen Press Emerging Trends Conference in Washington, DC.
A couple years ago, I got myself into big trouble when I wrote a blog post about the emerging phenomenon of “library bypass.” Library bypass is a publishing strategy in which a publisher that has traditionally sold most or all of its products to libraries begins to find ways to sell things directly to individuals, some of whom may have been library patrons. This is a form of disintermediation: the library sits in the middle between the publisher and the reader or end-user, and the publisher tries to go around the library. The point of that post was that publishers were developing library bypass strategies because libraries were telling them that they were out of funds; they could not buy what the publishers wanted to sell.
One document that was widely circulated from a library purchasing consortium advised publishers to lower their prices and not to introduce any new products.
Think about that for a minute: No new products.
New research in universities would continue, but there would be no new publishing products to accommodate that research. No one should be surprised that a publisher would read that consortium’s statement and call a meeting to discuss solutions to this. Library bypass is a rational response to tightened economic circumstances.
After I wrote that, the feedback pulled in two different directions. On one hand, I was accused of hating libraries and worse; publishers were accused of greed and a complete abdication of their civic responsibilities. On the other hand, publishers contacted me to discuss ways to sell books and journals directly to consumers. In some instances the discussion became emotional. The fact is, though, I was not advocating a line of action; I was simply describing what was already taking place. Publishers continue to seek new ways of reaching customers whether I write about it or not. This is a rational response to a challenging situation: if you have something to sell, you look for someone to sell it to. Publishing has this in common with politics, that if you want to figure out what is going on, the best practice is to follow the money.
I should not have been surprised to stumble upon resistance to library bypass strategies. Not too many years before, a similar agitation arose when a prominent library association called for the creation of institutional repositories. These repositories were to store all the intellectual output of a university and make it available under an open access formula. It was proposed that over time these repositories would take the place of publishers. Indeed, with the World Wide Web and the growing effectiveness of search engines, publishers would have no role to play — an author would simply upload a paper to a repository and Google’s search engine would do the rest.
You may recall the outraged response to this among advocates of the traditional publishing model. The combination of institutional repositories and open access would destroy the economics of publishing; the absence of a well-articulated business model would mean that these services would not be sustainable; the elimination of the publisher from the process would undermine peer review; and the collapse of peer review would reach to the very foundations of the university itself. In this critique, traditional publishers set themselves up as the Keepers of Civilization, and woe to anyone who chose to reconsider their priestly role.
It’s worthwhile to take a step back and take a look at what is actually happening in scholarly publishing today. Despite the clamor about library bypass, libraries continue to constitute about 85% of the revenue for academic journals. Obviously, that figure varies with particular publications, but few journal publishers have stopped thinking of libraries as their primary customers. In the academic book world, that percentage is lower, perhaps around 25%; but I have never met any publisher who was willing to turn away from 25% of its revenue. On the other side of the argument, we now see that almost every university has set up an institutional repository, but sales of publications continue to creep up year by year, allowing for a hiccup during the global financial crisis. If libraries and publishers are being disintermediated, it is not on a revolutionary basis. What we are seeing is a publishing system that is mature and not experiencing the robust growth it once did, but there is a big difference between middle-aged maturity and death.
Disintermediation may be the new area for growth, but it is starting from a very small base.
For many people, it’s surprising that disintermediation is not more widespread. The core of this argument is that the Internet changes everything, as indeed it does. Once upon a time the difficulties of physical distribution of printed matter mandated that there be large organizations called publishers that saw to it that material found its way into the marketplace, and another kind of large organizations called libraries that stored this material and made it available to readers. This circumstance reminds me of A. J. Liebling‘s famous remark that freedom of the press is guaranteed only to those who own one. Now, on the other hand, a high school kid can set up a Web server and go eyeball-to-eyeball with Rupert Murdoch and Arthur Sulzberger. You don’t need a press to become a pundit; you don’t need a big warehouse to store your books. There is no inherent reason why an entomologist working in Brazil cannot publish directly on the Internet and hope to reach an interested reader in Tallahassee or Tokyo; it’s only a matter of keystrokes. Once the material is online, people can find that material through search engines, social media, or perhaps the recommendations of colleagues. In this scenario, there is no role for the traditional publisher or the traditional librarian.
What’s taking so long? Why hasn’t disintermediation already happened?
I am going to get to some practical instances of disintermediation in a minute, but first I want to spend a moment on why the habit of traditional publishing is so hard to shake off.
An economic value chain begins at the beginning and ends at the end. The first link of the chain — or the “upstream” component, in the mixed metaphor of business jargon — is the author, who creates the work. Note that I don’t say that the first link is with the researcher. Research and the publication of research are different things; it is the publishing value chain under discussion here. The author creates value by originating an idea and marshalling information to support that idea. That idea is then shaped into more or less readable prose, with various tools such as citations to help support those ideas. The author then passes his or her contribution to the next link of the chain — or “downstream.” Depending on the individual author and the kind of publication, that downstream link could be an editor, a colleague who serves as an editor, or perhaps, especially for book authors, a literary agent. As the material continues to flow downstream, it arrives at a publishing house, which adds value through the process of editorial selection, editorial refinement and production, manufacturing, sales, and marketing. Nowadays some publishers also get involved with adding value through the participation in preservation services such as LOCKSS and Portico. But a publisher’s biggest job is to help create a market for a book or journal.
Publishers look downstream, too; depending on the kind of publication in question, the next link could be a subscription agent, a wholesaler, or in some instances, a library. Each of these downstream links adds more value to the process. I am particularly fascinated by subscription agents. To an outsider, it is hard to explain the role of a subscription agent. Why can’t a publisher handle its own subscriptions by itself? Of course a publisher can handle its own subscriptions, and in most cases, publishers would probably want to. The value of the subscription agent lies with libraries, who don’t want the administrative burden of having to deal directly with thousands of publishers. The same thing is true for book wholesalers, who add considerable value to libraries by making it possible to acquire books from a single source. Each link in the value chain adds something to the process. This is why we call it a value chain, because each link adds value.
When we get to libraries, the added value is obvious. Libraries are selective; they help guide readers to materials of higher quality. Libraries have purchasing power, which saves money for readers. Libraries provide a suite of tools for organizing publications and helping readers find what they are looking for. Libraries provide so much value that most people want them to be bigger. The call by Robert Darnton to build a national digital library is neither original nor surprising. Anyone who uses a library wants that library to include everything and at no direct cost to the user. I can’t think of many institutions that most people would like to be bigger and cover a wider range. Do you want your phone company to be bigger? How about your health insurance company? If you send your kids to a small school, you might wish that it were bigger so that it could offer a wider range of courses and activities; but at some point the big school becomes too big, too impersonal and bureaucratic. For many institutions, scale does not scale. Libraries are different from that. With libraries it is taken for granted that bigger is better. The only reservation most people have about libraries is that they don’t want to pay for them.
When we stop and look closely at each link in the publishing value chain, we begin to understand why disintermediation is so hard to bring about. Each link adds value; to disintermediate that link potentially destroys that value. Sometimes the value that gets destroyed is obvious. For example, direct publishing by an author of an article bypasses the publisher, but at the cost of editorial selection and refinement. Of course, some authors resent publishers’ editorial work, but not all readers agree. Less obvious is that a publisher has a network of relationships with wholesalers, libraries, review media, and so on, all of which serve to bring a publication to readers’ attention. Disintermediate the publisher and you lose more than access to a printing press and a warehouse. You also lose a variety of marketing relationships. Ironically, an open access publication could come to the attention of fewer people than a toll-access publication because the traditional forms of publishing are embedded in a value chain that looks to publishers to assert the merits of their publications.
Not all links in the chain are made of super-strong titanium, however. Let’s go back to the subscription agents. Agents add value through the aggregation of materials purchased by libraries, thereby reducing libraries’ administrative costs. But what happens when the publications are open access, as many of them now are? What is the point of aggregating toll-access publications if the publications don’t carry a toll? An open access publishing process would clearly disintermediate subscription agents. Disintermediation is possible when the value of a particular link is no longer required. The important issue is in identifying what value we want to retain and what new value we want to create.
Open access publishing provides a fascinating example of how disintermediation can take place, but not always in the way you would expect. At the outset of the open access movement, the general plan was to make research materials available to readers at no cost to those readers. Institutional repositories were set up to provide a home for these materials. Traditional publishers were going to be disintermediated. Libraries were going to take on a publishing role.
What is evolving with open access is very different from this, however. While most universities do indeed have open access repositories, by and large the repositories have not been as successful as their supporters had hoped. This in turn has given rise to mandates for researchers to deposit their articles into repositories. This is a curious development, is it not? No one has to be mandated to buy an iPhone or an iPad; no one has to be mandated to create a page on Facebook; but a mandate is deemed to be necessary to get researchers to do what open access advocates say is in the researchers’ own interests.
Open access services are thriving, however; they just aren’t thriving in libraries. BioMed Central, a commercial venture, created an author-pays model for open access that has now been widely imitated. BMC, of course, was subsequently acquired by Springer, among the largest publishers of scientific research. The Public Library of Science has established a highly regarded open access service, and they have done it entirely outside of libraries. Now PLoS has attracted many imitators: Wiley Blackwell, BMJ, and, in the social sciences, SAGE. Most intriguing is an open access service from AIP, which seems likely to create competition for the library-sponsored physics arXiv at Cornell. What we have seen with open access publishing is that publishers, rather than being disintermediated, are learning how to coopt it. With open access publishing, libraries have succeeded in disintermediating themselves.
In the book world, there are signs that publishers are indeed being disintermediated; the question is how exceptional are these instances of disintermediation. An established mystery writer named Joe Konrath decided to move his books over to Amazon’s self-publishing service because of the promise of earning higher royalties. I doubt that there is a trade publisher in the world who has not been following Konrath’s career closely, praying that he will fail. Even more fascinating is the case of a young woman named Amanda Hocking, who came to self-publishing with no prior publishing experience. Her young adult novels earned her a small fortune, attracting the interest of major commercial publishers, one of which has now signed up Hocking to a million-dollar contract. One emerging pattern seems to be that publishers are initially threatened with disintermediation, whether through open access or self-publishing services, and then find a way to reinsert themselves into the value chain. Having a big checkbook helps.
Rather than think of disintermediation as the collapse or shrinking of the value chain, we should probably think of it as a new way to re-create the value of the value chain. Hocking, for example, understood that traditional publishers create demand for books through their marketing arrangements; she then set out to create her own marketing network through an aggressive use of social media, which just happens to be the favorite form of communication among her young readership. In the open access world, publishers know that authors enjoy the value of peer review; thus rather than set up repositories that provide no rigorous editorial filters, the new open access services are determined to retain strict editorial procedures. We can disintermediate certain links in the chain, but it is very hard to eliminate the value embodied in those links.
Too often talk of disintermediation lends itself to a version of technological determinism. Because the Internet makes it possible for an author to have a direct connection to a reader, therefore, it is assumed, authors inevitably will connect directly to readers. This ignores the role of human agency. People have to think about things, assess options, and make decisions. When SAGE recently announced an open access repository for the social sciences, it was not because the Internet sent a text message to the CEO. SAGE studied PLoS very carefully, sifted the information, explored other programs that it might want to initiate, developed multiple scenarios of how the service might play out, and took into account how the service would impact SAGE’s other business activity. The SAGE Open Access service was very much a deliberative decision by the company’s management team. SAGE is intent upon having the Internet work for it, not the other way around.
What this implies is that disintermediation is not an abstract force “out there” but the outcome of a series of executive decisions. Those decisions are not made without a context; they are clear and tangible and they reflect the interests of those making those decisions. Disintermediation, in other words, is an outcome; but the input is management strategy. Rather than talking about disintermediation, we really should be talking about the things that affect management decision-making and strategy and to take control of them.
I want to digress for moment with an anecdote about an experience I had while working with Encyclopaedia Britannica in the 1990s. I was brought to Britannica in 1992 from its dictionary subsidiary, Merriam-Webster, to develop an electronic encyclopedia. This seemed like an incredibly cool thing to do and I was eager to get started. What I had not realized was how many obstacles I would find along the way, none of them technological.
At that time the encyclopedia was sold in 32 print volumes, mostly by a direct sales force, the encyclopedia salesmen of legend. There were about 2,000 of these sales representatives in the U.S., about a thousand overseas. Together at the company’s peak in print, they sold about 175,000 sets of the encyclopedia a year at an average price of $1,600. It was a big business, and it became big because of sophisticated sales and marketing. Britannica was indeed a sales and marketing company with a brand name that smacked of the academy. A stranger animal I have never seen in my entire career.
The question was, “Would an electronic encyclopedia be sold by this sales force or through some other means?” This is not a trivial question. Let me provide some analogies. For years auto insurance was sold through local agents, but then Geico came along with its telemarketing. We used to buy music in record stores; now we buy it online, if we buy it at all. In the 1970s and early 1980s, personal computers were primitive devices sold mostly to energetic hobbyists, but by the mid-1980s, PCs were primarily being purchased by corporate officials, who then made the machines available to employees. And the story of Amazon is too familiar to retell. Products are sold in various ways and sometimes those distribution methods move around, disrupting the value chain. Of course, everyone likes to be the disruptor, no one wants to be the disruptee.
At Britannica the historical success of the sales force meant that virtually the entire company was dominated by sales executives. The editorial staff had no voice in company decision-making. When the management spoke, it thus spoke for the interests of the sales force. This meant that an electronic encyclopedia should be sold in exactly the same way as a print encyclopedia. Try to picture this: a CD-ROM version of the encyclopedia, even an Internet version, would be sold door to door at a very stiff price. No matter that most of the sales force did not even use computers.
When I made the point at a Britannica management meeting that when the product went digital, the sales and marketing would have to go digital, too, I was given a lecture about our customers. “Our customers,” I was told, “have to be shown the value of the encyclopedia.” This meant making an appointment with a prospect and sitting down at the kitchen table with extensive sales materials describing the benefits of owning an encyclopedia, especially if you wanted your kids to do well in school. When I heard this, I said, We have to stop talking to our customers.
This is counterintutive, is it not? We are always being told to get close to our customers. But in fact it depends; it depends on what you are trying to do. If you have an established business that is continuing to grow, it is a very good idea to get close to your customers. You want to know what they need; you want to become their primary supplier. But suppose you have a business in a mature industry; suppose the number of new customers is not growing and even the amount your existing customers are spending is beginning to drop. In such a case, talking to your customers is about the worst thing you can do. You cannot afford to let your strategy be tied to the markets of the past. You need new markets, you have to find new customers.
I advise publishers and librarians alike to be wary of the temptation to think that their conversations with one another represent a means to determine a long-term strategy. These conversations are important, but they can at best solve today’s problems. Growth for publishers is going to come from outside libraries, and that is where the new conversation must take place.
Similarly, professional society publishers need to be thinking beyond their currect membership. Would a revamped publishing program entice more professionals to become members? What kinds of things are non-members looking at today and how can the society’s publishing program evolve to intersect with those needs in the future?
If I may be permitted another anecdote, not long ago a friend reported to me a conversation he had with a Board member of a professional society. My friend asked the Board member to describe the customers for the organization’s publications. The Board member replied: “We are our customers.”
What the Board member implied is widely held, that there is a circle for research and publication, with each member of this community serving both as author and as reader. This model does not take into account all the people working in the field who are not members of the professional society; the professionals in adjacent areas who may at time read the publications, but are not likely to write for them; the students who will directly or indirectly benefit from the publications; and the general public who become aware of new areas of research as the publications are mediated by consumer media — such as, for example, a scientific paper that gets reported on in the New York Times. Rather than think of the authors and readers of scholarly communications as working in a closed circle that is defined by their professional associations, it’s better to use a metaphor of a spiral, with researchers sitting at the center, and the publications they create spinning out to different categories of readers. When you talk to your customers, you have to talk to all of them, including the ones you don’t know about yet.
It would be silly to suggest that professional societies should attempt to disintermediate their own membership, but it is meaningful to ask how the work of a society, as expressed by its publications, fits into the larger context of scholarly communications and what value it brings to each segment of readers. This may prompt some societies to begin to rethink some of their marketing objectives in order to bring in a new class of readers; and this in turn may suggest reasons to create new products to serve those new audiences. On one hand, this could be called a form of disintermediation, but it seems to me to be more reasonable to think of it as the creation of a new value chain, with the stress being given to the new value that is created.