ANTONIO VILLARREAL / DATOS: JESÚS ESCUDERO / elconfidencial.com Las revistas científicas tienen el mejor modelo de negocio del mundo: los científicos pagan por publicar…
Joseph Esposito / scholarlykitchen.sspnet.org
The Journal of Electronic Publishing (JEP) has a very good issue out now on publishing economics, which I recommend to everyone. (The issue is labelled “Volume 19, No. 1: Economics of Publishing,” a useful atavism.) There are a number of good articles here, including the very well-written introduction by the editor, Maria Bonn. I looked at a couple of the articles with special interest because, well, I had an interest: the piece by Paul Courant and Terri Geitgey on free riders in monograph publishing draws in part on data developed by Karen Barch and myself on behalf of the Andrew W. Mellon Foundation (Karen and I expect to publish a report on that data on the Scholarly Kitchen by the end of the year), and I was an early reader of Nancy Maron’s piece. Taken as a collection, though, all the pieces here make an assumption about what publishing is, one that is not aligned with what publishers actually do.
Fuente original: via How Much Does Publishing Cost? | The Scholarly Kitchen.
The best expression of this assumption appears in the excellent piece by the aptly named Scott Smart and his colleagues. Here is the passage that set me a-pondering:
The charge of our investigation was not to develop a “profit-and-loss” statement for the press; in other words, revenue and other revenue-related considerations (such as royalties, which are a function of revenue) were identified, considered, and ultimately removed from the final analysis, keeping in mind that our goal was simply to determine, on a per-monograph basis, the “true” cost of publication.
The scare quotes around the word true serve to make the point that accounting is not an unambiguous science but is ultimately open to interpretation. In Smart’s formulation, the cost of bringing authors to the list don’t count as publishing. Smart reaches this conclusion for a very good reason, namely, that royalties are a variable cost (the more copies of the book you sell, the more royalties you have to pay): If you want to get to the theoretical “first-copy cost,” you don’t include all the costs, or the revenues, that come after the first copy is created. As I said, Smart’s analysis is very well done. The problem is that Smart (and Nancy Maron in her paper) are not catching publishing costs; they are capturing productioncosts. This is not just semantics; production and publishing are different things (production is a component of publishing in the same sense that a pound of flesh, say, is a component of a human being). I hasten to add that knowing the cost of production is a very important thing and between Smart and Maron the community is well served with the analyses on offer. But if you want to know what it costs to publish something, you have to look elsewhere.
Whenever someone talks about the cost of publishing, the conversation seems to me to take place in a vacuum. Step inside a publishing company and ask this question: Where is the greatest amount of energy expended? The answer is in finding the best authors. Publishing, in other words, is about the relentless pursuit of the best content for a particular program, and the best content comes from the small number of people who can create it. Publishing is brutally competitive, but the competition takes place not so much downstream in sales and marketing but upstream, where editors step into the ring every day, determined to knock out their rivals.
There are a number of ways editors attract the best authors. Most obviously (and particularly true for trade publishers) is the payment of an advance and royalties. Even scholarly publishers, including small university presses, will at times pay advances (some advances are fairly large), and just about all book publishers pay royalties when a book has sold a certain number of copies. But there are other ways to entice authors as well. University presses, for example, may trade on their reputation — the reputation of the parent university, the reputation of the press itself, or perhaps the reputation of a particular program at the press. Thus there is a university press that is regarded as the place to publish on the history of the American South, another for international studies, and two or three that battle for the mantle of best list in anthropology. As I write this paragraph I imagine that the staff of some presses may be reading with special attention: Is he talking about us? I won’t let the cat out of the bag, but I will say that if I were publishing in architecture, I know where I would send my manuscript; and the same goes for American history (several candidates here, but one sits above the others) and economics. Yes, press-people, I am indeed talking about you.
Presses also attract authors through their marketing presence (“We attend MLA every year”) and their networks of authors, advisors, and editors. Persistence helps, too: many a book has been acquired because of the tenacity of an editor pursuing an author over a long period of time. Sometimes the only way to get rid of the editor is to submit the manuscript. It is a source of amusement to me to observe how the highly-educated, bookish, ultra-polite editors at university presses sometimes talk about their rivals. There is as much focused energy here as there is on a football field or on Wall Street.
None of this is captured by Smart; his analysis gets at everything except what makes publishing publishing; and I want to repeat that his assumptions are implicitly or explicitly shared by all the articles in this issue of JEP.
So now we know how much it costs to produce a scholarly monograph, but how much does it cost to publish one?
Publishing is best viewed as having two complementary parts: frontlist and backlist. The frontlist is the books that you are publishing this year; the backlist is the list of books that have been around for a year or more. Some publishers qualify this a bit (e.g., the frontlist is said to cover two years, and the backlist may be divided into recent backlist and “deep” backlist), but the principle is the same. Every book on the backlist was once a book on the frontlist and, with rare exceptions, every book on the frontlist has a future on the backlist. Thus a snapshot of a press’s program at any point in time shows us the life cycle of an average title. Smart’s analysis does not capture life cycle costs, though it does capture much of the frontlist cost. The backlist cost is entirely missing. This is important because the backlist almost always operates at a profit; and that’s true for the smallest presses as well as the largest. By not including the backlist in the analysis, the alleged cost of publishing to the parent institutions is vastly overstated.
But to get at the cost of the backlist (as well as some of the frontlist costs), you have to include the costs incurred to generate revenue. Those costs are largely in sales and marketing, but also include the cost of printing and binding and online hosting. And of course you have to include whatever you pay to authors and advisors. In fact, you have to include all the costs of the press.
This is a complex calculation. In order to accomplish this, you need some sophisticated tools: the back of an envelope, a pencil, and five minutes. With these tools in hand, you now look at the total costs incurred by a publisher. Let’s say that a press has costs of $11 million a year (and revenue of $10 million, for an operating loss of $1 million a year). This press publishes 200 books a year. You then divide $11 million by 200 and find that the cost of publishing at this press is $55,000 per title. That may seem like an enormous amount of money, but bear in mind that the press earned back $10 million of the total cost in the marketplace — and you are assessing life cycle costs. Another way to do this calculation is to take the operating loss of $1 million and divide it by 200 new books per year. This leaves us with $5,000, the cost to the press’s parent institution per title. Smart’s and Maron’s figures are much higher than this because they don’t include the offsets to the institution’s costs from money earned in the marketplace.
Let’s not stop there. Imagine a larger press, with costs and revenues both running at $20,000,000 a year. The number of new titles is 300, giving us a cost per title of $67,000. But the cost to the parent institution is zero because this press operates at breakeven. Or we might have a smaller press, with $5 million in costs and $4 million in revenue. This press publishes 150 new books per year. This press has a cost of $33,000 per title. The cost to the parent institution is $6,700 per title per year.
You can do calculations like this all day long and what you will find is that the numbers are all over the place. And they should be: the cost of publishing varies with many things, including the overall costs and revenue of the press and the number of books published each year. The costs to the parent institutions vary as well: the better the sales and marketing of a press, and the larger the backlist, the lower the cost of publishing to the parent. And in many cases the cost to the parent is actually a negative number because the press is profitable.
What Smart is doing is what I call the “Shylock analysis.” He is looking for a pound of flesh, but the real cost includes the blood. The cost of producing the first copy of a book cannot be analyzed outside the context of the press’s overall operations. At least it can’t if you are talking about what it takes to publishsomething.
Production, on the other hand, is a different matter. It can be done for much less than Smart says: all you have to do is give up the cost of attracting the best authors. That would allow you to scale down your editorial effort, as you don’t need to invest much in editorial if all you are doing is operating the digital equivalent of the campus print shop. We see this in the world of Gold Open Access, where PLOS ONEchanged everybody’s understanding of costs by reducing the scope of editorial review.
In my view, we don’t have to decide which method is better: viewing publishing as simply a production process or viewing it as a matter of careful selection, whose success is measured by marketplace demand. Over time these two processes will either eliminate the other or each find its own place. But let’s not confuse one with the other. That would be neither true nor smart.
I am a management consultant working primarily in the world of digital media, software, and publishing. My clients include both for-profits and not-for-profits. A good deal of my activity concerns research publishing, especially when the matter at issue has to do with the migration to digital services from a print background. Prior to setting up my consulting business, I served as CEO of three companies (Encyclopaedia Britannica, Tribal Voice, and SRI Consulting), all of which I led to successful exits. Typically I work on strategy issues, advising CEOs and Boards of Directors on direction; I also have managed a number of sticky turnarounds. Among other things, I have been the recipient of grants from the Mellon, MacArthur, and Hewlett Foundations, all concerning research into new aspects of publishing.