Funding Open Access
June 27, 2006
The recent announcement by PLoS that they're increasing some of their fees has provided plenty of fodder for those on all sides of the issue. On liblicense-l, Peter Banks uses the occasion as proof that the PLoS business model is unsustainable. Jan Velterop, commenting on the Nature article, comes to quite a different conclusion, pointing out that it typically takes seven years for a new journal to break even anyway, so that the adjustments at PLoS are just part of the normal evolution of things. It's good to have some additional facts about the money, but at this point facts are obviously not going to change many minds.
Librarians are trying to figure out what their role is in this. While I've heard some urging that libraries provide more support to PLoS or BMC or to other OA initiatives, it has been somewhat depressing to hear more voices complaining that the memberships "are just another subscription fee." This stems from what I have always considered the misguided notion that OA was going to provide the solution to librarians' funding woes. I've never seen any analysis to show why this should be the case, but as I have said repeatedly, I do not consider this a reason not to support OA initiatives.
Many librarians seem happy to get on the OA bandwagon, as long as somebody else is going to pay. Take it from funding agencies, or take it from the research budget, or take it from a society's membership fees -- just don't take it from my budget! OA is a great and wonderful thing -- as long as somebody else is paying for it.
Frankly, I don't understand this. It's not my money, it's not the library's money -- it's the institution's money. My university gives me authority to spend a considerable amount of cash in order to advance its work. My particular specialty is supposed to be in helping the institution manage scholarly information -- to get access to what they need, to use it as efficiently as possible, to provide support for collaboration and the advancement of knowledge. I hire people, develop programs, manage public computer access, participate in association activities, buy some books, pay subscription & license fees for access to information. If I consider the work of PLoS to be important -- essential, even -- then why is it less reasonable for me to contribute $10,000 to its operations that it is for me to send $10,000 to one of the commercial publishers for access to a single title? The lack of enthusiasm for this notion from some librarians leads me to the suspicion that their interest in OA is really at root the hope that they'll be able to quit funding scholarly communication themselves, and that all the talk about the greater good is a bit of a smokescreen.
The argument is made from time to time that if the funding for scholarly communication shifts from the readership side (subscriptions) to the production side (author's fees, or funder's fees, or however you want to characterize it), the larger research institutions are going to bear a greater burden than they do under the current system. Peter Suber, whose work I quite admire, attempts to skewer this argument by pointing out that most of the OA journals in DOAJ don't charge any publication fees whatsoever. The implication is that if authors published in those journals, we could have OA without any increase in overall costs.
I suppose this might be so, but those making the argument are using a narrower frame, and looking just at the question of what would happen if we replaced subscription revenue with publication side revenue. The assumption here (contrary to the assumption underlying Peter's rebuttal) is that the overall cost of publishing doesn't change. If that's the case, then obviously some institutions will pay less and some will pay more. And those paying more will be those producing most of the articles.
But I don't see a problem here (theoretically -- the political problems are huge, of course). Why shouldn't the big research institutions pay the additional costs involved in disseminating high-quality, peer-reviewed, copy-edited and value-added research literature? We won't want to, of course. (Nobody actuallys wants to pay for open access). It doesn't seem like any great leap from the current situation in which research universities provide the bulk of the subscription revenue -- the purpose is still, ultimately, to fund quality publishing.
It may be, as Peter's rebuttal seems to suggest, that in an open access world we could actually do quality publishing for much less than the current system requires. The latest numbers from PLoS argue the other side, however. Their new fees are much closer to the figures that the commercial publishers were suggesting were necessary back when BMC was charging $550 per article and the OA partisans were bashing the publishers for coming up with unnecessarily inflated numbers for what publication really costs. The PLoS numbers suggest that those figures may have been on the mark after all.
If that's the case, and it turns out that OA publishing (no matter how it is funded) is not really going to be significantly cheaper overall than the subscription-based system, librarians are going to have to fairly ask the question -- if you want Open Access, are you willing to pay?
The question I still have is whether OA removes obscene publisher's-shareholder profits from the picture. If so, then there's a good argument to be made that OA will inevitably reduce publishing costs; a big sucking vacuum on the system goes away.
My opinion is that OA will reduce publishing costs by making scholars and researchers aware of them for the first time. They don't listen to librarians about what things cost; now they'll have to see it. I look forward to the first faculty member in my office sputtering "Why does [Elseviley Verlag Journal X] cost three times what [PLoS Journal Y] does to publish in?"
"Dunno." Adding with an evil smile, "Why don't you ask them that?"
Posted by: Dorothea | June 27, 2006 at 08:41 AM
This discussion is long overdue.
What is the business model?
Who funds it? From which budgets?
Who wins? Who loses? There always winners and losers in an economic shift. It is inevitable.
Peter’s rebuttal is interesting; if only costs could magically disappear. Unfortunately, they have a habit of reappearing at the most inconvenient times.
Peter, sorry, it is an undeniable fact. If the cost of publishing is transferred to the producers (authors at research institutions) from the users (readers – libraries in general), it is almost statistically impossible that the cost for the author’s research institutions would not go up. There are a few hundred research institutions. There are many thousands of libraries. If the cost of publishing (whatever the model) is shifted to the few from the many (even if there are considerable savings) then the few will pay more.
I have said this many times before. OA is merely a different business model. It isn’t necessarily a mechanism to reduce price (not costs – there is a difference). I find that librarians continually confuse price and profits. Profits, such as Elsevier’s are a function of two separate economic data; 1) the revenues (price) they generate from sales; and 2) the cost of producing their product. The two streams are mutually independent. They have no relationship to each other. Prices are set by the general market forces of supply and demand. Costs are a function of the efficiency of an individual organization.
It could very well be that Elsevier enjoys high profit margins because it is more efficient then its competitors. If the OA movement were to force Elsevier to close its doors it is possible that the cost (price) of publishing could go up – not down. Why? It could very well be that PLoS and friends are less efficient then Elsevier. In economics you can make less profits and still charge more.
Dorothea, you are half right in your comments. You can remove ‘obscene profits’ and still see an increase in the cost of publishing. But you are dead on correct in your second point. The problem with pricing in this business has always rested in the academic system itself. Why is scholarly publishing so expensive? Because, like healthcare, there is a third party paying. Ultimately, the end user pays nothing. How does this effect price? Third party payment pushes the demand curve to the right. Meaning, there is more demand for the product because the people using the service (the community the library serves) pays nothing for its use. This is true for both the traditional publishing business model and for OA. Neither business model will have any impact on the demand for the service. Why? Because, in both models, the cost to the user is zero.
Yet, there is still hope Dorothea. OA could have a positive impact (reduce price – not costs) in that if it increases the supply of content. Market price is a function of supply and demand. If the OA movement moves the supply curve to the right then the market price for scholarly publishing will fall (regardless of what impact it has on costs). But there is a danger here. I think there is already evidence building that the OA movement has moved the supply curve to the right. However, that is not entirely a good thing. I suspect that Elsevier is so profitable because it is very good at controlling costs. Unfortunately, many of Elsevier’s competitors are small societies. They are notoriously bad at controlling costs. It could be that the OA movement will force small societies out of publishing leaving you with the Kluwer and Elsevier conglomerates and a few OA publishers. In that event, it is anyone’s guess what will happen to price.
So that leaves you with the demand side. Dorothea, you have hit on a truth. If costs are shifted to the authors (who are also the users) it might make them painfully aware of the cost of scholarly publishing – which would result in a reduction of demand. Now I do admit this this outcome would fly in the face of the librarian communities stated desire to increase availability and usage of scholarly data – but then that is another discussion altogether.
Posted by: MarkD | June 27, 2006 at 05:42 PM
Peter Suber has some good responses to my comments on the Open Access News blog here: http://www.earlham.edu/~peters/fos/2006_06_25_fosblogarchive.html#115142966821975161
Posted by: T Scott | June 27, 2006 at 08:19 PM
Yes, I read Peter’s points. But Peter makes the same mistake most librarians make. The cost (to the publisher) of producing scholarly literature is irrelevant. This number, whatever it is (OA or traditional), has no relationship at all to the price you pay for scholarly literature. His citation of lower costs for OA publishing is therefore beside the point.
The market price is set by a combination of demand and supply. If OA drives down the cost of producing scholarly works there are two possible outcomes in economics; 1) more cost efficient producers will lower prices thus forcing down the general price; or 2) all publishers traditional and OA become more profitable.
It is not a given that lower production costs mean lower prices for the users of this content. If indeed OA lowers the cost (and I have seen no actual data to prove or disprove this thesis) price will only fall if the lower costs lead to a lower barrier of entry in scholarly publishing. If the barrier to entry is lowered, more publishers will enter increasing the supply and thus lowering the price. Librarians imply this exact scenario when they talk about freeing content and making it available to all. However, there is another side to this coin. Librarians also talk about increasing demand – making it available to everyone implies more use. Increased demand implies higher prices not lower – whatever the production cost.
What do I believe? I think there is evidence building that the OA movement has had a positive impact (from the buyers point of view) on the price of scholarly publishing. But your wife would be better at answering this question because EBSCO tracks industry price increases each year. So tell us Lynn, has the rate of inflation in scholarly publishing moderated over the past few years? If the answer is yes, then you may be seeing the impact of OA on the price of scholarly material.
Then again, other factors could be at play as well. Unlike the laboratory. It is almost impossible to do controlled trials in the marketplace. But there are econometric models that can predict (with some accuracy) the impact of a single event on an industry. Does the University of Alabama have an economics department? Maybe someone there can investigate this issue for you. It would be an interesting paper.
Posted by: MarkD | June 27, 2006 at 08:50 PM
In his posting above, MarkD argues classic economic theory. This works fine for widgets, but not for scholarly publishing. I would first state that cost *does* have a relationship to the price we pay for journals. If a journal costs X to produce, a publisher will charge X+Y, where Y is the profit. Y cannot stand alone, it must be considered along with X. If a publisher wants to realize a $2 million profit this year, then Y=$2 million. Subscribers will pay X+Y. If costs are low, than X+Y is lower than if costs are high. There is an inescapable relationship.
Second, classic supply and demand theory does not work well with journal publishing. Every journal is a monopoly. "Competition" is an entirely different animal in publishing, with little similarity to the industrial world. Neither libraries nor scholars can substitute a different journal for the journal desired merely because "it is just as good, but cheaper." If classic economics applied here, publishers would be lowering prices in order to compete. They haven't. Finally, the concept of "demand" when talking about moving electrons over a wire can hardly be compared to classic demand with production limits. Unlike a diamond, which can be owned, kept, and not shared, information can easily be shared and distributed at an amazingly low price. Increased use, or demand, of information does not imply higher costs.
Posted by: The Other Mark | June 28, 2006 at 09:30 AM
Mark (the other). We aren’t that far off in our respective observations.
Scott, the worst thing that could happen to your blog would be a protracted debate about the dismal science – so I will try to keep this brief and swear on a stack of library books to make this my last entry on this subject.
You are right Mark. The x (cost) + y (profit) = p (price) equation is central at the micro level (individual company). But at the macro level (the economy as a whole or an industry as a whole) the x (cost) of an individual company has no bearing on the market price. In actual fact, your equation is correct to a point. A more accurate presentation of your equation would work as follows: it is p that is given and the equation should read (at the macro level) p (price) - x (cost) = y (profit). I know this for a fact. I am a small businessman. I have have no power to affect price. My profits are a function of the price I can get minus my costs. If I set my price above market rates my customers will go elsewhere. They would not care that my costs are higher then my competitors.
At this very moment throughout the publishing industry you will find that there are many equations for many different publishers.
In my equation p-x=y you would find many variations throughout the industry. One publisher might work under this equation:
P-(x*1.4) = y another may be working off of this equation p-(x*0.5) = y. (where ‘x’ equals average cost) My fear is that Elsevier would work under the second equation while society publishers would work with the first.
Where we disagree is your statement about supply and demand not working well in publishing. On the contrary, publishing cannot escape the laws of supply and demand. I do not disagree with your observations about the industry however. You are dead on about monopolistic power. But publishers do not have the power to negate the forces of the market. You have merely described a market situation where the demand curve is inelastic.
I also don’t buy your case about electrons. They are in fact no different then widgets in the market sense. Every widget, every electron has a price. They can indeed be owned. They are owned. If money can be made from them, I assure you, people will find a way to make that money.
You are in fact making the case that electronic media may shift the supply curve by virtue of the fact that it is much easier to create an electronic journal. You may indeed be right on that point. I suspect that is true. The evidence of this would be in the numbers. Has scholarly publishing inflation moderated over the past few years? If it has, you may be seeing the first impact of the electronic medium.
Now to get back to my original point. We started this discussion with Scott talking about the conflicting goals of OA. Is it about more access or is it about lowering price? My point was that it is not a given that OA will lower price. OA merely represents a change in the business model from user pays to author pays. Let us take the case of the New England Journal of Medicine – for example. Let us say that NEJM announced today that they were changing their business model from user pay to author pay. What impact would this have on the NEJM’s ability to set a high price? Mark you are right the NEJM does enjoy a monopolistic opportunity. It cannot be replaced by another journal. The community is fortunate in that it is owned by a society. The MMS does not use the full power of its monopoly position in its pricing. You can be sure if it were owned by a commercial publisher it would be far more expensive. That is true whether the model used is subscriber pay or author pay.
Ultimately, that is my point. With an OA model the NEJM monopoly power is not reduced. If you are a researcher you would give quite a bit of money to get published in the NEJM. Its reputation and quality would allow it to charge an author fee far higher then any other journal in publishing. In short, the demand curve for the NEJM is inelastic in both models. If the average market price for author fees were P the NEJM would still have the power to charge P *2 or 3 0r 4 or even 5, and it would still turn articles away. The NEJM’s cost of reviewing and editing those articles is irrelevant in setting this price. In fact, I suspect that its costs are much higher then they would be for for-profit publishers.
So back to Scott’s original entry. Is OA primarily a cost cutting measure or is it something else? The answer, it all depends on how the movement impacts the supply and demand for scholarly work. It always goes back to that simple relationship. Always. The electronic age does not change the basic laws of economics.
Sorry Scott, I failed to keep it brief.
Posted by: MarkD | June 28, 2006 at 06:09 PM
Replying to Mark (I think)
Librarians do not want to increase demand in the sense that
they desire a greater number of scientific journals (or articles).
We want to increase demand by increasing the ability of readers to read the published materials, and OA is the way to do it. --David G
Posted by: David G | June 28, 2006 at 07:34 PM
Being managers of the "institution's money ... not the library's money", as Scott argues, reminds us as librarians that we should remain ideologically neutral when it comes to publishing models. We should be leery of medieval alchemy that professes to change an expensive publishing system into a cheap one, and leave behind the romantic notion of the Philosopher's Stone to the Philosopher's Blog.
Posted by: Phil Davis (international man of leisure) | June 29, 2006 at 11:09 AM
As a student I find it very frustrating that my university cannot afford to subscribe to all the journals, as often a search brings up articles which are highly relevant to my studies, but only the abstract is available to be read.
Surely the point is that the results of research should be published and freely available. So called peer review is only a mechanism used by publishers to cut back on the number of articles published.
The fact that NEJM charges authors to publish automatically means that good research done in a poor institution or country will not be published.
By charging a relatively large fee for each article some publishers are using publicly funded research as a cash cow.
I don't mind paying a little bit, say 10% of the cost of photocopying in the library, but $10 for an article is quite expensive, since working from home I already pay the cost of internet access and printer consumables.
In my view open access charges paid by libraries will occur in either one of two ways. They will pay a fee to someone else to host web pages, or they will become the publisher for their institution with reciprocal rights to other open access providers.
Posted by: Denise Tzumli | July 20, 2006 at 01:32 AM
... It's good to have some additional facts about the money, but at this point facts are obviously not going to change many minds.
Posted by: sohbet | October 12, 2007 at 06:31 AM
agla kalbim aglamazsan sekerim :)
Posted by: sohbet chat | October 12, 2007 at 06:34 AM
This is a very nice posting, thank you very much. Cheers.
Posted by: Vladlena Nyzhnik | April 29, 2008 at 11:43 AM
Replying to Mark (I think)
Librarians do not want to increase demand in the sense that
they desire a greater number of scientific journals (or articles).
We want to increase demand by increasing the ability of readers to read the published materials, and OA is the way to do it. --David G
Posted by: aşk şiirleri | March 29, 2009 at 12:41 AM
It's good to have some additional facts about the money, but at this point facts are obviously not going to change many minds.
Posted by: sohbet | April 05, 2009 at 05:28 AM