This is astonishing. Harvard is one of the best and one of the wealthiest universities in the world but last week its Faculty Advisory Council* announced that it can no longer afford to maintain its subscriptions to academic journals.
The announcement was made online by the Council as a message to the academic staff at the university. I have taken the liberty of quoting it in full below.
The message is notable since it bears out many of the factors — in terms of costs — that have been highlighted by the Elsevier boycott (though no particular publisher is mentioned in the communiqué).
Perhaps this announcement is simply part of a negotiating strategy (one would expect Harvard librarians to be a clever bunch). But, be that as it may, what is particularly striking is the advice offered to staff regarding what they can do to help the situation, which chimes with what I and many others have been arguing for as ways to bolster open access publishing (and was writing about only yesterday). To pick out just two key phrases (with my emphases in bold):
Consider submitting articles to open-access journals, or to ones that have reasonable, sustainable subscription costs; move prestige to open access.
If on the editorial board of a journal involved, determine if it can be published as open access material, or independently from publishers that practice pricing described above. If not, consider resigning.
This sends an important message (one is tempted to the hyperbole that it may even be a ‘shot heard round the world’): if one of the most prestigious and richest institutions in the world cannot afford its journal subscriptions, then there is a serious problem in academic publishing.
Well done to Harvard’s Faculty Advisory Council for coming up with concrete proposals for what academics should be doing to improve matters.
The announcement:
To: Faculty Members in all Schools, Faculties, and Units
From: The Faculty Advisory Council
Date: April 17, 2012
RE: Periodical SubscriptionsWe write to communicate an untenable situation facing the Harvard Library. Many large journal publishers have made the scholarly communication environment fiscally unsustainable and academically restrictive. This situation is exacerbated by efforts of certain publishers (called “providers”) to acquire, bundle, and increase the pricing on journals.
Harvard’s annual cost for journals from these providers now approaches $3.75M. In 2010, the comparable amount accounted for more than 20% of all periodical subscription costs and just under 10% of all collection costs for everything the Library acquires. Some journals cost as much as $40,000 per year, others in the tens of thousands. Prices for online content from two providers have increased by about 145% over the past six years, which far exceeds not only the consumer price index, but also the higher education and the library price indices. These journals therefore claim an ever-increasing share of our overall collection budget. Even though scholarly output continues to grow and publishing can be expensive, profit margins of 35% and more suggest that the prices we must pay do not solely result from an increasing supply of new articles.
The Library has never received anything close to full reimbursement for these expenditures from overhead collected by the University on grant and research funds.
The Faculty Advisory Council to the Library, representing university faculty in all schools and in consultation with the Harvard Library leadership, reached this conclusion: major periodical subscriptions, especially to electronic journals published by historically key providers, cannot be sustained: continuing these subscriptions on their current footing is financially untenable. Doing so would seriously erode collection efforts in many other areas, already compromised.
It is untenable for contracts with at least two major providers to continue on the basis identical with past agreements. Costs are now prohibitive. Moreover, some providers bundle many journals as one subscription, with major, high-use journals bundled in with journals consulted far less frequently. Since the Library now must change its subscriptions and since faculty and graduate students are chief users, please consider the following options open to faculty and students (F) and the Library (L), state other options you think viable, and communicate your views:
1. Make sure that all of your own papers are accessible by submitting them to DASH in accordance with the faculty-initiated open-access policies (F).
2. Consider submitting articles to open-access journals, or to ones that have reasonable, sustainable subscription costs; move prestige to open access (F).
3. If on the editorial board of a journal involved, determine if it can be published as open access material, or independently from publishers that practice pricing described above. If not, consider resigning (F).
4. Contact professional organizations to raise these issues (F).
5. Encourage professional associations to take control of scholarly literature in their field or shift the management of their e-journals to library-friendly organizations (F).
6. Encourage colleagues to consider and to discuss these or other options (F).
7. Sign contracts that unbundle subscriptions and concentrate on higher-use journals (L).
8. Move journals to a sustainable pay per use system, (L).
9. Insist on subscription contracts in which the terms can be made public (L).
*Update (23-4-12 23:35): The original version of this post may have given the impression that the announcement was made by Harvard University. I am grateful to @Medicalskeptic for clarifying that the memorandum comes from the Faculty Advisory Council. Nevertheless, it is worth noting that the council membership is drawn from senior academics, most at professorial level, from all the schools of the university and is likely to be influential.
That’s an interesting approach, but I can’t help but feel it’s tempered by the fact that Harvard appears to have absolutely no trouble at all raising enormous amounts of money through its alumni fundraising. Still, if they were to set a precedent this way, it might have a very interesting landslide effect.
I’m not sure how Harvard spends alumni funds but keep in mind that a few years ago Harvard established a policy that every student accepted to the university would be given scholarship funds to the extent that it was needed; meaning that no student would have to decline acceptance owing to financial consideration. Harvard has a huge endowment but money spent does have to be replaced. I suspect that at least some donations go into the scholarship fund, others go to capital improvements, and athletics. I doubt much alumni funding goes towards the purchase of journals, which probably come directly from operating expenses. The simply fact is journal costs, especially those published by for profit corporations, are outrageously costly. Harvard’s move is hopefully a first step in a more general academic protest over the cost of peer reviewed journals.
I agree there’s an element of posturing here. But what makes it interesting is the stance adopted and the particular recommendation to “move prestige to open access”.
Now, there are problems with this. As pointed out by a commenter over on the Disorder of Things blog, the shift to OA is easier to realise in life sciences than in other fields such as sociology where there are many fewer OA options. Moreover, considerations of prestige at the level of hiring committees also have to move away from basic reliance on impact factors — though I suspect that many hiring committees are more sophisticated that they are often given credit for.
It’s worth noting that option no. 1 is for Harvard academics to submit their papers to their green open access repository DASH. What’s good for Harvard academics should apply to those at other institutions too!
Harvard has actually been arguing that journal prices are too high for several years now, but this is the first time they have released the information in this way. See this video of Dr. Stuart Shieber from 2010. http://www.youtube.com/watch?v=xQBuXlW1gxc As of 2004, Harvard no longer tried to constantly increase their serials budget to maintain subscriptions. Shieber noted that “the technical name for that is ‘giving up’.” See his discussion 10-12 minutes in.
Thanks for the link to that video. Clearly Harvard has been thinking creatively about this problem for a long time.
It seems that Harvard is taking that first of the last steps necessary to convert itself from being an institution of learning and knowledge creation to museum where knowledge is an artifact and learning is an ancient tribal ritual long-abandon for more “civil” and enriching pursuits.
Not quite sure what you mean by that… can you re-phrase?
First, let me say up-front that I’m purely an observer of the central issues regarding research, the publication thereof, licensing and use, accessibility and researchers’ control of the content. I’m not an academic, I don’t even play one on TV.
I have, however, very familiar with Harvard and the academic/research grind – again familiar not expert. I grew up in W. Cambridge in the 1970s and many of my school chums went to Crimson or are now employed there (staff/faculty).
OK, caveats out of the way.
It seems to me that, in the case of Harvard, the budgetary “priorities” are skewed toward institutional brand defense rather than “mission” in this matter. My views on this are informed by many sources, but most recently by analysis of the entire “research publication” issue in general I found at boingboing (Cory Doctrow).
I’ll flesh-out this addendum later once I address today’s “To Dos”.
Thanks for asking, and prompting me to be more clear (with myself included).
Look forward to further enlightenment. I saw BoingBoing’s analysis — perhaps you can enlighten us as to who ‘Henry’ is…?
The letter from the Council bears reading. It makes no definite policy change, merely announces their finding that the current model for journal subscriptions is unsustainable. As a Harvard employee I found the argument made here far more convincing than recent explanations of our ongoing “transition” with it’s announced “involuntary staff reductions”.
Harvard can’t “afford” these subscriptions? Let’s see… what’s the word I’m looking for here? Oh yes: BULLSHIT.
Technically, probably true but that misses the point that the university has looked at the escalating costs of journal subscriptions and decided they no longer represent good value for money.
The segment of the video that Joe Kraus links to above is pretty informative on this score. Stuart Shieber presents an analysis that shows commercial publishers are much more expensive that non-profit publishers on a cost per page or a cost per citation basis. Hence the decision to take action and to promote — though not insist on — open access options to the faculty.
Look at it another way: every academic institution is being gouged by journal providers. The smaller institutions don’t have the clout that Harvard does. Harvard could quite likely watch and wait as less-wealthy institutions make decisions to limit access to journals, making Harvard ever more desirable. But by taking this stance, Harvard is supporting academia around world, as well as looking after its own finances.
Remember back when America was able to do stuff? Like send people up into space, or build roads and bridges, or subscribe to journals?
America still has a can do attitude and Harvard’s ‘can do open access’ is a good example of that… 😉
Can we please distinguish between different areas of scholarly publishing–among them journals in the sciences, mathematics, and economics and, say, the humanities? For the cost of a single subscription to some science journals, you can get access to thousands of journal volumes available through, for example, Project Muse.
I got the following commentary from my former postdoc advisor, Prof Jim Hogle, who is a senior member of the faculty and knows Harvard very well. It provides some excellent background and context to the post:
“The recommendation is not hot breaking news. Harvard has been pushing hard for public access to all scholarly work published by Harvard Faculty. Between the pressure from the University and the insistence that all work sponsored by NIH grants (including work published by postdocs or students who have been appointed to training grants of NIH fellowships at some point during their stay) be publicly available on PubMed Central (as documented by a PMID on any publication) the science faculty are making work available with a max 6 month embargo. Harvard has also established its own system DASH as an alternative. What’s new here is the committee pointing out the unsustainability of the current model of subscribing to journals. This committee has come under fire of late from labor groups because of its role in a recent University report recommending reorganization of the Harvard Library system, in an attempt to keep the system functional in the face of sky-rocketing costs. What I like about the link you sent is that it provides some options that both increase public accessibility and create ways to saving money that don’t involve decreasing library services or laying-off experience librarians (not so much an issue for us science types, but a big hit in the humanities and social science where libraries are your research platform).
The Library Faculty advisory committee is a very senior group. The Chair is the University Librarian and a named chair University Professor (Harvard’s highest faculty position, the ladder is asst. prof., assoc. prof., professor, named chair professor, university professor). Most of the rest of the Committee are name chair full Professors. For balance there is one assistant professor, one associate professor, and one full professor who does not (yet) been awarded a named chair.
When you think about it the journals have a very strange arrangement set up. We give them the copyright to our intellectual property, pay them page charges to publish it, then pay for subscriptions so that we can read our work. In the meantime we have so seriously bought into the ‘impact factor” myth that we are unwilling to break the strangle hold of the increasingly small number of publishers that publish large suites of journals by submitted to open access publication forums. Although the publishers have been allowed to merge into super publishing groups, the colleges and universities are not permitted to negotiate with them as a consortia, because the courts have said that such an academic consortium would violate federal antitrust laws. What a racket”
Overpricing of subscriptions is indefensible; however, it’s critical to distinguish the Open Access issue from the pricing issue. There are strong arguments that Open Access is both desirable and inevitable. Whether it will cost academia less overall is much less clear, and where it has been successful (e.g. PLoSOne) the costs involved are simply transferred from subscribers to authors.
It’s been calculated that highly productive research institutes/universities like Harvard would pay significantly more in author fees than they do in subscriptions. This may not come from the library budget but it has to come from somewhere. Most often the solution proposed is funding agencies, but, as an earlier commenter implies, all disciplines are not equal, and those outside biomedical research do not have typically have large grants they can use to pay author fees.
Librarians should pursue OA because it is the right thing to do in a digital age not because of a misguided notion that it will make publishing costs disappear. Indeed if cost saving is the real goal in an OA world with no subscriptions to be managed, deans and administrators may also begin to wonder whether they need libraries or librarians…
I do agree that “Librarians should pursue OA because it is the right thing to do in a digital age not because of a … notion that it will make publishing costs disappear.” In the end, the moral issue is more important than the economic one.
But I don’t agree that the economic notion is a misguided one. In this article at The Scientist I calculate — admittedly rather naively — that the total cost to Academia of publishing and article behind a paywall with Elsevier is eight times that of publishing open to the world in PLoS ONE. No doubt some of the numbers or assumptions are a little off, but even if I missed the mark by a factor of four (which I think is very unlikely), I reckon halving the world’s scholarly publishing bill would be worth doing.
Mike, I’m really confused by the numbers you cite in that article. First, you are combining numbers from reports of revenue in 2010 and numbers of articles published in 2011. Second, you claim that “In 2011, 78 percent of Elsevier’s total revenue, or £1,605 million, was contributed by journal subscriptions.” In the linked article, the only place I can find that percentage is on page 18 (slide 36), which clearly states that this is not total revenue, but instead the percentage of “S&T Revenue” derived from research journals in 2010 (£1.0 Bn total revenue, not £1.6 Bn). You then use the number of S&T research articles published from a 2011 document.
Even more confusing, in document one, they separate out “Health Sciences” revenue from “S&T”, but don’t define exactly what that is. In the other document they state that “Elsevier publishes over 240,000 new science & technology research articles each year…”. This is under a heading of “scientific and medical information”. Does either of these include medical/health sciences journals? And what exactly does “over 240,000” mean?
So in reality, you’re only looking at a subset of articles published, and a subset of revenue. Elsevier has an extensive journal publishing program in the arts and humanities, in business and management, in economics and finance and in something they call “decision science”. So a vast amount of their research journal publishing program is wholly excluded from these numbers, both in terms of revenue and articles published.
I really have no idea what to make of this, no clue if your suggested number is in any way accurate. If you can break down the numbers a little more clearly, I’d appreciate it. (crossposting this on the original article if you’d rather discuss it there)
Also on a later slide they note that >80% of that 78% of S&T revenue comes from subscriptions, again it’s unclear how much more than 80%, but the overall numbers should be reduced accordingly. Journals do produce other revenue streams like advertising.
Also, apparently 16% of that 78% comes from corporate sources. Should they be included here and given a free ride on the academic research community’s back? I suppose if you just want to know the overall cost worldwide that should be kept in, but it muddies the question of how much revenue is really coming from grant funding.
As usual, it’s a complex picture that doesn’t reduce down into one easy soundbite.
Hi, David, thanks for chipping in.
It’s been amazingly hard to get figures for the calculations you’re referring to, and I’m sure you’re right that they’re not perfect. I’m very keen to reach a more accurate and better referenced assessment, so any help you can provide will be more than welcome.
The figure that 78% of Elsevier’s S&T revenue is from research journals is taken from a 2011 investor report — the only place I have been able to find such a percentage. You are correct — I didn’t spot this — that it’s in a pie-chart (page 18) that uses 2010 data. However, I wasn’t using the 2010 revenue amount (which I imagine changed significantly in 2011), but only the percentage (which I imagine stayed much the same). So in that respect, I think my figures are good for 2011.
A much more serious issue with the numbers is that I have not been able to find a similar proportion-of-revenue-from-journals percentage for Elsevier’s Health Sciences division. Elsevier’s revenue is divided 50-50 between these two, so the percentage in the HS half is as important as that in the S&T half. In the absence of any number, I just made the only assumption I could — that the percentage was the same, at 78%. Applied to the total 2011 revenue of £2058M, this yields the figure of £1605M that I used for journal revenue. If anyone knows a better number for the percentage of Elsevier Health Science revenue that is from journals, I would like to know it, and see the reference.
You point that only “more than 80%” of journal revenue is from subscriptions is well taken. I’d missed that completely.
The most important thing to take away from all this, though, is that even on the most optimistic assumption — if we assume that Elsevier Health Sciences makes nothing from journals, and that S&T takes only 80% of its journal revenue from subscriptions — then that would still gives us a total journal-subscription revenue of £642M. Across 240K articles, that averages to a cost of £2675, which is about $4320. So even turning all the dials down as low as possible, the very lowest figure I can get for publishing with Elsevier is still well over three times that of publishing in PLoS ONE, and close to five times the cost of publishing in an average OA journal.
It is a bit like trying to find a series of needles in a series of haystacks. I have no idea what Elsevier’s “health sciences division” revenue consists of, how much of that is journals and what else is in there? And are their humanities/business journals on the same sort of price structure? You may be grossly underestimating things, or you may be wildly overestimating things, I really can’t tell.
I was also thinking that to accurately represent things, you’d want to exclude all revenue that came in for print versions of journals. From all the Elsevier journals I’ve looked at recently, there is an option to purchase just the online version, just the print version, or to pay for print and online together. If you’re comparing to PLoS ONE, which is only available online, to keep things equal you’d exclude this optional revenue that subscribers are choosing to pay for the physical object.
Regardless, I’m not really sure what to do with any number generated. I know Elsevier’s rates are not in line with those of most university presses, so the numbers are not representative of all of publishing. And there’s so much unknown in there that it’s hard to get a sense of how much they represent big corporate publishing.
Yes, it’s hard. But I’d much rather have a go at the calculation, acknowledging the possible sources of error, than just throw up and hands and walk away. To determine the best way forward, we need numbers like these. By publishing my best estimates, however flawed they may be, I’m at least putting something out there — something for all of us to refine, and for Elsevier to refute, if they can.
It would be great, by the way, if someone could dig out analogous figures for Springer, Wiley and T&F. I’d be interested to see how they compare, but my intuition is that they’d likely come out pretty close.
I think this is a disagreement we’ve had in the past as far as approaches go. I don’t think basing important decisions on flawed data is a good thing to do. And given the enormous amount of assumptions that have to be made in order to come up with this figure, it’s as accurate as a number randomly pulled out of a hat.
If you were peer reviewing a paper and someone’s data was collected with the same level of rigor, even if they admitted the likely errors, would you approve the paper for publication? How much weight would you put on scientific conclusions based on similarly fudged data?
As such, I’m not sure what the appropriate action to take would be here. The message is already clear– the biggest commercial publisher makes lots of money from journals, money that could instead be kept in the community and put to better use. I’m not sure how much putting out a questionable per-article dollar number for that one publisher advances the argument.
For what it’s worth I agree that the calculations are problematic. However, I think the difficulties that emerge from the process of calculation highlight an important point: that we are not able to find out what publishing in Elsevier costs us. I understand from my librarian at Imperial that she is not even permitted to tell me what they have paid for their Elsevier bundle because the publisher has put a confidentiality clause on the agreement. As I have written elsewhere, this is not the action of an organisation that truly sees itself as a ‘partner’ of the scholarly community.
However, the incontestable figure, which is the one that currently has most resonance with people, is the 35% profit margin which, as Cameron Neylon insightfully pointed out in your interview with him, is an indicator of ‘market failure’ (see last question in the interview).
Open access is a complex issue with unresolved problems but one potential advantage is that, by moving charges up front — right into the faces of authors — it creates greater cost transparency and that should improve market efficiency.
David Crotty says “I don’t think basing important decisions on flawed data is a good thing to do.”
Neither do I. But, David, the alternative at the moment isn’t “basing decisions on good data” — which I would love to do. It’s basing decision on no data at all, on pure supposition.
I am forever reading statements from barrier-based publishers that say things like “there is no evidence that open access will be cheaper”. Well, now there is evidence. As I showed in my last comment above, even giving Elsevier the benefit of the doubt in every single number that goes into the calculations, the absolute best case is that publishing with them costs the community more than three times what it costs to publish in PLoS ONE. Three times. Best case. The true figure is almost certainly much more than that, but let’s be as charitable as possible.
“And given the enormous amount of assumptions that have to be made in order to come up with this figure, it’s as accurate as a number randomly pulled out of a hat.”
No, it’s not. It’s a simple calculation, that anyone can follow, based on published numbers, and with the assumptions laid out. I simply can’t see any wiggle-room for arguing that Elsevier publications cost less than 3.2 times as much as PLoS ONE publications. That is an important number, and needs to be heard, if only to annihilate the “no evidence that OA is cheaper” nonsense.
I’ll say again that if we can find better data that lets us make a more accurate and precise estimate, I will be delighted. As Stephen points out, Elsevier themselves make this hard by keeping so much secret. But at the moment, the numbers I’m putting out there are the state of the art. Please help me improve them. But don’t ask me to pretend that they’re uninformative. They’re not.
John Houghton’s report in 2009 looked in detail at the costs of publishing. Of course the publishers hated it and said it deeply flawed, and everyone else thought it was about right. I haven’t followed all the debate on the report since it came out.
Alma Swan published a report in 2010 on the topic too.
Frank – don’t know how I managed to overlook this comment on Friday morning (though work may have had something to do with it).
But thanks for turning up these important resources — what a great librarian you are!
Thanks! The old memory does work sometimes.
As a former employee of Elsevier Health Sciences and a survivor of 20 years in the medical publishing field (now a teacher), I can say that the big problem with the pricing of these subscriptions is monopoly practices.
I worked for a non-Elsevier medical publisher that Elsevier wanted to buy in 2000. The Clinton Justice Dept. had lawyers crawling all over our offices and it looked like the sale was going to be blocked as a monopoly violation. Elsevier set up a PAC in Florida (!) that heavily contributed to the Bush campaign. Even before we were bought, my company’s employees were “encouraged” to contribute to help the sale go through, presumably for our own good.
Bush “won.” The lawyers disappeared. The sale went through. And 80% of my department was let go in the name of “eliminating redundancy.” I walked out on my own terms as quickly as possible thereafter.
Reed Elsevier has its hands in everything from medical publishing to arms dealing (which led to a boycott of Elsevier by health professionals in England years ago). They’re purely for profit, even when that means stifling the flow of information that publishing, especially health sciences publishing, exists to promote.
I applaud Harvard for drawing a line in the sand and hope that enough other institutions follow their example to put some hurt on Elsevier and make them rethink their practices. With that said, I won’t be holding my breath until Elsevier abandons the dark side.
@Former Elsevier Employee That’s an extremely stimulating and enlightening comment. Thanks for publicly sharing it !!
@Frank – Ah yes, the Houghton Report. This came up during a CRC & Key Perspectives Workshop I took part in in 2010.
From my old blog http://mcblawg.blogspot.co.uk/2010/07/cost-of-open-access.html
As the Workshop involved creating new raw unpublished data, we were advised not to disclose the results. As such, one had to “artfully” word parts of my post and not live-tweet it 😉
I last followed this up a few months ago.
I think I am allowed to say that whilst there was no “published” conclusion per se, I can mention that “he aim of the workshop series was for Institutions to calculate the benefits of OA for them. 30 – 40 people in total participated from the UK”.
At least based upon the one that we took part in, the results of replicating Houghton’s model with real live data from three UK Institutions were of an extremely positive nature.
Very much in link with the Harvard FAC memo, the Scholarly Publishing and Academic Resources Coalition (SPARC) has produced its own guidelines for academics who are looking for ways to take practical steps to promote open access modes of publishing. Well worth a look, I’d say.
When I have started my academic career (beginning of 90s) it was really difficult to get information apart from academic journals. As a PhD student, I had to move to several places (in Europe) to get physical copies of the journals I needed. Nowadays it’s really different and I cannot understand why we should pay money/attention just to what a (very) few publishers are organizing. In the perspective of a private company this is normal: profit of the editorial company (possibly in the short run) is the main goal.. but this is not precisely the goal of a university professor.
Several years ago I sent to a colleague a copy of a preprint for an encyclopedia entry I have contributed – inviting him to leave it at disposition of his students. I believe that this is still the best way for anyone to reach my (modest) contribution to science, as any google search on the (quite specific) topic ( at least since the last five years I think) does bring the link in the first results. This does not cost anything to me and, I believe, not that much more to the colleague who is hosting the paper.. and still is very efficient in the dissemination of the paper itself..
That’s great news. However, publishing open access is expensive. Basically not affordable for every publication. As an example, in germany total funding per year for faculty professor working with theory is 5000 euros. DFG grants, the german standard to get money for research, account for a total of 500 euros per year for publication costs. Now I just payed my 2200$ to publish in Plos Comp Biol. Fine but I cannot expect to do that for all papers. What’s missing?