Harvard can’t afford science journals

Harvard University, where I’m headed to speak next week, has the largest endowment of any university in the world: 27.6 billion dollars (in 2010). But they have a policy of not using that money to finance independent units of the university: their unofficial motto is “every tub on its own bottom.”

I’m not sure whether that applies to the libraries, but the Guardian has a new article implying as much, “Harvard University says it can’t afford journal publishers’ prices.

Exasperated by rising subscription costs charged by academic publishers, Harvard University has encouraged its faculty members to make their research freely available through open access journals and to resign from publications that keep articles behind paywalls.

A memo from Harvard Library to the university’s 2,100 teaching and research staff called for action after warning it could no longer afford the price hikes imposed by many large journal publishers, which bill the library around $3.5m a year.

The extraordinary move thrusts one of the world’s wealthiest and most prestigious institutions into the centre of an increasingly fraught debate over access to the results of academic research, much of which is funded by the taxpayer.

A lot of the squeeze comes from the rip-off publisher Elsevier, which I’ve discussed before and am boycotting (see this Guardian piece on that boycott):

More than 10,000 academics have already joined a boycott of Elsevier, the huge Dutch publisher, in protest at its journal pricing and access policies. Many university libraries pay more than half of their journal budgets to the publishers Elsevier, Springer and Wiley. . .

See Elsevier’s lame defense of its policy: its price increases have been “among the lowest in the industry for the last several years, averaging around 5%.” Yeah, because their prices are bloated to begin with, and 5% of a higher price still represents a larger dollar increase.

According to the Harvard memo, journal subscriptions are now so high that to continue them “would seriously erode collection efforts in many other areas, already compromised”. The memo asks faculty members to encourage their professional organisations to take control of scholarly publishing, and to consider submitting their work to open access journals and resigning from editorial boards of journals that are not open access.

The profits of these gouging journals are exorbitant, and they’ve gotten away with it simply because they can: libraries were flush, and scientists need the journals.  But the profits are way out of hand:

The memo from Harvard’s faculty advisory council said major publishers had created an “untenable situation” at the university by making scholarly interaction “fiscally unsustainable” and “academically restrictive”, while drawing profits of 35% or more. Prices for online access to articles from two major publishers have increased 145% over the past six years, with some journals costing as much as $40,000, the memo said.

Forty thousand dollars!  I’d love for some reader to give me the name of that journal. It’s absurd.

And it’s unsustainable.  Libraries can’t afford it, and scientists are mad as hell and aren’t going to take it any more.

h/t: JP


  1. Ray Moscow
    Posted April 26, 2012 at 4:22 am | Permalink

    You can read or download Elsevier’s annual report here: http://reporting.reedelsevier.com/

    Looks like a pretty healthy profit-making venture to me.

  2. Posted April 26, 2012 at 5:11 am | Permalink

    The most expensive journal that I know of is Journal of Comparative Neurology, which costs $35,489 per year for both the print and online version (http://ordering.onlinelibrary.wiley.com/subs.asp?ref=1096-9861&doi=10.1002/%28ISSN%291096-9861).

    I seem to recall a chemistry or physics journal that cost over $40,000 a year, but it was split into sections – an A, B, C, and so on – so it’s debatable whether it was one journal or several.

    • Dominic
      Posted April 26, 2012 at 5:29 am | Permalink

      I think you are right – it would probably be a physics journal. Publishers charge institutions extra because they want to network the journal to multiple users.

      There are also issues with archive access if you go electronic only. We no longer take the hard copies of Elsevier journals (by diktat from on high). There is however an issue with pay to publish journals – that the budget comes out of research money instead of library money. It seems to me that libraries are on the way out – the attitude is that it is ‘all online’ when it is not, and certainly not cost free. Tie in the fact that while publishers increase subscription prices by ‘only’ 5% a year, with the fact that library budgets are static or dropping. Therefore we cut something else – maybe a journal subscription where the journal is less popular or more obscure (I mean not ‘less important’, but ‘less appreciated’).

      • Voltaire 2
        Posted April 26, 2012 at 10:50 am | Permalink

        What are these journals made of, solid platinum?

        I know they are not aimed at Joe Sixpack, but they also smack of intellectual elitism.

        When the barbarians are at the gates, lowering the price will be much too late by then.

        • Posted April 26, 2012 at 12:06 pm | Permalink

          The cost of Journal of Comparative Neurology is almost understandable because of two things: (1) it still prints paper copies, and; (2) there’s inertia from the pre-digital costs.

          Journal of Comparative Neurology publishes several thick issues each month. And it’s on glossy, acid-free, archival paper: top quality stuff.

          Because it’s an anatomical journal, there are lots of high resolution pictures. And they do a good job; Journal of Comparative Neurology has always very high standards for image reproduction.

          Because of the economics of printing paper copies, small print runs (like academic journals) always have a higher cost per unit, because so much work goes into setting up the presses.

          I could understand why that journal was one of the most expensive 20 years ago, before digital typesetting and photography. I think the “legacy” of printing paper copies is part of the reason why the cost has stayed high.

      • gillt
        Posted April 26, 2012 at 10:42 pm | Permalink

        that the budget comes out of research money instead of library money.

        Doesn’t it always cost to publish? Also PLoS says it makes exceptions if you have no money.

        • Posted April 30, 2012 at 1:27 am | Permalink

          No, it doesn’t always cost to publish, at least not directly to the author or his institute. Many journals are funded by subscriptions (which is a flat fee) and even non-subscribers can publish. Some are funded by national research councils. Of course, there is no free lunch, and someone pays for it, but it is good not to have a conflict of interest: an author shouldn’t worry about what it costs him to publish and a journal’s revenue shouldn’t increase by accepting an article which it shouldn’t.

          There is a lot of double-speak going on. “Open access” can mean many things, but some people have, probably intentionally, hijacked it to mean “the author pays”.

          • gillt
            Posted May 2, 2012 at 4:45 pm | Permalink

            I don’t find what you say to be true for many biology journals: it costs to publish. It even costs per colored figure.

  3. Reginald Selkirk
    Posted April 26, 2012 at 5:22 am | Permalink

    their unofficial motto is “every tub on its own bottom.”

    They should put it in Latin, it would probably sound more impressive.

    • Dominic
      Posted April 26, 2012 at 5:31 am | Permalink

      Google translate would render it thus –
      omnis dolium in sua deorsum!

      • Dominic
        Posted April 26, 2012 at 5:32 am | Permalink

        I will check with my latin teaching chum!

      • Necandum
        Posted April 28, 2012 at 5:09 am | Permalink

        Latin student here. A reworking of the google translate to get something crude but more or less grammatically correct would be:
        omnia labra in suis fundis

        All tubs on their own bottoms.

  4. morkindie
    Posted April 26, 2012 at 5:52 am | Permalink

    I am not a scholar or scientist, but occasionally I will search on google/scholar, and it seems like every article costs more than a book.

    It’s ridiculous. I thought this was the age of knowledge!

    • Sajanas
      Posted April 26, 2012 at 7:06 am | Permalink

      I have the suspicion that those $30 per article prices aren’t there to actually sell articles, but to force people into buying subscriptions. But they’re ignoring the fact that selling lots of low priced articles will make them more money than selling no high priced ones…. as an employee of a small company, I simply can’t use any of their electronic stuff, and if I need something, I have to go to our university library instead.

      The thing is, if they used micro-transactions, then they couldn’t bundle crap with the good. Its the De Beers diamond selling model, and its only good for the seller… its time everyone stop putting up with it.

      • Kevin
        Posted April 26, 2012 at 7:44 am | Permalink

        Not exactly.

        There’s a wide range of folk who pay the $30 because if we were to buy a subscription for each journal that we need one article from, we’d be broke in a week.

        I’m a medical writer. I use up to 50 citations for the average CME lecture I create on various topics. They don’t all come from the same journal (in fact, it would be appallingly bad research on my part if that were to happen). Most of the time, it’s one article from journal A, one from B, two from C, etc. Completely random. You can’t possibly buy a $300-$3,000 subscription to a journal that you’re going to use maybe once or twice a year.

        But do the math. 50 x $30 for each lecture I create. And I do a LOT of lectures each year. And I’m just one small cog in a very large machine of medical writing.

        It’s nothing more than extortion. There are times when I refuse to pay the bucks to get the entire journal article when all of the data I want are in the abstract. But that’s rare. Most of the time, I pay full freight.

        Multiply that times the thousands of people like me and you’re talking about a HUGE cash cow for the publishers.

        Now, the good news is that many journals are opening their files to open access after a certain interval — 6 months to a year. And some journals are freely available online from the start.

        But Elsevier titles? Never. They’re the worst of the worst.

        I encourage the boycott. And yes, it’s because my ox is being gored. But it’s also because it affects my ability to create high-quality educational programs for health care professionals. Our budgets are totally out of whack because of the greed of Elsevier (and Springer and a few other bad actors).

        • Ray Moscow
          Posted April 26, 2012 at 8:48 am | Permalink

          This is not a solution to the underlying problem, but for people based in the UK or elsewhere in Europe, keeping a student status in The Open University gives one access to their on-line library, which is excellent. Just take a course now and then (or like me, all the time).

          • Voltaire 2
            Posted April 26, 2012 at 10:53 am | Permalink

            I see that Ivory Tower just acquired barbed wire and machine gun turrets.

            Are those peasants with pitch forks and torches that I see in the distance?

  5. Pray Hard
    Posted April 26, 2012 at 6:37 am | Permalink

    Universities should have never adopted the “corporate model”.

  6. Posted April 26, 2012 at 6:49 am | Permalink

    A short opinion article on this by Michael Taylor in the latest issue of ‘The Scientist’:


  7. Sameer
    Posted April 26, 2012 at 7:36 am | Permalink

    Is there something equivalent of arXiv for biology? I think arXiv already has quantitative biology. I don’t know if the existence of freely available preprints makes any dent in the rise in journal subscription costs.

    • Sameer
      Posted April 26, 2012 at 7:38 am | Permalink

      Oops! Bad link for arXiv.

      • Sigmund
        Posted April 26, 2012 at 8:03 am | Permalink

        The closest is perhaps the PMC database that contains free full text publications available for download.
        Most new publications are, however, only available to subscribing clients.


  8. Schenck
    Posted April 26, 2012 at 8:10 am | Permalink

    Here is the blog of a Harvard reseacher on just this topic,


    Particularly impressive is the post on the Journal of Machine Learning Research, which, obviously though parochial, is free, open access, charges no submiter fees, and operates as a non-profit. It’s also indexed as one of the top journals in it’s field with high impact for it’s scope:


    Really impressive story. They’ve been doing this for 12 years. The biggest cost was at start up, and that was around $3,000, which is a pittance, my campus (actually our union) provides that amount to attend a conference.
    Imagine, instead of a conference, I coulda started a journal! Silly me!

  9. ThyroidPlanet
    Posted April 26, 2012 at 8:34 am | Permalink

    The mistake Elsevier made is obviously making some MBA’s hack together that letter – replete with mind-numbing words like “value added” and “business model” – when the audience is scientists. Major pitchforks-and-torches turn-off.


    what is the argument against the fact that many of their big titles – including Cell Press titles – are free – that’s right , FREE as in freedom and beer – after 12 months?


    look, here’s a random article from January 2011:


    the PDF looks like its from the journal itself.

    so – what is the argument against their after-12-months-free model?

    • gillt
      Posted April 26, 2012 at 10:50 pm | Permalink

      Because a researcher cannot get by on a diet of year old articles, especially in fields like epigenetics or all of cancer research.

      • ThyroidPlanet
        Posted April 27, 2012 at 5:38 am | Permalink

        OK, but scientists at Anywhere University don’t have that problem.

        So ….

        • gillt
          Posted May 2, 2012 at 9:39 am | Permalink

          What are you talking about?

          • ThyroidPlanet
            Posted May 2, 2012 at 12:32 pm | Permalink

            How many cancer research or epigenetics articles – or what publishers- could you personally not access within a year of an article’s publication date?

            Or, name five e-gen or cancer journals. Why doesn’t your library have them? Or if they do, then what is the problem?

            The reason I ask is because I cannot understand if the protest is against not having Elsevier articles for free within the first year of article publication.

            If not, then you must bring MacMillan into the debate as well because Nature is not ever free (from what I have seen).

            • gillt
              Posted May 2, 2012 at 4:50 pm | Permalink

              What I’m saying is you cannot keep pace in rapidly advancing fields, such as most of human genetics. It would be unreasonable for universities to ask their researchers to wait around a year to read an article.

              • ThyroidPlanet
                Posted May 2, 2012 at 5:03 pm | Permalink

                >you cannot keep pace in rapidly advancing >fields

                I understand this – what other field is there?

                > It would be unreasonable for universities to >ask their researchers to wait around a year to >read an article.

                I understand this.

                but again – if I sit down at a computer I can get any high eigenfactor/impact factor journal article I want because someone somewhere is paying. this is what I mean by it not being a problem.

                If I try to get some Wiley or Springer articles for instance, I might have a harder time. but this doesn’t hold up me or the field.

                that is why I asked for a few titles and what not.

              • gillt
                Posted May 3, 2012 at 11:16 am | Permalink

                I understand this – what other field is there?

                What other field besides human genetics? That could be read as sarcasm since it covers an awful lot and besides that’s where most of the money goes.

                It’s seem obvious to me that not having immediate access to all the data being published in your field is a research and career obstacle: you’re no long competitive.

              • ThyroidPlanet
                Posted May 3, 2012 at 11:43 am | Permalink

                No I certainly do not mean to be sarcastic – clarity/wording is my error – but it’s besides the point. But since I am now on the spot:

                I mean to ask – rhetorically – that yes, every field is undoubtedly advancing – but what fields are not *rapidly* advancing? Who would claim that their field is not rapidly advancing and therefore doesn’t need immediate access to as-soon-as-published articles? Who wouldn’t access to articles at will all the time?

                … So now that I tried to erase the illusion of my sarcasm, I can only hope my meaning is more clear.

              • gillt
                Posted May 9, 2012 at 9:44 am | Permalink

                Of course your’re right about everyones bias to their chosen field, and I suspect, based only on my own limited experience, that staying relevant, especially in the interent age, means having immediate access to the latest research in most of the sciences. How can you be at the cutting edge in your field if you can’t read what’s been published in the last six months? Attending conferences is an expensive alternative I suppose, but one you should be doing anyway.

  10. Posted April 26, 2012 at 8:55 am | Permalink

    It’s easy to blame the big bad capitalist publishers for this, but unfortunately, you are to blame! Not you, personally, but your fellow scientists and academics who prefer to publish in the best name journals, must of which are owned by the big publishers like Elsevier and Wiley. Andrew Odlyzko, an economist at the University of Minnesota, did a lot of fascinating research on the ‘journal crisis’ estimates that the average cost of publishing an article is $4000, and a typical journal will publish about 200 articles a year. That cost has to be recovered somehow. Since academic journals is a closed market (their only customers are libraries and a few specialist in the field), the average subscription for many obscure titles can be as low as 100. If you do the math, you find that the lower the subscription, the more expensive the journal, and it doesn’t help when libraries keep canceling journal titles. (By the way, libraries account for 3 to 4% of a typical universities budget, and 1/3 of library’s budget goes to books and journals and that has not changed much in decades.)

    So one alternative is free open source journals, but like Grandma always says, there is never a free lunch. Guess how must open source journals recover their publishing cost: they charge the author. I hate to you a religious expression here, but this a classic case of robbing Peter to pay Paul.


    • J.J.E.
      Posted April 26, 2012 at 7:09 pm | Permalink

      Sure… Academics are to blame. Right. If academic scientists want to remain academic scientists, they must get funding and tenure. However, tenure review committees and grant review committees, as a rule, tend to value papers published in certain commercial journals higher than papers in even the best open access journals. And remember, there are only so many slots in the good open access journals anyway (not everyone can publish everything they do in PLoS Biology for example). In my field, if you can, you should try to publish in Nature, Cell, Science, Trends in *, Genome Research, PNAS, Genome Biology, PLoS Genetics, and PLoS Biology. Only 3 of those journals are open access, and if everyone were to try to publish only there, then there would be a serious supply/demand issue. Not to mention, defectors would be richly rewarded.

      So, no, this isn’t something that can be solved merely by the type of collective action that groups of people never ever engage in. The system needs to change, and that can happen if the funding agencies and the libraries throw their weight around.

  11. Gordon
    Posted April 26, 2012 at 2:04 pm | Permalink

    It’s not only science journals rorting the system. The Cambridge Law Journal want 20UKP for an article published in 1949 – or I can “rent” it for 4UKP.

    I can also buy a recent e-copy of a detective novel from an online site for about 2UKP

  12. MadScientist
    Posted April 26, 2012 at 2:17 pm | Permalink

    I think that’s an excellent move from Harvard. Now we just need others like Princeton, MIT, Carnegie Mellon, the SUNY, Tex A&M, UC, Stanford, ASU, and others to do likewise.

    Overall I think money would be far better spent maintaining government-funded datacenters to host open journals. That would also help a lot with the mandate that federally funded civilian research be publicly accessible. I think we need another agency akin to the NSF to provide those services. The next step could be one which I had been suggesting for over a decade: a repository for free high quality electronic books for all levels of schooling. Don’t let Don McLeroy and the publishing houses dictate what goes into books.

  13. rmbyoung
    Posted April 27, 2012 at 6:42 am | Permalink

    The Elsevier subscription rates for 2012 are here:


    The most pricey single journal subscriptions following the ‘standard’ journal article format seem to be Tetrahedron, which goes for ~$20k, followed by Journal of Chromatography A (~$18k) and Chemical Physics Letters (~$17k). There is one called Adverse Reactions Titles which is ~$31k, but that seems to be a reference publication.

    There are several ‘packages’ more expensive than these; by far the most expensive is Excerpta Medica Full Set with 774 issues per year at ~$103k.

  14. Posted April 29, 2012 at 2:21 pm | Permalink

    The institutions whose faculty, students, and alumni provide journal articles should — at the very least — be receiving free subscriptions to every single one of those journals.

    The current situation is like charging a cow for milking her, and then charging her calf for feeding it!

    • Posted April 29, 2012 at 2:52 pm | Permalink

      As for overcharging everyone, institutions included, here’s a quaint old saying: “Pigs get fat. Hogs get slaughtered.” Elsevier et al have proven themselves hogs.

%d bloggers like this: