Scientific article download costs

By Grant Jacobs 17/07/2010

A brief look at the range of costs of downloading a scientific article.

I never really take that much notice of per-article fees. It’s not as if I ever pay them.

As a diversion from writing, I did a very lax survey of sorts of a small number of biological science journals. (I’m a biologist, so I’m sticking to what I know.)

What tipped me off was that while writing an earlier article I noticed that the cost to ’rent’ one of the papers was $US0.99. (This diverts to the DeepDyve website, which enables you to read the paper for up to 24 hours. One subscription option gives 25 articles for $US19.99. The first three rentals are free.)

By contrast the cost of purchasing the articles in my earlier blog article directly were:

The Carbohydrate Research article is only 3 pages long, which would work out at $US 10.5o per page.

On top of these fees would be credit card fees and current exchange costs for those outside the USA.

Looking at more biological science journals – leaving aside open access journals, which cost nothing to read – the wide range in costs of access to articles continued. I’ve picked these from randomly selected articles in the latest editions (sorted by cheapest option):

Clearly you couldn’t buy articles one-by-one on a regular basis unless you are rather well off!


Most journals now have some articles free. Variations on this are noted below for the journals looked at.

* Many of the commentary articles are also free.

** This is the formal, aka ’correct’, abbreviation. It’s usually known by the acronym PNAS (most people prefer to say this letter by letter, rather than as a word for fairly obvious reasons).

*** All research articles are free but review and commentary articles are by subscription or payment. These are the estimate costs the service offers. A 30-day free trial is available.

Other articles in Code for life:

What famous writer do you write like?

Temperature-induced hearing loss

Royal Society publishing free to read, 1665 – today

Oliver Sacks on Hallucinations

What is your relationship with your research notebook?

0 Responses to “Scientific article download costs”

  • Interesting stuff. I’ve written a quick post on this, which I’ve scheduled for Monday (this is too science for the weekend!). I’m surprised more journals haven’t tried selling paper for $1 – I’m sure they aren’t getting much money with the present pricing scheme.

  • Look forward to seeing your take on it.

    I’d love to see stats for the sales for the single papers — and the rationale behind the pricing schemes!

    There should be one, they’re business and I’m sure that they’ll have thought about it, but I doubt they’d expose their business plans to the public, though…

    I can imagine at $1/paper they’d get more interest from people would otherwise try get them by asking the authors for a copy or a getting a photocopy from the local library or whatever. There’s the immediacy of the thing for one thing (no waiting on the authors or having to trek to the library, etc.) and time is money to anyone who is busy.

  • One note: Nature started making news pieces freely available without login.

    Aside from that, the ranges and rates are astounding. Reasons for high rates are doubtlessly varied, but here are a couple I’ve considered. First, “rental” is problematic b/c it’s pretty easy to store a copy of the paper once you have access–either by downloading the PDF or using a program like Evernote to archive the webpage view.

    Second, I wonder if the high rates for single articles are intended to promote full subscriptions, especially for Nature and Science, which cost $200 to $300 US per year. More subscriptions=higher circulation numbers, which provides some leverage with advertisers.

  • BB,

    One note: Nature started making news pieces freely available without login.

    Thanks, that’s part of I meant when in the first asterisked footnote. I wasn’t being very clear, to say the least! It’s good though, eh? 🙂

    it’s pretty easy to store a copy of the paper once you have access

    I wondered about that (it’s partly why I stuck rent in double inverted commas). I haven’t tried, so maybe you’re right. Mac OS X also allows you to “print” to PDF, which might be another option, too.

    On your second point: one of my thoughts too, but when I tried thinking it through a bit more it started to make less sense. I ended up unsure if my reasoning made for sound logic so I left it out. I have a suspicion it’d take an industry insider spilling the beans to make sense of it properly.

  • “It’s not as if I ever pay them.”

    At he risk of asking a stupid question, why is this?
    Or do you mean you don’t pay per single article?

  • rainman,

    Or do you mean you don’t pay per single article?

    The “them” in “It’s not as if I ever pay them” is referring to the closing passage of the preceding sentence: “… of per-article fees.” 😉

    Universities, etc., get institutional subscriptions that cover a wide range of journals. Good institutions cover most of the journals their scientists use.* As a result not having access through the university (CRI, etc.) library should be the exception rather than the rule.

    Researchers or departments can — funds and rules permitting — allocate some of their research funds to subscribing to a few journals related to their specialist interests that the institute doesn’t carry. (‘A few’ reflects budget limitations.)

    When there is no subscription for a particular journal, readers can interloan copies of articles from another library that carries the journal, or write to the authors and ask for a copy. (I usually do the latter as I prefer to have PDF copies of the articles I use.)

    * A few journals seem to slip through the cracks and some are dropped if their fees seem excessive for how much they’re being used.

  • Of course, academic and corporate librarians know that cost per download is (by chance) lower under package subscriptions …

    I know that real cost (calculated on annual fee/total of downloads) for institutions would be around 2-5 $US per article: most of the time this kind of information is undisclosed for commercial reasons.

  • Maybe this is a record? (via twitter)

    Jess Palmer (@bioephemera) wrote:
    A (defunct, I think) journal just offered me 48 hours of access to an article for $208. For heaven sake! I don’t need to cite it that badly.

    (That’ll be US dollars.)

  • I was looking at the stats for the BES journals, and people do buy single papers – not many in a year, but still there are people who haven’t heard of the #icanhazpdf hashtag.

  • Ha! I can’t win. To get your comment to show up, on the top page at least, I tried escaping the apostrophe with a backslash. The backslash doesn’t show on the top page, but it does here. What gives?

    I wonder how the returns compare to the overheads of maintaining the payment system, assuming a reasonable fee (not the $US200+ fee Jess ran into!)

  • The journals are all published through Wiley/Blackwell, so I assume their set-up is the same for all journals. I guess, then, it does cover costs: the BES journals were bringing in a few thousand euros/dolalrs/pounds, so multiplying that up over all journals suggests a decent chunk of money, albeit minor compared to otehr sources of income.

  • So now you’re ‘Bob O apostrophe H’ 🙂 You’ve got to laugh, eh? It makes you wonder if there is anyone working in the WordPress development team with an apostrophe in their surname – I guess not… (Or not enough people letting the developers know.)

    Anyway – thanks for the thoughts. It was just that I got to wondering that the high costs might be reflecting trying to match low sales against the overheads.

  • My current theory is that the price is high because they don’t want to undercut institutional subscriptions. But I have no evidence for this. If i can corner the right person I’ll ask them about it.

  • That will be a factor, I’m sure; you may be right that it comes down to that (and “to hell with those outside of institutions who aren’t rolling in the green stuff”).

  • The Guardian ran a series of stories on the system of peer-reviewed papers in high value journals. One of these is:

    The Guardian then published this editorial using how the Welcome foundation is responding to the closed market that a few publishers hold:

    As the paper in the NZ Science Review, Vol 67, number 4, pages 106 to 113 explains (see Box A, page 107) funding depends very much on your research output. And the quality of your research output depends to a great extent on your “journal impact factor” (see Box B, page 108). This paper can be downloaded (for free!) from this page of the NZAS website:

    So the science funder, the NZ government, does not fund publication or peer review through research grants. By default libraries of the research institutes have to pay what the publishers charge and individuals (such as retired gentlemen online commentators) have to either get nickel and dimed to death or not get access to the papers. Publication and peer review seem to be yet another a part of the overheads that got debated at the emerging scientists conference and subsequently in Sciblogs.

    And, in a nice irony, John Wiley (US scientific paper publisher) is suing an American firm of patent attorneys for copyright infringement. The attorneys supplied to the US patent office copies of papers published in a Wiley journal as “prior art” – something they are obliged to do under US patent law. A recent chapter of this ongoing story is at:

  • Doug – quick tip in case you’re not aware of it: if you include more than two URLs, the comment will get flagged as potential spam and held up for approval. If you’re wondering why your comment got held up – there’s the reason.

    While the big publishers likely do have a fair bit of leverage, I wouldn’t quite go as far as saying “the closed market that a few publishers hold” myself. Perhaps you mean that the clients are captive, not that the market is closed? (New publishers can and do get established.)

    As the paper in the NZ Science Review, Vol 67, number 4, pages 106 to 113 explains

    A fair chunk of this readership has first-hand experience of this one!

    So the science funder, the NZ government, does not fund publication or peer review through research grants.

    I hope you can forgive me for putting this aside and making a few tangental remarks instead.

    I presume you’re adopting the line of the Wellcome Trust’s (two l’s!) initiative, which I’ve previously written about (there’s also an update more recently, but I’m out of time to locate it). Their idea of having research funding tied to “free” publication in a journal hosted by a/the funding agency is interesting, but not without it’s issues and debates!

    As you say, peer-review isn’t funded – by any route. That’s true world-wide (I believe there are a few exceptions) – rightfully or wrongly it’s considered ‘part of the job’ of being a scientist. You are right that the costs are beholden to the publishers, they’re an independent industry after all, and that those outside of the large institutions can find access to some material difficult or at least harder than it might be (networking helps).

    If the Wellcome’s efforts interest you I’d recommend the discussions over at Occam’s Typewriter and elsewhere – one is linked within the article I linked to above. Stephen Curry (at OT) has also written a fair bit on this topic.

  • H/T’s to Mike Taylor (U. Bristol), the Right to Research coalition and Siouxsie Wiles for the previous example (via twitter).