The Economics of JSTOR

Aaron Swartz was arrested a few days ago for the unauthorized bulk downloading of files from JSTOR, the academic journal archive. According to the indictment [pdf] , he faces up to 35 years in prison and a fine of up to $1 million. Alongside charges of “unlawfully obtaining information from a protected computer”, said indictment also makes outlandish claims of wire-fraud; the hyperbole is further ramped up by the Department of Justice’s Press Release [pdf]. Oddly, JSTOR have put out a statement saying they did not want to prosecute Swartz.

At the moment, there’s more heat than light in the commentary around this case. We don’t know what Swartz intended to do with the documents he downloaded, nor why he obtained them the way he did. He has a strong track record for opening up information and analysing it: witness his liberation of court records, and his analysis of funding academic writing. But this case led me to think about JSTOR, what it claims to do, and the way it acts as a gatekeeper of academic knowledge.

When you visit the JSTOR about page, you are greeted with the statement:

JSTOR is a not–for–profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive of over one thousand academic journals and other scholarly content. We use information technology and tools to increase productivity and facilitate new forms of scholarship.

Sounds good! Who could possibly have anything against a not-for-profit helping spread knowledge? Well, if you don’t have an institutional affiliation, if you’re one of the general public, reading the journals is an expensive business. There is no option to subscribe to the archive as a whole, and the pricing of individual articles is astronomical.

(Academics and students generally have access to these archives through their institutions; for them it’s seemingly free, but the University is paying. How much I don’t know, although an indication can be had from JSTORs price calculator. According to AP, “Its annual subscription fees can cost a large research university as much as $50,000.”)

Take, for example, the journal 19th Century Music. A single issue from the publishers costs $18 for an individual, a subscription for 3 issues $50 (discounted to $28 for students and the retired). The latest issue has 6 full articles, as well as various supporting material. On JSTOR, each article is priced at $12. That’s $72 for the substantive content of just one issue. JSTOR are even charging $12 for the two page preface. This is by no means exceptional: another University of California publication The Public Historian – oh the irony – is $17 for a single issue direct, but $12 an article on JSTOR, again even for two page editorials. Or again, a standard individual subscription to Radical Teacher is $24 – $28; a single article is $18 at JSTOR, even just for the front matter. One more time: $14 for an article from the American Historical Review when joining the AHA starts from $40 and gets you the journal plus many other benefits.

As far as I can tell, these prices are around the average but some are much higher. Rogers’ article on the Black Act, published in the Historical Journal in 1974, is $34, which is $4 more than even Cambridge U.P.’s own inflated price.

Some articles are being sold even though they have been legitimately released for free. For example, A Descriptive and Phylogenetic Analysis of Plumulaceous Feather Characters in Charadriiformes is available from JSTOR for $15 but can be obtained for free at the Searchable Ornithological Research Archive. Other material, like Law and History Review, doesn’t seem to be accessible at all to the public, yet there is no indication that the volumes for 1999 to 2009 are archived for free at the History Co-operative. This is not aiding the discovery and use of scholarly content.

(And while we’re talking about L&HR, Cambridge UP are flogging off articles at $30 / £20, again even those legitimately available for free at the History Co-operative.)

Then there is the murky area of those materials that are out of copyright. The William and Mary Quarterly from 1899? $9 an article. Likewise, it’s $14 for a single article from the Journal of Political Economy for 1892. This may be the fault of the publishers rather than JSTOR, but it points to a relationship between the two that excludes academic and public interests. This also raises the question of publicly-funded research being mined for profit by private institutions, as pointed out by Glyn Moody.

It is not just a case that JSTOR is charging too high a price, whether in relation to other vendors, with comparable products or the pockets of those who would like to read this material. In economic terms, they could increase sales and revenue by lowering the price to hit that sweet spot where they maximise returns. My feeling is that they are deliberately trying to put the public off, restricting access through their pricing mechanism.

I also have the feeling that charging these prices is directly contradictory to their status as a non-profit. I have found nothing on their website explaining how they price materials, or how they work with publishers. I find it extremely unlikely that they have absolutely no say in the prices they charge individuals for individual articles. Nor is there any information on how much revenue goes to the publishers. (It’s well-known that the authors don’t get paid through the current academic journal system.) Furthermore, I have found no financial information as to JSTORs revenues or outgoings, or whether it has, as a non-profit, tax-exempt status; nor have I found any information for Ithaka, with which JSTOR merged in 2009. Accountancy is not one of my skills, but I feel such information should be easily accessible on the website of a non-profit. This lack of transparency and openness I find disturbing.

JSTOR is not living up to its declaration. It is obstructing the dissemination of scholarly, publicly-funded and public domain writing. It clearly has a commercial aspect, yet is shy of stating this. I can’t even be sure it is a not-for-profit in any meaningful, legal sense. It is a contributor to, not a solution for, the crisis in academic publishing.

You can support Aaron Swartz via

UPDATE: The Philosophical Transactions of the Royal Society freed from JSTOR and the Royal Society, bittorrent at Pirate Bay!

This entry was posted in edubiz and tagged , . Bookmark the permalink.

16 Responses to The Economics of JSTOR

  1. Pingback: Royal Society proceedings, 1600-1923, now available! « David Gerard

  2. As a non-academic researcher I find it incredibly frustrating to google some obscure subject, find an article that sounds like just what I’m looking for, and then realize I have to pay for it (and can’t afford to). Of course you could say the same about having to pay for music or a book, difference is that at least there some of the money goes to the creator of the content (musician or author). What kind of royalties are these academic authors getting from the sale of their works? I suspect very little if any – deal seems to be that you’re supposed to be grateful for getting printed in a peer-reviewed journal because it’s good for your career. Academic journals were necessary in the past as the only means to share research findings, they had to be expensive because of the small print runs. In the age of the internet why not cut out the publishers and make all articles available free via online publications? Or is the articifical scarcity managed by JSTOR also a means of maintaining a monopoly of knowledge?

  3. Ben O'Steen says:

    @History is Made at Night
    “What kind of royalties are these academic authors getting from the sale of their works?”

    Simple answer:

    Please, if someone finds this not to be true for any academic journal publishing agreement, please let me know as I have yet to hear of any reimbursement scheme for any journal.

    I have heard of the other – where the author *pays* for their work to be made Open Access.


  4. Andrew says:

    I did some digging this evening and found the IRS filings; they are indeed a 501(c)(3), registered in New York, but they don’t publish the details very clearly.

    Money-wise, the “capital fees” (the sum paid when an institution joins) approximately cover the direct cost of scanning; out of the ongoing annual fees, ~30% goes to the rights-holders and ~70% is the overhead of running JSTOR – the offices, staff, servers, etc.

    The really interesting part, I found, was that they have an income line for “pay per view”. It’s astonishingly low compared to subscriptions – $145k, or 0.35% of their income in 2008, and $80k in 2009.

  5. Pingback: Aaron Swartz versus JSTOR « Jean-Etienne's blog

  6. Todd says:

    Where were all of you in 1995? Sure, the content is available and in the public domain. However, it was trapped in ink on paper for decades, if not centuries. It took vision, millions of dollars, tremendous technical expertise, and an extremely skilled team to create the system people take for granted today.

    People who expect that information should be free should find their own resources to *properly* scan, OCR, clean, tag, enrich, host and guarantee the long-term availability of content in the way that JSTOR has done, with their OWN money.

    I’ve never seen a groundswell of people or organizations willing to stand up and do what JSTOR has done. The few that have tried haven’t been able to do it as well, as efficiently or in as quality a manner. In fact, most people rightfully praise JSTOR for having done something that no one organization was willing or able to do. So now that a majority of the really difficult work has been done (except for preserving the content in perpetuity – which is also part of JSTOR’s mission and no small feat in itself), a few seem to think that we can make the files free and all will be well. Anyone who thinks this is naive.

    For all of the validity there may be in defending of the public domain, we can not dismiss the real costs of undertaking the work that JSTOR has done and is committed to doing into the future, well beyond our own lifespans. Figuring that JSTOR has done the digitization and we’ll just throw the files up on BitTorrent is a solution belittles the entire publication process. If you’re a true researcher or scholar, you would be a fool to believe we’d be better off with a system of files posted to the network, without peer review, without editors, without discovery tools, without persistence, without preservation. These are all things that the publication system provides and files posted to the network never will.

  7. Pingback: Sunday Reading « zunguzungu

  8. johnl says:

    The comment above apparently comes from Todd Carpenter, Managing Director of NISO, the National Information Standards Organization, judging from the email & IP addresses. Ithaka/JSTOR/Portico is a voting member of NISO.

    The trouble with this ‘it costs money’ line is that it dispenses with any ethical considerations, presumes that JSTORs way is the only way, and ignores that by their pricing policy for non-academic researchers, JSTOR not only betray their mission to disseminate, but also undercut their own revenue.

    And BTW, JSTOR has nothing to do with peer review, which is generally unpaid academic work.

  9. Pingback: Free Swartz! (In the Event that He is Imprisoned!!) |

  10. scann says:

    all that this folk tod is saying is absurd. for example, these guys: have made home-made scanners, and some of them costs no more than 20 bucks. of course, scanning well is an expensive activity… and you know why? because the technology that makes incredible scanners is enclosed with patents, patents, patents, and even more patents. just like the way copyright works. yeah, researching is expensive if you let private business to let in and enclose all knowledge in order to make profit, without taking in account that you may be duplicating research, for example. so, the problem of price is a problem that copyright, patents and in general all intellectual monopolies have created.

    there are a lot of examples that probes tod’s wrong. has millions of books in their database, and they all seem to be properly tagged, OCR and scanned, and there a lot of examples more. the only problem that this sites has are copyright lawyers, nothing else.

    pardon for feeding the troll.

  11. cymru says:

    I have a good deal of sympathy with this. It is a mistake to think that only independent researchers are affected. I am at a “Russell Group” institution but my access to academic journals is severely limited. There are *many* articles I find that I cannot read. My institution subscribes to *some* of JSTOR so I can access *some* of their articles but by no means all.

    Even when an article is available, it can be extremely time-consuming and quite unobvious how to access it. Just because article A is available through provider P, does not mean that article A is available through provider Q even though Q does have A and article B is available through Q though not through P although P does offer B. Often there are three, four or more providers of any given article. Things are also divided up so perhaps access to years prior to 1990 is not available to members of my institution; years 1992-1995 are available through provider P; years 1996-1998 are again unavailable; years 1999-2003 are available through Q; years 2002-2004 are available through R; and years 2005- are not available. But it may be that 2005-2007 was available until last week or that last year 1999-2003 were available through R. And so on. It is amazing that anybody ever manages to actually find time to *read* the articles at all… [To be fair, my institution’s IT could be better, too…]

    The costs of journal subscriptions to institutions are deadly. Even though institutional funds are the primary source for the research, writing and reviewing (and often support the editing, too), the published content costs are astronomical. Not just in straight terms of subscriptions, although these are absurd, but in terms of the costs of IT systems required to properly identify and authenticate users in order to ensure that licence agreements are adhered to. This requires different categories of users with access to different content and complex systems for managing off-campus access. Problems with these systems are hard to track and resolve because IT support are, of course, on campus and cannot “see” problems only visible from the outside.

    But I am not sure to what extent JSTOR is to blame and to what extent publishers and the system as a whole are to blame. And it *is* true that the mere fact that something is public domain does not mean that it costs nothing to digitise and serve.

    I do know that researchers in some countries can effectively access JSTOR content for free by affiliating through a national organisation. This applies to (certain?) majority world countries where JSTOR provides access at no cost. Perhaps other providers do this as well?

    One thing which is starting to happen in the sciences but not, that I know of, in other areas, is the requirement that research funded by various organisations be made available through open access routes. Perhaps that will gradually alter the situation by incorporating the costs of dissemination into the upfront costs of research. At the moment, the costs of meeting the requirements are significant because publishers charge authors for the right to make their work available through open access repositories. But maybe that will change…

  12. Pingback: A Move Towards Open Access: JSTOR Launches “Register & Read” Program | Information in Social Context

  13. Pingback: Locked Out « Adventures in (Post) Gradland

  14. Pingback: Academic paywalls mean publish and perish | Ezymagazines

  15. Pingback: Paywalls Académicos significa publicar y perecer – Opinión – Al Jazeera en Inglés « Jestoryas's Blog

  16. Pingback: #IcanhazPDF – Om vetenskaplig fildelning – Filip Svensson om muntlig, skriftlig och visuell kommunikation

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.