Blog Post

Open Access and its Costs

Harvard's recent (Feb 2008) Open Access publishing mandate for its faculty makes me think about whatit would look like to make a national, interoperable open archive,across all of the libraries, across all the electronic archives, andthat took into consideration production as well as consumption in itsgeneral business plan. Here's the link to the mandate: At the end of this blog posting, in the Comments section, you'll find the full text of the Harvard press release, "Harvard to Collect, Disseminate Scholarly Articles for Faculty."


I 'heart' Open Access. I wouldn't be writing this blog every day if I didn't. I wouldn't have co-founded HASTAC back in 2002-03 if I didn't. I begin by stressing that point because, whenever I worry aloud about a sustainable business model for open access electronic publishing, invariably someone accuses me of not being a true believer.


Actually, I believe that only if we have a sustainable economic model for open access to scholarly publishing will we ever achieve full, free, and open access. That is, I'm such a True Believer in open access that I'm determined to get us all, together, to think, together, about what would be required to make open access not a pipe dream but something that actually works. It would be too ironic if the Information Age went forward with the current lock-box model of dissemination for scholarly research. We have to find a better way.


What about Harvard's mandate that all faculty publications be held in an open access repository? Whenever I read the Harvard press release, I stumble over some of its implicit assumptions. It's clearly intended to be a visionary policy (I applaud that 100%), but how transferable is it from Harvard to every other institution? And is the model we want for every university to have an open repository for its faculty? What Harvard is doing is a first step, but the real goal is to have all scholarship in one vast, searchable, open repository. So my concern is how you go from this one-institution model to the grand scale of true open access. There are still some big gaps we have to fill in to make that leap.


Maybe I spend too much of my life worrying about this stuff. Maybe I've spent too much of my career studying the whole publishing cycle, from author to publisher to distributor to reader, and all the points in between. I know (in the eighteenth century as in the twenty -first) that you cannot push any part of that cycle without having some dramatic impact on the other parts. So if the end product is going to be free--if the "distributor" is the internet and anyone in the world can download any scholarship or any other writing for free, in the same way that some people file-share music illegally--then there is no motivation for the publisher to be producing it. (The current state of the record business provides us with many examples, not all of them applicable to our realm of nonprofit scholarly publishing, but with some rather dire warnings from which we should be learning in many different directions: this comparison will be the subject of a future posting, I promise.)


In any case, publishers have to have some motivation, some guarantee not that they will make money (very few scholarly publishers make money) but that they won't lose it. The for-profit publishers (and there are many scholarly publishers such as Wiley that do a healthy, for-profit business) will need some form of compensation. The scholarly non-profit publishers (largely but not exclusively university presses) will need some method for offsetting any revenue lost from open access. Unless that bottom-line is figured into the equation, there is no motivation for someone to be a publisher. Without publishers, then authors have to be in the business of self-publishing. That's fine . . . but, as I've written about many times in this blog, self-publishing isn't free. And it isn't easy. And, most definitely, it isn't "professional."  Finally, authors should be in the business of writing (which is hard enough); we should not also have to be publishing our own writing and doing all that is required to have it published in a polished, formal way (i.e. not in the casual and spontaneous format of a blog but as refereed, responsible online scholarship).  


There are so many issues, partly because "open access" sounds like one thing but, in fact, it is the end product of many different forms of publishing, each with its own particularities and its own contingencies. The Harvard mandate plucks one set of authors (Harvard faculty publishing scholarly journal articles) from the huge world of publishing and puts their work into an open access archive. But what are the wider implications, both for this system and for open access publishing as a whole?


Yesterday, before writing a blog posting about the wider implications of the Harvard Open Access mandate, I wrote to my former Princeton colleague, one of the most brilliant historians of the book, Robert Darnton, who also now happens to be the head of Harvard's Libraries and the mastermind behind the open access mandate. The mandate reads, in part: "Harvard will take advantage of the license by hosting FAS faculty's scholarly articles in an open access repository, making them available worldwide for free. The faculty member will retain the copyright in the article, subject to the University's license. The repository contents can be made widely available to the public through such search engines as Google Scholar. Faculty members may request a waiver of the license for particular articles where this is preferable. The new legislation does not apply to articles completed before its adoption."

I wondered what this meant in economic terms. If Harvard republishes in an open access archive an article originally published in a subscriber-only scholarly journal, would Harvard be paying the journal for the right to make the article open to all?  Or would they just be doing it, by fiat?  The economics of simply making closed work "open" concerns me.  Not for commercial publishers.  I don't care much if Harvard (or anyone) makes an article in an Elsevier journal open to all without paying extra fees. Elsevier is a commercial publisher that charges such exhorbitant prices for its science and technology journals that, increasingly, libraries can't afford subscriptions for anything else and have had to radically decrease the number of books they purchase, too. Elsevier is doing fine, thank you, at making money from the business of scholarly scientific publishing. But what about some tiny classics journal, barely squeaking by? Or what about those university presses that are able to publish scholarly books because their journal program supports the book losses?

I asked Bob Darnton those questions and he said that his budget (which, like most library budgets, comes from his university's central administration) "provides for submission and publication fees, so that the author can help the journal cover its overhead."


You could hear my sigh of relief when I read that. Finally, a plan for open access that includes compensation to those who stand to lose revenue from open access. It's not clear how much is paid, on what model, but even a recognition that scholarly publishing doesn't just "happen," that it has many associated costs, is a significant gesture.  It may seem small but it isn't. Harvard's policy shows an awareness that production isn't free and that, if you are going to open the distribution process, then you must also subsidize the production process in some other way.


I doubt that the fees Harvard pays to publishers are sustaining but even thinking in those terms, about the ecomonic implications of the larger process, is a step in the right direction. Interestingly, I haven't read any reports about the Harvard mandate that underscore this feature of the program. So I want to emphasize the point here because I believe it is crucial if we are ever going to have truly open access for all of our scholarly publishing, especially in the social sciences, humanities, and in various scientific fields (some areas of mathematics, for example) that operate more like the humanities than like grant-driven science.


Most grant funding in the sciences covers the cost of publishing so scientists have addressed this issue, in large measure. And for a long time their reprints and prints have been available on line in open source forms. It makes sense that they are allowed to disseminate their articles for free if they have had to pay to have their work published; some science journals charge thousands of dollars to publish one article. And the subscription fees for the journals can also be exhorbitant. It would be immoral not to have an open access version of such articles.


In the humanities and social sciences, more and more universities are understanding such a subsidy is necessary for books, even if there isn't such an awareness yet about journal-publishing. But in the humanities and social sciences, we have to recognize, scholarly book and journal publishing are intertwined economically. There are far more universities that depend upon publication of a scholarly book as the gold standard for tenure than there are universities with presses and virtually all university presses lose money on their scholarly book operation. These presses have to subsidize the book publishing in one way or another. It used to be subsidized through journal publishing--but open access for journal articles threatens that economic model. It used to be that publishers covered their losses by charging libraries more for a library-bound (cloth) covered version of the durable book. But libraries now regularly buy paperbacks and sometimes even share one book in a consortium.


The point is there has to be a way to pay for those scholarly books and journals that don't pay for themselves. If journals go open access, it disturbs an array of complex economic relationships. That's why I applaud the Harvard fee model--it is a model of how we have to think about these things in tandem, all the parts of the process at once and as part of a greater whole, contributing to the greater good of open access.


We are in a transitional moment. The scary thing about transitions is that some things win, some things lose. And it isn't always the best thing that wins. Beta tapes anyone, to use an example of an old technology where a far inferior one succeeded (largely because of distribution and content issues, incidentally.) What wins usually has the best marketing strategy and most sustainable economic basis.


So how can we make an open access model of information that works? There are lots of possible ways and all require some form of subsidy, systemically. What if as part of every start up package for a professor a subsidy for articles and books was mandatory? What if the professors publishing their work had subsidies to help support it, with the understanding that open access would be the end result? Universities are always talking about "knowledge in service of society." What better way to exemplify knowledge in service of society than making the research of its faculty available to all?


Of course, all of these forms of subsidy require universities to put up funds for scholarly publishing. Many (especially those that already support university presses) will balk at this since most universities these days are also strapped for cash and face projected enrollment declines as well as a recession. Some will say this simply shifts the burden of the problem. There is no quick fix, no one solution, that much is clear.


Here is another idea. Some universities have such enormous endowments that they are whole economies unto themselves. I once heard that Harvard's endowment made it something like the twentieth largest independent economy in the world. I'm not sure if that is true and a quick Google search just now didn't confirm it. So don't hold me to that. But suffice it to say that many of the nation's universities would not have to charge a penny of tuition and, every year, they would still grow larger, just from the interest on their unspent endowments, than the total endowments of almost all of the other universities. Recently, these great universities have come under attack for not spending enough of their endowment. What if, as a gesture of support for knowledge, some of those endowments went into a huge public open access trust designed to make all the scholarly publishing in all fields by all scholars open to all? What if the great foundations, such as the incomparable Mellon Foundation (the Foundation that has most aggressively thought through and pushed the boundaries of electronic publishing), were to not just speak about electronic publishing and open access but work to create an endowment for open access that any publisher could use to support the open access portion of its operation? Pie-in-the-sky, I know . . . but some kind of drastic infusion of capital might help make open access sustainable.


Or something simpler: some specialized journals don't make money at all. Their subscriptions don't cover their costs. They have limited readership. Since there already is no workable business model but these enterprises are already sustained for the sake of knowledge, why not also digitize their products and make them available free to all? If the business model for those journals is already philanthropic, then it makes sense to make the user-base wider, to make the fruits of so much uncompensated labor to a greater number of reasons.


The point is that there are many specific circumstances that need to be attended to but, in the end, a workable model has to evolve that addresses both the specificities of production and the end-goal of universal access. Right now, too much of the open access rhetoric makes it seem as if scholarly publishers are the "bad guys," greedily hoarding knowledge. Wagging the finger at scholarly publishers, as if they have the capacity to just hand over their revenue sources for the "greater good," is unrealistic. University presses typically operate as a nonprofit within a nonprofit. How can they support open access? The goal will never be achieved until economic realities are factored into the equation.


Earlier I said you cannot push any part of the publishing process without having an impact on every other part. So let's say that we have this fantastic Open Knowledge Endowment supporting open access to the highest quality scholarship in such a way that does not bankrupt scholarly publishers. We don't want Elsevier-style price-gauging either. Just quid pro quo. Everyone breaking even. Then the next step has to be pushing at the distribution side.


A downside of the Harvard mandate is that the publishing by its faculty goes into an open access repository that resides in one place. What we need is not ten or twenty or a hundred or a thousand or two thousand separate institution-based open access repositories. We need interoperable repositories that, for the user, are seamless: I type in "dyslexia and digitality" and I suddenly have on my screen a list of every single scholarly article that not only uses those keywords but uses folksonomies of related words, hundreds of articles. I click on the bibliographic entry and, voila!: there on my screen is the article, free and open and right there for me to read.


Oh, utopia. That is open access.


But to get to that place requires all of us involved in the publishing process--scholars, readers, writers, publishers, university presses, university central administrators, distributors, electronic publishing experts, scholarly and commercial publishers, search engine experts, web 2.0 collaborative communications experts, libraries, foundations, philanthropists--realizing that all of our intellectual lives are interconnected. You cannot starve one person in the food chain and expect to have a banquet at the end.


You can't be pious about open access without thinking who is sacrificing what to allow YOU to have all the knowledge you want, for free, at your finger tips. It's the easiest thing in the world to say, "I want that!" But I also happen to want a new iPod. That doesn't mean when I go to the Apple Store tonight they're going to hand it over to me. My students want courses. That doesn't mean I'm going to teach without benefit of salary so they can have all the courses they want for free. Why in the world would we think open access to scholarly publishing would be our right? That just cannot work, my friends. Nor should it. If open access is about creating a community of scholars who can share ideas, we all must contribute. Me doing the work and giving my labor to you for free is not all of us doing the work. That's not many to many. It's me giving to you. It can't work. It can't be sustained.


Personally, I am not particularly happy with any of the current subsidy models and hope that we can move to a better, more far-reaching model that may end up being more a consortium approach across publishing and across libraries and online distribution systems. I don't know what that model is yet and, at this point, no one does, although lots of smart people are trying to think this through, with all the various publishing and distribution models in mind.


I keep thinking that Mellon or some other visionary foundation needs to get about twenty people in a room representing the most flexible, smart, collaborative people from each of the different parts of the publishing cycle. Two days later, those people come up out of that room with a plan. Of course, there has to be a funding pot o' gold as part of that plan. And the plan has to be based on the idea that we seek an interim solution, not a final one. Interim solutions tend to be more visionary, riskier, more entrepreneurial. Besides, transitional moments shouldn't have final plans. They need transitional ones that work for now. We try it. We see how it works. We see who gains, who suffers, and what the bottom line looks like for everyone in the process. Then, three years later, twenty more people disappear into a room and emerge with a different plan that includes fixing everything that went wrong with the first one. We need not just vision but collectivity and an equitable, collective solution. One reason I like the Harvard mandate is that it is exactly this kind of "let's try it and see what we learn" visionary approach. Harvard has the resources and the prestige to be trying such experiments and it is great that they have leadership bold enough to recognize the importance of an experiment at this transitional moment.


Open access can't exist without an open discussion of how everyone profits and what it costs in order for everyone to profit. We have to stop pitting libraries against university presses,proprietary lock-box archives one against the other. We needinteroperable and open access archives and open access publishing. Thatrequires cooperation across the entire spectrum of the publishingprocess. It's time, finally, for us to think about how this all works and time for us to make it work. Not just for faculty at Harvard, but for all of us, everywhere, who are concerned with open access, with making scholarship public, and who understand (and, again, this is part of the Harvard plan that has not received the kind of attention it deserves) that public scholarship is a public good and, like all public goods, it needs to find a way to be funded that is sustainable and equitable.







And here's the Harvard University Libraries Press release from February, 2008:


Robert Mitchell

Harvard To Collect, Disseminate Scholarly Articles For Faculty

Legislation Designed To Allow Greater Worldwide Access

Cambridge, Mass. - February 12, 2008 - In a move to disseminate faculty research
and scholarship more broadly, the Harvard University Faculty of Arts and Sciences voted today to give the University a
worldwide license to make each faculty member's scholarly articles available and to exercise the copyright in the articles,
provided that the articles are not sold for a profit.

In proposing the legislation, Professor Stuart M. Shieber said, "There is no question that scholarly journals have
historically allowed scholars to distribute their research to audiences around the world. But, the scholarly publishing
system has become far more restrictive than it need be. Many publishers will not even allow scholars to use and distribute
their own work. And, the cost of journals has risen to such astronomical levels that many institutions and individuals have
cancelled subscriptions, further reducing the circulation of scholars'


maybe of interest:

Keeping the fruits of knowledge within reach
The Berkeley Research Impact Initiative explores alternatives to
traditional publishing models

By Barry Bergman, Public Affairs | 27 February 2008

At a time when budget-strapped university libraries nationwide are
struggling to keep pace with rising subscription costs for scholarly
publications, Berkeley is exploring alternative routes to the
unfettered pursuit of knowledge.

For the next 18 months, campus researchers can obtain funding to
publish their work in open-access journals, which typically charge
authors a fee to make their work available online immediately upon
publication at no cost to readers. Funds are also available for
publication in traditional, fee-for-access journals offering an
open-access option to researchers willing to pay for the privilege.
The subsidies will pay for publishing costs not covered by grants or



Thanks for posting this. I'd heard about it but had never read the details. I think it is fantastic that we're all pushing the system, trying out alternative ways. Something will come of this. One important thing we need to do is educate our colleagues. We need them to understand why this is important, why it needs their attention.


Hi Cathy

First let me start with a procedural gripe. This <expletive> software forced me to go through all the rigamarole of settting up an account that will probably not be used again just so I could comment on your blog post. If you find you are not getting a lot of comments, that's probably why.

Now for my comment. Obviously you have a deep professional involvement with the issues of scholarly publishing whereas I am merely an onlooker - not even an academic (although there was a time in my life where if things had gone otherwise I might have become one). But is strikes me that a lot of the cost of delivering scholarly publications could be eliminated if paper deliverly was not the default mode.

I find this an awkward argument to make, because I am ordinarily of the view that virtual (i.e. on-screen) publications will never replace books and magazines for a variety of reasons, both emotional and ergonomic. That being said, I think the arguments against paperless publishing may apply less to the academic realm than to other areas. Basically - academic research is generally more work than pleasure. One has no particular expectation of curling up comfortably with ones favorite journal.

In a paperless model journals would still exist as distinct entities, but the default mode of delivery would be electronic. I think a paper option can and should exist, but this would be exceptional and the costs would be borne by the end user (or that user's sponsoring instution if they have one). This could range from informal laser printing of short articles for transitory use to some more elaborate type of print on demand to produce bound volumes of work where the circumstances justify it.

I would see "publishing" in this model tending to gravitate away from commercial publishers (although they might find a niche) and toward publication sponsored by academic institutions. I admit that by about halfway through your blog post I was thinking "let the university do it" (This is influenced by the fact that I graduated from Michigan State - a large state-supported institution with a variety of existing in-house publishing operations.) Looking at your idea for a consortium, it seems that that could work just as well - or perhaps a mix of the two.

The important point of my proposal is that electronic delivery as the default mode can reduce costs to such an extent that the economic assumptions of academic publishing can change radically.

Just a thought,



Hi Steve,


Thanks for this. And, over the smmer, we are totally redoing the HASTAC site to be more interactive, more Web 2.0. Right now, we are such a skeletal crew that moderating is extremely difficult so we have a relatively high level of entry for control. We hope to have some student interns helping us in the future. The problem with all-voluntary (ha!) is that it often means a lot of one person's labor for no remuneration.


Enough whining!


I agtree completely about the cost of paper. It's a real issue. The flipside is preserving all-digital materials is difficult and costly and ecologically as bad as the paper costs, when extended over time. And some forms of writing (booklength narrative designed to be read rather than searched) is, for some people, still preferable in a book version, as shown by numerous experiments of online v. downloadable versions published simultaneously.


Thanks for contributing to this conversation and helping all of us think about this together. There are no easy slutions and we need the wisdom of crowds.


Incidentally, I taught at MSU for a decade and I still bleed green. I love MSU.