BEWARE: your web browser may be out-of-date...



Background


Over the past fifteen years or so, consolidation in the academic serials publishing industry has led to consistent price increases of around eight percent per year. This is at a time when, ironically, many journals are e-only so that the distribution costs should be going down. In case anyone wonders where this money is going, Elsevier recently posted over thirty percent profits (see page 6 of their financial report), meaning they could afford to give us all a twenty percent discount and still make a remarkably good profit margin.

Because of my position in the past as Chair of our University Library Committee I have become more involved in this issue. Our library budget has been generally flat, meaning that we have had to cut our collection, year after year. Such a process is not only depressing and ironic, as information is supposedly "more free" now than it ever has been, but is now getting to the point where faculty and students are regularly noticing what is missing. So over the past few months I attended two different conferences on academic publishing, one with librarians and one with administrators. I recommend that a few more faculty start doing the same, since faculty, librarians and administrators absolutely must work together on these issues, in some cases overcoming basic lack of understanding or mistrust of one-another.

Things can change for the better when we do put in the work. In my subfield, topology, twelve years ago there were three premier journals, all with private publishing firms who were then just at the beginning of their price increase binge. Some of the very top people in the field - faculty from Berkeley, Stanford, Columbia, etc - saw the writing on the wall and formed low-cost high-quality electronic journals, generally freely available electronically with print versions available as well. Typesetting mathematics is done by individual authors through TeX, and the editing and refereeing of papers has always been done by other mathematicians, so the tools to publish a journal electronically (additionally, archiving, bandwidth and basic support staff) were not difficult to acquire. The difficulties were more achieving the perception of high-quality - the all-star editorial boards helped - and convincing libraries to pay for something which was freely available. The latter is a significant matter and still sometimes a problem: if libraries do not pay, how do journals generate the small amount of revenue needed for operations, except through relying on some society or campus for support? But it is a hard sell to use money for a freely available journal when it could go elsewhere whose value is immediate, especially when library budgets are under enormous pressure. Indeed, some readings of the law in some states would forbid a University from purchasing an e-subscription for something freely available.

Those issues have been overcome (unfortunately, I do not know all the details for all cases), and now instead of having "Topology", "Topology and Its Applications" and "K-Theory", we have "Geometry and Topology", "The Journal of Topology", "Algebraic and Geometric Topology", "Homotopy, Homoogy and Applications" and "The Journal of K-Theory," all of at least the quality of their predecessor (in some cases fairly equivalent, as editorial boards resigned and new ones reformed with many of the same faculty) and collectively of much greater quality, at somewhere between one-tenth and one-half the price. Those who like metrics should note that "Geometry and Topology" has a remarkably high eigenfactor. Note that some editors of "Geometry and Topology" and "Algebraic and Geometric Topology" have started a publishing house called Mathematical Science Publishers so that more titles could be produced in low-cost venue.

By cutting those older titles and supporting the newer ones, our library is saving thousands of dollars each year. Math is well-positioned to take a lead in such issues, but I have heard similar stories of editorial boards starting up high-quality low-cost titles, sometimes after the editorial board of a similar high-cost title resigned, in many different fields, which gives me hope that we can address this problem effectively on a larger scale. But that hope is coupled with urgency. My University's collection and thus my access to some of the best current work are already being damaged, which impedes my ability to do research. But revenue models that can sustain such efforts on a large scale will require wide-spread faculty, library and administrator buy-in of such models. All of these groups need to think seriously about economic models which can lead to affordable access and sustainable library collections, then engage in serious dialogue to try to agree on and finally act to implement such models.

To encourage the first step I offer a few models which I have heard about or thought up myself. Note that some of these can and are being tried in tandem. In this current draft, I will not offer any suggestions about the transition to such models, which should be part of their evaluation, but I hope to soon. I will be changing this page to reflect feedback which I get. In fact, I would like to post a page based on parts of some of the replies which I am getting. If you do reply to me, please let me know whether you would like me include your comments anonymously, include them with your name and affiliation, or not include them at all. The default is to not include them at all, though they may be significantly paraphrased as in "some faculty feel that...".


Faculty-driven production of low-cost high-quality titles

This first model has a great amount of overlap with models below, and can be viewed as an key element of just about any (combination of) solutions. Moreover, it is essentially the model which led to the creation of titles which has happened so far. The need is for this to be much more widespread, which means being far more organized and in particular needing to get full librarian and administrator buy-in, which has frustrated some of the efforts so far, as well as of course having editors fully behind this. Universities, through their governance structures, and professional societies will play a key role.

Advantages This is already what is happening, albeit too slowly, so it is much easier to envision how this would happen than other models (we "just" decide and then do it). If we were conscientious about it, this is one of the surest ways to meet our needs.

Disadvantages As faculty who have gone this route will quickly tell you, paying the bills can be difficult, especially if you want material to be freely available. Relying exclusively on this model thus means in some cases placing a high burden on some of our editor colleagues. Some universities might be slow or unable to pick up new titles (because of budget freezes and the like) even if they are essentially replacing older ones. If in competing with these titles, the quality of our older ones goes down to the point that they are no longer widely subscribed to, then we would lose some historical continuity (we are currently paying a lot of money for historical continuity).


The author-pay model

This seems to be favored by some of the most knowledgable people about the serials crises. The idea is simply to have journals funded by fees paid by an author when the article is accepted instead of having libraries pay by subscription. Basic economics predicts that the journal market would become much more "efficient." Libraries' ability to manage their costs is limited by the fact that journal subscriptions are not substitutable. It does no good to buy two copies of a subscription of X instead of one of X and one of Y, if X is a lower-cost comparable journal to Y. On the author hand, if authors were to pay, they would naturally gravitate towards publishing - repeatedly if they liked - only in places which they felt were a good value. The market would become much more price sensitive.

Advantages Elegance - setting up the market to be efficient (which it is far from currently) and then being able in theory to let things run their course. Many journals, including by private publishers, are already offering authors the option of paying a fee to have their work openly available. Letting a properly functioning market work will allow for more continuity.

Disadvantages Very foreign concept in some fields, with a common objection "Why should I have to pay to publish something which I produce?" Would give authors with access to resources even more advantage in being able to publish wherever they wished. Would represent a significant shift of who ultimately pays for the publication of our research, being much more concentrated at elite universities (some might consider this an advantage, but it is at least a challenge for getting buy-in at such institutions.) Without accommodations for authors from developing countries this would be a disaster for research in these areas. Some early career mathematicians could be vulnerable.


The shareware model

This model complements the faculty-driven production model. In the software industry some software is freely available with the understanding that someone who uses a piece of software is morally, if not legally, obligated to pay for it. We could try to formalize such a model. The idea is to collectively guarantee a market for high-quality academic publications which may be distributed openly by having colleges and universities sign on to an agreement which could be called something like the "Sustainable Publishing Protocol." What this means concretely is that if for example a high-quality journal is formed at an open venue (thus freely available electronically) and the publisher and editorial board ask universities whose members use the journal to pay a reasonable price for a subscription, then as signatories we agree to do so if we indeed use the journal, even though we are "getting nothing" by paying.

Advantages Would guarantee a market for low-cost high-quality journals, meaning more faculty and their supporting institutions would take the plunge to bring our best work to lower-cost venues. Easiest option for authors, whose role would not change. Could lead to lowest reasonable prices through cooperation. Variations on this idea are being implemented in high-energy physics, through the Sponsoring Consortium for Open Access Publishing in Particle Physics and astrophysics.

Disadvantages Might lead to anti-trust issues depending on implementation. Some state universities might not be able to sign on if the agreement conflicts with their procurement procedures. Most natural fit is with open distribution model, which is not desired in some fields such as chemistry or psychology. Some professional societies might worry about a race to the bottom of the publishing market when they rely on journal revenues for their activities (though we must ask: why should our libraries underwrite the activities of some societies to a much greater degree than others?).


The author's institution/ funding agency pay model

While this is abstractly similar to the author-pay model, in which a good number of authors would end up having their institutions pay, its emphasis would give a completely different feel. Universities - or university systems, consortia, or funding agencies - pay in advance for the right for their faculty to publish. You can promote buy-in by making it less expensive to pay ahead of time than to pony up page charges when an author from a university which hasn't bought in tries to publish. You could also set incentives by asking the AAU and other bodies to consider the number of journals which are pre-paid in this way. You can think of this as a subscription, but the money is for something different.

Advantages This would probably feel most like an idealized version of our current model. Institutions pay an amount which would be on the order of what they would pay for a subscription now, but in a more efficient market setting. Authors would be less involved than in author-pay model. Costs would be spread widely.

Disadvantages Would potentially upset the balance between university and funding-agency financial roles, an obstruction to potential buy-in. With both universities and funding agencies potentially paying, would need to figure out cost-dividing if an author is "double covered." As in author-pay, absolutely crucial to make accommodations for authors from developing countries. As in author-pay, authors from some colleges or universities with less resources could feel shut out from some venues.


Mandatory archiving

This is the model being implemented now in the biological sciences, where according to federal law all papers based on NIH-funded research must be posted (in final form, but not the printed version) to the NIH's server PubMedCentral after an embargo period which could be up to 12 months. There is talk of doing the same for NSF-funded research. The existing law and new proposals have been vehemently resisted by for-profit publishers and are viewed suspiciously by some professional societies dependent on journal revenues for their operations.

Advantages There is movement in this direction already. Having such a mandate would mean that any faculty at colleges and universities which cut their serials collection would lose less access to the literature than they would otherwise.

Disadvantages Sustainibility is dependent on federal law. In most fields the percentage of research supported by federal grants is much less than the percentage in biological and medical sciences, so depending on such an archive for access would mean large gaps in access. Saving money through cutting titles also hurts morale, and relying on this as a predominant cost-saving mechanism could make it all the more difficult for institutions to achieve excellence (especially in some metrics) on limited budgets. Citation can be a problem for those with access to the archived but not the published versions.


Change in university budgeting model.

This model is offered in a bit of tongue-in-cheek sense. Economics is failing to rein in prices because faculty are the primary consumers of periodicals, but the money comes from library budgets. The idea is to add a subject's library budget to its general budget and let a department spend as it sees fit, pocketing the difference (or some fraction thereof) for the department's general budget if it doesn't spend the whole amount.

Advantages It certainly would get faculty thinking about the issues in a different light, and would as stated get the economics aligned. In particular, there would be incentives accross the board to support low-cost venues.

Disadvantages If such a system were imposed, most if not all faculty would chafe at being manipulated, at having to weigh their research resources against other budgetary pressures, etc. Such a system would definitely need to be faculty-initiated rather than administration-initiated. Faculty would also be suspicious about any cuts to their total budget which came after a department chose to make significant cuts in periodicals in order to use the money for other things.