HARVARD UNIVERSITY
OFFICE OF THE PROVOST

January 21, 2010

Dr. Diane DiEuliis
Assistant Director
Life Sciences Office of Science and Technology Policy
725 17th Street, NW
Washington, DC 20502

I write on behalf of Harvard University in response to the White House Office of Science and Technology Policy request for information “Public Access Policies for Science and Technology Funding Agencies Across the Federal Government.” In summary, I strongly support White House action to require and enhance public access to government-funded research. I provide our general recommendations, as well as more detailed responses to several of the nine particular questions that were called out in the RFI below. However, I emphasize that decisions on many of the detailed issues under discussion here and in the other responses to the RFI are secondary to the general principle of requiring public access.

I endorse the view that every federal agency funding non-classified research should require free online access (”public access”) to the peer-reviewed results of that research as soon as possible after its publication. There are three simple yet powerful reasons to take such a step. First, taxpayers deserve access to the results of taxpayer-funded research. Second, public access makes research as visible and useful as it can be, maximizing the return on the public’s enormous investment in the research. Third, public access accelerates research and all the benefits that depend on research, from public health to economic development.

The United States already recognizes the public interest in amplifying the impact of publicly funded medical research. A strong public-access policy has been in place at the National Institutes of Health (NIH) for more than a year (since April 2008). But the same interest calls on us to amplify the impact of publicly funded research in every field, from alternative sources of energy to public safety to American history and culture. The NIH policy has been good for professional researchers, good for lay readers, good for medical professionals, good for patients, good for the NIH, and good for taxpayers. The same principles should be extended across the federal government.

While legislation on public access, which I support, is currently pending in Congress, the executive branch can uniquely take direct action to provide for public access, and can nimbly respond as new technologies and conventions are adopted through applications of the fruits of research keeping new policies current. Even Harvard University, whose library is the largest academic library in the world, is not immune to the access crisis that motivates much of the campaign for public-access policies. In fact, the Harvard library system has gone through a series of serials reviews with substantial cancellations, and further cancellations will undoubtedly occur in the future.

With respect to some of the specific questions posed in the request for information, I provide our recommendations below.

1. How do authors, primary and secondary publishers, libraries, universities, and the federal government contribute to the development and dissemination of peer reviewed papers arising from federal funds now, and how might this change under a public access policy?

A public access policy will not reduce the need for publishers or the services they perform. It will just prevent the fruits of our large public investment in research from being locked up by publishers that provide access only to paying customers. The NIH policy, for example, does not bypass publishers or peer review. It provides public access to peer-reviewed articles accepted for publication by independent (i.e. private-sector or non-governmental) publishers. Other research-funding agencies in the federal government should follow the NIH policy in this regard.

If publishers believe they cannot afford to allow copies of their articles to be released under a public-access policy, they need not publish federally funded researchers. To date, however, it appears that no publishers have made that decision in response to the NIH policy. Hence, federally funded authors remain free to submit their work to the journals of their choice. Moreover, public access gives authors a much larger audience and much greater impact.

2. What characteristics of a public access policy would best accommodate the needs and interests of authors, primary and secondary publishers, libraries, universities, the federal government, users of scientific literature, and the public?

The public access policy should (1) be mandatory, not voluntary, (2) use the shortest practical embargo period, no longer than six months, (3) apply to the final version of the author’s peer-reviewed manuscript, as opposed to the published version, unless the publisher consents to provide public access to the published version, (4) require deposit of the manuscript in a suitable open repository immediately upon acceptance for publication, where it would remain “dark” until the embargo period expired, and (5) avoid copyright problems by requiring federal grantees, when publishing articles based on federally funded research, to retain the right to give the relevant agency a non-exclusive license to distribute a public-access copy of his or her peer-reviewed manuscript.

There are two compromises here to support publishers: (1) the embargo period or delay before the government releases its public-access copy, and (2) the use of the author’s manuscript rather than the published edition. For the length of the embargo period, publishers will have the exclusive right to distribute the peer-reviewed text, and for the full term of copyright they will have the exclusive right to distribute the published edition of that text, sometimes called the “version of record” (copy-edited, formatted, paginated, and so on). Of course publishers remain free to allow other kinds of distribution as well, for example, by allowing authors to “self-archive” their articles without delay.

These are compromises because there is a strong public interest in immediate or un-delayed access to the results of publicly-funded research, and a strong public interest in access to the final, published edition of the researchers’ results. For more on both, see Questions 6 and 7 below.

3. Who are the users of peer-reviewed publications arising from federal research? How do they access and use these papers now, and how might they if these papers were more accessible? Would others use these papers if they were more accessible, and for what purpose?

Public access helps professional researchers as well as lay readers. It helps professional researchers who don’t have access to the same literature through their institutions and it helps lay readers who generally don’t have any access at all. (Public libraries rarely subscribe to peer-reviewed research journals.)

It doesn’t matter whether many lay readers, or few, are able to read peer-reviewed research literature or have reason to do so. But even if there are many, the primary beneficiaries of a public-access policy will be professional researchers, who constitute the intended audience for this literature and who depend on access to it for their own work.

If the United States extends a public-access mandate across the federal government, then lay citizens with no interest in reading this literature for themselves will benefit indirectly because researchers will benefit directly.

Just last month, the Research Information Network released a report showing that researchers can discover new research more easily than they can access or retrieve it, and that access barriers slow their research, hinder collaboration, and “may well affect the quality and integrity of work produced….” <http://www.rin.ac.uk/our-work/using-and-accessing-information-resources/overcoming-barriers-access-research-information>

That is the primary problem for which public access is the solution.

6. What version of the paper should be made public under a public access policy (e.g., the author’s peer reviewed manuscript or the final published version)? What are the relative advantages and disadvantages to different versions of a scientific paper?

The policy should require public access to the final version of the author’s peer-reviewed manuscript. This makes the publisher the sole distributor of the published version, unless of course the publisher has consented to allow author-initiated self-archiving or some other form of distribution. The NIH allows willing publishers to substitute the final published edition for the author’s peer-reviewed manuscript in the government repository.

The published edition contains the final (often copy-edited) version of the language. It also contains authoritative pagination. For these reasons, the published edition is preferable for citing a paper or quoting from it. However, for the purposes of advancing research, it is sufficient for researchers to have access to the peer-reviewed manuscript. Allowing publishers to be the exclusive distributors of the published editions will help protect their business models, and will not harm research; hence, it is an attractive compromise.

7. At what point in time should peer-reviewed papers be made public via a public access policy relative to the date a publisher releases the final version? Are there empirical data to support an optimal length of time? Should the delay period be the same or vary for levels of access (e.g., final peer reviewed manuscript or final published article, access under fair use versus alternative license), for federal agencies and scientific disciplines?

The embargo period (time between publication and public-access) should be as short as possible. Because the public has a strong interest in immediate access, any delay is a compromise, and we should not compromise the public interest any more than absolutely necessary.

The NIH policy allows a 12-month embargo. However, I recommend a maximum six-month embargo, not only for other agencies but for the NIH itself. This is also the position taken by FRPAA. The NIH is the only medical funder in the world with an open access policy allowing an embargo longer than six months. Every other agency without exception caps the embargo at six months: the Arthritis Research Campaign (UK), British Heart Foundation, Canadian Breast Cancer Research Alliance, Canadian Health Services Research Foundation, Canadian Institutes of Health Research, European Research Council, Cancer Research UK, Chief Scientist Office of the Scottish Executive Health Department, Department of Health (UK), Fonds de la recherche en santé du Québec (Canada), Fund to Promote Scientific Research (Austria), Genome Canada, Howard Hughes Medical Institute, Joint Information Systems Committee (UK), Michael Smith Foundation for Health Research (Canada), National Cancer Institute of Canada, National Institute for Health Research (UK), Vetenskapsrådet (Swedish Research Council, Sweden), and the Wellcome Trust (UK).

When the European Research Council adopted its public-access mandate (December 2007), it adopted a six month embargo but added: “The ERC is keenly aware of the desirability to shorten the period between publication and open access beyond the currently accepted standard of 6 months.” <http://erc.europa.eu/pdf/ScC_Guidelines_Open_Access_revised_Dec07_FINAL.pdf>

When the European Heads of Research Councils (EuroHORCs) — representing all the major funding agencies in 24 European countries — recommended that its members adopt public-access mandates (April 2008), it also recommended that they “reduce embargo time to not more than six months and later to zero.” <http://www.eurohorcs.org/SiteCollectionDocuments/EUROHORCs_Recommendations_Op enAccess_200805.pdf>

When the Canadian Library Association recommended that Canadian libraries support public-access policies (May 2008), it wrote, “If delay or embargo periods are permitted to accommodate publisher concerns, these should be considered temporary, to provide publishers with an opportunity to adjust, and a review period should be built in, with a view to decreasing or eliminating any delay or embargo period.” <http://www.cla.ca/AM/Template.cfm?Section=Position_Statements&Template=/CM/ContentDisplay.cfm&ContentID=5306>

If embargo periods should be no longer than necessary, then how long is necessary? This is a difficult question, chiefly because it is difficult to choose the most appropriate criteria. Publishers have often suggested that public-access policies should use embargo periods no shorter than those that publishers voluntarily adopt for themselves. This, at least, is the wrong criterion. Either it presupposes that publishers are already trying to minimize their embargo periods, or it presupposes that funding agencies have the same interests as publishers. Both are untrue.

I am not ready to propose the exact length of the shortest necessary embargo. But I urge that the decision be made in light of this principle: the public interest in shortening delays and the private interests in lengthening delays must each give up something. The public interest bends by allowing some delay (any delay). If we allow publishers the same delays they voluntarily adopt for themselves, then there is no compromise; we simply subordinate the public interest to private interests.

9. Access demands not only availability, but also meaningful usability. How can the federal government make its collections of peer-reviewed papers more useful to the American public? By what metrics (e.g., number of articles or visitors) should the Federal government measure success of its public access collections? What are the best examples of usability in the private sector (both domestic and international)? And, what makes them exceptional? Should those who access papers be given the opportunity to comment or provide feedback?

Among the metrics for measuring success, I can propose these: the compliance rate (how many articles that the policy intends to open up have actually been opened up); the number of downloads from the public-access repositories; and the number of citations to the public-access articles. As we use different metrics, we must accept that we will never have an adequate control group: a set of articles on similar topics, of similar quality, for which there is no public access.

Publishers sometimes cite downloads from public-access repositories as evidence of harm to them. But downloads are not cancellations, and so far publishers have not shown that increased downloads from public-access repositories correlates with increased cancellations. I recommend that increased downloads be regarded as a sign of success, among other signs. It is a sign of meeting previously unmet demand.

Sincerely,
/signed/
Steven E. Hyman
Provost

In the case of the current Harvard policy, I’ll explain why for the time being at least we are not underwriting hybrid fees.  My comments extend the arguments I give in my PLoS Biology paper.

Hybrid fees would be worth supporting to the extent that (as their proponents claim) they provide an approach to smoothly transitioning to an open-access publication fee business model.  The argument is that as uptake increases in payment of the hybrid fees, revenue smoothly shifts from the reader side to the author side in a revenue-neutral way.  At the end, the journal is openly accessible, with hybrid fees (now become just publication fees) providing the revenue.  To the extent that the hybrid system works this way, there is no “double dipping” — using hybrid fees as a way to increase revenue rather than transitioning to OA in a revenue-neutral manner.  I’ll call such a model “true hybrid”.

There are two problems with supporting hybrid fee systems.  First, they may not be set up in the way just described as true hybrids, and determining whether this is the case may in fact not be possible.  Second, even if they were set up in this way, game-theoretic problems may inhere in the system that prevent them from realizing the transition.

Whether “double dipping” is going on is independent of when the hybrid option is set up.  In particular, whether a journal is hybrid from the start or a hybrid option is added well after the journal is founded is irrelevant.  What matters is the dynamics of how hybrid uptake affects subscription fees and revenues.  To be a true hybrid model, hybrid fees must be set at a rate no more than the true average revenue per article, so that universal uptake of hybrid payments is not a means to revenue enhancement.  Further, uptake of hybrid fee revenues must be directly offset by reductions in subscription fee revenues, so that incremental uptake of hybrid payments is not a means to revenue enhancement.

True hybrid journals must therefore have a degree of transparency that is difficult to imagine achieving.  The journal would have to demonstrate linearly decreasing subscription fees in direct proportion to the hybrid fee uptake.  Determining hybrid fee uptake is straightforward.  But determining subscription fee reduction is not.  Publishers practice price discrimination, bundling, and price changes over time, which separately and together make it impossible to tell what a subscriber’s costs would have been absent the hybrid fee discount.  The issue of transparency is not all or nothing.  Certainly some publishers, Oxford University Press for instance, may be more transparent than others in their hybrid discounting practice.  But determining whether appropriate reductions are taking place is difficult at best.

Even if there were some way to determine that a journal were a “true hybrid”, receiving no incremental revenue from its hybrid program, there is a more fundamental problem with the normal way in which hybrid programs are implemented, viz., that subscription revenue reductions are shared among all subscribers.  As institutions decide whether or not to underwrite hybrid journals, they are confronted with a kind of prisoner’s dilemma.

In a prisoner’s dilemma, a set of agents can either cooperate or defect.  If they all cooperate, all are better off.  But if any defect, all are worse off except for the defector.  In the case at hand, a cooperating institution underwrites hybrid fees, a defecting institution does not.  Given a status quo ante (as we have) of all institutions defecting, all would be better off if they cooperated, since they would pay the same amount (true hybrids being revenue-neutral) but would gain open access.  However, at the margin, an institution cooperating sees increased costs from the hybrid fees, but only a share (and a small one at that) of the subscription revenue reduction.  Defectors see a reduction in their subscription costs relative to the status quo ante at no cost to themselves.  Though both fully defecting and fully cooperating are equilibria,^[1] there is no transitional path from the former to the latter.  This is an inherent flaw in the hybrid system as traditionally conceived.  Empirical evidence accords with this analysis; uptake of hybrid fees has been exceptionally low.

It may be that institutions can be convinced to act temporarily against their financial interests in the expectation that others will join them and the cooperating equilibrium will be reached.  Especially in the current economic climate, I’m not sanguine about this prospect.  And without such temporary action against self-interest, the hybrid model is no transitional model at all.  If however, evidence accumulates that the transparency problem can be addressed and true hybrids can be identified and institutions appear increasingly willing to pay these fees against temporary self-interest, I can imagine that the Harvard fund might then change its policy for support.  Similarly, if changes to the hybrid model were made that eliminated the prisoner’s dilemma — I believe that this may be possible, and hope to talk about it in a future post — then again hybrid fees ought to be supportable for true hybrid journals.

Open-access journal publishing has been criticized on a whole range of grounds as being unsustainable, unfair, or ineffective.  Perhaps the starkest criticism is that open-access journals amount to a vanity publishing industry, and will exhibit a “race to the bottom” in which journals compete to lower editorial standards to capture the revenue for publishing articles.  Is open-access journal publishing prone to the problems of a vanity press?

There are both theoretical and empirical arguments that the concern is unfounded. From a theoretical point of view, the prerequisites for vanity press are not found in scholarly publishing.  From an empirical point of view, current open-access journals display a pricing structure that does not indicate a vanity press industry, as we demonstrate below in a new analysis of OA publication fee data.

What is a vanity press?

In traditional publishing, such as trade book publishing, authors want to be able to sell their writings to the public as an income source. Publishers provide services to authors to make this possible. Publishers have editorial and production expertise and marketing ability, and most importantly can provide authors entry to distribution mechanisms that make trade publishing scalable, and that are otherwise difficult or impossible to access.  There is thus a natural trade that can go on, authors contracting publishers to provide these services.

How much should the publisher be paid for this transaction?  Since high-quality books sell better than low-quality books, publishers have an incentive to use their efforts on the best books they can acquire. The better the book, the less the author will have to pay the publisher to take on the project.  Indeed, if the book is good enough, the author may arrange to pay the publisher negative fees.  These are called royalties.

In such a system, where publishers are attempting to maximize revenues, you would expect that different publishers might establish different quality standards, some pursuing high-quality books likely to sell well but requiring high royalties to acquire rights, others providing lax standards in return for high author fees.  This is in fact what occurs.  The latter type are just the vanity publishers.

In summary, the higher the publisher’s quality standards, the lower the fee the author must pay for services.  Conversely, if an author is willing to pay enough, some publisher will be willing to take on the publication.  This is the genesis of vanity publishing, and its hallmark is the inverse correlation between publishers’ quality standards and the fees they charge.

Since subscription journals are paid for by or on behalf of their readers, not their authors, there is no useful notion of correlation between a journal’s quality standards and publication fees, as publication fees are constant (and zero).^[1]

Open-access journal publishing, however, is likely to have to rely on article-processing fees for the bulk of its revenue.  (In fact, article-processing fees have a lot going for them as a funding model. In particular, they scale extremely well with the cost basis of open-access publishing, since the vast bulk of non-fixed OA publishing costs are first-copy costs.)  This raises the issue of whether open-access publishing might have the negative properties of a vanity press.

Why vanity journals might be a problem

The existence of vanity open-access journals would not in and of itself be a problematic state of affairs, just as the existence of vanity book publishers does not infect the entire trade book industry with suspicion.  The trade book industry offers a wide range of publishers along a spectrum from highly selective publishers who limit their catalog to books meeting only the highest standards to vanity presses that will publish anything for a fee.  In fact, it is well known that closed-access journals already exhibit a huge variation in quality standards, with some accepting most all submissions and others a tiny fraction; it is conventional wisdom that any article can be published some place. (Some journals, such as Elsevier’s Applied Mathematics and Computation, have even been willing to accept computer-generated nonsense articles.)

Rather, the worry is not about vanity journal existence but vanity journal domination: A profit-maximizing open-access journal can only increase revenues by increasing the number of articles published or raising the fee per article, both of which will require lowering its quality standards.  Other journals, seeing lower-standard competitors steal articles from them, will have to lower their standards in response to maintain their article flow.  A spiral of lowering standards will result. This race to the bottom could lead to a situation in which there is no longer a wide range of quality standards among journals, but only the low-standard vanity journals will be left standing.

You see this worry expressed in various forms.  Here is a typical example (via Open-Access Journals Flourish):

But if the researchers pay, doesn’t it turn journals into servants to authors, like the vanity-press publishers who publish anything for the right price? “In our capitalist society, one of our basic tenets is who pays the fiddler calls the tune,” said Dr. Jeffrey Drazen, editor in chief of the prestigious New England Journal of Medicine, at the national meeting of the Association of Health Care Journalists in Chapel Hill, North Carolina, on April 1.

It’s possible, Drazen said, that an open-access journal would find itself in deep financial trouble and loosen its standards about the papers it accepts.

The argument implictly relies on the inverse correlation noted above. With a positive correlation, a journal would need to raise its quality standards to raise its fee per article.

Peter Suber has recapitulated a broad range of counterarguments against the worry that OA journals have intrinsic quality problems. I concentrate here on a particular set of questionable economic assumptions that underly the worry about vanity journal domination.

Counterarguments

There are multiple assumptions at work in the argument that OA journals would lead to vanity journal domination.

First is the assumption that journal publishers are purely profit maximizing.  This is true for large parts of the journal market published by commercial firms, but non-profit publishers, including scholarly societies and independent journals, are set up not to maximize profit but to serve their scholarly constituency.

Second, even if a publisher is profit-maximizing, processing fees need not be publication fees.  Journals are free to charge for other aspects of their services.  In particular, it has been shown that the use of submission fees can eliminate any incentive to lower standards to improve revenue, even for profit-maximizing publishers.

Third, and most important, is the assumption that authors are willing to pay more for journals with lower standards, as they are for trade book publishing.  Whether you accept this assumption depends in large degree on what you think authors are “buying” when they pay a processing fee. To first approximation, scholarly journal publishers provide four kinds of services:

• management of article processing, including peer review
• production (copy-editing, typesetting, etc.)
• branding and imprimatur
• distribution

An author’s willingness to pay should correlate with the quality of these services, and all of these services (with the exception of the last in the case of open-access journals) should correlate with the standards of the journal.  Primary among the services is branding, which by definition correlates with standards.  Journals have a good brand exactly because they are selective in publishing only the best articles.

If you think that authors choose journals to publish in based on the brand of the journal (as a signaling mechanism to demonstrate the quality of their research), then authors should be willing to pay more for higher standards, not less.  In other words, journals should show a positive correlation between processing fee and standards, and the worry about vanity journal domination is ill-founded.

Now any academic will tell you that imprimatur is exactly the reason that authors select journals.  Academics must publish or perish, but not all publishing is equal.  Publications in high-quality selective journals weigh a lot more than publications in bottom-feeding journals.  Any journal that lowered its standards to raise short-term revenue would soon find its pool of submissions getting thinner and of even lower quality.  It seems obvious to many academics that journals ought to be able to charge more for higher standards, not less.

Still, it’s an empirical question. Which force is predominant in the journal publishing industry — the short-term profit-maximizing force that generates a negative correlation and vanity journal domination or a brand-payment force that generates a positive correlation and a range of journal qualities?  We answer this question shortly, but first a digression.

The importance of editorial independence

You hear less of this kind of charge that publication fees lead to vanity press behavior now that major scholarly publishers all have instituted “hybrid open access” charges, where if an author pays a substantial fee (typically $3000 to $3500), the publisher makes the article freely available.  Elsevier, Springer-Kluwer, Wiley-Blackwell, Sage, Taylor and Francis, and dozens of other publishers now offer this option.

One might worry that the availability of this option would lead to a journal willing to sell lower standards for authors willing to pay the hybrid fee.  The worry is unfounded for exactly the same reasons that OA journal fees need not lead to a vanity press downward spiral: any lowering of standards would reduce the brand value of the journal that it depends on for its market position.

The ability to keep considerations of short-term financial gain from affecting quality standards depends on independence of the editorial process.  Hybrid charges don’t lead to vanity press worries because the editorial decisions can be systemically separated from the financial decisions, with no knowledge of whether a person will be paying the fee leaking into the editorial process.  Publishers take pains to point out this separation in their descriptions of their hybrid OA programs.  (Says Springer, “There is no difference in the way that they are treated between Springer Open Choice articles and other articles among the well over 100,000 that Springer publishes annually.”) Similarly, it is crucial that editorial decisions not be affected by page charges, color figure charges, and any other author-side charges that a publisher might institute.  Nor should editorial decisions be affected by advertiser interests.  To the extent that subscription-based journals can maintain editorial and financial independence, open-access processing-fee journals should be able to do so as well, and are motivated to do so for the same reasons. The founding of the Open Access Scholarly Publishers Association is a sign that OA publishers appreciate the importance of sound practices to their authors and their readers.

Temptations to short-circuit editorial independence for financial reasons exist for subscription-based journals just as for open-access journals. Elsevier was discovered to have published several faux journals on behalf of and paid for by pharmaceutical companies, and many journals generate substantial revenue from advertising that might skew editorial decisions if not for strong editorial independence.

Gavin Yamey, an editor at PLoS, is quoted in Wired on just this point: “As for the vanity-press charge, Yamey said his company makes exceptions for authors who can’t pay. Editors aren’t in the loop on those decisions, however, ’so that cannot influence our decisions on which papers to publish.’”

An empirical study

I conducted a study to answer the empirical question that I raised before — whether open-access journals present the positive correlation between standards and publication fees indicative of a quality press or the negative correlation indicative of a vanity press.  As a proxy for quality, I used the Eigenfactor-based Article Influence metric, as the Eigenfactor methodology is likely to lead to metrics that are more comparable across fields.  Of the Eigenfactor-based metrics, Article Influence is appropriate to use as it, like publication fee against which it will be correlated, is a per-article metric.

Eigenfactors are calculated using citation data from Thomson ISI.  We therefore extracted from the Directory of Open Access Journals all journals that had a Thomson ISI Impact Factor, and extracted Article Influence scores for each one.

We also tracked down the fee structure for these journals by examining the journals’ web sites or contacting the journals by email or phone if necessary. We were able to acquire data for most, but not all, of the journals. (I am indebted to Tim Credo, Elmer Soriano, and Thomas Dodson for aid in acquiring this data.)

This data was used to calculate the correlation between publication fees and Article Influence.  For simplicity, we excluded in the analysis those journals that charge a submission fee, as that would require modeling the effect of submission fee as well.  As many have noted, submission fees in theory mitigate the vanity-press motive in any case, so the exclusion of these cases is a conservative step in the analysis.

A scatter plot of the data is shown in the following figure. In the figure I also show best fit lines for all of the journals (the green dotted line) and for those charging a non-zero publication fee (the blue dotted line). (The point in the upper right-hand corner is PLoS Biology.)

Graph of Article Influence as a function of publication fee for all open-access journals in DOAJ with publication fees and impact factors. (Click to enlarge.)

The correlation coefficient between standards and publication fees for those journals that charge them as calculated in the study was .70, a quite high positive correlation.  Even if we include the journals that don’t charge publication fees, the correlation is .47, which is typically thought of as a medium to large positive correlation.  The study clearly demonstrates that OA journals show a positive correlation between fees and standards indicative of a quality press industry, not the negative correlation characteristic of vanity publishing.

For those interested in the details, I provide (as previously) the data that we compiled, as well as the computer program that calculated the correlations and generated the plot. All journal, pricing, and currency exchange data were acquired in January 2009.

Vanity journal existence is not a problem in the OA world

The positive correlation that the study demonstrates implies that a publisher of an OA journal will be disinclined to reduce standards merely to acquire more fees, since a journal with lower standards will only command lower fees. As it turns out, there is a way for an OA publisher to solve this quandary. Rather than lowering standards for a high-tier journal, it can set up a new journal with lower standards and lower fees to publish the type of articles that the higher-tier journal will not, thereby obtaining the revenues without diluting the brand of the higher-tier journal. In fact, this is the approach that subscription journals have taken for years: increase revenues by founding more journals. The availability of this aproach is again consistent with having a healthy range of selectivity in OA journals; it provides a further argument against a race to the bottom.

Based on this study, then, one should expect that open-access journals will have a broad range of quality standards, from the most selective and expensive to bottom-feeding journals that will publish anything for a low fee. And this is exactly what we find.  The premier PLoS journals are as high quality as journals come.  On the other hand, Bentham Science Publishers journals are apparently willing to publish most anything, including the occasional computer-generated article.  Scientific Journals International, at a $99.95 processing fee per author per article, shows signs of similar questionable standards.

Does the existence of bottom-feeding journals pose a problem?  Ironically, such journals are far more problematic when they are subscription-based than when they are open-access. Though different journals have different average qualities, even journals that tend to publish lower quality articles trip over a decent one every once in a while.  And when they do, if they use a subscription-fee business model you’ll need to be a subscriber to read it.  At least if they use an OA business model, you can read it without having had to subscribe all along “just in case”. The existence of bottom-feeding open-access journals costs no one except for the authors gullible enough to submit their articles to them. Not so for the low-standards closed-access journals that bulk up bundles and place the very occasional read-worthy article behind a paywall.

Conclusion

Vanity journal domination is not occurring, nor is it likely to occur, among OA journals.  Vanity journal existence will and does happen among both OA and subscription-fee journals, but at least for OA journals is a benign phenomenon.  As subscription-fee journals more and more charge author-side fees, including hybrid open-access fees, one can only hope that the baseless vanity press recrimination against open-access journals will fade away.

We the undersigned universities recognize the crucial value of the services provided by scholarly publishers, the desirability of open access to the scholarly literature, and the need for a stable source of funding for publishers who choose to provide open access to their journals’ contents. Those universities and funding agencies receiving the benefits of publisher services should recognize their collective and individual responsibility for that funding, and this recognition should be ongoing and public so that publishers can rely on it as a condition for their continuing operation.

Therefore, each of the undersigned universities commits to the timely establishment of durable mechanisms for underwriting reasonable publication charges for articles written by its faculty and published in fee-based open-access journals and for which other institutions would not be expected to provide funds. We encourage other universities and research funding agencies to join us in this commitment, to provide a sufficient and sustainable funding basis for open-access publication of the scholarly literature.

MIT provost Rafael Reif says “The dissemination of research findings to the public is not merely the right of research universities: it is their obligation. Open-access publishing promises to put more research in more hands and in more places around the world. This is a good enough reason for universities to embrace the guiding principles of this compact.”

These universities realize that in the long run, underwriting processing fees for open-access journals is “an investment in a superior system of scholarly communication”, as Peter Suber says and as I have argued previously. As more universities sign on to the compact, joined by funding agencies as well, fee-based open-access journals may become an increasingly viable alternative to subscription-based journals.

Full details about COPE are available at http://www.oacompact.org/.

Prime Minister Gordon Brown has apologized on behalf of the British government for the appalling treatment of Alan Turing, who was obliged to undergo chemical castration for the crime of being gay. Prime Minister Brown’s statement in the Telegraph follows an online petition drive that enlisted over 30,000 British citizens and residents, and a follow-on global petition with over 10,000 signatories worldwide.

Much has been made in the discussions surrounding the petition efforts and in the Prime Minister’s statement of Turing’s code-breaking efforts at Bletchley Park, which directly contributed to the allied victory in World War II. Less mentioned, but also central to his legacy, are Turing’s seminal contributions to computer science. It is no exaggeration to say that Alan Turing was the progenitor of computer science, in his brief career providing building the foundation of theory, hardware, systems, artificial intelligence, even computational biology. His death at 42 as a result of the British government’s misguided “therapy” constitutes one of the great intellectual tragedies of the twentieth century. I commend Prime Minister Brown for his prompt and complete apology.

August 24th, 2009 at 2:04 pm

I think you missed Phil’s point, Stuart.

What Phil was saying is that libraries can’t control the disbursement of open access fees precisely because of academic freedom, which makes these fees more susceptible to unchecked growth and possible abuse. If they establish an OA fund, librarians will be on the horns of a dilemma — allow unchecked spending or violate academic freedom. Since they won’t violate academic freedom, their only option will be to allow unchecked spending.

I think you’re responding to a misinterpretation of Phil’s post.

via » Three good things: open-access funds, fiscal responsibility, and academic freedom The Occasional Pamphlet.

Yes, that’s exactly how I understood the argument, so perhaps I didn’t make my own view clear. Let me try to clarify it. (The issue is important enough and my response long enough that it makes more sense to devote a post to it than to place it in the comments thread.) Just so that my conclusion doesn’t get lost below, here is what I am saying: What libraries will do is check spending but without violating academic freedom. The dilemma as stated is a false one.

Kent says “libraries can’t control the disbursement of open access fees precisely because of academic freedom”.  The premise here is that any method an OA fund uses to control disbursement must if effective necessarily cause a change in behavior of authors, for instance encouraging them to publish in less expensive journals over more expensive ones ceteris paribus. This much is true. Furthermore, there is an implicit assumption that any such policy that causes behavioral changes in where authors publish is coercive and a violation of academic freedom. They are not “free” to publish in any location because some are financially more attractive to them than others.

But no. Academic freedom means that faculty can study what they want, and publish the results where they want.  It doesn’t mean that the university must cover all costs for doing so, nor does it mean the university cannot cover some costs and not others in ways that redound to what the university sees as the benefit of its constituencies.

Let me provide some examples in the area of “studying what they want”. Harvard has policies about what kinds of grants its faculty are allowed to take.  They can study what they want, but they can’t take money to do so if the grants violate university rules. For instance, university rules disallow grants that must be kept secret. (See Principle 2 of the Principles Governing Research at Harvard.) Nor can grants restrict investigators from publishing their results (Principle 4). And so forth. The fact that the university disallows investigators from taking certain grants is not a violation of academic freedom, even though it affects the financial support of certain research activities.

As a second example, the university itself provides funds to some researchers (but not others) to pursue research projects. (See examples here, here, and here.) It is itself, in a sense, a research funder. By playing favorites, and choosing among the applicants for these funds, is it violating academic freedom? No. No one brings up the horns of a dilemma between the university  funding every single application for research funds or violating academic freedom.

Faculty must be allowed to publish where they want, but it is not a violation of academic freedom to monetarily support some venues over others. As a third example, universities have been systematically supporting subscription fees over processing fees for decades by paying the subscription fees, but not processing fees (page charges, figure charges, etc.), through their library budgets. Universities could have provided funds for these additional charges and have generally chosen not to.Some journals charge only subscription fees; others charge both kinds of fees. It has not constituted a violation of academic freedom that authors are out of pocket $0 for the solely subscription-fee journals but must pay the additional processing fees to publish in those journals that charge them.

A final example: My colleagues in the History of Art and Architecture department publish articles that typically contain large numbers of figures reproducing the artworks that they are studying, the reproduction rights for which can come to many thousands of dollars per article. These charges are not reimbursed by the university, and grants to cover them are hard to come by. The fees typically come out of the author’s pocket. Some topics of research end up, by happenstance, to require very low rights clearance fees, others quite high fees. The differences in fees among these research topics may, for all I know, cause researchers to vary their research topics. But the fact that the university does not pay these fees does not constitute a violation of academic freedom. Scholars are free to study what they want, and to pursue funding to cover expenses to do so as available, and the university is free to support that research how it sees fit.

The fact that funds are differentially available for this or that purpose around a university may seem wrong, or unfair to one or another constituency, or well-balanced, or whatever, depending on where you sit. But the current apportionment is not fairer just because it’s been around a while. And raising the specter of violation of academic freedom in these cases belittles the importance of this crucial concept. If you want to see violations of academic freedom, read the AAUP’s Committee A annual reports.

Of course, if a university capriciously forbid publishing in this or that journal, that would be another matter, but no one is suggesting that. The suggestions, rather, are for support for certain kinds of charges for certain journals. Authors can avail themselves of these funds or not as they see fit, freely and without prejudice.

Just to make the OA fund proposals more concrete, let me list a few methods that allow for controlled distribution of OA fund fees without restricting academic freedom. (I’ll couch them in terms of the amount of reimbursement an author receives from a fund to cover processing fees for an article in an open-access journal, but funds may choose to disburse funds in other ways than reimbursement of authors.)

1. Cap the total reimbursement per paper.
2. Pay all but a fixed amount of the fee, that is, institute a “co-payment”.
3. Cap the total reimbursement provided to a single author per year. This method, discussed in my PLoS Biology paper, has the advantage that it forces authors to trade off the services that they receive from the publisher (including journal imprimatur) for the fee that they charge, providing exactly the economic market signal that the current system lacks.
4. Pay all but a fixed percentage of the fee, a percentage-based co-payment. This similarly provides an economic motivation to the author at the cost of placing them out of pocket for some costs for every article.
5. Steve Shavell, an economist at the Harvard Law School, proposed to me the following clever scheme: For each article, reimburse up to x dollars. If the fee f is less than x, some percentage p of the excess will be paid to the author (that is, p(x—f)), perhaps as an augmentation of a research fund. This provides a direct motivation for an author to search out high-value journals for each and every article, at the cost of increasing the total cost of running the fund.

None of these methods violates academic freedom (nor does removing all restrictions, as we’ve essentially done for subscription journals, but this could lead to hyperinflation of processing fees mimicking the subscription case). Several of them have the potential to introduce a self-correcting market signal into the payment scheme, thereby leading to controlled costs, rather than the runaway costs we see in the subscription model. I am not suggesting that all of these methods are equally good or effective or will solve the problem of sustainability of OA processing fee costs. They do not exhaust the possibilities; I am sure there are many other mechanisms that could be explored for setting up a properly functioning funding market for scholarly articles.  The only limitation is our creativity.

Here is a dilemma for you: Do we continue the status quo, which involves only supporting a business model known to be subject to uncontrolled inflationary spirals, or do we experiment with new mechanisms that have the potential to be economically sound and far more open to boot? To me, this seems like an easy choice to make. Academic freedom, thankfully, doesn’t enter the picture.

This new post, however, is so transparently spurious that it makes one wonder what Davis’s agenda is. Nonetheless, at the risk of giving the post more credence than it deserves, I’ll succinctly respond to the argument such as it is.^[1]

In summary, the claim is that open-access funds will either require vast amounts of additional funds, making them fiscally irresponsible, or will require aggressive filtering of claims on the fund, violating academic freedom. In fact, open-access funds are neither fiscally irresponsible nor contrary to academic freedom.

Open-access funds are not fiscally irresponsible

Open-access funds consistent with the “open-access compact” will require tiny amounts of funds in the short term (on the order of tens of dollars per faculty member per year based on extrapolations of current experience). My article on the compact discusses the issue at some length and I won’t reiterate it here.

Once open-access funds require large amounts of money (if ever), it will be because many journals have switched to a fee-based open-access business model, thereby freeing up subscription fees. We know that in aggregate there have been in the past sufficient funds in library budgets to pay journals for the services they provide. Moving the funds from subscription payments to publication-fee payments doesn’t change the macro situation for the worse.^[2] To the extent that the publication-fee model doesn’t have the manifest market dysfunctionalities that the subscription model has, it is in fact likely to improve the situation.

Open access funds don’t violate academic freedom

Davis argues that “Authors will view these [open-access] funds as a free lunch, and certainly much more appetizing than spending one’s own money paying those pesky page charges to non-profit society journals.” He assumes libraries will respond by filtering requests based on publication venue, and that will constitute an arbitrary imposition on where scholars can publish that violates their academic freedom.

Academic freedom requires that faculty be allowed to publish what they want, where they want. It does not require that universities pay arbitrary moneys to make that possible. That’s why it’s not a violation of academic freedom that universities don’t pay the thousands of dollars per article of page and figure charges that some journals charge. To the extent that universities add subsidies for some costs, that may change the incentives as to where to publish, but certainly doesn’t decrease anyone’s freedom. Thus, even if funds restricted disbursements, this would not constitute a violation of academic freedom.

Further, there is a variety of ways to set up an open-access fund that does not have the moral hazard that Davis imputes. The key is to make sure that funds for open-access charges are not fungible, as I’ve discussed elsewhere. It simply is false that open-access funds inherently can’t be set up in such a way that a reasonable market for publication charges ensues. On the other hand, we know that the existing market structure for the subscription-based model is broken for just the moral hazard reasons that Davis worries about.

Summary

It beggars belief that providing funds to make open-access journals more accessible to authors decreases their academic freedom, and only a lack of creativity limits setting up funds for that purpose in economically sustainable ways. If you want to worry about problems of open access, these are not the ones to worry about.

^[2]This macro argument has to do with costs in aggregate. Micro estimates based on extrapolating costs per article over the set of published articles are more problematic. The best known estimate, reported in a Cornell study, makes specific assumptions that indict the conclusions. But more generally, this type of analysis makes a range of assumptions about how contingent economic facts will be maintained despite the hypothesized wholesale shift in business model, for which there is no basis. (See for instance, this discussion about how moving from the subscription-fee business model to the publication-fee model turns complementary goods into substitutable goods, for just one example of how the markets completely differ.) For that reason alone, the sketched macro argument, which makes no such assumptions, is considerably more robust.

Scholars write articles to be read—the more access to their articles the better—so one might think that the open-access approach to publishing, in which articles are freely available online to all without interposition of an access fee, would be an attractive competitor to traditional subscription-based journal publishing.

But open-access journal publishing is currently at a systematic disadvantage relative to the traditional model.

I propose a simple, cost-effective remedy to this inequity that would put open-access publishing on a path to become a sustainable, efficient system, allowing the two journal publishing systems to compete on a more level playing field. The issue is important, first, because academic institutions shouldn’t perpetuate barriers to an open-access business model on principle and, second, because the subscription-fee business model has manifested systemic dysfunctionalities in practice. After describing the problem with the subscription-fee model, I turn to the proposal for providing equity for open-access journal publishing—the open-access compact.

We first took advantage of this possibility with an agreement with the American Physical Society. The APS wanted clarity on some issues regarding how the Harvard open-access policies would be used in providing access to APS-published articles before they could see themselves clear to fully supporting the open-access policies. The university was happy to provide that clarity in that our plans were wholly consonant with what APS wanted. The result was an agreement that was a win-win for both APS and the university. APS agreed to acknowledge the policy and not require addenda to their publication agreements (much less waivers of the OA policy). In return, Harvard made clear that for articles covered by the OA policy it would

1. Refrain from using facsimiles of the publisher’s version of the articles unless the publisher permits;
2. Not charge for the display or distribution of articles;
3. Cite to the publisher’s definitive version of the articles and link to them where possible;
4. Authorize others to use the articles only subject to these same restrictions.^[1]

The formal agreement and announcement are available at the web site of the Harvard Office for Scholarly Communication.

We’ve now concluded a large handful of such arrangements and have started listing the publishers and journals that have been supportive in this way in a listing of publishers who are “easiest to publish with”. The listing provides a resource for our faculty to let them know which journals they can publish in without waivers or addenda.  Already, we have affirmations from scholarly societies (APS, American Mathematical Society, American Economic Association), non-profit publishers (Public Library of Science, Berkeley Electronic Press), commercial publishers (BioMed Central, Hindawi Publishing), and university presses (Duke, Rockefeller, and University of California Presses). We expect more to be added soon.

Publishers interested in being added to the list of “easiest to publish with” should contact the OSC. We’re happy to work with publishers to simplify compliance with the open-access policies.