Rumors have been circulating in recent weeks of an impending US Executive Order focusing on public access to federally funded research and open data. It appears that there is indeed a document making the rounds of Federal Funding Agencies for comment. The order has apparently been in the works for a while now, emanating from the US Office of Science and Technology Policy (OSTP), which has been tight-lipped about the the existence of the order. There seems to be little concern over the fate of non-profit and society publishers. Among the likely recommendations appears to be that of a zero embargo on published journal articles. Essentially, this means that articles from researchers who are federally funded will be freely available immediately following publication.

If you add this to the Plan S initiative from Europe, you may be forgiven for predicting the end of academic publishing as we know it. At the very least you may imagine the forthcoming discussions that will ensue, with hackles raised on all sides and little empathy shown for differing viewpoints.

The White House

Here I want to explore the environment. It may be useful to provide insight into what a zero embargo could do to the publishing landscape, as well as how researchers may respond. First though I thought it may be useful to understand exactly how an Executive Order works here in the US, especially for those who may be reading in other parts of the world.

Perhaps the first thing to understand is the nature of an Executive Order. There are three branches of government in the U.S. that supposedly balance power so that no one branch can rule the roost in isolation. The Legislative Branch of government makes the laws – this is Congress, including the Senate and the House of Representatives. The Executive Branch enforces the laws, and the Judicial Branch, interprets the laws. The Executive Branch concentrates power in the President of the United States, who also happens to be Commander-in-Chief of the armed forces. The notion is that independent agencies of this branch then enforce laws enacted by Congress. As head of the Executive Branch, The President – yes, take a deep breath and say the words “Donald Trump” — appoints heads for the many federal agencies.

In addition, specific offices may be set up as a part of the executive branch, and an example of this is the Office of Science and Technology Policy, currently headed by Kelvin Droegemeier, a political appointee. This office has a broad mandate to advise the President on the scientific, engineering and technical aspects of the economy, national security, health, foreign relations, the environment, and many other areas. Heads of Federal Funding Agencies, such as the National Sceince Foundation (NSF) are appointed by the President. The President can manage the Executive Branch through Executive Orders, which essentially orders Federal Agencies to abide by rules set out by the President. We are seeing a significant number of Executive orders flowing from our current President on a wide range of issues. There is arguably a blurring of lines between branches of Government, and some would say, abuse of the power of the Executive Branch with elements of democracy at risk. An Executive Order is not open for public discourse. The details of such an order are often not revealed until it is announced for implementation. In this case, Federal Funding Agencies have been asked to comment, though their concerns may or may not be considered. OSTP is central to the development of such an order.

Let’s assume we now live in the post-apocalyptic world of a zero-embargo policy for federally funded research. What does this mean for the publishing of scholarly articles and their authors?

On the one-hand, it seems natural that if an institutional subscriber is faced with freely available content, they may quite justifiably decline to pay a subscription fee. Whether or not this occurs will be highly discipline dependent. In a field such as immunology, a majority of a journal’s articles may be fully, or partially funded by federal monies. In mathematics, the field in which I work, this is likely to account for roughly 25% of articles published. Part of the problem is that journals have to be financially supported to thrive. Whether it be subscriptions, article processing charges (APCs), donor funding for Diamond access etc., one way or another the costs of publishing a journal and its articles need to be borne. If a journal is effectively made freely available, the journal will potentially no longer be able to exist. One can argue that in enacting an ideal of openness, the medium of expression will wither.

One area that I sense policy makers do not grasp well is that different fields have widely varying cultures around publishing academic articles. In the biomedical science, competition to be first is fierce. Content usage may well be high initially, but  will likely fade as new research is published. The article is likely a report on an experiment, rather than the article being integral to the research itself. In mathematics, at the other extreme, the article itself is the research. The elegance of a proof, the way an idea is expressed etc., matters. In math, as for many humanities disciplines, there may not be heavy usage, but usage will continue for generations. The mathematics remains correct, relevant, and useful.

Another issue that will be faced by all stakeholders, be they researchers, institutions, funders or publishers is the not so “civil” war between Gold and Green open access (OA). US funding agencies are currently agnostic to Green and Gold OA, with a 12-month embargo being the minimum requirement for Green. If there is a zero embargo, Green essentially dissolves away. Publishers of all stripes will be forced to look for publishing revenues in other ways, hiking up Gold APCs, or charging for Green. Journals may close up shop. There is grant money in the research ecosystem. The balance of access to these monies is the problem. In the end the burden will fall on a researcher’s budget. Those who have will survive, and those who do not may lose their jobs. What does this say about the very differing routes that Europe and the USA are taking? The rest of the world is likely looking at this rather ornery discussion with a bemused smirk.

Here in the US, it does seem hard to imagine that President Trump has a deep understanding of public access mandates. The real question here is why the OSTP is driving policy in this way? Are those drivers, engaging with researchers themselves? Researchers are already confused. Researchers are not equipped to disentangle such complexity – nor should they be.

Robert Harington

Robert Harington

Robert Harington is Chief Publishing Officer at the American Mathematical Society (AMS). Robert has the overall responsibility for publishing at the AMS, including books, journals and electronic products.


59 Thoughts on "Politics and Open Access"

A few comments.
1 “Publishers of all stripes will be forced to look for publishing revenues in other ways, hiking up Gold APCs, or charging for Green. Journals may close up shop.” To this one might add: increased costs for toll-access subscriptions which are now required to publish immediate OA for some significant portion of the articles in any given year. (Here I am assuming that the proposal on deck is that there is no provision by the government to pay for the immediate OA content. Government subsidy of publishing is a road we definitely don’t want to go down, to repeat an earlier point. Use of govt grant funds is already imo very problematic in funding OA.)
2. “In the end the burden will fall on a researcher’s budget.” Yup. And/or on libraries now expected to make up the balance of significantly increased subscription costs or APC costs to make up for the loss of revenue on the content posted immediate OA to fulfill the EO.
3. “In mathematics, the field in which I work, this is likely to account for roughly 25% of articles published.” This assessment is based on the number of federally funded publications in math journals? I don’t dispute the number, but prima facie I would have guessed that it was lower than this.

Hi Brian — some thoughts:
1) IIRC, at least for the NIH, publication costs are an allowable expense under grants. If publishing OA is a requirement of those grants, one would assume that this practice would continue and grant funding would be paying APCs for all federally-funded research.
2) Given the current state of library budgets, I would suspect the money to pay for this is coming out of research budgets. For a lab that’s relatively productive, this will mean a few postdoc/tech salaries that can no longer be paid or equipment/reagents that can no longer be purchased, as the funds will instead be going to APCs. Reducing the amount of research being done is a likely unintended consequence.

Two points:
1. I have asked my librarian colleagues in various contexts if they’ve even thought about setting an OA percentage threshold for cancelling journals (eg “would you cancel a paid journal subscription if the content was 20% OA? 50%? 80%?). I never get a fixed answer but I get the sense as a serials librarian that the figure is much closer to 80% than 20% so if your 25% federal funding figure is correct, you have no worries for your own society at least.
2. I don’t know about your society, but other major scholarly societies have admitted that they price their institutional (aka library) subscription prices high enough to subsidize the cost to their own members of attending their conferences and other services. As a serials librarian with a shrinking budget, I find this deeply offensive – our budget doesn’t exist to subsidize your members. Maybe you could live with lower revenues if you were actually charging us according to your publication costs. If AMS isn’t one of these, consider the “you” here to be the ones that do, not literally “you”.

I can never quite fathom the argument that libraries don’t “exist to subsidize your members.” Publishing a journal (or a book for that matter) is difficult, time-consuming and often very tedious work. Surely there should be some reward offered for taking on that work. I wrote about this at length here:

Research societies are, like many charitable organizations, ambitious, and often their goals exceed the pocketbooks of their members. Most researchers are not particularly wealthy, and it is worth remembering that most grant funds cannot be used to pay for memberships…And so these organizations will turn to other activities and services in order to raise funds. The local high schools where I live often hold bake sales or car washes to raise money for their class trips. Patient advocacy and disease awareness groups hold marathons or other social events to bring in funds to do work on behalf of their communities. Through journals, research societies are putting their members to work, taking advantage of their expertise and authority to provide a service that can help raise funds to be put toward scholarship, training, education, advocacy and the many other positive things that societies offer.

If one stipulates that a society can only charge break-even prices for their journals, then many (most?) societies will lose their motivation for taking on the costly and enormous workloads involved. And so what you’ll be left with is just the commercial publishers. Given that society publishers (according to several studies) charge lower prices and offer higher quality publications than their commercial counterparts, and that all surplus is returned to the community (unlike commercial publishers where profits go into the bank accounts of shareholders), this does not seem like a good alternative.

In the end, libraries are purchasing products and services. For each product/service, the questions should be around whether the pricing is fair and the value provided is high. If it truly matters whether the money going to purchase that fair-priced, quality product is then used to pay for researcher education and advocacy or for a new yacht for a company’s CEO, it strikes me as odd that librarians would favor the CEO over the researchers at their own institutions.

We don’t favor the CEO and we do cancel commercial subscriptions whenever we determine that the price is in excess of the value we’re getting. I’m responding to the special pleading I and other librarians keep hearing from the non-profit/scholarly society publishers who ask us librarians to have more lenient (in terms of cost per use) policies regarding cancellations than what we apply to commercial publishers, and which is now being extended, as in this blog post, to the larger picture regarding OA pressures. Your “surplus being returned to the community” is disingenuous at best. Let’s take AMS as an example, assuming it does have a surplus on its scholarly journal costs. That surplus is funding math professors having to pay less to go to their own conferences, while those extra funds are reducing our ability to purchase books for the other departments like English and History that rely much more on books. So underneath it all, the STEM fields’ societies’ surpluses are usurping more than their “fair share” of our budgets at the expense of the research needs of their HSS colleagues. That’s the reality on the ground for many university library budgets. And where are the math faculty when the library comes begging to the Senate, VP-Academic, or whoever controls the library budget, that we need more money? I don’t hear them speaking up on behalf of their book-using colleagues – they’re getting their precious journals and their low conference fees. I may be a little bitter this week – I’ve just been doing budget predictions for next fiscal year, and it’s really very painful. We expect for the third year in a row to start the fiscal year with absolutely no funds to purchase books thanks to the price increases in STEM subscriptions that more than eat up whatever meager budget increase we might get. Librarians at public universities all over North America are facing the same situation.

Strategic pricing usually includes a surplus; simple ‘at-cost’ pricing measures are not usually effective. Something you seem to be frustrated about is that researchers at other locations (lets say, AMS members/researchers at other universities) benefit from your dollars spent on society publications. As a former researcher, the way I see it is that you are using library dollars to invest in a society and a product that you consider to have value; it means a lot when the library has shared interest with researchers and are willing to use library dollars for those publications. In my opinion, part of maintaining that value (in publications) is an investment into the continual education of its researchers. If a society like AMS is gracious enough to subsidize conference costs for paying members/researchers, that is fabulous. large corporate publishers are not doing that (although they have other ways that they give back to the research community).

I am sure that your library budget either has at one point or currently still does include a budget for conference travel (I know many library budgets that are currently struggling but still have this line item). There could be stakeholders above you who are allocating money for the library budget and could make the same argument as you: “we are not here to subsidize your librarians”. But, wouldn’t you agree that a conference budget and subsidies would be incredibly beneficial for your work as a librarian? Does it not contribute both to the value of your library as well as the library research community as a whole? If we stop valuing this whole piece of community knowledge, then the meaning of publication will fade dramatically regardless of the status of OA.

The library budget is not supposed to be used for “the continual education of its researchers” – we have professional development funds for our faculty for that, and not a penny of that goes to the collections budget. You are mixing up the larger institution’s role in supporting researchers, and the role of the library budget in particular. If the VP-Academic (or Provost or whatever your equivalent is) wants to spend money to help support the scholarly societies to which the faculty belong, I have no problem with that. But pretending the library is “willingly” participating in this when in fact we have no choice as we are forced to do so in order to subscribe to the copyright-monopoly-controlled content that our researchers require seems naive. If scholarly societies want to start offering us two tiered pricing options, one with a cost add-on to “invest in the society” and a lower one without it, I’d support that so we can all see how many libraries are truly willing participants in this.

Does the library have similar policies about how the money it spends on office furniture or copy paper is used by those you purchase it from? Does it really matter what is done with the money as long as you receive good value for your spend? Isn’t having that money put toward the benefit of your own researchers (assuming if you buy a journal you have researchers on campus in that community) a win:win, and a much better value than just spending the same money and having it go to some investor’s bank account?

I was actually referring to specific line items in library budgets that go towards the librarians’ conference travel. I have witnessed that in the US, at the very least. I agree that the library budget does not go toward the PD of researchers outside of the library; however, some of that money can go for PD of librarians depending on the institution.

I think what is surprising for me to read in your post is that you don’t seem to be okay with being forced to participate in paying a price for a good where the price of that good is higher than what it should be and the extra money goes to things beyond our control (in this case, subsidies). That is typically the case with all purchased goods and tiered level pricing is never offered as an option, nor should it be. The price is what it is and if you don’t want to pay that then you don’t receive the good. We see it in our own lives – when we go to a restaurant, a meal may cost $20 but we didn’t receive $20 worth of actual food. Yet, we don’t refuse to pay for the cost of the good or offer to pay less (well, in some cases we might). There is an understanding that extra costs are ‘hidden’ in the price, some of which go toward training or education, etc. Libraries and the purchase of publications as goods want to avoid those hidden costs due to severe reductions in budget, but the issue really is not why there are hidden costs and what they go toward from a philosophical perspective. The frustration lies in the lack of value in that the university recognizes in the library budget and your increasing pressure to get creative in working to get the same amount of goods for less money. You don’t seem to have a philosophical issue with the societies subsidizing conference costs; it is only under the current budget pressures that now being forced to spend that money is an issue.

I don’t understand the expectation that revenue models for information good resemble that of material goods. I mean it would be awesome to have a car that all of my friends could use simultaneously and that’s utility I could access at any point throughout my lifetime and pay the same price as a normal car. Unfortunately, material goods don’t work that way and are priced accordingly.

I don’t understand why societies should only be charging to cover their publication costs. Big publishers like Elsevier and Wiley certainly aren’t charging to only cover their publishing costs, so why are societies held to a different standard?

Re: Kristina and others about hidden costs in everything else including for-profits: the point is that we librarians keep hearing society/non-profit publishers complaining that pressure from library groups and governments re OA is going to make them have to stop publishing their journals because they can’t afford to publish those journals if they lose (apparently any) subscription revenue, unlike the for-profit publishers who have lots of profit slack and can handle those added revenue stresses. I’m pushing back on that by pointing out that many society publishers do in fact have excess revenue in their journal subscriptions as well, revenue that isn’t being spent on the cost of publishing the journals, so it’s dishonest to claim that they are on such tight budgets that they can’t afford any revenue lost and still actually publish the journals. No doubt it will put pressure on OTHER parts of the societies’ overall financial situation, forcing them to pay for their other activities with other revenue instead of continuing to slide along on the backs of library budgets to pay for those other activities. It’s kind of a “marginal cost” argument in a way – I’ll have sympathy for the scholarly societies’ worries about Plan S or the new US Govt proposal that was the basis for the original blog post (remember that?) on the day that they demonstrate to me that the journal subscription revenue is already break-even for the journal production/publication costs and NOT also subsidizing other society activities.

There is no such thing as excess revenue or excess profit. A publishing enterprise, even one set up as a not-for-profit, is not simply a cost center. It invests capital and seeks a return. Your quarrel is with capitalism.

Having recently spent a significant amount of time consulting with the editors of a scholarly society journal (which only breaks even right now), I am wondering what it would take to demonstrate this. And, how would that happen? Some sort of registry? Melissa, I do hope you would you allow that there should at least be some reserves in order to prepare for one-time capital expenses or sudden emergencies (e.g., there is a flood in the editorial offices building and all of the equipment and furniture in the editorial offices has to be replaced) and the like?

This list of the percentages of income of societies gain from journals, may be helpful. It is a mixed bag. Some are definitely profiting.
The letter signed by 125 publishers and societies [and yes, Wiley and Elsevier signed too] to oppose the end of embargoes is disingenuous and written in Trumpian language. I have little sympathy.

Others can explain better the limitations of the data sources that are being relied upon to calculate the percentages here. I’ll just mention that one should be certain to take note of the dates for the information. I spot checked a couple of societies I am familiar with and there have been shifts since this was compiled.

Interesting data – I think adding conference attendance costs/income may show a more complete picture. Those costs may be more than the publications.

How could publishers ‘charege for Green’? Green OIA is self-archiving by authors. Publishers are not involved.

Publishers are involved because publisher-author contracts can (indeed, do) stipulate all sorts of things. Including no green and embargoed green. So, it isn’t a stretch to imagine, “no green unless you pay this fee” …

If a publisher were to upload a copy of the manuscript to the author’s institutional repository, I can imagine that service would be worth something to the institution.

Re. society use of their publishing revenues, there is imo a distinct issue that is more concerning than the question of how societies utilize these revenues.
To what extent is society pricing of journals in excess of whatever reasonable metric of inflation one should use in this context? To know the answer to that question, we need to know more about the intrinsic costs (which are significant) of publishing. The question is, could unreasonable year over year increases in society subscription costs be handled by tighter editorial policies, so that less stuff is published?
Everyone benefits, especially researchers, who now have less stuff, and of higher quality, to wade through just to stay abreast of even a sub-sub-niche of research. And there would be less need for so many genius-hours devoted to doing peer review.

Do you serve on your institution’s promotion and tenure review committees? Are you willing to lower your “publish or perish” standards for scholarship production quantity in order to reduce the pressure on publishers to accept as many manuscripts? We’re all part of a complex interconnected ecosystem of incentives and motivations, some of which are about profit, some about prestige, and some about professional survival. On top of that you have enormous populations coming “on board” in the world academic community that used to be insignificant “third world” consumers and are now major producers of new scholarship (obviously China, but others too). “Less stuff” is absolutely not what the future holds for academic publishing and the “more” that is coming is not necessarily of lower quality (although of course some of it is). Just look at how much more China is spending on science R&D in recent years: and a lot of that funding is going to lead to high quality publications that benefit the whole world to have access to.

The question is, could unreasonable year over year increases in society subscription costs be handled by tighter editorial policies, so that less stuff is published?

I’m not sure the two are directly tied — every article you reject carries some costs, and depending on the individual journal and field, getting to that point of rejection may be the most costly part of the process. And as Melissa points out in her comment, the funding/career advancement/hiring criteria require authors to publish in quantity, so if Journal X cuts its volume by half, those rejected papers are going to go to Journal Y which you’re now going to have to buy as well.

Indeed, it does cost to reject articles and if a quality journals cut in size, the demand side for quantity of publishing means that any number of other journals will publish their material. These are economic realities. Editorial operations cost money.
But where are the societies in clamoring for new standards in tenure and promotion expectations that benefit integrative articles, rather than piecemeal announcements of one small stage of a research endeavor? The latter fuels the surfeit of publishing, and correlatively, of numbers of journals.
Related to this, journals have come more and more to adopt the traditional role of the “letter” format in the sciences (cf. the preprint in today’s world). Namely they’ve become places for piecemeal publication of findings that are best published in a format that befits announcements of disclosures of new ideas, in those fields where this is a priority, even a necessity for professional survival.
The role of journals should be much more properly integrative and synthetic. (Well written review articles are among the finest achievements of the human intellect.) Of course, there is a “letters” genre in journals, but for certain STEM fields, notably physics, preprints arguably do the necessary work.
A faculty member in social sciences I talked to recently recalls a day when four or five components of their research would have been published in *one* journal article. Another STEM professor talked about how much stuff gets published that merely “rebottles” old research. An exercise for TSK readers: talk to your favorite scholars and ask them if this is a problem. Ask them if they are similarly overwhelmed by the quantity of stuff published. The vast majority of them will say yes, and agree that transmission of interesting research has suffered as a result.
Those scholars are members of societies. Are their societies concerned about a serious problem that their members would immediately acknowledge?
Societies (how many of them?) are now strong advocates of OA journal publishing. Wouldn’t it be as important for them to start advocating new standards of publication, not just new formats (OA) of publishing? And for development of new standards for tenure and promotion.

Meant to say: “But where are the societies in clamoring for new standards in tenure and promotion expectations that promote integrative journal articles?”–and less publication of sliced and diced research.
I concede of course that the large society publishers, at least the big ones, have entrenched economic interests that discourage them from advocating a contraction of the journal space. Societies of course have important functions to support, such as conferences. However, one would like to see discussion by societies of two topics: the tenure and promotion and grant-funding practices that encourage so much publication, and how societies in some future time can find new revenue streams to replace whatever is lost with contraction (not disappearance) of the journal space should those practices be transformed somehow.
It might seem that this outside the purview of societies, or for that matter libraries, on the basis that “we should not tell our researchers how to publish”. I disagree. Look at the activist role societies and libraries now play in advocating and supporting open access publishing. The transformative publishing schemes are now gaining a foothold in the U.S. Regardless what one thinks of these schemes (and there are serious concerns and long-run consequences that have not been confronted, imo), societies and libraries (on pains of inconsistency) cannot deny their now activist stance and interventions in how researchers go about communicating their findings.

“There seems to be little concern over the fate of non-profit and society publishers” – the post’s framing around society’s and nonprofits strikes me as a bit disingenuous, since the big commercial publishers aren’t even mentioned—and yet they’re the ones to be most affected. The closing point, with the gratuitous Trump reference and the point about not-confusing-the-researchers, does come across as pretty condescending.

No one dismisses the challenges of the OA transition for nonprofit society publishers, who of course currently rely on tolled-access revenues. But many learned societies are working through the issue with a variety of strategies, all of them on fascinating display in the recent Transitioning Society Publications to OA (TSPOA)/Society Publishers Coalition webinar series. Without dismissing the difficulties, the core point has to be front and center: The societies exist to serve their scholarly community, not to preserve publishing revenues. We shouldn’t let the tail wag the dog.

– the post’s framing around society’s and nonprofits strikes me as a bit disingenuous, since the big commercial publishers aren’t even mentioned—

I’m not sure why it would be disingenuous for the leader of a not-for-profit research society to write an opinion piece from the point of view of a not-for-profit research society. We do not have anyone working for a commercial, for-profit publisher writing for us, but would certainly welcome guest posts from commercial publishers discussing the impact that OA mandates may have on their business.

“The societies exist to serve their scholarly community, not to preserve publishing revenues. We shouldn’t let the tail wag the dog.”

Except that in most cases publishing revenues are what’s allowing the societies to serve their scholarly community. I’m not quite sure how many societies could continue to function as they do without a money-making program to subsidize other areas (conferences, awards, travel grants). Even though these societies are non-profit, their money has to come from somewhere. If your suggestion is to increase membership and conference fees, that means becoming a member or attending a conference becomes prohibitively expensive, and if this is the case can it really be said that the society is still serving its community?

This is exactly the discussion I mention, in general and in the webinar series that wrapped up last week, in which a number of major nonprofit publishers detailed their plans to transition (with attention to revenues) to OA, some considering subscribe-to-open, others various read-and-publish or tiered models. Many, many nonprofit societies (including the one I am serving on an OA task force for, the International Communication Association) are actively working through financial models for the transition that everyone agrees is coming.

Fair point. Not everyone agrees that it’s coming. But it is coming.

So Plan S combined with immediate OA in the US presents the question just who is going to publish the research? The only thing we know is that the very wealthy and the government enjoy free lunches while the rest of us have to pay!
I wonder if there is a market for journals that only publish articles which receive no government funding.
Robert will you be able to split out articles that receive funding and publish those in a special journal called: The Free Mathematical Journal of Funded Articles or is the society ready to simply reject funded articles.
Of course, there is the argument that with the internet there really is no place for journals which charge either the library or the author. After, all Google Scholar can be the indexing and archvers and readership can provide post reviewing. We can have a body of scientific literature that is based on polling. In fact, I would enjoy voting either thumbs up or thumbs down on something I know nothing about. Think of the sudden revival of alchemy and astrology as bona fide science not to mention the efficacy of miracle water!

Just some thoughts:

Math could thrive without journals at all, because everyone is posting on arxiv. Whats needed is some sort of symbolic capital-granting/refereeing mechanism on top of arxiv, maybe some sort of oa mathscinet layer. Those terrible venues like Annals of math should die off, all they do is symbolic capital distribution. Publication delays there are outrageous, and they are not OA, this means they are not working to provide scholarly communication, they work to provide scholarly reputation, and that could be done differently

Businesses (for that is what journal publishing is) are often disrupted by changes in their environment: they either adapt or they die. I see no particular reason why scholarly publishing should be exempt from this reality. Either scholarly publishing continues in a changing environment to deliver value that can sustain it as a viable business, or it can’t. Will research and education ‘come to an end’ if many journals cease to exist? I doubt it. Sure, many would miss journals – not least RPT committees which would have to turn to other ways to justify their decisions. I miss a number of things that are no longer economically viable, like milk and newspapers being delivered to my door, but it doesn’t mean I don’t get news or have milk to put in my tea. This leads me to a question that’s been bothering me for some time.

Let’s compare two data points. First, arXiv costs ±$10 per preprint to run. Second, a typical journal charges ±$3000 to (re)publish a preprint as a peer-reviewed paper (worth remembering that the number of changes between preprint and final paper is often negligible). Here’s the question that’s been niggling me: is the *value* returned against the ±$3000 it costs to (re)publish a $10 preprint in a journal – often many months after the preprint was posted – justified and/or worth it? (I emphasise the word value because this is at the heart of my question, not cost or price). If it is, then logically it doesn’t matter if there is no embargo, the value added to a paper being in a journal will keep the journal alive. If it isn’t, then an alternative world will emerge that might have far fewer journals and, like the milk in my tea, scholcom will find another way to deliver the value that’s actually needed to publish scholarship in a changed environment.

Toby, the proof is in the pudding. If a journal charges $3,000–or $50,000–and someone buys it, then it is worth it. If no one buys it, it is not worth it. These are not matters of opinion. The impersonal workings of the marketplace determines who is to live and who is to die. These are not moral arguments; there is no “should” in the economy, only what is.

(Not for the first time) I think we’re agreeing with one another. My feeling is that APCs have made the ±$2990 difference between preprint and article visible for the first time (before it was hidden behind subscription prices and big deals) and I wouldn’t be surprised if many in the economy will conclude that the low-cost preprint model will suffice.

I reject that. There has been a ‘should’ debate in academic publishing for decades, and especially since Robert Maxwell realised Pergamon could profit from huge markups, and things started to get out of hand. And a lot of us academics will refuse to publish in journals and with publishers that we do not approve of. If they wither and die, it is not the impersonal marketplace that does this, but ethical decisions driving behaviors. We also support libraries with their budget problems.

You have just described the impersonal workings of the marketplace perfectly. If something does not have value for you, don’t buy it. Now we can all go home.

In this landscape, will the old adage, “content is king” still ring so true, or will money be made on other areas like technology, workflow, analytics, etc.? I think we know the answer to this already and many players have moved their business away from content as a sole source of revenue. Finding problems to solve and new ways to deliver value will be crucial.

“The argument that immediate self-archiving risks subscription revenue does reveal an implicit irony especially where archiving of postprints is concerned. If the value publishers add to the publication process beyond peer review (e.g., in typesetting, dissemination, and archiving) were worth the price asked, people would still be willing to pay for the journal even if the unformatted postprint is available elsewhere. An embargo is a statement that in fact the prices levied for individual articles through subscriptions, are not commensurate to the value added to a publication beyond organizing the peer review process.”


I find it a bit odd to entirely dismiss the value in the peer review process, which takes the most time, effort, and cost for publishers. If it is so valueless, then why require the postprint? Shouldn’t a preprint suffice?

In other words, what have the Romans ever done for us?

I see a complementarity between preprints and peer reviewed journals, with the latter serving as guideposts for navigating the preprint space, as well as the antecedent peer reviewed journal literature.
This is all taken up in my arXiv-ed preprint about preprints. The third version of the latter will distance the “model” it espouses from the overlay journal model a bit. Also, it will make a more clarion call for peer review, traditionally understood. (I have to thank TSK and liblicense, among many other sources, for helping me –hopefully– make the arguments in that preprint more subtle.)
I think the preprint and journal models properly serve different but complementary needs in scholarly publishing. A project for next year is to create an online exhibit, in conjunction with the Special Collections department here at Lehigh, that displays how this longstanding theme has played out since the late 17th century in scientific publishing, through to the present.

Over in the social sciences, we don’t really use preprints much. The urgency to get out a new finding is not really there: a lot of work is not time-sensitive and the standard refereeing process for a journal publication seems to be accepted.
SSRN got some traction, but it was taken over by Elsevier. Today I know nobody that uses it.

Preprints are used by economists, probably because of long peer review periods. As for the overall trends in usage for economics preprints, I don’t know. Perhaps in that space working papers contends with the preprint format, as it probably does for engineers, who rely on conference proceedings.
As for other social sciences, see for example the uptake on PsyArXiv and SocArXiv here:
The verdict is still out on the long run uptake of preprints outside physics.

Though that’s changing. The nonprofit SocArXiv has, at last count (if I’m not mistaken) close to 10,000 preprints.

Economics is different. Not sure about psychology. Preprints and working papers count for little in careers for sociologists, anthropologists, human geographers, etc. – it is all about published refereed articles and a few books. A few drafts may appear on researchgate and academia .

Preprints are probably most suitable for disclosing results quickly in contexts where several research groups in a STEM area are potentially on the brink of making the same discovery and need to stake a claim to priority, or where there are t and p expectation of productivity in STEM areas, pending full publication. Or where there’s a need to float a trial balloon for an unusual idea.

I assume much of the work in such areas as sociology and anthropology is not of that kind. The research agendas of the latter are more unique and there are fewer dollars/grant incentives driving the research.

There is no sense, in Procrustean fashion, in imposing a mode of scholarly communication inappropriate to a field for which it doesn’t make sense. I suppose a question though is whether in the fields you mention there a habit has emerged that shadows STEM areas, of publishing piecemeal components of an unfolding research agenda. If so, it would be better to rectify that tendency through editorial policies than to advocate use of preprints. The latter in my view accommodate long standing needs for communication in the sciences, previously played by letter-writing or the “letters” format in journals. The need in at least some STEM areas for rapid disclosure is not going away, probably ever. That need is probably not universal to all fields of inquiry.

Preprints are probably most suitable for disclosing results quickly in contexts where several research groups in a STEM area are potentially on the brink of making the same discovery and need to stake a claim to priority

Recently had a conversation about this with a PI from a top research institute who insists that “publication is precedence”, that if everyone cites the Nature paper that announced a discovery and you’re arguing that instead they should go check biorXiv to see the timestamp on your preprint, then you’ve already lost that argument. Will be interesting to see how this plays out over time in the research community.

I am an early career publisher with a degree from the University of St Andrews. I was thoroughly disappointed by my experiences in Academic publishing and the utter disconnect that publishers had with the Academics, researchers, and members we were meant to be supporting. I decided to leave my role to volunteer in Greece at Action for Education, a human rights emergency education charity, and was absolutely infuriated by the lack of educational resources and support despite the fact that every university has a department researching exactly what I was experiencing first hand. I thought that research in psychology, mental health, human development, medical research on trauma, learning theory, international relations, and other academic disciplines could have helped me and the organizations I worked with save lives, yet Academic publishers, Higher Education Administrators, and professors are too busy squabbling over how they will make money and Impact Factors instead of how they could help people. I needed access to the research, but I could not afford the subscription fee – nor did not know where to start in reading the thousands of human rights papers that exist online. I also thought that the grassroots activists could have been creating invaluable reports and research as they were there everyday working with the people that Academics were studying from 4,000 miles away. I am thankful that this OA is getting pushed through. I wish that administrators and publishers would think of the next generation as well as developing countries and the citizens who have a right to access the research that they are paying for with their own tax money. Academic Publishers need to collaborate with Facebook, Google, and Netflix, get out of the ivory tower and make a real impact. I am trying to get involved and find out how to make sure that grassroot organizations and small charities have access to cutting edge research, especially if their country is already paying for the research. If you have any advice, suggestions on how to get involved in OA and scholarly publishing, or even if you think I am too hippie for scholarly publishing, I would greatly appreciate your comment or feedback.

Does anyone know if immediate OA would be required on final article (Version of Record) or on final peer-reviewed manuscript (post-print)?

If it is on final peer-reviewed manuscript without publisher formatting, the proposal would simply join the Belgium new law. It may still give space for subscription journals, Gold-OA, and other new business models to develop. Subscription journals would survive, taking money to finance new online services for customers-researchers (peer-review online system, post-reviewing system, corrections and retraction system, indexation by databases). Money would not go article itself, but for digital new services. Immediate OA on institutional or subject open archives would allow fair access to research (and still keeping) for everybody.
Summary: gold, diamond, subscription fees for online services by publishers, content access for everybody but without or with less digital services for researchers.
In this configuration, OA advocates and publishers should agree that proposal is interesting, and not only a storm.


I wonder what subject area the PI you mention is in and whether he/she can substantiate this claim more widely, which is based on n=1. Also, the fact that if a final settled version is more often recognized as establishing precedence should not be confused with the question whether researchers (at least those with integrity) –and editors– have an unspoken code of honoring the time stamp on a preprint. In fact, one wonders whether the Nature editor who vets a paper in your PI’s example would not honor a preprint time-stamp in deciding whether to publish a result from group X as opposed to group Y, in cases when both groups have submitted the same findings to Nature. One would hope that they would honor the first on plate. It would unethical for them not do so.

As I’ve argued, there are now natural experiments unfolding (involving European and now U Cal discontinuation of big deals) in which survey data would help assess whether the claims of this PI, or something different, actually reflects the views of most researchers, and whether the truth of the matter differs widely in different subject areas. We just don’t know enough at this point to make systematic or generalizable claims about the role of preprints.

My hunch is that precedence in physics is routinely established via preprints and that there are perhaps unspoken sanctions on someone who lacks the baseline integrity of recognizing a claim to priority as disclosed in a preprint. Maybe for purposes of deciding who gets a Nobel prize or a large grant there is cache placed on the final published version, but that is distinct from the question whether a preprint discloses priority of discovery, in actual practice. (How to explain the frenzy of activity to post to arXiv? It may have multiple motives but on the other hand, surely establishing precedence is an important motivator.)

I wouldn’t know about biology.

Anyhow, the question remains outstanding about the difference between the way things are done, and how they should be done.

One may argue that the market will decide this. That relies imo on too stark a devaluation of the role human will plays in the market. After all, look at how all the commitment to OA is creating a huge impact. This is at least partly human agency at work in attempting to change the face of publishing, even if one can identify market forces that also motivate this phenomenon. This is not an argument for OA, just an argument that activism in the publishing space can effect changes. (I personally think a lot of the transformative OA inttiatives will create huge unintended effects and that libraries and societies do not understand them. Things are moving too quickly on the OA front.)

In your post you state: “In the end the burden will fall on a researcher’s budget.” <– This isn't necessarily true. We'll have a chance to discuss this further soon. I, for one, can't stand gumming up scholarly communication with arbitrary transaction costs that have to be dealt with at the researcher level. Better business models that are less expensive all-around exist.

My concern with the transformative agreements is that they will look like good arrangements–at the outset. But then, per the historical dynamics of toll-access publishing, and the dynamics of APC charges, prices will increase after these models have established themselves. Who will foot the bill? If it is libraries, they will face opportunity costs in their abilities to buy things other than journal subscriptions.
I will wax cynical and even suggest that the publishers will be happy to participate in these schemes as a deliberate part of their business strategy–ride the wave for a while, then start elevating prices at the opportune time They are, after all, businesses.
I agree there are indeed new ways of doing things but they will take huge concentrated efforts to implement, . In my view, it will take decades.

Comments are closed.