At the end of February, the University of California (UC) system announced that it had ended a long period of negotiations with Elsevier. Those negotiations had been undertaken as the system’s collective agreement for access to Elsevier’s complete list of scholarly journals (a.k.a. the “Big Deal”) was coming to an end, an inflection point at which UC hoped to create a completely new kind of agreement with the publisher. The close of those negotiations leaves UC without a deal, though not yet — as we shall see below — without access. Jeff MacKie-Mason, University Librarian and Professor of Information and Economics at UC Berkeley, graciously agreed to answer a few questions for Scholarly Kitchen readers about the background of UC’s decision and its plans and expectations going forward.

Sather Gate, UC Berkeley

Can you give our readers a very quick summary of what you were asking for from Elsevier?

From the start of our negotiation with Elsevier in July 2018, we publicly stated our two goals:

  • An integrated agreement that covered access to Elsevier journals as well as default open access publishing for all UC corresponding-authored articles in Elsevier journals. That is, a “publish-and-read” or “offsetting” agreement.
  • An overall cost reduction commensurate with the value that we believe Elsevier journals offer.

For more detail, we are making every effort to be transparent, and we have posted a public summary of what we and Elsevier offered each other.

Why do you think they finally weren’t willing to agree? Were you surprised?

Of course, I can only hypothesize why they didn’t agree to our requirements for a contract. But since their offer (summarized in the document linked above) came close in several regards to what we proposed, with the main difference being price, that appears to be the sticking point. Elsevier sees open access publishing as a second source of revenue – they keep referring to it as a separate, second “service”, and that they should be paid for both services (reading via subscriptions, and open access publishing via article processing charges [APCs]).  So, they want a lot more money to publish our articles open access. We keep pointing out that historically they provided both publishing and reading for a single charge (publishing was free, and we paid subscriptions to read), and that we’re simply proposing reversing that (we’ll pay for publishing, let us read for free), but they say they want to be paid for both.

Were we surprised? Not really: in our experience (and the experience of many others!) Elsevier has been intransigent on this issue in the past. We hoped that it realized that times were changing, and that they might be ready to move forward, and so we engaged in good faith, very frank and energetic negotiations for eight months before giving up.

At any point, did you consider just negotiating for a price reduction instead of for a change in access/publishing model?

No. We set out both goals from the beginning and we never wavered.

Since the announcement of your Big Deal cancellation, there have been strong public statements of support from UC’s President and Academic Senate. Can you also tell us how you went about consulting with rank-and-file faculty before the decision to cease negotiation with Elsevier, and how they’ve responded since the announcement?

Good librarians are always communicating with faculty (not just elected representatives); for us, that communication has included open access for some time. And the communication has been two-way: it was the faculty who wrote and passed the first UC open access policy, in 2013, and the faculty have been leading activists on this front for well over a decade. (Yes, formally the policy was passed by the Academic Senate, but in UC the faculty senate is very powerful and is very highly regarded by faculty in general, and so there is a lot of engagement and communication throughout the ranks: the senate is not just a puppet debating society operating on its own!)

When we, together with the Senate committee on libraries (UCOLASC) and the UC Provost’s library advisory committee (SLASIAC) collectively decided in winter 2018 to change the way we negotiate with journal publishers, we began an extensive two-way consultation operation to engage faculty. On all 10 UC campuses we – librarians and faculty – held open town halls, met with the library committees, sent out open letters to all faculty and graduate students, published web pages describing what we were doing, and informed our departmental liaison librarians who then reached out to their particular faculties.

Reaction generally has been strongly positive: UC faculty care about open access, and they understand the very high and increasing prices that for-profit publishers have been charging, which have been cutting our ability to spend on other scholarly resources.

What’s the timeframe for enacting the cancellation decision? Will it leave any UC campuses with current access to any Elsevier journals?

We did not renew our contract, which ended 31 December 2018; since then we’ve had no legal right to access, via Elsevier’s website, its new (2019) publications and a small amount (by usage) of historical content (we have perpetual access rights to most historical content). However, as of this writing at the end of April, Elsevier has not yet turned off direct access.

When the cutoff occurs, no UC campuses will have access to current issues of any Elsevier journals: we had a single collective “Big Deal” license, and we are not entering licenses for individual titles.

What are the UC Libraries’ plans for the $11 million that will be freed up by the dissolution of the Elsevier deal? (And has the university administration committed to letting you keep that money for other uses?)

That we suddenly have $11 million of spendable money on our hands is a myth that’s been floating around; I’m happy to try (again) to squash it. It’s just too early to tell, for a few reasons. First, our scholars (faculty and students) will still want to read many articles published by Elsevier. We are prepared to help them obtain alternative, legal access. That will require extra staff effort, and in some cases, extra spend (for document delivery). Thus we will need some of the funds to pay for alternative access, but we’re not sure yet how much. The second important unknown is what the future holds for our relationship with Elsevier. Elsevier has made clear to us (and in public statements) that it still hopes we can reach an agreement, and we are open to that, as long as the agreement satisfies our goals. Thus, at some point we may resume negotiations and eventually reach an agreement. If so, such an agreement might include payment for perpetual access to articles published during the time we were out of contract, so we need to reserve some of the $11 million for that purpose as well.

Should we never return to a Big Deal contract with Elsevier, there will undoubtedly besome savings (though alternative access will always require some amount of funding). As to that, the funds come from each of the 10 campuses and the California Digital Library separately, and each will have to decide what to do with any savings. But speaking for my campus (Berkeley), and noting that 8 of the 10 campuses have signed the OA2020 Expression of Interest, we have committed to “converting resources currently spent on journal subscriptions into funds to support sustainable open access business models.”

What do you estimate will be the cost of providing alternative access to the Elsevier content to which you no longer have licensed access?

As I mentioned above, since direct access via Elsevier’s website has not yet been terminated, we don’t have hard data for this. We have internal estimates, but we’d rather not share our forecasts before we have data. We do plan to measure costs, and as with all of our work on open access and journal negotiations, we plan to share what we learn publicly.

Do you expect UC faculty and researchers to stop editing, reviewing for, and submitting articles to Elsevier journals?

That is up to them. The libraries are not taking any position on what engagement faculty and other scholars should have with Elsevier journals, because we think that is their prerogative. It is notable that there are already two faculty petitions published related to this, one of which is an explicit call for faculty to stop editing, reviewing and submitting (the other is a petition from editors of Elsevier journals calling on Elsevier to change its stance, indicating that many will resign if Elsevier does not). And those who have signed have done so before Elsevier cuts direct access – many expect that the numbers will grow rapidly if and when Elsevier cuts access.

UC’s “Alternative Access to Elsevier Articles” guidance indicates that “The UC Libraries do not endorse using Sci-Hub for article access.” Are the UC Libraries willing to go further and say that you actually _discourage_ the use of Sci-Hub?

That is not for us to say: the faculty and students don’t report to us, and just as we will not take a position on whether they should refuse to provide free labor to Elsevier during the dispute, we won’t tell them how to behave vis à vis copyright laws. All 10 campuses provide copyright guidance to faculty and students, and we are unequivocal: it is our understanding (though we are not attorneys!) that Sci-Hub is in violation of U.S. copyright law. We will not advise nor help anyone to use it. We do get lots of questions about it, so we provide the statement you quoted as a FAQ: we don’t endorse its use.

The Pay It Forward study concluded that a shift to open access would actually end up costing institutions like UC more than paying for access does. If your negotiations with Elsevier had been successful, do you anticipate the net result would have been a cost savings or a cost increase for UC?

That’s an overly simplified, and thus misleading characterization of one of the Pay it Forward conclusions. The analysis to which you referred investigated what would happen to the distribution of the cost of scholarly publishing if APC price levels were unchanged, and if other funding mechanisms did not emerge. That is, that analysis was completely static – in the world as it is today, what would happen if authors or their institutions paid full APCs? But if the entire business model of the scholarly publishing industry changes, then many aspects of the flow of funds will change, and, I am confident, also the level of prices. First, at the same time we are working to get publishers to switch from pay-to-read to pay-to-publish, there are a variety of efforts to develop mechanisms for routing publication funds to publishers. Many of these involve direct payment from funding agencies – rather than from authors and their institutions – to publishers. That is the model, for example, being followed by Plan S countries. The Gates Foundation has already run a pilot of this model with Science, and it has also signed on to Plan S. We hope that in time US funding agencies will recognize that they should directly fund scholarly publishing, rather than through the current mixture of indirect cost recovery and direct grant funding flows to universities.

Second, prices will adjust: as we economists say, the market equilibrium will change. In a pay-to-publish open access world, to the extent that authors pay any portion of the publication cost through their research grant funding, they will have an incentive to consider submitting articles to the journal with the appropriate prestige level and topic coverage that offers the lowest price (there is never just one journal that is acceptable for any article – there are always at least a few journals in a quality-and-fit equivalence class). That creates price competition between publishers to attract submissions, and that is the lifeblood for the publisher – without quality article submissions, publishers have nothing. And faculty sitting on editorial boards also want top-quality submissions, and will push the publishers to lower prices.

So, we expect that much funding won’t come from authors and their institutions at all, and that prices will fall so that publishers are making respectable, but not monopoly (e.g., 40% or higher) profit margins. And as the new equilibrium is reached, I think it most likely that even the most research-intensive institutions like the UC will see cost reductions, even in a pay-to-publish world.

I’ve written about this issue in more detail in a blog post I wrote a few years ago.

You have publicly said that the dissolution of your Elsevier deal won’t actually deny UC faculty and students access to Elsevier content, but will only make access less convenient. And yet it’s a fundamental principle of the open access movement that charging for access really does make content inaccessible to researchers. Is promoting open access really just about making access more convenient?

I’m an economist, and it is instinctual for us to see everything as a matter of degree. And that applies here. All Elsevier content is accessible without a license: it’s just a matter of how much time and/or money. First, a very high fraction of content is available in author accepted manuscript form (what we use to call “preprint” in the days of print!): much of it on open access repositories (like our institutional, or on subject-focused repositories like and PubMedCentral), or at ResearchGate or, or finally, directly from the authors (we like to be read – I’ve never denied a reader request for me to send one of my articles). Those all involve some time (often, just a few seconds, given how easy it is to find repository copies, e.g., via Google Scholar). Another approach is to ask the library to get a copy via interlibrary loan: again, a bit of time (typically somewhere between a couple of hours and a few days). Finally, one can purchase any Elsevier article as a single document, either via Elsevier’s website, or via a document delivery service. So, with some time or some money, everything could be available.

I can’t speak for the (mythical) “open access movement”, but our open access objective in the UC is for all published scientific research to be available immediately, to anyone, at zero cost, in its final published form. Time delay and cost are “inconveniences”, but they can be serious inconveniences, slowing the spread of knowledge and hindering the advance of societies through education and further research. And the final published form is important too: the author’s accepted manuscript (AAM), post peer-review, is generally accurate, but in the final published version there may be some copy-editing differences, and final page numbers: thus the AAM is not the version of record (e.g., we cannot cite to the final page number), which also gets in the way of the best and most efficient progress of science.

Is UC contemplating taking similar steps with other publishers at this point?

The 2018 statements by the UC Academic Senate, the UC Provost’s advisory committee, and the UC Council of University Librarians were calls to change the way we negotiate with scholarly publishers, in pursuit of our twin goals of reduced cost and universal open access. We did not single out Elsevier. Indeed, we just recently signed a transformative publish-and-read agreement with Cambridge University Press. And we have regularly stated that we are negotiating with other publishers as well.

Rick Anderson

Rick Anderson

Rick Anderson is University Librarian at Brigham Young University. He has worked previously as a bibliographer for YBP, Inc., as Head Acquisitions Librarian for the University of North Carolina, Greensboro, as Director of Resource Acquisition at the University of Nevada, Reno, and as Associate Dean for Collections & Scholarly Communication at the University of Utah.


38 Thoughts on "The University of California and Elsevier: An Interview with Jeff MacKie-Mason"

Jeff, afew comments, that may reflect some misunderstandings.
1. This stance presupposes that funding agencies should or will fund journal publishing. Doesn’t this just shift the burden to the taxpayer? Why should hard-working taxpayers fund journal publishing? Won’t this just perpetuate high pricing?
2. The stance seems to assume that elasticities will enter into author publishing charges. Thus far journal publishing has been quite inelastic. Can you go into more detail about why you think the inelasticities will not replicate themselves in APCs? Will a competitive market for journals arise with respect to APCs in a way that it has not among in the subscription funded toll-access model? RE: “And faculty sitting on editorial boards also want top-quality submissions, and will push the publishers to lower prices.” This hasn’t happened for the most part, in the aggregate, thus far with respect to toll-access publishing. While these faculty now be incentivized to push for lower prices because they are more directly impacted? (After all, it was the library paying before.)
3. I’ve been following scholarly publishing for a long time and have the sense that OA is a movement. In fact, it has become a mantra in the library world. OA is inherently good, but in my view has distracted attention from the economic dynamics that you are addressing and about which I have the questions above. One thing that OA advocates have clearly not focused on is the *glut* in publishing, which makes inter-generational transmission of knowledge much more difficult, as well as findability and synthesis of research results.
This glut drives high prices. Why not address that demand-side variable–namely, the drive (perpetuated by the tenure and promotion system) to reward huge output of articles?

For question 1, funding agencies are already funding journal publishing. Indirect costs are a significant portion of the funding for my library’s (UCSF) collection budget.

For question 2, several things – first, there is a hope that APCs are a transitional model and other funding models will arise; second, there is a hope that if faculty are indeed paying APCs from their grants or other funding, they will be responsive to pricing when selecting publication venues.

In regard

Re. 1., this must vary significantly by library and university. (Incidentally, it would be good to see how many libraries across the land use their dollars for this purpose.) In any case, there are in my view plausible questions as to whether this use of public funding *should* be the practice, and whether the practice should be expanded. I assume the UC system would have no problem with it expanding, but please disabuse me if I’m wrong. Concerning expansion of the practice, should taxpayer dollars go toward fixing a broken system that universities themselves should attempt to fix, demand side, by making efforts to contract the stock of journals for which universities are willing to pay, and by attempting to influence tenure and promotion criteria? Related to the them of public funding, elsewhere you mention that “The UC Libraries are committed to paying the full cost of APCs for our authors who are unable to provide their own funding, for whatever reason.” Is this a good use of taxpayer dollars, relying as it does on a theory that this strategy will work to bring down prices?
Re. 2., yes, this is the hope, but I have not seen sufficient reason to think that this microeconomic assumption that faculty will respond to APC pricing in a way that will bring down overall prices for journal publishing (either APC charges, or tolled access) is at all correct. It seems that this assumption about faculty preferences and future behaviors is now the key argument of the UC system. But why wouldn’t publishers on the supply side merely “up” the prices for erstwhile low cost APCs, in response to faculty demand to publish in the journals that have low APCs? Unless you can assume that there is inelasticity of supply, but that again is entirely an assumption.
I don’t mean to be difficult. All of us in libraries want prices to go down. And OA is an intrinsically good thing. It’s just that the proferred arguments (in my view) are not convincing. Ultimately we are talking about stewardship of public dollars.

I sense that you’re taking something of a free-market approach in wanting the “customer” (universities) to fix this system through demand/sales pressure and wanting to leave “government” out of it. The problem is that this problem is caused by government, namely the granting of unregulated monopoly power to publishers through the form of “copyright” law. I won’t get into a debate as to whether copyright is necessary or not in this market, but because it does exist and unlike other government-granted monopolies is fully unregulated as to pricing, there is a role for government to play in solving the mess that public policy has wrought.

Yes to the first sentence. Partly this is philosophical, growing out of a commitment to the principle of subsidiarity. We don’t want government meddling with publishing. Dangerous for many reasons. Also, government funding is unreliable, even if one didn’t have these philosophical concerns.
I’m not averse to busting trusts or monopolies, or averse to government regulation of some aspects of the economy. But that’s a far cry (in my view) from actively perpetuating a dysnfunctional system in deus ex machina fashion. Wholly aside from the worries about government encroachment mentioned above.

When I replied, I hadn’t seen Jeff’s very detailed response below. I would refer you to that regarding the source of funding for library collections.

In regard to question 2 about APCs driving researcher decisions, the reason we proposed a three year trial was to get empirical evidence that would allow us to answer the questions we and you are asking. We don’t have that now.

A few additional comments that occurred after posting. (Usual comment: any comments reflect my own views, not those of my employer.)
1. Why isn’t the UC system just focusing on contracting incrementally the number of journals to which it subscribes, toll-access or otherwise? The following suffices in itself as a rationale for challenging bundled pricing schemes: create lower demand for journals (many embedded in bundled pricing schemes) by refusing to pay for so many subscriptions–Bradford’s law could help here–with the consequence that prices go down. In tandem, don’t get into the practice of funding APCs as a way to fix the pricing issue. The approach of not subscribing to so many journals is not a solution that will work immediately, as publishers may then increase their pricing to maintain their pound of flesh. But one could see long run price decreases as a result.
Also, start promoting, on the part of university and society publishers, a reconfiguring of the goal of peer-reviewed journal publishing to discourage salami-sliced publishing of so much peer-reviewed stuff.–along lines that accord with the practices of scholarly publishing going way back. Also educate the faculty about how t and p practices contribute to this bloated publishing monolith.
2. Certain things have worked for centuries in scientific communication; let’s learn from them. The physicist Steven Weinberg suggested a Burkean approach to publishing, certainly correct, but in my view consistent with some re-calibrations to fix problems in the current system. For some ideas that will not be realized any time soon, and that are at most useful in–perhaps–suggesting an analytical or heuristic framework of diagnosis, see my preprint about arXiv on ResearchGate, now under revision. It argues for a complementary system of journal publishing alongside reliance on preprints for physics. (The focus is on physics–the verdict is out on whether this scheme would work for other areas of science.)
3. My initial reply to your posting mentioned “inelasticity of supply”–would be better phrased as inelasticity in APC prices imposed (supply side) by publishers. I suspect they are much more elastic than the UC system appears to presuppose.
4. There needs to be much more emphasis by libraries and librarians on the extent to which high prices–whether for APCs *or* for toll access publishing (which is not going to go away anytime soon) impose a barrier to access. The constant refrain of “immediate open access” overlooks the fact that many institutions cannot access high priced toll-access journals, placing a burden on ILL and helping indirectly to sustain illegal operations like SciHub (a dysfunctional response to a dysfunctional system.) The “immediate OA” refrain also, in my view, has harmed the credibility of the librarian profession. It has turned into an ideology, but that is another story.
5. There are things that universities could do to create a new direction for scholarly publishing, but I doubt very much that this will happen any time soon. Far more likely is the gradual implosion of the system purely because of the market dynamic of so many universities being able to support run away costs. If and when that happens, the conditions will emerge under which new approaches will come into their own. Will be interesting, whatever happens.
6. The large commercial publishers are amazingly agile, and more than happy to go along with the immediate OA imperative. This is not an anti-business statement, just a reminder that decreasing their grip has to be met by academia with a cold economic realism of the kind that animates the commercials. The societies have somewhat different incentives than the commercials, but they too have strong economic incentives to fund society activities.

I think a very good use of the $11 million in savings from the cancellation of the Elsevier deal would be to give it to the University of California Press, which could use it to invest in new programs.

Jeff, I was surprised to see you characterize the open access movement as “mythical.” The Max Planck Institute believes it’s a real thing (and, in fact, claims explicitly to be a “co-founder of the open access movement”). The Free Knowledge Institute also thinks it’s a real (and successful) thing, as do the Right to Research Coalition and Jon Tennant. Robert Jan-Smits talks about the “movement” as if it’s a real thing that real people are a part of. Are you using the word “mythical” to mean something other than “existing only in the imagination”?

I also think so ( it is real thing difficult to attain but REALLY REAL !!!)

I think the funding questions are extremely important here, largely because we’re talking about a zero sum game. As far as I can tell, no Coalition S member has pledged to increase the existing amounts of funding to universities and researchers in order to pay the additional costs of author-pays OA, and certainly, given the recent statements from the US Government about OA (, it’s not likely that any agencies there are going to increase their funding levels to cover these costs. So what’s being talked about here is money that’s already in the system — either coming directly out of research budgets or out of indirect funds from the university. Have UC researchers with funding pledged to cover their own costs? What about those at UC without grants, where will their funds come from? Is the UC willing to devote a portion of indirect grant funds to fill the significant gap that will be created? It’s one thing to have faculty and administration voice support for a policy, it’s another to have them back that up with money out of their own existing budgets. What is the current level of faculty compliance with the UC’s OA policy?

I also think it’s somewhat contradictory to argue that “much funding won’t come from authors and their institutions at all” and at the same time suggest that the economics of the entire system will undergo a major shift because authors will be paying out of their own pockets and that will introduce pricing pressure.

Another question I’d pose is whether taxpayer funding of journal publishing doesn’t erode already scarce public research dollars. Therefore hindering the advancement of science.

Curious what rationales are being offered by other universities for their recent expressions of opposition to the Big Deal. What weight are these other institutions placing on pricing as the problem, versus provision of OA? What economic mechanisms do they think are at play that could achieve their goals?

It’s worth noting that many (most?) research grants from public agencies include a flat rate “indirect costs of research” dollar amount that goes to the institution outside of the particular lab/project doing the research work. Some of that money in turn is directed to the library budget to help pay the costs of the journals that those researchers use as part of their research. So the tax payers are already paying some of the journal publishing costs through the research grant dollars, just much more indirectly.

How common is this channeling of funding dollars to subscriptions, however indirectly? I assume it is up to the discretion of an institution, but don’t know.

Yes, each institution decides how to allocate the indirect costs funds throughout their own budget.

Yes, many funders (the NIH included) allow grant funds to be used to cover publication costs. I think the important question is whether those funds are additional (as is the case for Wellcome) or if they just come out of already existing funding (like the NIH). How many funders are pledging to increase their current spending levels to cover these additional costs? Or does the money just come out of the same pool of funds that a researcher would otherwise use to buy reagents, equipment, and services or pay salaries for lab members?

The question is how all of that gibes with the concurrent demand for decreased spending, given existing market conditions and UC’s own study that showed that following this path would result in increased spending levels.

I’d refer you to Jeff’s comments about the limitations of the Pay It Forward study, which you can find below.

Maybe this implies that APCs induce elasticities in pricing. I don’t know. Maybe, maybe not.
Has any of the academic work by economists on the dynamics of journal pricing addressed this issue?

I don’t know the answer to either question. But I think it will be interesting to continue to see how faculty respond to policies that both restrict their activities (as the Plan S response letter from Lynn Kamerlin and other chemists notes) as well as taking what they see as their own money out of their laboratories (as is suggested here). We’ve long wanted the actual researchers to join the conversation, which has long only included publishers, librarians, funders, and advocates. So a really positive result of these recent steps toward regulation of researcher activities is that it may drive them to engage and help better guide future policies.

In point of fact, many UC faculty are refusing to edit, reviewing for and submit to Elsevier Journals. I have heard this from my own sister and some of her colleagues in Engineering and Computer Science at UCI. She was asked to edit a book for them, and refused, saying she wouldn’t do any work until the matter was settled. They asked if she would recommend someone else, and she replied that doing so would amount to doing work for them. This is probably easier for a senior full professor, but the tide is definitely turning.

But is this a pro-author-pays-OA tide, or an anti-Elsevier tide that’s turning?

I’ll respond to a few of the questions above.

– “This stance presupposes that funding agencies should or will fund journal publishing.” First, they already largely are; as Melissa Belvadi pointed out, a large fraction of research library budgets already go into the calculated indirect cost recovery rate. Don’t get confused (as so many do) by the red herring that “that money isn’t earmarked by the university for library budgets” — money is fungible. The indirect is part of the university’s central resources: it’s a pile of dollars, mostly doesn’t matter where it comes from. And the university decides to give some of those resources to the library for journal subscriptions. And (usually), the more the library spends on journal subscriptions, the more the university gets in indirect revenues.

Put another way, where do you think research library funds come from if not tax payers (including tuition payers)? Manna from heaven? We get some philanthropy, but most of that *is* restricted as to use, and it’s not for subscriptions (and for most research libraries it’s a small part of the budget anyway). Most is from state taxpayer money (if public), research agency funding (indirects), and (taxpayer) tuition money.

I think it would make sense to redirect some of those monies so that the portion that funding agencies are paying be paid more through direct research funding than indirect, but our strategy doesn’t presuppose this.

By the way, though the cost of subscriptions is a high (and growing) fraction of library budgets, it is a *tiny* fraction of total research spending. Look at slides 34 and 35 of our CNI presentation ( paying APCs to publish *all* articles open access would cost only about 1% of all US federal research funding. That’s hardly going to have a massive negative effect on research — that’s less than one year of inflationary increases (that is, roughly every 6 months the purchasing power of the federal research budget goes down by more than this, but we manage to increase the budget enough so that US research isn’t collapsing). (Slide 35 shows that under our proposal, which still has the libraries paying most of the cost of publishing, the *direct* research funding cost would be about half a percent.)

– Is there “elasticity”, or responsiveness to changed incentives? Authors are paying part of the cost of publishing directly, then, when they are deciding to which journal to submit a particular article, the price the journal charges might be one factor they consider. Yes, prestige may continue to be the top factor, but for anyone given article authors virtually always consider a few options — however good you think the article is (we don’t all submit all of our articles to Nature) there will be a few roughly prestige-equivalent journals to consider. I’ve only been a librarian the past 3.5 years; I’ve been a regular faculty member for 33, and have published about 80 articles. I can’t remember a single one in which I (and my co-authors) didn’t discuss multiple factors in deciding where to send the paper: prestige, typical turnaround time, closeness of fit to the topic of the paper, etc. If we were paying part of the cost and one journal charged more than another, we would have considered that too — knowing that if we saved some of our research dollars — even just a few hundred — that might pay for another conference trip by one of our students, etc.

Brian Simboli said he hasn’t seen authors showing much elasticity before — but then he answers his own question: libraries were paying the subscriptions, not authors, and choosing to submit to a different journal would have zero impact on the author’s research budget (and the library’s since, that single article decision wouldn’t change which journals the library subscribed to). No incentive, no effect on behavior. That’s one of the core problems with scholarly publishing: the authors have the content — the *value* — but when they decide who to give their copyright to, they don’t consider price (quite reasonably!). That’s one of the things we’re trying to change (modestly).

– David Crotty thinks it is “contradictory” to say that in the UC plan much of the funding for publishing would still come directly from the institution, yet authors paying part would affect the economics of the industry. Not contradictory at all: if authors have *some* skin in the game, it will have some effect on their behavior — and since the publishers’ market power comes entirely from authors (transferring their copyrights), changes in author behavior *will* affect the industry. It’s not contradictory, just a matter of degree. If only 5% or total publishing costs are paid through direct research funds, the effect will probably be tiny. If 100%, the effect will be enormous. We’re proposing something in-between.

– David Crotty asks, “What about those at UC without grants, where will their funds come from? Is the UC willing to devote a portion of indirect grant funds to fill the significant gap that will be created?” Yes. This is core to our proposal, and we’ve been very clear about this (see. e.g., our CNI slides). In our model, if authors don’t have research funds sufficient to pay the about $1000 per article that would be their share, then the library will pay.

– Brian Simboli asks “One thing that OA advocates have clearly not focused on is the *glut* in publishing….Why not address that demand-side variable–namely, the drive (perpetuated by the tenure and promotion system) to reward huge output of articles?”

Actually, I’ve almost never been in a major discussion of the future of OA that *doesn’t* discuss this problem. There are recommendations strewn across the landscape that evaluation (e.g., for promotion and tenure) change to reduce unwarranted pressure to increase quantity rather than quality, and to put undue weight on “prestige” journals (rather than on high quality *articles*).

However: libraries don’t have any meaningful influence on faculty evaluation policies (and shouldn’t). Heck, even provosts have precious little influence over these issues: faculty evaluation in academia (especially in the US) is driven by peer review — it’s faculty promotion committees and department recommendation processes that decide what criteria to use in evaluations.

We in the UC libraries are pushing on strategies with which *we* can make a difference. Faculty at UC (and elsewhere) are simultaneously pushing on strategies to improve evaluation processes (e.g,. the Computing Research Association’s recommendation that computer science evaluations be based solely on the most important 5 publications that the candidate identifies from the review period, to raise the focus on quality vs. quantity, and to limit the volume so the review committee and external reviewers can *read* the articles and pass judgment on them, rather than simply rely on journal impact factors).

– “Mythical” probably not the best word choice. Just meant that “movement” is a bit overused: there’s no formal organization, no clear leader, not a lot of agreement on specific objectives (or even what the terms mean). But this is beside the point — not relevant to the UC strategy, and if folks want to call it a movement, fine. Let’s not get distracted by that.

This sentence (or its equivalents in other presentations/statements) always makes me wonder how this all might play out: “if authors don’t have research funds sufficient to pay the about $1000 per article that would be their share, then the library will pay.” Unless this is levied as an actual tax on the funds by an external determinant of fund availability rather than relying on the faculty member to self-report, given the fact that research funds always have multiple pulls on them, why wouldn’t a faculty member say “sorry, I have no funds to contribute” (e.g., because I have allocated them to hiring assistants, funding conference presentations etc.). Leaving the library to pay 100% of the publishing cost and relieving the faculty of whatever consideration they might have to consider price? Also, to the degree that faculty may be considering price in their own considerations now and choosing no or lower priced APC options, might not this scenario potentially lead some faculty to choose higher priced APC options that are currently outside what they can afford because they would be free to choose something more expensive because the Library will pay the remainder?

Jeff (if I may),

A few replies. We have some disagreements, but I want to emphasize that it is wonderful that you are engaging in these debates. (I should mention that any views expressed are mine solely and not those of my employer, in this or any other fora.)

I’m not convinced that significant elasticities will enter into APCs. Significant, in the sense of making a dent on journal pricing in excess of the CPI. I understand your argument, which relies on microeconomic assumptions about consumer preferences and behaviors. But let me ask: Have APCs had any impact on journal pricing yet? Is there any empirical evidence? Is the idea that they will, if there is a critical mass of participation in the APC market, where that critical mass has not *yet* been reached? And that when we reach that critical mass, then we will start seeing an impact? I don’t know. I think this is entering a speculative arena, and speculation is not necessarily a good basis for policy.

A bigger question: has the availability of preprints or other modes of OA research disclosure had any impact on journal pricing? I haven’t seen it. Of course, what makes that assessment difficult is that logically, we’re now dealing with counterfactuals about what the world would have been like had OA not been such an initiative. Maybe journal prices would be even bigger now. But how would we know that?

In my view we need thinking about other strategies–e.g., very active attempts to contract the journals market, per my preprint–under revision–about this.


Your comment about libraries not having a meaningful influence on faculty evaluation seems to contradict itself. You mention that UC libraries are pushing for strategies in which they can make a difference, and then your comment follows–seamlessly, without pause–about the role of faculty at UC trying to improve the evaluation process. So I take it you see some role for libraries in this process?
If not, why not? Libraries make available the Journal Citation Reports and have played a large role in training researchers how to use databases and assess their citation counts. Librarians are interested in the bibliometrics literature and can play a role in educating faculty about new metrics for evaluating research that go well beyond the much-criticized impact factor.

One problem right now with the bibliometrics industry is that the ratio of bibliometric research to practical, operationalized action is very high. Libraries and librarians should be very active participants in discussions about how to *operationalize* new and practically useful bibliometrics, esp. ones that focus on the quality of individual pieces of work.

Sure, it’s not clear how much influence libraries can have in actually changing evaluation of research for t and p or for grant funding purposes. Yes, it’s the faculty themselves who have to drive the actual changes in evaluation practice. But on the other hand, I can’t agree that librarians don’t have a whole lot to say about these matters, and more than that they don’t have a whole lot to say about the whole publishing ecosystem. True, development of *policies* about t and p evaluation will have to emanate from faculty primarily–and their societies. But this is not an argument for librarian quiescence about these matters, again any more than libraries should not have anything to say about how to transform the scholarly publishing system. (The U Cal system has decided already that it will try to impact the latter, of course.)

I am (among other hats) our serials librarian. In 10 years, I have never, ever, had a faculty member ask me what a particular journal costs our institution as part of their decision as to what journal to submit their manuscript to. I strongly suspect that if you ask your own serial librarians that question, you will find the same response. Some might in recent years have had the joy of being asked to help identify a suitable OA journal from a particularly “woke” faculty member. As a side note, because of the “Big Deal” journal subscription packages, depending on the journal I might actually have some trouble answering that question directly, although I could certainly provide the “retail price” if we were to subscribe to that journal individually. As an amateur student of economics, I absolutely support initiatives which bring into a single person’s head the decision where to publish with the burden of the financial cost of that decision. In reference to another question, I have also seen no direct effect at all of APCs and OA on the cost of subscription-paid journals, although it’s entirely possible that those are putting pressure on the publishers to accept lower price increases when they negotiate with the consortia through whom I get our Big Deals. Certainly those increases, still higher than my budget increases, are lower than the ‘retail’ price increases that I would have to pay if not for the consortial deals.

Great point on the “retail price”. So much of the analysis done on journal prices is based on the retail price for a print subscription where in reality most journals are purchased online as parts of heavily discounted packages. It would be really interesting to see a study of the increases in Big Deal prices over the years (though this may prove difficult as some publishers require NDAs around such deals).

As a former researcher, the calculation that is done when choosing a journal to submit to is all about finding the right home for the work which will provide maximum reputational and career benefits. Given the current funding and career advancement structure that’s in place, I suspect this will still be the deciding factor, even in an OA world. Each researcher will consider the cost/benefit of the journal — a publication in Nature that costs $25K may be a really smart investment for someone just about to hit the job market or if it results in significant further funding. And this will further enforce the Matthew Effect (, as well-funded labs will have better options open to them than those without funding, and hence will have a better shot at career advancement and funding, and the rich will get richer.

Which raises another point to consider David – when are individual scholar interests and preferences aligned with institutional interests and preferences and when might they conflict? Both currently and in the flipped-to-OA world. Is that very high APC (should one like that come to exist) a good investment for the institution if the scholar is about to depart? Are investments better made in early career or long-established researchers? And, now I’m wondering if anyone is negotiating an APC fund as part of a recruitment/retention package!

Individual scholar interests have been conflicting with the institutional (aka library budget) interests for decades, which is how we got into the “crisis” that OA is designed to solve. Any time the person choosing the service provider gets to hand the bill to someone else to pay, you’ve got a conflict. Compare to the problem of health insurance in the US and why so many insurers responded by forcing patients to use only “in network” providers. I can just imagine what the journal publishing industry would look like if librarians/admins could force faculty to only publish with “preferred” publishers! I understand some countries have point systems tied to published-article journal ratings that connect back to eligibility for govt grants and such, but as far as I know neither the US nor Canada has tried that. I would be interested in learning more about those systems, and what librarians and research faculty think about how well it works.

I could in theory do a price increase analysis of our own such deals, notwithstanding NDAs (price increase isn’t the same as revealing the actual prices), but it’s a nightmare to do over more than one year’s increase, because the big publishers are constantly selling/trading/acquiring titles which affects the price of the package. Teasing out what portion of the price change is due to content change compared with price increases is incredibly complex. For instance, if the price goes up just 2%, but you also lost 5 STEM titles to other publishers (one of which was part of your base price so that drops) but got 10 SSH titles into it (3 of which you had prior subs so now get added to your base price), how do you make any sense of the price change if your point is to determine the “value” for your price? Do you find some way to include your usage on the lost titles, and on the gained ones if you already had them, or turnaway data or ILL data on them if you didn’t? You see what I mean about “nightmare”. No wonder many people just punt and look at retail print prices – it is so much simpler, like the drunk looking for his keys under the streetlight “because the light is better here” (my favorite joke).

My favorite joke as well. I first encountered it in a Bazooka bubble gum cartoon when I was a kid. It is also a famous Zen koan.

Without delving into the particulars of research funding by government entities, just a few comments:
a. it should not be their role to rectify problems that cannot be resolved in other ways. This is a straightforward application of the “principle of subsidiarity”. The *amounts*–however small their size–already spent by the government in supporting journal publishing infrastructure, however indirectly, are not material to the argument that this is a poor use of taxpayer monies. By way of analogy, one could extend the point to subsidies already provided by government entities that help perpetuate an already bloated higher education system, which burdens hard working parents and send their kids into debt, the latter with very major sociological impacts. In both cases universities have to work on transforming their own systems without government intervention. (Again, subsidiarity is the operative social principle here.)
Your point might in response be that government entities should provide a fix to a publishing system beset with problems–in the form of dollars. But isn’t this the very manna from heaven attitude that you otherwise criticize? And that seems to animate so much OA rhetoric? (With notable exceptions–again, it is terrific that you are addressing the economic dimensions of these matters.)

Apparently Elsevier now has begun limiting access by German universities that have canceled their subscriptions the last few years as part of Project Deal, many of which have been without a contract with Elsevier for quite some time now, but Elsevier has kept the access open for.

That is likely a dry run on what is to happen with UC, too.

If this is true that Elsevier is limiting access by German universities (what is it scale, if ture), this occasions the question: why not use this as a test case to discern how a contraction of journal access plays out? Perhaps this will provide a useful laboratory to test how well researchers can find alternative means of accessing research. Hopefully German universities will actively discourage reliance on SciHub.
I’m curious how this is playing out:
…in terms of burdening interlibrary loan.
In any case, we may be entering a salutary phase of Schumpeterian creative destruction, precipitated by inability of universities to continue sustaining the current system, economically speaking.
“May you live in interesting times.”

Comments are closed.