Open@VT

Note: This is the first in a series of Open@VT blogposts that will appear over the ensuing months focusing on Virginia Tech’s “Big Deal” contracts with commercial journal publishers. As the University Libraries’ contracts with Elsevier, Springer, and Wiley come up for renewal in 2-3 years, we will have to decide whether to renew or cancel these contracts. We look forward to engaging the VT community in a conversation about the best path forward.

Dominoes falling (Photo by aussigall. CC BY 2.0)

On February 28 the University of California announced that it was terminating all of its journal subscriptions with the scholarly publishing giant Elsevier. The news sent shock waves throughout the world of higher education—not just in America but globally. Why? Because Elsevier is the world’s largest publisher of scientific research and the University of California (UC), with its ten-campus system, is one of its largest customers. The impact on Elsevier was immediate: its parent company, RELX, saw its stock drop nearly 7 percent in the aftermath of the UC announcement—and its value still has not yet recovered.

In Virginia we are paying special attention to the situation because our own research universities, including Virginia Tech, have a similar journal subscription agreement with Elsevier that is set to expire in two short years. Millions of dollars are at stake in Virginia. Globally it is in the billions.

What’s the Problem?

At the heart of UC’s dispute with Elsevier is what is known as the “big deal.” A big deal is a contract between an institution (often a university library but sometimes a business or government) and a publisher to purchase access to a large bundle of the publisher’s journals. Think of cable TV bundles in which customers get hundreds of channels at a lower per-channel rate. Many of the channels, however, go unwatched, all while customers’ bills continue to rise. The same is true with big deals. Elsevier publishes more than 2,500 journals. Many are invaluable to their fields and frequently used and cited. Many, however, are used infrequently, and yet libraries still have to buy them as part of the bundle. All the while, the price of the bundle goes up and up. Over the last thirty years library journal budgets have risen by a staggering 500 percent (see chart), which inevitably leads to cuts in other areas of library budgets. UC was paying Elsevier more than $10 million per year for its big deal. Altogether, the publisher’s revenue in 2018 surpassed $3 billion and its profits exceeded $1 billion, resulting in a gaudy profit margin of 37 percent.

Universities are understandably tired of big deals. Not only have big deals meant runaway prices, they also perpetuate an outdated business model from a time when subscriptions were an efficient way to pay for the cost of printing and distributing journals. Today subscriptions are inefficient for the simple reason that journals can be published online for immediate access. Publishers like Elsevier, however, have an interest in keeping the old system alive. This is why they continue to invest in expensive publishing platforms that restrict access to only the wealthiest institutions. There must be a better way.

The solution proposed by the University of California is to do away with the big deal concept and replace it with what is known as a “read and publish” agreement. A read and publish agreement (RAP) is a single integrated contract that enables a library to pay a one-time, up-front charge for the right to read all of a publisher’s content and to publish in any of that publisher’s journals under an open access model. The first RAP agreement in North America was announced last year, between the MIT Libraries and the Royal Society of Chemistry. Ultimately, the goal of RAP agreements is to transition scholarly publishing to a universal access model.

Momentum Is Building

UC is by no means the first university to stand up to Elsevier, but UC has special clout because of the sheer size and research output of its ten-campus system, which accounts for nearly 10 percent of the nation’s research publications. Meanwhile, governments and national research funders are increasingly demanding open access to their researchers’ articles, even imposing concrete deadlines. Sweden’s government is calling for OA by 2026. Norway’s goal is 2024. The initiative known as Plan S is even more ambitious. Originating in Europe, Plan S calls for all publicly funded research to be published in open access journals by 2020. The Bill and Melinda Gates Foundation was the first North American foundation to sign on to Plan S.

As more universities and governments push for open access, the more it seems that Elsevier is destined to lose this battle. But this does not mean that it will lose the war. Elsevier is shrewd enough to adapt to (and even shape) whatever new business model emerges around open access publishing. Perhaps anticipating this change in business model, Elsevier has skillfully and steadily turned itself into one of the world’s largest publishers of open access as well as toll-access journals. It has also been diversifying its business portfolio to the point that it no longer even refers to itself as a publisher but as a “global information analytics business.” In other words, Elsevier is not going away anytime soon.

Implications for Virginia

Virginia will soon be in UC’s shoes. In 2004 seven Virginia research universities including Virginia Tech negotiated a big deal agreement with Elsevier. (The other schools are George Mason University, James Madison University, Old Dominion University, University of Virginia, Virginia Commonwealth University, and College of William and Mary.) The number of journals in that big deal was 1,800 and the total cost to the seven universities was $27 million over five years. The license has been renegotiated several times since then, and we are now in the third year of a five-year contract covering 2,278 journals at a total cost of $46 million. This contract will expire at the end of 2021.

Not surprisingly, these universities are already looking ahead to 2021 and considering the possibility of walking away from Elsevier big deal as UC has done. (See, for instance, the University of Virginia.)

Here at Virginia Tech, the University Libraries, under Tyler Walters’s leadership, will be engaging the campus community in an ongoing conversation about how Virginia Tech can confront this scholarly publishing crisis. On this, we sincerely want your feedback. Please watch for Library-sponsored events that provide a forum for discussion. In the meantime, feel free to reach out to our librarians and engage them in conversations. Or let us know what you think by replying to this blog post or to future Open@VT blog posts. You can also find up-to-date information at the Library’s Open Access-Open Knowledge website.

In my last post, I mentioned that I archive my journal articles in VTechWorks because I really dislike encountering paywalls for scholarship, and I don’t want anyone looking for my work to encounter one. Two new projects could make the scope of the access problem much more apparent.

The first, the OA Button, will use a map to show instances where people from around the world are hitting paywalls, but it will also help users find an open access version of an article. The button, in prototype with a beta version coming soon, is a bookmarklet that you drag to your toolbar and click when you are denied access to an article. If you allow your location to appear, it will be included on a world map.

You can also say why you need access, which could result in compelling advocacy for article archiving. Check out the blog, an interview with the creators, and a hack day project page.

The second project involves Wikipedia, which encourages those who create or edit entries on the site to document their claims through the use of references. However, when a Wikipedia user clicks one of these links, there is no way of knowing whether it will be accessible or not. The WikiProject Open Access/Signaling OA-ness aims to add an icon next to these links to show users which are open and which aren’t. This will be a convenience since you will know about paywalls before you click, and it will show at a glance what portion of research isn’t accessible by everyone. It would also be interesting to build on this and show how many references across all of Wikipedia aren’t accessible.

What is out of sight is out of mind. These projects will help make the openness of research obvious to all.

As the open access movement advances, more faculty become interested in providing access to their work. But there aren’t enough open access journals yet, especially in some fields. Self-archiving is very spotty- a few do it, but most don’t. So faculty associations in universities, colleges, and departments are adopting open access archiving policies.

One policy debate that sometimes arises is whether archiving should be voluntary or mandatory. Though some open access issues are unsettled, this one isn’t. Voluntary policies don’t work and never have.

Stevan Harnad made his “subversive proposal” to archive articles on a Virginia Tech e-mail list nearly 20 years ago. Recently he said:

If one thing has been learnt from the slow, sad saga of OA (now at least two decades old) it is that mandating OA works, but encouraging it doesn’t.

The faculty votes (many of them unanimous) that have taken place in the last few years have been for mandatory policies (there is no list of voluntary policies). The purpose of a policy is to increase the amount of openly available research online, and mandatory policies do a far better job of that:

Open Access (OA) Self-Archiving Percentages for Institutions With Self-Archiving Mandates Compared to Non-Mandated, Self-Selected Controls (Gargouri et al. 2010, doi:10.1371/journal.pone.0013636)

About 60% of journals now allow archiving of the author’s final version (the post-print), but only 10-20% of articles are archived. That’s why a mandatory policy is necessary. We need that nudge. We need that collective expectation that research will be open, not paywalled.

The word “mandate” evokes negative reactions even before policy details are heard, but it’s an inaccurate term in two ways. First, there is always an opt-out, so it is not a mandate in the strict sense of the word. Second, a policy is approved democratically by those directly affected by it. It’s not a directive from the provost or a dean. It’s a statement that says we think making research openly available is important enough to require it of ourselves.

As authors, we know that open access serves our own interests. More people can read our work if it’s not behind a paywall. Numerous studies confirm that openly available articles are cited more often. When everyone has access, promotion via social media is far more effective and raises the researcher’s profile. Even for those who don’t actively promote their work, the simple act of providing worldwide access generates views and downloads month after month.

We know that providing access to our research helps a lot of people, including those in the developing world and independent researchers. So we can produce a great public good (and advance our own careers) just by setting aside 10 minutes to archive one of our articles.

Let’s make policy decisions based on evidence. No faculty group can singlehandedly repair our dysfunctional system of scholarly communication, but we can begin taking steps to better fulfill the university mission, and that of land grant institutions. If the journals in our field put our research behind a paywall, let’s ensure worldwide access for our articles by taking 10 minutes to archive them.