The People vs. Elsevier?

The appointment of Elsevier as a subcontractor for the Open Science Monitor caused some heated discussion about Elsevier’s role in Open Access and Open Science. As a believer in the Open Science principles, I’m biased towards one side of the conversation. However, I think there are some points on both sides that are overseen by the other party, which need to be mentioned in more detail. I had some great discussion last night with mrgunn on Twitter, which got me thinking about some of the aspects.


It all started way before the recent events, however, let us just review these for simplification of that matter. Elsevier got awarded a tender as the single subcontractor for the European Open Science Monitor. Given the standing of Elsevier in the Open Science Community, it didn’t take long for a response by one of the most outspoken persons in the community, Jon Tennant. Followed by a statement from Elsevier and a second, to date unanswered, response by Jon and some heated discussions on Twitter.

All of the articles are worth reading. However, I don’t want to go into detail about the single points made in these articles and discussions. It all basically centered around the role and business practices of Elsevier as the biggest publisher for research literature in Open Science.

Is a tool that is free to use and helps researchers in their daily work (irrespectively of its use to promote Open Science) a tool that can be considered open? The answer is no, just because there are clear definitions of what is considered open. An open license under which the software/tool is released would be the minimal requirement. These definitions are important for the community because they fought hard for them and they serve an essential purpose. However, can this tool still be useful to the research community? Sure. I will talk about this in more detail in a later post.

Open Science being on its way to mainstream and a hot topic right now in the research community, it is easy to just claim a tool as being open even though it is not and get some positive feedback. Getting this bandwagon feeling? Realizing this kind of method is not easy for most researchers, which do not care so much about Open Science. Now, imagine the largest company in the business doing something like that. Their tools might be useful, however, they aren’t open by definition but advertised as such. Understanding the frustration here is important.

Elsevier, on the other side, is publishing over 400,000 submissions every year, around 25,000 of which are Open Access (Ref). As such, it is contributing to the opening of science but has apparently history and business practices that put them in doubt about their true mindest regarding Open Science (Ref). I’m not sure if Elsevier is the biggest Open Access publisher in total number of publication as I don’t have any numbers for other like PLoS or BioMed Central for example. But the question that was raised in the discussions was if the total number is better than the percentage of Open Access publication compared to the total number of publications. Depending on your own assessment, Elsevier is either an important Open Access publisher or has still some room for improvement. I’ll have a short and probably inaccurate calculation about a possible scenario in the end.

What is right and what is wrong? Arguing from inside your sphere. Picture take from https://clairechandler.com/tag/point-of-view/

Who is right and who is wrong?

mrgunn pointed me to the idea of Monkeyspheres last night, which in the first place confused me as it felt like he was trying to show me that the Open Science community and I were arguing unfairly even though he was himself arguing from within its own sphere.

But I think it is in an important concept for further discussions on that matter. From within my sphere, it is easy to see Elsevier as the main problem for several reasons mentioned by the Open Science community, however, as Will said there are people working at Elsevier that also care a lot about Open Science and I believe that. This is probably not true for all Elsevier employees and maybe even not for the ones that make the big decisions in the end, but maybe there is a way… It seems that at least for these people the goals between Elsevier and Open Science might align, so it should be possible to find a solution. As will said:

So maybe we can start by claiming only these tools as open that are really open and do not confuse the users.

What would it cost Elsevier to go All-In on Open Access?

Coming back to the discussion about Elsevier’s role as the biggest Open Access publisher. With 25,000 publications in 2016 and 27,000 publications in 2017 of its over 400,000 publications Elsevier might be the biggest by number, but definitely not the biggest by percentage as they only publish around 6% of their papers as Open Access. But who is to blame for that? (I’m bad with metaphors, but I’ll still try something here) Elsevier, as the doctor how subscribes steroids to his athlete, or the researcher, as the athlete, who publishes Closed Access instead of investing a little bit for Open Access?!? This is another topic for another post, but let’s just look what Elsevier would gain or lose when they would go All-In on Open Access. The following calculation is probably way too simple and doesn’t take a lot of things into account, but it is best I could come up with in 5min googling and should serve the purpose of showcasing an option to look into in the future.

Elsevier publishes worldwide around 400,000 submissions per year (Ref), 25,000 are already Open Access (Ref), meaning that 375,000 submissions are still Closed Access, which can potentially be paid APC’s for. Let’s say for this example an average APC at Elsevier is 2000€ (I haven’t done the calculation, but here is a list of all Elsevier journals and their APCs).

According to the EUA report, Section 4 “Big Deal” subscriptions cost around 380 million € per annum (conservative estimation) with Elsevier making 65% of these contracts. That means Elsevier makes around 247 million € per annum in Europe only.

375,000 closed access publication that can be open by paying 2000€, makes 750 million € per annum in revenue through the new All-In Open Access system. That is three times as much as they make right now in Europe by subscription deals. If we subtract those revenues we’ll end up with 500 million €. I don’t have any data on “Big Deals” in the US, but I assume it is a little bit higher than in Europe. Just from this quite estimations going All-In on Open Access seems to come at no cost for Elsevier if the revenue on subscriptions is not significantly higher than estimated. What they would lose is the revenue in their hybrid journals, because they are usually Closed Access and thus in those subscription deals but have an Open Access option meaning those can be paid for twice. Maybe this would be a way to show their actual intent to opening science.

I’m happy for any comments and valuable discussions.