Wednesday, November 04, 2020

Community Action Publishing: Broadening the Pool

We are today seeing growing dissatisfaction with the pay-to-publish model for open access. As this requires authors (or their funders or institutions) to pay an article-processing charge every time they publish a paper it is felt to be discriminatory, especially for non-funded researchers and those based in the Global South  (see, for instance, here, here and here).

PLOS' Sara Rouhi

As a result, various alternative approaches are emerging intended to move away from APCs, including crowdfunding and membership schemes. Here institutions are asked to commit to paying an annual fee to a publisher, with the aim of pooling sufficient funds to cover the costs of making all the papers in a journal open access.

One of the more successful implementations of this model is the Open Library of Humanities (OLH), which has operated what it calls its Library Partnership Subsidies scheme since 2015.

Clearly, if the costs of the annual fee are to be viewed as reasonable by those asked to take part a sufficient number of institutions need to sign up.

The inherent weakness of the model is that a small group of community-minded institutions could end up paying all a publisher’s costs and everyone else would be able to “free ride”. In effect, those who join a membership scheme could end up shouldering the costs for everyone to have free access.

A key question is how many universities are willing to join an OA membership scheme. OLH co-founder Martin Eve has estimated that there are only around 300 libraries who will do so.

Nevertheless, we should not doubt that there is a real wish to move beyond APCs and many have come to believe that membership schemes are the best way of doing this. To be successful, however, they will need to broaden the pool of those willing to take part.

Annual Flat Fee

With this aim in mind, in October the Public Library of Science (PLOS) launched what it calls its Community Action Publishing (CAP) initiative for its two selective journals PLOS Biology and PLOS Medicine. The hope is that the journals can be gradually moved away from having to charge APCs and that the publishing costs can be shared as widely as possible.

Specifically, PLOS is inviting universities to pay an annual flat fee that will give their faculty unlimited publishing opportunities in the journals (there are separate “communities” for each journal) without the need to pay an APC each time. However, authors of institutions who do not join the scheme will be charged a non-member fee (NMF) that will increase in price over time, as below.

The rising cost of the NMF is intended to encourage universities who have not joined to do so in order to avoid their faculty having to pay to publish.

PLOS hopes that the annual fees will be low enough to attract a sufficient number of institutions but that the pooled funds will eventually be adequate to meet all the costs of publishing the journals. In the interim, the NMF means that there will be two separate revenue streams coming in and the free riding issue will be avoided.

The hope is that the NMF can be discontinued when the pilot period ends (it lasts from 2021 until 2023). However, says PLOS Director of Strategic Partnerships Sara Rouhi, that will be dependent on all the large institutions that PLOS is targeting joining the scheme. “If we don’t have a stable membership by then, we might have to offset lack of participation with more NMFs”, she told me. 

Tiers

The annual fee institutions will be asked to pay will be calculated by looking at their faculty’s publishing activity between 2014 and Q3 2019. Based on that they will then be assigned to a fee tier, as below.

To help broaden the pool, the annual fee will be calculated on the publishing activity not just of the institution’s corresponding authors, but of their contributing authors too. The aim is to ensure that “the cost of publishing is distributed more equitably among representative institutions.”

In other words, PLOS will calculate how often during the historical publishing activity period an institution was associated with both corresponding and contributing authors and then assign a weighting to each type of author. Specifically, contributing-author articles will be weighted at half that of corresponding authors.

As the PLOS FAQ explains: “Publishing activity is counted by determining the number of times an institution was associated with a corresponding author and the number of times an institution was associated with contributing authors. Papers where the institution is affiliated with the corresponding authors are weighted as 1 article and papers where the institution is affiliated with the contributing author are weighted as ½ article. (Multiple contributing authors from the same institutions are counted only once).”

Universities whose faculty have never published in the journals can also opt to pay the tier’s lowest fee in order to “insure” themselves against the possibility that one or more of their faculty might publish in the journal within the time period of the scheme (Jan 1st 2021 –  Dec 31st 2023).

In addition, says PLOS, Research4Life countries are automatically members of each community so researchers in those countries will never be subject to fees. And authors unable to pay non-member fees can apply for waivers as per the standard fee-waiver mechanisms offered by PLOS.

To signal its own commitment to the community PLOS has set targets for each journal and will cap its margin at 10%. Revenue exceeding the community targets will go back to members at renewal.

Time will tell how successful the scheme will be in broadening the pool beyond the 300 libraries who normally join OA membership schemes. Certainly, the challenge will be that much greater given that it is being launched at a time when many libraries are signing expensive transformative agreements with legacy publishers and universities are facing the financial impact of the pandemic. Some have also suggested that, as selective journals, PLOS Biology and PLOS Medicine do not publish a sufficient number of papers to make it worthwhile for many institutions to participate.

My initial thought was that PLOS will be targeting three separate types of institution (corresponding-author institutions, contributing-author institutions, and “insurance” institutions). Sara Rouhi suggests that this is not the way to view it. Below is an exchange I had with her.

Levers

RP: I understand the PLOS Community Action initiative will target 3 types of institution:


1.       Corresponding-author institutions.

 

2.       Contributing-author institutions.

3.        Insurance institutions (those institutions whose researchers don’t currently publish with the PLOS journals but who might want to insure against the possibility that one of their researchers will do in the future).

I guess there will be some overlap between 1 and 2 but I assume you will have done some calculations on the numbers of potential institutions for each category. If so, can you share them with me?

SR: While we initially grouped institutions this way, libraries and consortia consistently fed back to us that this was more complicated/confusing than it needed to be. So rather than think of different types of “institutions,” it’s easier to think of different publishing activity “types” – publishing as a lead author, publishing as a contributing author, or publishing with affiliations in both designations.

Given the selectivity and niche of these journals, some research-intensive universities publish little to not-at-all, and smaller institutions publish more frequently as lead authors. Simplifying the participation criterion to just publishing activity elides the traditional distinction of “research intensive” vs. “teaching” institutions etc.

I don’t want to add further confusion by grouping institutions as you’ve indicated. We purposely eliminated this distinction for clarity.

RP: Ok. I ask this question in the context of Martin Eve’s estimate that there are only 300 libraries that will support OA membership schemes. (I should add that he said this in the context of a new initiative from COPIM that aims [like PLOS] to “broaden the pool” of universities willing to join a membership scheme).

SR: As you say, Martin Eve identified about 300 libraries that will support OA membership schemes. There’s some nuance worth flagging, however. 

The TL;DR is that Martin’s statement is true if you assume institutions that are participating for largely altruistic reasons – because it’s the “right thing to do.” The PLOS CAP model couldn’t be predicated on that as a buying motivation. So it helps to go back to “collective action” basics. Credit to Dr. Kamran Naim for elucidating this at the Basel Sustainable Publishing Forum last week.

For collective action schemes to work there are generally two levers you can pull: 

 

1.       Encourage “pro-collective” behaviour by leveraging group affinity and social incentives

 

2.       Appeal to economic self-interest (aka “private benefit”)


From my perspective, SCOAP 3 and OLH lean heavily on the first lever. Participating in them is the “right” thing to do so libraries support them.

The challenge with only pulling on this lever is sustainability over time. SCOAP3 relies heavily on CERN’s ongoing support and has the challenge of large beneficiaries like Russia, Brazil and India not participating while benefitting from the open content.

OLH memberships are very small financial commitments that maintain OLH but make growth of the program difficult. Martin talked about this on a webinar we did for UKSG earlier in the year.

A collective model that focuses primarily on the second lever is Subscribe2Open. The primary benefit is getting the community to agree to make content open in exchange for a discount. If all members do not opt-in, the content stays behind a paywall. The combination of a private benefit (the discount) with the public good (making the content open) is a strong motivator for participation and a big part of why the model is so successful.

PLOS Community Action Publishing attempts to pull both levers. The pro-collective/group affinity/social incentives relates to the moral imperative – especially in a time of pandemic – to make biomedical research open to read and open to publish. Jeff Kosokoff’s comment that “Open Access is social justice,” speaks to this. Indeed, we have many commitments based on this mission-aligned priority for libraries. However what has interested partners from mere interest to actual commitment is the equitable fee structure (and relative affordability given the current budget crisis).

That said, without including appeals to economic self-interest, the PLOS CAP model would suffer from the same sustainability issues as other collectives. Hence our implementation of a secondary, “back up” revenue stream – non-member fees (NMFs) for authors from non-CAP members.

These NFMs are not meant to penalize authors but rather to encourage libraries to join rather than expose their authors to fees. For many institutions, the NMFs per article will be higher than the annual fee the library would pay to join one or both collectives.

Benefits

The NMFs help offset slow uptake of institutions who are not able to find the funds to support a model like this immediately. Thanks to those fees coming in, libraries can take more time to join and we can offer flexibility for institutions that need to leave the collective.

There are several benefits to this:

Enlarging the pool of institutions that would participate in this collective gives us the flexibility to create a much finer tuned fee structure with a long tail of low dollar fee tiers to accommodate institutions that do not publish frequently in either journal.

Those low fee tiers allow institutions strapped for funds and/or infrequent publishing institutions to participate because they support the moral imperative of equity and inclusivity that the model promotes. It simply doesn’t cost them that much to “do the right thing.”

Research intensive institutions then derive benefit from the “research light” institutions participating since they offset some of the cost burden. More institutions of varying publishing intensity means lower fees for everyone.

So, Martin is right if you’re counting only relatively wealthy institutions that want to do the “right thing.” If you start adding other considerations, especially private benefits (aka, we don’t want our authors to see non-member fees) and recognition of contributing author affiliations, you get a much larger pool and set of incentives to motivate ongoing participation.

Targets

As for our target institutions:

For PLOS Medicine there are 909 total institutions with some publishing history (some combination of lead and co-author affiliations) in the time period we evaluated.

44 of those are in Tiers 1-4

80 are in Tiers 5-7

The remainder are in Tiers 8-12

For PLOS Biology there are 1,557 total institutions with some publishing history (some combination of lead and co-author affiliations) in the time period we evaluated.

45 are in Tiers 1-5

152 are in Tiers 6-8

The remainder are in Tiers 9-12 

As I say, I wouldn’t necessarily correlate rough groupings of similar institutions as “corresponding author institutions, contributing author institutions, and insurance institutions” especially since it does not include the thousands of organisations with no publishing history who might want to participate. 

However, it’s probably fair to say that the highest tiers have institutions that published a lot in both designations. The middle tiers were a combination but at lower volume, and the lowest tiers were low volume and mostly in the contributing author designation.

Threat

RP: In a webinar you gave in September you said that if transformative agreements flourish there probably won’t be sufficient money left in library budgets to support schemes like PLOS Community Action. How big a threat do you think there is here?

SR: This, of course, comes down to how you define “transformative agreements.” If you mean the standard RAP/PAR deals that integrate historic subscriptions with open access publishing that stays revenue neutral, then yes. If these are negotiated with large commercial publishers first, we’re looking at new kind of “big deal” that locks in library monies with subscription publishers of hybrid open-access journals. Non-profits, small societies, and native-OA publishers may very well not make it out the other side of this transition (especially if “read” institutions’ subscription monies exit the system).

It’s hard to know how big the threat is as it ultimately comes down to choices libraries make. Many appear to be balancing both imperatives – looking at where they publish and spend the most and evaluating how mission aligned those outlets are. It’s not easy.

RP: So, what are your expectations about the choices that libraries will make with regard to the PLOS Collective Action Publishing scheme?

SR: It’s totally dependent on region, funding structures (block grants in UK versus how the US does it) and how radical libraries want to be about cancelation and re-negotiations.

I have personally been overwhelmed by the number of cash strapped organisations that are pushing to support this model and find the money.

The December commitment update will tell the tale!

RP: Thank you. And good luck!

 

 

 

1 comment:

  1. The APC (article processing charge) model of open access publishing seemed inherently flawed to me from the beginning. Much like the so-called gig economy and "sharing economy", it seemed to shift the burden of costs from institutions to individuals while presented as a standard bearer for democratizing knowledge and virtue signaling a socialist aesthetic. Open access has led to vanity journals, predatory journals, and publishing for those who can pay to do so. I know that you are no fan of peer review, but pre-print servers and publishers such as PeerJ require researchers, expert practitioners, academics, and other participants to ensure quality to control. It is possible, but even a few bad actors are enough to undermine the effort.

    This fascinated me:
    The inherent weakness of the model is that a small group of community-minded institutions could end up paying all a publisher’s costs and everyone else would be able to “free ride”. In effect, those who join a membership scheme could end up shouldering the costs for everyone to have free access.
    Why? Because it has a precedent from my world. The precedent led to a sad outcome for all. If open access community participants were wise, they would avoid such a scenario.

    I'll tell you about the precedent. You have no doubt heard of Standard & Poor's the financial securities rating company. As of about 1960, S&P was a specialty publishing company who sold its rating reports directly to customers, be they individuals or businesses. S&P had worked under this business model since its founding in the 1870s. S&P decided which securities it wanted to rate, requested the cooperation and access of the issuing companies, then S&P's analysts and presses produced and published the reports, available under various subscription plans or on a single-item basis. There were no conflicts of interests, i.e. that the company issuing the bond paid S&P to rate that bond, as rating agencies including S&P do now. (Compromised rating agency reports contributed to the 2008 financial crisis.)

    So, this worked well for decades. Securities rating agencies were credible and helped inform investors. In the mid-1960s, the Xerox company developed a remarkable technology, the high-speed copying machine. Instead of purchasing S&P reports, some less successful or less ethical asset managers and speculators started buying the reports, making photocopies, then reselling them for half the price or less than what S&P charged for its original work. Information wants to be free? Everyone wanted the valuable research and analysis from S&P ratings reports, but only a few users were willing to pay for them. S&P was acquired by the big textbook publisher McGraw-Hill in l970. S&P began operating under a new business model: If a company or municipality wanted its bonds to be rated, then they (as the bond issuer) would need to pay S&P for its work. It wasn't quite a tragedy of the commons, like open access or The Fens, but there are resemblances.

    ReplyDelete