Annual Reviews is an eighty-six-year-old independent, not-for-profit publisher with a mission to synthesize and integrate knowledge for the progress of science and the benefit of society. It publishes 50 review journals across the physical sciences, life, biomedical, physical, and social sciences. The reviews capture current understanding of a particular topic, including what is well supported and what is controversial; set the work in historical context; highlight the major questions that remain to be addressed and the likely course of research in upcoming years; and outline the practical applications and overall significance of research to society.

Annual Reviews is keen to pursue open access (OA), believing its content to be of wide general interest, but does not consider APCs to be a viable route. The reviews are invited, and it can take three or four months to write them, so the publisher feels that it is not appropriate to present the author with a bill/invoice to publish. In addition, Annual Reviews authors rarely cite funder support, closing off the most obvious source of APC-based funding.

Instead, the staff of Annual Reviews want existing library subscription payments for gated access journals to be leveraged and then retained to convert and sustain the journals as OA. They are putting the idea to the test with a program called Subscribe to Open.

In this post, I interviewed Richard Gallagher, President and Editor-in-Chief of Annual Reviews, and Kamran Naim, Director of Partnerships and Initiatives, about the organization’s rationale for pursuing OA and details of their Subscribe to Open approach.

Open padlock with green OK checkmark

Annual Review journals are not subject to specific funding mandates, so why are you pursuing OA so deliberately?

There is both a push and a pull. The push is the OA movement itself. As an organization that is run for researchers and by researchers, Annual Reviews is very aware of the support for removing barriers to access for all science publications, and we want to participate. It is a fit with our mission to contribute to the progress of science. The pull comes from the other component of our mission, benefiting society. A functional democracy requires that citizens have access to the knowledge and wisdom of the world’s leading researchers. Annual Reviews authors address many of the issues that society is grappling with today — climate change, aging, mental health, food and water security, crime and justice, and so on. Making Annual Reviews freely available would be a powerful public benefit and we are experimenting with ways to make this shift a reality.

We have also developed Knowable Magazine as a complementary approach to opening up our content. It presents journalistic coverage of the real-world significance of science. All the content is published OA CC-BY-ND and the html version of relevant Annual Reviews articles are opened up for readers who want a deeper dive.

Do you have evidence that Annual Reviews content will be widely accessed when open?

We do. The Robert Wood Johnson Foundation provided funds to publish the 2017 volume of the Annual Review of Public Health open access and at the same time we made the back volumes freely available. There was dramatic increase in usage. In February 2017, there were 26,000 downloads from the Annual Review of Public Health; in February 2019, there were 144,000. When it was a subscription-only publication, researchers and students from 2,000 subscribing institutions from 57 countries had access. In just the first six months of 2018, the Annual Review of Public Health was accessed from more than 9,900 institutions across 132 countries. This included a lot more universities and corporations, but also government and state agencies, NGOs, schools, hospitals and even prisons.

What is your strategy to transition to OA?

It is premised on the notion — advanced by groups such as OA2020 — that if libraries were to transition their current subscriptions to support OA publishing, there are sufficient financial resources in the system to support a full transition.

Subscribe to Open will succeed if all of our current customers continue to subscribe, and we are offering a 5% discount as a financial incentive. If we don’t get full participation, the project will not succeed, in which case institutions that wish to receive the content will need to have a traditional, gated access subscription at the list (undiscounted) price. There is no opportunity for subscribers to free-ride. In future years, non-subscribing institutions that regularly access the open access titles will be invited to become subscribers.

Because Subscribe to Open is contingent upon the participation of all subscribers, it is effectively a subscription. It can’t be interpreted as a voluntary donation and should not, therefore, violate state procurement policies. This allows libraries to utilize their existing subscription spends to support an OA program without the need for new OA funds.

Numerous intellectual and practical contributions have guided us to this model, for example that of Martin Paul Eve, and we also acknowledge guidance from many librarians, funders and publishers as we developed Subscribe to Open.

Will you set a public financial threshold to reach before you switch content to open?

Annual Reviews operates on a balanced budget – revenues closely match our expenditure – so we need essentially all subscribers on board for Subscribe to Open. For this reason, we will not be setting a public financial threshold, as it would introduce a group dynamic in which some institutions may delay their decision or attempt to free-ride if they believe the threshold will be met by other subscribers. Instead, the Subscribe to Open offer remains very much a subscription: the only way an institution can guarantee access to the journals is by participating, and they are further incentivized with the discount. Subscribe to Open is therefore a self-interested economic decision that should be taken by individual institutions.

Is the approach penalizing existing subscribers by letting others free-ride?

We don’t believe so. Our existing customers value Annual Reviews. We are asking them to continue to pay for access (at a slightly reduced price) while leveraging their collection budget in a dramatic fashion. We have discussed Subscribe to Open extensively with librarians who subscribe to Annual Reviews and their reaction has been very positive.

Does this lock libraries in to subscribe in perpetuity?

As with any subscription, institutions will participate in Subscribe to Open for as long as they value access to the content. There is no lock-in risk.

Do you have concerns about a loss of support after an initial surge of support dies down?

It is something we will be watchful for. However, we also see the opportunity to attract support from institutions that do not currently subscribe, if we can present them with impressive usage statistics. Also, while Subscribe to Open as currently conceived does not rely on additional income sources, we will look at mechanisms to top up funding if it should wane.

How does Subscribe to Open play with Plan S advocates?

We had already set our course on Subscribe to Open before Plan S was announced and it (indeed the open access movement in general) has not given much consideration to reviews. Nonetheless, we have engaged the leaders of Plan S on three occasions and each time they have been supportive of Subscribe to Open. They realize that there is need for diverse approaches to open access and that we are not the only publisher that can’t readily implement APCs. We view Subscribe to Open as a realistic and immediate route to open access.

When will you be launching the Subscribe to Open program?

We intend to pilot Subscribe to Open with five titles in 2020 and are currently introducing these plans to our customers. If all subscribing libraries participate, we guarantee that next year’s volumes will be published OA (with CC-BY licenses), all back volumes will be made freely available, and we will make usage statistics generally available so that subscribers and others can judge the effectiveness of Subscribe to Open.

What role do you think Subscribe to Open can play in shaping the future of OA?

Every publisher has a unique set of factors to juggle, but it is possible that Subscribe to Open will be a valid approach for a range of publishers, particularly society and nonprofits that enjoy collegial relations with their customers. We are happy to share our experience with Subscribe to Open with others who are exploring OA options, just as we hope to learn of alternative approaches from others.

Ann Michael

Ann Michael

Ann Michael is Chief Transformation Officer at AIP Publishing, leading the Data & Analytics, Product Innovation, Strategic Alignment Office, and Product Development and Operations teams. She also serves as Board Chair of Delta Think, a consultancy focused on strategy and innovation in scholarly communications. Throughout her career she has gained broad exposure to society and commercial scholarly publishers, librarians and library consortia, funders, and researchers. As an ardent believer in data informed decision-making, Ann was instrumental in the 2017 launch of the Delta Think Open Access Data & Analytics Tool, which tracks and assesses the impact of open access uptake and policies on the scholarly communications ecosystem. Additionally, Ann has served as Chief Digital Officer at PLOS, charged with driving execution and operations as well as their overall digital and supporting data strategy.


22 Thoughts on "Subscribe To Open: Annual Reviews’ Take on Open Access"

Some questions about this model:

Assuming it is successful, does this put a permanent cap on revenue growth for the journals? If the required spend is for all current subscribers to continue to subscribe, then the journal goes OA, why would anyone beyond that group start a new subscription? That would mean that all growth must now come solely from price increases on existing subscribers.

How much can they increase the price by every year in order to keep up with costs, incorporate new technology and other developments, launch new titles?

How does this model address libraries’ desire for reduced spending?

How will they handle churn (as each journal sees some gains and some losses in subscriptions each year)? How many subscriptions can they lose before the journal goes back behind a subscription barrier? If they’ve created a mindset that one doesn’t need to subscribe to the journals to gain access, how easy will it be to sell new subscriptions?

Once a volume (and back content) comes out as CC BY, it is now OA forever. So essentially, are they moving to just selling one year’s content at a time? Does this affect the price you can charge or a library’s willingness to pay? Would it be cheaper for a library to just PPV this year’s articles rather than subscribe?

“How does this model address libraries’ desire for reduced spending?”

Reduced spending for libraries needs to come from the large, exploitative, for-profit publishers. Libraries are looking for ethical publishing agreements for their scholars, not to pressure publishers interested in fair deals.

I suspect that many not-for-profit publishers who are seeing their journals dropped from libraries as a cost-savings measure would suggest that this is not an accurate depiction of the market as a whole.

Thanks for your questions David.

“Assuming it is successful, does this put a permanent cap on revenue growth for the journals? ”
We believe that there is a possibility of attracting previously non-subscribing institutions and organizations that engage significantly with the content to become subscribers, indeed we have already received such a query. The majority of users will continue to be in academia, and this is likely to be the leading source of new subscribers, but readers will also come from corporations, schools, hospitals, NGOs, and other institutions, and we will make the case to them to become subscribers also. Annual Reviews is a non-profit publishing organization that works on fine financial margins. As long as we have a critical mass of institutions that continue to fund publication costs, we will be able to deliver OA content.

“How much can they increase the price by every year…?”
Price increases will be proposed in the same way as currently. This has not been raised as an issue by the librarian community.

“How does this model address libraries’ desire for reduced spending?”
A 5 percent discount is offered on the Subscribe to Open journals. Beyond that, libraries will not need extra funds to cover OA fees such as APCs.

“How will they handle churn…?”
We have some churn in our subscriber base already, however it does not result in a net loss of revenue. If we do see a significant net loss, then the program will have failed. It is essential for subscribers to understand that they do need to continue to subscribe to gain access, either using Subscribe to Open, or via gated access should we not attract sufficient support. Our objective is make it crystal clear that that one does need to subscribe to the journals to guarantee access.

“…are they moving to just selling one year’s content at a time…?”
The price is not affected by the move to Subscribe to Open. We currently provide permanent data rights for any paid subscription year; so effectively our subscription model charges existing subscribers only for new content.

As a general point, we have found librarians to be enthusiastic about building a viable mechanism to move Annual Reviews to Open Access.

“…are they moving to just selling one year’s content at a time…?”
The price is not affected by the move to Subscribe to Open. We currently provide permanent data rights for any paid subscription year; so effectively our subscription model charges existing subscribers only for new content.

That’s true for subscribers who were subscribing in past years. New subscribers don’t get perpetual rights to years where they weren’t subscribed. Here, once that back-content becomes OA under a CC license, it is OA and CC forever. If a library is considering subscribing, and they see that most of their usage is of older articles, that may be a hard argument to make.

I think my main concern with the model though, is that once a library stops subscribing, that money is out of the budget forever, and really hard to replace. We see this with society journals moving to a new publisher and leaving a Big Deal ( The library is still on the hook for paying the same amount for that Big Deal, but no longer has access to the society journal, and has to find new money to pay for that subscription (or new Big Deal top-up) with the new publisher. Very few libraries have spare money on hand to cover such expenses. Here you may have some libraries dropping their subscriptions as they can get the journal OA. Then if it drops back into subscription access, it will likely be a struggle for them to re-up (they’ll have to drop a subscription to some other journal). If the model works, it’s great, but if there’s a year where it doesn’t work, it’s gonna be tough.

These not-for-profit publishers are being dropped as a direct result of the gouging from the for-profits. Unfortunately, its easier to drop the smaller publishers even though they are more trustworthy vendors.

Agreed, but it would seem to add some nuance to the statement that, “Libraries are looking for ethical publishing agreements for their scholars, not to pressure publishers interested in fair deals.” Libraries aren’t looking to pressure the smaller publishers who often have better prices, but in practice, that is exactly what they are doing through cancellations.

“Does this affect the price you can charge or a library’s willingness to pay?”

I would think there would be a long-term impetus for universities to scrutinize subscriptions (as always) in their library budgets and ask new tough questions such as, “Why are we paying to make this free to everybody?”

So librarians will need a convincing answer to this question, and nonprofit publishers will need a convincing plan to add subscribers willing to subsidize things for everybody else.

I just have one question, which arises from this:

“If we don’t get full participation, the project will not succeed… There is no opportunity for subscribers to free-ride.”

I can’t imagine this means that if _any_ current subscriber cancels, the project will dissolve — after all, some customers will cancel (and others will newly subscribe) in the normal course of business, for reasons that have nothing to do with a desire to free-ride, and I’m sure that kind of normal churn is expected. So if the project is designed to withstand at least some degree of churn, can you tell us roughly how much net attrition it can handle and still remain viable?

Thanks Rick. As you say, we do experience a normal churn of a few percent of subscribers, and we could handle a reduction in subscribers on this order. With the content becoming freely available, there will be opportunities to attract new subscribers, possibly contributing at lower levels, to cover this regular attrition.

The incentives on the part of the funder aren’t there for this model, and it certainly isn’t sustainable. The longer you are successful in publishing OA content, the weaker your sales pitch to libraries, universities, corporations, government and state agencies, NGOs, schools, hospitals, and prisons becomes. Furthermore, new customers will be introduced to your content by getting it for free. Ask online news sources how easy it is to ask for money after years of giving away content.

You will need to start looking for alternate revenue streams for this to work long-term.

Sorry to be cynical, just my 2 cents. I do think someone needs to pay for OA, but I don’t think it’s libraries.

Why not the libraries Tammy? They already have budgets to purchase scholarly content and expertise in collection curation, and the great majority of librarians support the principle of open access. Subscribe to Open provides the opportunity for them to support it in practice. If they choose not to, as you believe they will, they are in the situation of having to have a gated access subscription.

Incentives? An OA goal realized, easier access, a service to the community beyond the institution, a lower cost and a much bigger impact for your spend.

It is a lower cost in the short-term. If your program is a success, you will then be asking them to pay for a large amount of content that will always be free to access. The longer your program is successful, the less incentive there is to keep paying.

I guess my larger issue stems from my ignorance about who really controls a library budget. Is it the university/medical school president? Is it a board of directors? If a librarian wants to support OA, great. But I have a feeling that those that control how much money a library has to spend won’t be quite so idealistic.

I agree, Tammy. This is what I was thinking about in my comment above. Libraries are famously under fiscal pressure from their institutions, How will librarians defend paying a subscription to make content free to the rest of the world?

Thank you for your questions. To be clear, the Subscribe to Open program remains a subscription, i.e. the only way an institution can guarantee having access to the journal is to participate. They are further being incentivized by a discount. It is only if we have all subscribers participate that we will reach a critical mass and be able to publish subsequent volumes of content open. Ultimately, this model is an institution’s own economic self-interest: the only way to get guaranteed access for their patrons is to pay (as in a subscription), and it is cheaper. If we do not get sufficient participation, the subsequent years content would be paywalled, and any institutions that they demand content would have to pay the higher undiscounted price to access content. So this model is not about paying for making content free for the rest of the world: it is about providing a subscription option, incentivized with a discount and therefore in the interest of each subscriber. Open access publication is an outcome that will remain conditional on participation each year.

I was glad to see this post because AR has been on my mind. It absolutely makes sense to me that AR is unique and cannot rely on APCs to fund OA.

But – in 2018 AR transitioned to a tiered pricing structure. On this campus the tier assignment and ramped up pricing have resulted in large subscription price increases. If others are seeing big changes in subscription prices that they cannot absorb, there may be cancellations. Having Subscribe to Open rely on a stable subscription base when subscription prices have gone way up doesn’t seem like a stable path to OA. But perhaps they haven’t gone up for everybody 🙂

I know one large research university that cancelled the AR full package and only subscribed to 9 or 10 individual titles after they introduced the tiered pricing. So, it has happened.

Thanks you for your comments Joy and Robert. I appreciate your understanding of our unique situation, Joy.

The introduction of tiered pricing served two purposes: it moved Annual Reviews to a sound financial position and introduced equity to the pricing. It has been accepted by the vast majority of our customers although in small number of cases, changes were made to purchases.

It is partly because of this strong affirmation from our customers, but also because of the goals that we share with librarians of curating knowledge and making it widely available, that we have pursued Subscribe to Open as our approach to OA. In the thread above, Kamran describes why Subscribe to Open is a continuation of the subscription to Annual Reviews for our customers.

Cheers to Annual Reviews for launching this innovative program and to the libraries who subscribe to support it. I suspect one advantage for libraries participating in Subscribe to Open would be continued access to COUNTER reports. Many APC-driven gold OA journal publishers do not provide libraries with any usage data. PLOS for example does not provide COUNTER reports. Libraries aren’t seen as the customer for APC-driven journals so why bother? The result is that librarians have a blind spot regarding OA usage. This program I think prevents that blind spot and assures libraries see the value of their subscription payment via COUNTER, where they can continue to compare cost for access etc with other publications.

Mark, thank you so much for your important observation. Certainly, subscribing libraries will continue to receive counter-compliant usage data under this new model; as well as regular reports from us on the broader impact of their investments. We aim to help libraries continue to make the case for subscriptions- to support usage at the own institutions, and beyond.

Comments are closed.