In this article I suggest that some society journal publishers may wish to consider moving their journal program to a Diamond open access (OA) model. Nice idea, right, but easier said than done.

Let’s first take a brief look at the OA landscape. There are many varieties of OA journal. There is no question that Gold OA has taken root as the primary model, but is less than a perfect model, and sits alongside Green OA emerging as a flawed alternative to Gold. And then there is Diamond OA. What is Diamond OA? Essentially, Diamond OA is a form of Gold OA that does not include a requirement for authors to pay article processing charges (APCs).

diamond lock

I am not going to delve too much into the gold versus Green debate here, as there are many fine advocates of both, able to floridly create arguments that sometimes border on fake news to promote their worth. I will turn to one recent, excellent, and balanced post in the Scholarly Kitchen from Rick Anderson, (The Forbidden Forecast: Thinking About Open Access and Library Subscriptions) who at a recent conference in London (Researcher to Reader Conference, February 2017) posed the question:

“The question at hand is this: is there good reason to expect that open access (OA), and particularly OA of the Green variety, is likely to lead libraries and other customers to cancel their paid journal subscriptions?”

The context for this question is in Rick’s observation that, “…Big Deals themselves may be losing their iron grip on library budgets.”

This is something that society publishers are keeping a close eye on, especially those that remain independent and are not included in commercial publisher big deals. In fact, many of us would love to see the back of the big deal. Rick goes on to suggest that Gold OA publishing is not a threat to library subscriptions:

“The emergence of, for example, a new Gold OA journal that competes with The Lancet does not lead our faculty and students to give us permission to cancel our Lancet subscription. What they will want us to do instead is add the new Gold OA journal to our online journal list, while continuing to support the Lancet subscription as well.”

Of course Gold OA is fine as a low risk means of starting new journals – just look at Springer’s blossoming program of Gold OA journals for evidence that this is a major money spinner for publishers. For a society publisher, especially one in a field where there is little funding and negligible author appetite to pay APCs, Gold OA represents a problem. Another flaw for a society considering how OA may affect its journal publishing is in the pervasive rhetoric that somehow OA is synonymous with free access. There are real costs to managing a strong journal, and real revenues received that fund the many programs that such societies provide to their academic community. I recognize that at this point the reader may think I am veering off into an anti-OA rant, but actually I am not.

In Diamond OA I am not including freely available alternative hosting arrangements, such as preprint overlay journals from Episciences, or Discrete Analysis. These are low cost operations for new journals that in my view are not definable as Diamond OA journals. The question for many societies, especially those whose profile is resolutely independent, is how to publish their journals effectively, given market pressures, and given that there is a complex and intriguing blend of business and mission that propels a society’s future. Societies are mission driven. For example my society, the American Mathematical Society’s mission is:

To further the interests of mathematical research, scholarship and education, serving the national and international community through publications, meetings, advocacy and other programs.

Society journals in many cases are important journals for the field, perhaps subject specific, or generalist journals, offering a wide range of sub-fields in the discipline. In face of the big deal, independent societies with this profile are experiencing significant subscription attrition. Societies are looking to innovate their business models, and yet do not necessarily want to burden their communities with APCs. One way to reimagine journal publishing at a society is to accept that journal publishing is in fact a program of the society, provided to the community as part of its mission. Moving journals to a Diamond OA model removes these journals from the journal subscription market, and fulfills the mission. The journals remain strong as established, branded, quality journals.

Revisiting the statement “easier said than done”, this approach requires significant funding. There is a big barrier to entry for a Diamond OA journal program in that it requires a large investment, perhaps greater than investments needed for starting a journal program from scratch. This is because annual funding needs to come from spendable income from investment earnings, and the 10-year outlook from many investment analysts is that 10-year earnings will be much lower than historical earnings, so spending rates are relatively low, requiring a hefty corpus of endowment funds.

A society may choose to use society funds, combined with donor funds to endow an operating fund for their journals, with the understanding that you would need to replace current revenues from subscriptions, recognizing that the surplus from publishing supports other programs. What you would hope for in endowment management is to earn the spending rate plus inflation through investment earnings. You do this by applying the spending rate to a four year average of invested fund balances so that ups and downs in the market don’t affect the spending too much. However, it does not allow for growth, it only allows for costs keeping pace with inflation. Growth in the program can only occur if more funds are added to the endowment through donations or other additions, such as earnings from operations. In summary then, with good financial management of endowment funds and appropriate spending policies, societies might expect to maintain a steady flow of income, keeping pace with inflation.

In a sense the Diamond route is a nod to future-proofing, recognizing that if you can act now, do so and look for growth rather than accept slow decline.

Diamond OA may represent a way for societies to mark out a unique path to success in a way that a commercial enterprise that needs to satisfy shareholder value would be unable to match. Perhaps this is part of the future for society publishers.

What do you think?

Robert Harington

Robert Harington

Robert Harington is Chief Publishing Officer at the American Mathematical Society (AMS). Robert has the overall responsibility for publishing at the AMS, including books, journals and electronic products.


21 Thoughts on "Diamond Open Access, Societies and Mission"

The funny thing is that diamond (free) open access seems to be easier for smaller organisations than larger ones.
First Monday ( and the Journal of Science Communication ( are two examples of diamond open access journals that have been free since they started.

I think that is the trick – start without any charges and that model is built into your operating plans. Once you become addicted to the revenue, it is very hard to turn off the money tap (as you point out).

Scale does seem to be an issue for many Diamond OA journals. If I recall correctly, the majority of journals in the DOAJ are Diamond, but they produce a minority of OA articles, where a smaller number Gold OA journals produce far more articles. When one is relying on volunteer labor, as Diamond OA journals often do, there are limits to the quantities one can process. Once one starts getting thousands and thousands of submissions, many of the jobs become full-time activities, and no longer something one can do in one’s spare time.

There’s an interesting notion here that lines up well with Malcolm Gladwell’s argument against carrying large endowments solely for the sake of the endowment itself (, as well as our own Kent Anderson’s thoughts here:

But it raises many questions:

Does this strictly limit the ability to publish a Diamond OA journal to societies with a significant amount of money to create an endowment? Is this a route only available to the rich? How would one ever create a new society/journal?

Once the journal is created, would there be any motivation for investment/improvement? Why bother to add new technologies/services when you could instead put that money away into the endowment for future rainy days?

If a society is reliant upon journal revenue to do its other services (meetings, education, training, advocacy, etc.), then would this require giving all of that up? At that point, If I’m a member, what am I getting out of my membership? There’s a free journal, but I get that whether I join or not. So that would likely mean an end to membership. What then, is a society with no members and no services other than a free journal? Would this mean abandoning being a society and transforming into a non-profit, subsidized publisher, basically the equivalent of a university press that is dependent on subsidy? Not a bad thing to be, but completely different from the purpose and activities of a society.

“In Diamond OA I am not including freely available alternative hosting arrangements, such as preprint overlay journals from Episciences, or Discrete Analysis. These are low cost operations for new journals that in my view are not definable as Diamond OA journals.”

Interesting. So a journal that is free for authors and readers counts as Diamond OA only if it is sufficiently expensive to run?

Indeed, this seems bizarre to me. Look at outputs, not inputs!

Well, that’s sort of encoded in the label Tim…

Diamonds are valuable only because they are expensive. With no utility and differing little from coal, diamonds would cost about the same as coal per mass unit, were it not for De Beers cartel control over supply and brand image for the past century.

I am sure there’s more than a few allegories there with academic publishing and academic value systems, but I’ll leave that be. My inner troll is all fed up already 😉

On David’s point. I think it depends on the profile of the society. It is all about how to best use and partition the funds you have at your disposal, and what your society brings to its membership. It may be that your content streams are diverse – books and databases for example, and the whole constitutes part of your brand. Perhaps you have a multiplicity of programs that serve the membership and the wider community beyond members – though you really want non-members to become members. What you do for the community, along with some actual benefits is why your members may join – but this is an important question for our time – how to grow membership? For some societies, while your journals are important it is not for these that individuals become members. One could argue that by pursuing this more open model the impact of a society may grow, and you may attract more members – the opposite really of what you suggest.

Many individuals do join for cheaper access to journals at membership rates, this is proven.

It can cost hundreds of thousands of dollars to launch a journal properly – how would such costs be recouped? If a society has a great idea for a journal that meets a need – it is more likely to launch it as a subscription journal to capitalise on the great idea. Societies also often launch journals that are free to access and free to publish in for a period of between 1 and 3 years — another period of expenditure where the costs need to be recouped. Diamond (or Platinum as it has been previously called) will just not wash its face without significant guaranteed perpetual support (a huge risk). eLife could have been considered to be diamond but was completely unsustainable as such and now charges a $2500 APC.

One possible source of endowment money that has not been much talked about is the estates of scholars. A society might encourage its members to include the society in their estate planning. At Penn State Press we actually had the longtime editor of one of our top journals make a substantial bequest to the journal in his will. Universities are always encouraging their alumni to consider a variety of ways of benefiting their alma maters through estate planning, so why should not societies take a cue from what universities are already doing?

Great Article Robert. The challenges to societies is indeed great they are fighting for members, being challenged by the web, and commercial houses are duplicating their efforts.
However, living on an endowment is not easy. Especially if you relied upon it in 2008!

It costs between $3500 and $4000 to publish an article ( You need a very hefty endowment devoted to just the publishing activities of a society to meet these costs.

Even ELife which was endowed by the Welcome Trust to the tune of 43M Pounds now has an APC of $2,500.

Some may pooh pooh the costs cited. And, if you have a small journal with three or four volunteers the costs do go down. But, the big question is what happens to the journal when the volunteers stop volunteering?
Lastly, it is expensive to publish scientific articles and as has been said: There is no free lunch!

There is no way that $3000 plus can be anything near the correct cost for publishing a typical article. Large publishers spend a lot of money on irrelevant “services” and inflated profit margins. If small journals are actually more efficient, why not just have more of them? What is the need to make journals large? One big problem with scaling a journal is that it attracts submissions from authors with little connection to the scholarly sub-community of the editors. This makes peer review harder. Also, the risk of an extreme bad event (eg legal action) becomes high enough that resources must be spent to defend against it. It seems to me that small scholar-run journals, with modern technology, and very small subsidies from library consortia, should be enough in most fields to do what we want journals to do – ensure quality of research and add some curation. “Publishing” in the sense of making available to the public to read has been fairly trivial for years. It is a shame that many “publishers” spend so much to keep the public from reading their work.

One way to have Diamond OA viable, affordable and sustainable is thru collective / common publishing platforms as we have been operating in SciELO Network in the last 20 years. SciELO Network operates decentralized national collection of journals. Each national collection replicates the platform. Today, there are 15 national collections (12 from Latin American countries, plus Portugal, Spain and South Africa), totaling more than one thousand journals from the different disciplines which publish about 50 thousand articles per year. The accumulated database hold about 700 thousand documents that serve daily over 1 million access and downloads of HTML and PDF files.
Journals maintain their editorial autonomy, editing rules and styles. SciELO platforms provides storage, publishing, preservation, retrieval, interoperability and performance evaluation.
Replication of the platforms was the best way to develop SciELO Network. Today an additional layer of reducing cost that we are considering is to move the operation to cloud computing.
Most of SciELO collections are funded by national research agencies. If public policies and funding are not available journal publishers may establish a common solution.

This is an excellent point. The article really only applies to the odd US structure for funding.

What you propose in this post is very close to what in Latin America we have called “Open Platinum Path”. For us in Latin America, it has been a tradition for universities and professional associations to fund academic journals as part of their mission – although in our case, not for profit.

It may be useful to consider the evidence from UK learned societies in the social sciences. Colleagues and I have a paper now on early view with Science and Public Policy which examines the finances and activities of 44 societies in 2012/13 Basically, no more than 5 could seriously consider the policy proposed by Robert Harrington – for most of the others that publish journals the consequences would be pretty devastating. With the stagnation of academic salaries and university conference and CPD budgets, there is no scope to increase other income streams to compensate. The societies that could go Diamond tend to be those that are also involved in professional accreditation that generates funds from outside the academic world. For the rest, we would be looking at the loss of a lot of functions that are often overlooked but important to the vitality of national and international research systems.

I would be interested in reading your article, but sorry, €33 for 24 hours access…..

I find this post and most of the discussion (if I understand it correctly) completely unrealistic. Even a relatively small-scale society-sponsored journal would require an endowment of at least $20M to generate enough income to keep it afloat. And that wouldn’t leave much for the Society to do enough to continue to attract members. I suspect only the largest societies could raise endowments of that scale (but the largest societies sponsor many journals, each of would require at least a $20M endowment). This just doesn’t make sense.

And yes, the cost of producing an article is about $3000 (give or take a few hundred).

You’d be surprised at how large an estate gift can be. The gift from the Penn State journal editor was in the $20M range, as I recall.

ACS Central Science is entirely open access, with no subscription fees or article publishing charges for authors. The Editors anticipate publishing no more than 200 articles per year, which might be explained by the fact that there are ~150,000 members. Perhaps ~$4 per member is a reasonable society expense for a sustainable ‘Platinum’ research journal.

I am curious how large the endowment would need to be to fund AMS’s journal publishing. It appears there are about 25 titles. Based on the $3,00 per article estimate, would $50 million do it? $100 million? There will likely be solutions to the publishing issues facing society journals, not a solution. I don’t see why this idea couldn’t work for some journals, for some societies. I really liked the mention of SciELO above.

Comments are closed.