According to recent reports, the advertising model underpinning digital media companies like Reddit, Buzzfeed, and others has been stretched too thin. There are simply too many companies trying to attract dollars that are finite and increasingly captured by Google and Facebook. The implications of these reports are captured in a recent article by Josh Marshall in Talking Points Memo, which lays out the case in convincing detail. Marshall is the editor and publisher of TPM. He writes:

What crystallized for me from this and a few other discussions I had yesterday, though, is that we’re actually in the midst of a digital news media crash, only no one is willing to say it. I’ve mentioned before that digital news media in the midst of a monetization crisis. But again, it’s a full blown crash.

Earnings reports for some digital media darlings led to the following tweet from Jeremy Owens, the San Francisco bureau chief for MarketWatch:

Buzzfeed’s missing IPO is another implication of the disappointing revenue reports, as venture funding, which can sustain new companies through initial revenue gaps and market headwinds, will now be less likely to materialize, if it comes at all. That means the end is nigh for some portion of these companies, once the VC funding is gone. And new companies looking to start won’t find ready funders. The boom in online news may be over, and all because there was an over-dependence on advertising revenue as the primary source of funds.

ice cream stand

The obvious lesson is about market saturation — there is only so much money to go around, and the herd will be culled when sustenance runs short. The less glaring lesson is about the need for variety and redundancy in business models, the need to diversify the risk to revenues — if each of these entities had other significant revenue streams to offset down-cycles in advertising, they may be able to weather the ups and downs of the markets more easily.

In our industry, running parallel revenue streams is sometimes called “double-dipping,” a term soaked in overtones of illicit behavior and corruption. But is it corrupt to charge a subscription for a hybrid journal while also realizing revenues from article processing charges (APCs)? My stars and garters, what kind of scoundrel would do such a thing?

The answer is, every business. I’ve written about this before, outlining how governments, universities, OA publishers, non-profit publishers, commercial publishers, and others charge multiple times for the same thing. In short, it’s normal, expected, and helpful.

There is a rich history of blended revenues providing incredible information access, allowing free or inexpensive subscriptions for thousands of physicians and scientists.

There is a rich history of blended revenues providing incredible information access, stretching back to the heyday of print advertising allowing free or inexpensive subscriptions for thousands of physicians and scientists, to small processing or optional fees for publication offsetting editorial costs and allowing lower prices for libraries and subscribers.

Using the phrase “double-dipping” casts aspersions where none belongs. Multiple revenues streams help consumers and purchasers by spreading costs and decreasing the burden on any single payer. This is what can be missed in discussions of such practices. We are currently seeing the negative effects of an over-reliance on library budgets. Our information economy has grown tremendously in the past 10-15 years, driven largely by both organic growth and the emergence of China in particular as a source of new research findings. Absorbing these inputs has left publishers scrambling. At the same time, online advertising has proven less lucrative and reliable than print was, while APCs are largely capped and also lack the long-term flexibility of licensing or other traditional syndication approaches (in fact, open access [OA] largely eliminates licensing or syndication revenues, further constraining publisher options).

Imagine having to front the entire fee each day for your Internet service at home — the depreciation on the hardware, the salaries of the support teams, the pensions of the service people, the land leases with the various communities allowing wire or tower placements, the office leases, the healthcare costs, and the licenses for hardware and software. You probably couldn’t afford a minute of Internet access each day. But with these costs spread over millions of consumers, and with other revenue streams — business Internet, security services, hardware rentals, and more — the prices for Internet services are high but tolerable, which is to say value for money. Is the ISP “double-dipping” because they are charging multiple consumers for bandwidth? Are the “double-dipping” because they charge you for the high-speed modem and your DVR as well as the underlying services?

Markets by definition have multiple parties seeking value — producers and consumers, typically. One significant change in our market is that producers (authors) are now explicitly seeking value via APCs. Does their desire to pay for value obviate the consumer value proposition? If not, then there is no “double-dipping” in the traditional sense. There is only charging each party what they are willing to pay. That’s not illicit or unethical.

A misapplication along these lines of the term “double-dipping” can be seen in a recent tweet from Richard Poynder, citing a practice from SAGE Publications involving a publication charge levied for certain types of content:

Publishing has a rich history of cobbling together multiple charges in order to generate revenues and keep other charges from going up too much. From page charges, color charges, and other publication charges to submission fees, APCs, and soon I’m sure data-handling charges and open data storage fees, small charges on the front end or back end of publication have been common for decades, and will continue in all likelihood. In my opinion, it’s entirely likely that preprint servers will soon be levying charges, assuming the form gains enough credibility, something that appears bound to happen in 2018 or 2019 (at the latest).

Multiple revenue streams make sense for any business. In our economy, if authors pay a small amount or percentage of fees, while institutions pay another larger share, and advertisers perhaps contribute some portion of revenues, while rights deals and other secondary revenues provide additional income, a publisher has a good variety of sources, some more predictable than others. For professional societies, a membership fee or allocation of dues can also provide another stable source of revenues. Is the publisher “double-dipping” by charging members and institutions a subscription fee?

Uncertainty of funding is a major issue for any business. For example, eLife recently secured another philanthropic revenue stream to support its underlying services, adding the Knut and Alice Wallenberg Foundation to its list of founding funders (Wellcome, HHMI, and Max Planck). Diversifying its funding options increases its likelihood of success. The German government’s infamous negotiations with Elsevier are another example of uncertainty, as are new sources of competition (the new OA journals from the Gates Foundation), changes in the macro environment (new EU approaches to OA, new pressures on US universities due to decreasing overseas applications and potential changes in the US tax code that could stifle post-graduate education and tax endowments), and simple bad luck (natural disasters, getting hacked). Businesses have to survive a lot of uncertainty. Organizations that put all their eggs in one basket are vulnerable to failure from one unlucky environmental or market change.

From a purely selfish perspective, libraries should be encouraging double-dipping, in fact, as it reduces the likelihood of either untenable price increases or cancellations of useful resources. If APCs, advertising, licensing, and other revenue streams mean the difference between a 2% vs. an 8% or 10% price increase, aren’t they helping relieve the burden? Aren’t they allowing more information to flow through the library and reach patrons?

One downside of effectively diversifying revenues can be that outsiders can believe that what they pay is the actual cost, after some back-of-the-envelope math. Let’s assume I’m paying a $1,500 APC for a hybrid OA journal, and they publish 1,000 articles this year. I might calculate that it costs them $1,500,000 to run this journal. However, diversification conceals the full costs by allowing the APC prices to be lower than the full cost. The journal actually costs $3.8M to run and return a decent margin. If fully supported by APCs, the charge to each author would more than double to $3,800. Diversification defrays costs to every participant in a business’ micro-economy.

Unfortunately, smaller non-profit publishers in our world seem more vulnerable to double-dipping objections, because they are less complicated organizations with smaller portfolios run by boards that are more likely to find appeal in the objections. The current focus on this certainly has a chilling effect on normal revenue diversification approaches for smaller non-profit organizations, causing them to talk themselves out of adding reasonable fees or subscription increases. I’ve heard a few of these conversations, and the imprecision of the objections can cause editors, boards, and publishers to quash new revenue streams simply to avoid being accused of double-dipping. Doing so leaves them with few options, which over a period of a few years can lead them to simply throw in the towel and contract with a large commercial publisher. These publishers, with their larger portfolios, global customer base, complicated fee structures, and vast existing diversification can shift revenue streams around more capably and without setting off alarms, appearing to avoid double-dipping while retaining their diversified revenue streams.

Double-dipping purists indirectly drive market consolidation in this manner.

Watching the larger information space, there are lessons to be learned about relying too much on a single revenue stream — whether a funder, a set of purchasers, a single payment approach. Doing so is risky to the point of being reckless. Calling prudent revenue diversification and the normal practice of stacking smaller fees up to get to sustainable revenues “double-dipping” is casting aspersions without considering how helpful diversified revenues are — to managing prices overall, to ensuring sustainability, to making information affordable, and to smoothing out revenue variations. It’s more akin to creating a safety net, so that a crisis doesn’t wipe out valuable scientific or scholarly outlets.

If we want a scholarly publications economy that works for more organizations than just a few large corporations, we need to allow smaller organizations to diversify their revenue streams, perhaps to the point of actively encouraging it. There is only good that can come from fostering diversified economics that work at the small and large scale. Otherwise, we’ll continue to move toward a commercial world where revenues concentrate. Such a world isn’t good for libraries or publishers. Worse, we would have helped this happen by portraying revenue diversification as an illicit and unacceptable reality, when in fact it’s exactly how any business, and smaller organizations in particular, survive.

Kent Anderson

Kent Anderson

Kent Anderson is the CEO of RedLink and RedLink Network, a past-President of SSP, and the founder of the Scholarly Kitchen. He has worked as Publisher at AAAS/Science, CEO/Publisher of JBJS, Inc., a publishing executive at the Massachusetts Medical Society, Publishing Director of the New England Journal of Medicine, and Director of Medical Journals at the American Academy of Pediatrics. Opinions on social media or blogs are his own.


25 Thoughts on "Creating a Safety Net — Why Double-Dipping Is the Wrong Term and the Right Approach"

Many smaller businesses require multiple sources of revenue, but the recent complaints about advertising as a support for content is that the concentration of distribution (mainly in Facebook and Google) is making advertising an illusionary goal for businesses that formerly counted on it. So it really isnt going to be a second or third or fourth pillar. That really is a problem for publishing businesses that have not anticipated the revenue slide. Subscriptions are a better bet, and should happily co-exist with other service charges.

I totally disagree. The problem is the business is built on entwining publishing research and evaluation of researches. Universities and institutions are forcing academics to meet the numbers in terms of papers p.a. and IF of the journals that they publish in. The bean-counters have taken over; academic publishing has no longer higher aims of simply publishing great ideas, but it is now devising ways for researches to collect brownie points at every opportunity. We are building ever-bigger haystacks, but are not taking care of ensuring that researchers are able to work freely in suitable creative spaces; hoop-jumping is the name of the game. All the while the large publishing houses are making more profits than virtually any other sector, and that on tax-payers money. Sure the smaller publishers are going to be having ever harder times. But we have to re-boot the system not keep it on life support with double dipping.

I would suggest you’re conflating two separate issues here. The problems you discuss are largely generated by academia itself, a result of the career advancement/funding structure that is in place. Scholarly publishing is a service industry, and its practices reflect the demands made upon it by those academic structures. We often see publishers blamed for the problems that these academic practices cause, but this is really a case of trying to treat the symptoms, rather than the disease.

And as noted in the article, allowing for experimentation and finding alternative revenue sources helps the smaller and independent publisher work against the hegemony of the big corporate behemoth. Licensing and advertising revenues, for example, offer relief from having to put tax-payer money toward publishing. Certainly in the US, a good deal (if not the majority) of subscription revenue comes from tuition and student fees rather than out of grant overheads. The more money we can bring in that doesn’t come out of the research community itself, the better.

All of the issues you’ve outlined as far as the academic reward system are taken. There is too much reliance on quantitative (and mostly flawed) measures of academic achievement via research reports. However, there are major benefits to the system, as well, including incentives to produce new findings, generate interesting ideas, and move research forward. Things have moved too far toward dumb numbers, however — I agree.

Taxpayers are supporting less and less research and academic budget than ever, especially in the US and the UK. Tuition, fees, and grants are having to fill the gap. The old model of taxpayer-funded research is less and less relevant. In the US, 70% of research and development is funded by private industry.

One of the unintended consequences of the last 20 years of attempting to rectify inequities in the system has been that larger publishing companies have gotten larger and become more influential, exactly the opposite of many of the stated goals of the disruptors. Casting aspersions on attempts to diversify revenue streams only further hampers the ability of smaller organizations to succeed. I do agree that rebooting the system is a good direction, but remember that the initial state of the OS was around subscriptions — which actually make it easier for smaller organizations to have better market success based on the value they offer, while decreasing the sway of producer-oriented (IF, metrics) measures of value.

We keep shying away from the subscription model, but it has many virtues and few downsides compared to everything else on offer.

Who is actually compensating the reseachers (mostly paid by public or private research institutions) who provide free labor as authors, editors and reviewers to the publishers?

By the way, not the German government is negotiating with Elsevier but a coalition of the research institutions – makes a big difference. And I cannot see an infamous behavoir if a customer demands an appropriate service for his money from a monoplistic provider.

The researchers are compensated by those paying their salary and offering them funding. Scholarly publishing is a service industry. Journals and books perform a long list of services for researchers, meeting important needs that they have for things like communication, career advancement, reputation building, etc. And doing these things, particularly doing these things well with rigor and care, costs money. Subscription journals provide those services to authors which they generally get for free, choosing instead to pass the costs along to consumers of the literature through sales of the content. One of the really useful things about Gold open access is that it makes this a direct transaction — the researcher pays directly for the services rendered.

And in my experience, the editors of journals are often compensated (sometimes well-compensated (

“If APCs, advertising, licensing, and other revenue streams mean the difference between a 2% vs. an 8% or 10% price increase, aren’t they helping relieve the burden?”

That’s quite the “if” statement. I’d welcome evidence that such a situation exists. I suspect many librarians think that it all those revenue streams and also an 8-10% increase. Wouldn’t a publisher, accountable to shareholders, want to maximize all revenue streams, not trade them off against each other?

This assumes all publishers have shareholders and a priority of maximizing all revenue streams. #notallpublishers #universitypresses #supportresearchsocietypublishing

Fair enough. But, for those that do have shareholders, which likely form very large items in a library budget, the question stands. (I think there are reasons to think that other types of publishers might face pressures to maximize revenue streams as well … they do have purposes for the profits they make.)

Again, that gets back to the community. I think Angela Cochran raised useful questions when she noted that, “These journals must be doing something right” to be making all that money ( If there are publishers that the community sees as exploitative, then the community should adjust its behavior to punish those behaviors.

But that’s still a separate matter from the usefulness of diverse revenue streams, the stability they offer, and the ability they provide for funding publication with sources outside of the research community itself. Like open access for example, even though some may abuse it, the concept has value.

You didn’t write the original post but the claim that librarians should support double dipping was followed by hypotheticals. If there is evidence, is like to see it. If not, I indeed expect librarians will continue efforts to punish these behaviors. It’s going to take more than hypotheticals to convince librarians to join the double dipping enthusiasts.

I’ll let Kent answer that question, but to me the problem is that things that aren’t really “double-dipping” get called that in kind of a knee-jerk reaction, when in fact, they can be helpful to all involved. Is the fact that eLife gets money for publishing both directly from funding agencies as well as from authors “double-dipping”? Is that a bad thing or would it be better if they just charged authors the full $7K per article it costs them? The same goes for hybrid journals — subscription revenue means that APCs can stay at the current market level instead of the much higher price they’d need to be to sustain publication for many journals. It’s a poor terminology and the idea of the value of multiple revenue streams (particularly ones from outside the library/researcher) is one that deserves more exploration and acceptance.

I’m not automatically against multiple revenue streams. But, I want the models evaluated on what actually happens… not “what ifs” that sound good but are counter to experience.

I agree that it is a problem if the term “double dipping” is used erroneously. But, that doesn’t mean all uses are erroneous.

Hypotheticals were used because businesses vary in many ways. But they are similar in that a) they don’t like to fail, and b) their best shot at not failing is often multiple revenue streams (Google is an anomaly). If you want evidence, read any annual report or review any 990. Or review your own university’s finances.

Also, part of what this post sought to clarify is that it is usually unfair to call rational pursuit of multiple revenue streams through multiple relationships with different value-seekers as “double-dipping,” which has illicit and unethical connotations. It is not unethical or illicit to divvy up the total charge between multiple parties.

“Markets by definition have multiple parties seeking value — producers and consumers, typically. One significant change in our market is that producers (authors) are now explicitly seeking value via APCs. Does their desire to pay for value obviate the consumer value proposition? If not, then there is no “double-dipping” in the traditional sense. There is only charging each party what they are willing to pay. That’s not illicit or unethical.”
The problem here is an imperfect market with monopolistic tendencies. As a librarian I have seen large commercial publishers (let’s not pretend that small publishers are the problem here) continue to increase their subscription prices while “diversifying” their revenue stream by charging APCs to authors who wish to open their articles readers. And yet, I see no concomitant decrease in subscription prices. In fact, you just made my e-resources librarian’s life more difficult because we used to be able to say our users had access to x volume, but not y. Now our knowledgebase has to somehow try to keep track of not just x and y, but y except for those articles that have been opened. Publishers have myriad ways of indicating that an article is open and knowledge bases aren’t — yet — designed to cope with this level of granularity. This makes us look bad. I could accept this as revenue stream diversification if large commercial publishers were adding value by making us — their customers — look good. But the reverse is happening. Thus I will continue to encourage faculty authors to submit to OA repositories were possible and not pay the double dipping fees.

All markets tend to consolidate. That’s not monopolistic, just consolidation-prone. Granted, recent philosophies around anti-trust have made consolidation easier, as has technology (and technology is very monopoly-prone), so there’s a lot going on in the world of the big getting bigger. One variable is that when there are fewer options for smaller entities, that also contributes to consolidation.

OA was something the academic community dreamed up, and publishers have now embraced. Again, publishers being blamed for the added complexity is certainly not unexpected, but it’s not accurate. They didn’t add the complexity except in response to what academics said they wanted, and OA advocates pushed through mandates, etc.

Also, a small corrective — libraries are not THE customers, but A customer. Readers, faculty, students, researchers, and authors are also customers. This is fundamentally why there can and will be multiple revenue streams.

The suggestion that APCs might reduce the subscription fees is misleading. There are no such things as research money versus library money. There is only institutional money. So, there is no double dipping either. It is just a confusing term.
The core sentence of the blog is, “The journal actually costs $3.8M to run and return a decent margin.” Institutions do not know (and often greet with skepticism) postulated running costs of $3.8M, and reject a standard annual profit margin of 35% to 40% as decent. They embrace the OA model for two reasons: (1) a moral one, i.e. open access to the results of publicly funded research, and (2) an economic one, where OA introduces competition between publishers making running costs transparent and, indeed, bring profit margins to a decent level. That’s all.
Publishers, big or small, who are prepared to meet these wishes are in an excellent position in an OA world.

There are no such things as research money versus library money. There is only institutional money.

I’m not sure this is true. Many grants come with restrictions on how the money can be spent. Many donations/endowments have similar restrictions. I suspect that overall, this varies quite a bit depending on the type of institution you’re in (public/private, teaching/research) and its location (country/state).

(2) an economic one, where OA introduces competition between publishers making running costs transparent and, indeed, bring profit margins to a decent level.

And yet all evidence so far points to OA leading to high profit margins (see Hindawi and every megajournal), and no apparent increase in competition. But the principle discussed in this post still stands, even in an OA world — the more revenue streams that can be found to support publication, the less burden would then be put on the individual author, which would surely be a good thing, would help further the spread of OA, and help eliminate some of the problems with strictly “pay-to-play” models.

Many OA publishers have additional revenue streams to complement their APCs — advertising, memberships, meetings, and so forth. Every organization seeks sustainability.

As for the characterization of publisher profits as 35-40%, they are generally lower (10-15%) for most publishers. Logically, any profit margin that is less than leaving money in an indexed investment fund is an irrational use of funds, as it adds nothing to the economy, and can actually detract. In 2016, most of these funds returned 14-22%, so any profit margin below these levels would be a poor use of capital. It’s always good to look at profit margins in this context, as institutions and others are realizing similar gains in their investment funds, as well.

“I suspect that overall, this varies quite a bit depending on the type of institution you’re in (public/private, teaching/research) and its location (country/state).”
I guess that’s more likely in the US than in Europe. In Europe subscription budgets and APC funds are becoming communicating vessels. And we also may have less donations and endowments.

“And yet all evidence so far points to OA leading to high profit margins (see Hindawi and every megajournal), and no apparent increase in competition.”
Sure. And you could have pointed to the newly settled OA deals in the Netherlands as well. All (but Elsevier’s) articles of Dutch corresponding authors are published in OA now, but at a price (ca 5%) higher than the subscription prices. Yet, the APCs are much lower that the list prices. E.g. the Springer APC is now €1300, whereas their website quotes (and others pay) €2200. Taylor & Francis went down from €2150 to €1500. Etcetera. So, competition is emerging. It is just a beginning and there is still a tough way to go.

“[…] the more revenue streams that can be found to support publication, the less burden would then be put on the individual author”.
Individual authors, like individual readers, never pay anything. As said, it’s all institutional money.

The major structural differences between US and European universities make it difficult to find one solution that works everywhere.

Individual authors, like individual readers, never pay anything. As said, it’s all institutional money.

Are you sure? No author has ever paid an APC out of their own pocket? We still sell a good number of individual subscriptions (particularly in the Humanities where journal costs are much lower than the sciences). And that’s not even mentioning the subscriptions that often come with society memberships, since many grants won’t pay for such memberships, so joining is a personal expense for many researchers.

If a company builds a motorway, it is paid by the state (=tax payers) for doing so (generally the case in the US and Germany). Another way of financing the work is to collect toll fees from the cars driving there (often the case in France and Italy) by the company building an d later running the motorways. Double-dipping is being paid by the state (to the full extent) and still collecting toll fees from drivers. You fancy that?

That sounds like pretty much every road I travel on. The state builds the road, then charges tolls to pay for maintenance. Or more relevant to my daily life, the state pays for the construction of the subway system, and I pay a toll to access it, which pays for upkeep and improvements.

The problem with your metaphor though, is that although the tax payers may be paying for some portion of the research being done, they aren’t necessarily paying for its publication.

I do not get the tax payer argument. There is nothing like a global tax payer. And not every country is reinvesting in science the same % of tax money. It is as weak as the argument that there is enough money in the system constantly repeated by OA advocates.

Institutions should outcry for more funding and not accept that money spent by their governments on science and research is flat or decreases year over year.
Apple, Amazon, Google and Facebook avoid paying taxes in countries they do business in ( and If research institutions arguing that there is enough money in their system, your governments will be very comfortable and not feel the pressure to ask those giant to settle their bills.
On the other hand, tech giants already dominate R&D our days ( With open access you even want them to read for free even they avoid paying taxes? This is naive social romanticism and kind of double-none-tax dipping. You fancy that?

In a next step tech giants and their family trusts show social engagement by donating a friction on the original tax savings back into public education/research/oa projects for what they get applauded by the open access community again. Another absurdity.

My learning point is that by now the whole discussion around Open Access moved from it originates and is just about the proof that others then publishers can foster scholarly communication cheaper. But as an OA advocate your argument needs to be that you can do it BETTER. Even for the costs of double dipping.

This spoof of the taxpayer argument from a few years ago (April Fools post) may tickle your funny bone:

Joanna Bryson recently urged people to return to denominated trade, not barter (your data for big tech’s services, for example). Make them pay. Pay for things. In her view, and I agree, this is how we close the income inequality gap. If you provide good and reliable information, what’s the problem with people paying for that? Knowledge is a more sustainable product than a car, to paraphrase Bryson.

Comments are closed.