In the first part of this three-part series, I discussed the intrusions of governments in scientific publishing, and how their leverage, incipient politics and bureaucracies, and concentration of power all create problems for scientific publishing as they become more entrenched. Yesterday, I discussed how funders are exerting power over scientific publishing policies and practices. Today, I’ll examine the tool both are using to get their way — money. Issues include how it’s used when not wielded by the actual users of the scientific literature; what it is buying; trade-offs and constraints; and whether there is enough of it around to sustain what some envision as a new way to publish scientific information.
Show Me the Money
Before we get into the future of money in a world of open access (OA) publishing that is heavily influenced, if not directed, by funders and governments, we have to acknowledge that Green OA — the notion that free pre-publication manuscripts can be deposited in repositories and made available in a practical and usable manner — has been largely discredited by the multi-million dollar and multi-year PEER study. Green OA is unlikely to work, is difficult and expensive to execute, and doesn’t meet the needs of scientists. Only ideologues will continue to argue that it can serve as a substitute for a robust and professional funded approach.
This is a first, hard lesson about OA — there are costs to managing, enabling, curating, and maintaining access to the literature, costs that were not anticipated by many when the movement started years ago, but which have now been quite thoroughly explored and understood.
Given that someone has to pay for publication in some manner, the question for pragmatists has become: Is there a new way to fund scientific publications? The subscription model has been criticized as increasingly unaffordable. This has become conventional wisdom despite clear indications that more value is being derived from subscription products each year. Even skeptical thinkers have absorbed this misguided talking point, as this line of reasoning in a recent statement from the American Historical Association, which is generally critical of OA, shows:
The subscription prices for many journals, especially scientific journals, have escalated to the point where almost no individuals and fewer and fewer institutions can afford to subscribe.
Yet, the true story of the subscription model’s troubles is one of deprivation and double-standards at universities — as tuition and fees and administrative costs at universities have increased, along with universities’ reliance on soft-money research grants, the libraries’ share of university budgets has decreased steadily for 30 years now. Among individuals, incomplete thoughts like “information wants to be free” and misconceptions about the costs of digital publication as well as extremely poor commercialization of news sites, email, and other digital goods, have undercut the notion of information subscriptions online, despite the success of the subscription model for mobile phones and basic cable, for instance.
The problem with the subscription market is not, as it’s so often portrayed, merely an expense issue. In fact, prices — once adjusted for inflation — have actually been increasing more slowly, and more recently have been falling, for scholarly information. The problem is more specifically how much scholarly information there is, which means the “crisis” is actually and more fundamentally a funding issue within a growing market. There are more researchers producing more research and wanting to read more research. Yet, the percentage of available funds spent on information needs has declined significantly and steadily at institutions, and individuals have a sense of entitlement to free information. This deprivation of subscription dollars is backing publishers into a corner as they seek to grow and meet the needs of a larger and more international research community. The subscription model is being soured. Demand-side funding is falling even as demand itself is increasing.
The subscription model has also been decried as exploitative of authors and reviewers — authors provide their content for free, and reviewers worked for free, and publishers make money off their donated labor. OA publishing has been trumpeted as some sort of solution, but has only revealed how benign the subscription publishing arrangement is — reviewers for OA journals still work for free, but now authors are paying in large numbers to have their works published. OA publishing moves money from the demand side of the economic equation to the supply side, with many significant consequences. But it still thrives because of the free work of academics. It now also takes money from academics, on top of asking for free labor. (Of course, these arguments about “exploitation” have never made sense — it’s perfectly rational economic behavior throughout.)
Rational or not, the conduit of author payments to pay for OA provides funders and governments with unprecedented leverage over how publishing works, what gets published, where publication can occur, and copyright and licensing issues. When governments get involved, politics gets involved, and the money of politics gets involved. We’ve already seen paid lobbyists trying to sway the government’s rule-making around OA. The frequency of this kind of petitioning for political favor will only increase as the stakes rise.
The author-payment model also provides a less risky and possibly larger — and therefore more seductive — source of revenues for publishers. This is a potential carrot for publishers, with mandates and funding threats acting as the stick for authors. Joe Esposito contemplated some of this in a 2004 essay:
Unleashing academic publishers from the user-pays paradigm is about the most remunerative activity I can think of. Currently publishers can make money only when purchasers (mainly librarians) apportion a certain amount of their budgets to them. Such purchasers don’t really have much skin in the game. . . . Authors, on the other hand, are acting out of personal impulse. . . .
The “personal impulse” for authors can be exploited by funders to drive behavior — funding can be cut, increased, or diverted if the funder is unhappy with adherence to publishing policies. This is exactly how funders and governments are enforcing OA mandates currently — by threatening funding if there isn’t compliance. Money is a carrot to publishers, a stick for authors.
The questions this begs are:
- Is there enough money to accomplish an OA world?
- Where will this money come from?
The UK is where government mandates are farthest along, and judging from early observations, the news is not heartening:
Academic freedom is compromised by a ‘pay-to-say’ system, because institutions and academics will have to bid for the funds to publish their work. This means for academics that unless they are rich enough to pay for the publication of their own research, they will have to convince non-expert committees of the value of pre-published work, and compete against other University colleagues for funds. They will be restricted as to what they can publish and where. It is clear that Institutional Publication Committees will have to ration funds in line with pressures for REF and impact, meaning that lots of potentially valuable work will go unfunded.
Where the funds come from, and the unanticipated expenses of enacting a system of funded publication, is also raising concerns:
There is a huge shortfall in the money being provided to kick-start open access, and the amount which would be required to fund current research outputs. Professor Tim Blackman, Pro Vice Chancellor (Research Scholarship and Quality) at the Open University has estimated that to get to the minimum suggested 45% level for ‘gold’ open access, his university would have to find an extra £740,000 per annum, possibly rising to £1.6m per annum. For all institutions, these will have to be found from already tight budgets, wasting money that could be spent on research itself or other scholarly activities.
Administering Institutional Publication Funds will also take up vast quantities of academic and administrative time, as non-expert committees will have to make impossibly contentious decisions about colleagues’ pre-published work.
Costs should be a major concern because OA publishing has an inherent drift toward scale. It tends to get big, because that’s how the economic incentives of author-pays works — the more you publish, the more you make. Author interests align with more publications — authors want publication, and the fewer barriers, the better. Editorial review, statistical review, rejection, resubmission — all these things are barriers. They are also expenses. OA reconciles these things neatly — eliminate the expenses, and you eliminate the barriers, leading to both a satisfied customer (the author, and through them, funders and governments) and a more profitable business. You also move to quantity-based publication, which is a trait of many OA initiatives, from mega-journals to fly-by-night operations that suddenly launch hundreds of journals. When the supplier is paying, there is a wholesale approach rather than a retail approach to publishing.
The risk of quantity-based publication is that it subverts quality. This can be hard to detect, because low-quality work in academia usually dies silently and alone, rather than being flagged as wasteful and irrelevant, especially if its publication in an OA journal satisfied everyone in the value chain who paid — the author, the funder, and the government. When no reader is paying for it, nobody complains about wasted money. It’s also well-known that if you don’t charge for something, it’s hard for people to complain about its quality. When readers are paying, standards have to be higher — users want their money’s worth, will complain when they don’t get it, and this virtuous circle improves quality. As Dan Fields wrote in the Huffington Post, the subscription model rewards quality:
. . . the quality of the journal was validated by its readers. If the journal consistently published important and accurate studies, subscriptions would rise, income would increase and authors would strive to publish in those prestigious journals.
In a model in which high-quantity matters, small and more selective journals are more likely to fail or be swept up. Consolidation will be more extreme. OA will be most profitable when it’s big. And OA will be big, if the carrots and sticks of funders and governments pound it into reality, as well as the bureaucracies to monitor the system of mandates and compliance. As Esposito wrote in the same 2004 essay:
The total cost of research publications will grow enormously, driven by the author side of the equation. It is important to emphasize this point: OA, through the range of new services it will provide, will increase the overall cost of scholarly communications.
Both funders and governments are incentivized to publish as much as possible, again aligning with the interests of the OA business model. More publication events can make their funding appear to be more productive of results, and therefore more efficient and better spent. For funders, this can drive further funding; for governments, this can provide political expediency.
When funders, politics, and money meet, the offspring can be downright demented. A recent story out of Texas shows this clearly. Lance Armstrong, the infamous former cyclist, and Rick Perry, the bumbling Presidential candidate and current governor of Texas, teamed up to create the Cancer Prevention and Research Institute of Texas (CPRIT), a $3 billion initiative to make Texas a center for cancer research. Scientists and academics were recruited to run the center, but resigned shortly thereafter once it was clear that political and business concerns were running things, and scientific merits were not being considered. One prominent academic involved in reviewing projects was quoted as saying:
I recently learned that at least two scientific reviewers who had given non-fundable scores to a commercialization project were asked by CPRIT to ‘reconsider’ their scores so that they would be in harmony with those given by the commercial reviewers, who were much more favorable. I am not confident that scientific quality and rigor will triumph over grandiose promises and hucksterism.
Missing emails, an insincere investigation, and more political shenanigans followed. The residents of Texas will be paying $300 million per year for 10 years, funding something that will likely produce no science.
Big buckets of money attract special attention. When governments and funders begin to dance together, the results can be hard to undo.
There is a lot of money available on the supplier side of scientific publishing — more than enough to entice authors and publishers to change their ways, as the following list of available funds or revenues illustrates:
- US National Institutes of Health (NIH) — $31 billion
- Wellcome Trust — $22 billion
- STM Publishing Industry — $19 billion
- RCUK — $18 billion
- HHMI — $16 billion
These major buckets of money are also highly aligned. As this shows, when the NIH and Wellcome Trust align their plans, as they did around the launch of eLife, there is enough money to influence behavior at many levels. Even their participation in scientific publishing can have a chilling effect on debate and dissent, as academic leaders with millions of dollars of grants at risk walk on eggshells around the policies these entities are beginning to enforce. Among publishers, there is less money, and it is spread around to a greater degree — Elsevier has an 11% market share, and the next biggest company has 3.5% of the market; overall, there are 100 different STM publishers making $40M or more, and small companies dominate the market in aggregate.
However, funder and government money that is being diverted to publishing was once earmarked for research funding. What is the long-term toll this takes? Let’s look to the NIH for an example. While we don’t know how much PubMed Central costs because nobody will tell us, a figure that is batted around is $2.5 million per year. These are NIH funds being used to create a competitive and redundant publication archive, and it’s been up for a decade. Simple mathematics shows that each year, 10 potential research grants of $250,000 each are being consumed by PMC’s expenses. Over 10 years, that’s 100 grants of $250,000 each. Spread into the UK and the EU, where millions are being set aside for OA fees, and we’re seeing hundreds of grants consumed by the pursuit of OA causes.
Publishers are unlikely to balk when they are given money, especially when the community seems to approve of it — the OA model provides potentially greater and definitely more stable revenues, taking what has been an expensive and risky business and turning it into a less expensive and less risky business. Some publishers are seeing the virtues of exploiting the “personal impulse” of authors and the trends in government and funder policies and tactics. They see the money on the table, and don’t worry over its intent or provenance. Taking it may undermine academic integrity, but it’s easy and safe money which, to a business, can trump all other concerns, and allow for non-virtues (quantity-based publishing) to be rationalized as virtues (“openness”).
OA has the potential to change very fundamental assumptions and the social role of scientific publishing. It shifts our world from “hands off” to “hands on” for funding and government bodies. It takes the power of the purse-strings away from readers and scientists and their proxies in the market and gives it to funders and governments, who have different priorities and a demonstrated willingness to leverage science grants in order to achieve political, organizational, and bureaucratic goals.
We have a user-oriented payment system that has been eroding for 30+ years. Its erosion has been used as justification to move to a model with very different players and priorities — governments and funders working through authors to publish more and change standards and business practices to suit their goals.
Publishers may find it easy to adapt to this new funding paradigm, as costs are lower and less volatile, and revenues can appear to be more predictable and obtainable. Yet, one risk being courted is still a funding risk, and it’s significant — a sizeable cut in government funding or a bad year in investments across a set of philanthropies could have a multi-year effect on available APCs and therefore revenues. And because funding is concentrated in fewer hands under OA, these risks are less diversified and therefore more potentially damaging.
Aside from funding issues, the involvement of funders, governments, and their money in scientific and scholarly publishing is new, and poses real threats to independent and painstaking evaluation of scientific reports. Editors and publishers exist to serve as checks on authors and their sponsors, as well-funded and independent arbiters of science for scientists. OA asks publishers and editors to align with authors and their sponsors. This shreds the part of the social contract we own, leaving it in tatters. These mandates and funder strategies are that fundamental and important.
As these trends are propelled forward in a world that has become “OA at all costs,” we would be well-served to wrestle with the existential questions these intrusions are begging, namely:
- Why does a journal exist? As a conduit for funds and papers (business purpose), or as a filter and arbiter (intellectual purpose)?
- What is the role of a publisher? To run a commodity-based business for authors, or to create differentiated value for readers and users?
- What is the societal and intellectual role of authors and journal editors? Providers of services to funders and governments? Or providers of high-quality information to scientists and peers?
As long as these and other questions go unanswered, or current answers go undefended in any substantive and active manner, governments, funders, extremists dressed as innovators, and business opportunists will be able to elide the questions to suit their goals — goals that are different from the goals of scientists and users of the literature. And I believe that if we are not careful, we may look back at inflection points like these with some shame, as bureaucratic, governmental, and funder interests are allowed to use their money to severely dilute a centuries-old attempt at an objective structure of scholarly output. As one observer noted after attending a meeting in the UK about the implementation of the Finch Report’s recommendations:
I left the conference myself frustrated at the lack of discussion space and dismissal of serious concerns, alarmed at the brevity of the ‘consultation’ period and fairly convinced, in the words of Robert Dingwall, that there was a ‘reasonable chance of this whole thing being a car crash’.
The lack of discussion and fast-tracking of decisions are both symptomatic of OA practices. Because of the intolerance to questions and concerns, the result could be an especially ugly car crash — with governments putting high-octane fuel into a car they don’t understand or own, funders willing to close their eyes to the dangers on the road ahead, and advocates pressing the gas pedal down while ignoring the increasing shouts of concern from the backseat.