Open science, publishing, and public research support: Could Trump have it right?

In the three years since Donald Trump was inaugurated as president, I have rarely supported anything he proposed. And for nearly 50 years as an academic researcher, I have almost always sided with the professional research establishment. Yet on one issue I now find myself siding with the president and opposing the scientific establishment.

In December, E&E News reported that the president was considering an executive action requiring that all federally funded research become available to the public immediately upon publication. After all, taxpayers paid for much of this research, which could enhance their health or quality of life, and it should become open science.

The current system allows publishers to collect substantial fees in order to provide access to articles protected by copyright, and blocks access to individuals and institutions that don’t pony up. Younger investigators and those from less-wealthy institutions are rarely able to pay these fees.

Within days of the rumor that Trump was considering the executive order, 125 of the most influential scientific organizations and publishers expressed their displeasure about such open science in a letter to the president. Scientific articles, they argued, are the property of the journals that own the copyright, not of the authors or the funding agencies, and certainly not of the taxpayers who paid for the research.

To be fair, the issue is complicated. Publishers and scientific societies devote considerable effort to reviewing and selecting the best research for publication. In addition, they contribute by providing copy editing, assist with graphs and artwork, do the layout, and distribute the ensuing reports. The costs of producing the publications, however, are tiny in relation to the substantial investments made by the sponsors of the research.

In reality, academic publishing is built on a pyramid of free labor. Although some journal editors are paid small stipends, most journals offer their editors minimal or no financial support. They depend instead on the peer-review system, which is essentially free labor from thousands of experts.

Most important, the authors of articles who spend years doing the research and writing the articles are not paid a cent. Instead, they are rewarded by recognition from their peers and by promotions from universities.

Journal publishing used to be expensive: There were costs for printing, paper, binding, warehousing, shipping, marketing, and more. In the internet era, those expenses have diminished dramatically. Yet the costs for academic journals continue to climb, making journal publishing an attractive business proposition.

What about the scientific societies? They are the good guys, right? Not so fast.

The mission for most scientific societies is to promote quality research and disseminate results to the public. Why would scientific societies challenge efforts to make taxpayer-supported research more accessible? The answer, in a word, is money. Journal publishing has become the cash cow for many research organizations. Take the American Medical Association. It had total revenues of $361.3 million in 2018. Of that amount, only about 10% came from membership dues while nearly 85% came from publishing and related activities.

The American Psychological Association, a very successful 501(c)(3) nonprofit, describes itself as a membership-based organization dedicated to advancing psychology “as a means of promoting health, education, and human welfare.” Its membership dues yield only about 7% of its $128 million annual revenues, according to its IRS Form 990. Two-thirds of the APA’s income comes from journal subscriptions, publication sales, and publication licensing fees for nearly 90 of its publications. The organization proudly boasts that nearly 70,000 editorial board members and reviewers provide service to the publications without compensation.

The issue of the rising cost of journal access came to a head in 2019 when the University of California cut ties with Elsevier, the largest academic publisher. The university system simply could not afford to pay that one publisher more than $10 million a year in subscription fees. While the conflict was brewing, Elsevier reported a profit margin of 37% and more than $3.3 billion in revenue in 2018.

At the end of the day, academic institutions bear the brunt of the high cost of journal access, while providing for free the lion’s share of the effort that goes into creating the articles these journals so desperately depend on. The costs of academic publishing have increased because, we believe, universities can afford it.

Academic institutions are in a bind because their researchers and students need journal access. And academic prestige and accreditation depends on the quality and completeness of library collections. The cost results in higher student tuition and become the burden of private school endowments, state legislatures, and families. Not only are taxpayers paying for the research they will have difficulty getting access to, but they’re also subsidizing companies that exploit students and faculty members.

There are nonmonetary problems related to commercial scientific publishing. For one, academic journals are snobby. Much of the research funded by taxpayers is never made public because the most prestigious journals publish only a highly selected subsample of completed research. Reports of research with less-stimulating results are delegated to the file drawer or to obscure journals. Studies with strong commercial backers, such as those supporting the value of an expensive pharmaceutical product, are not only published but are selectively spun in advertising. Consumers and taxpayers get exposed to findings that may not be representative of the range of completed research studies.

There are solutions that need to be debated. One of them is to become less reliant on the academic publishing system. For centuries, scholars and researchers freely communicated their results through informal channels. Through the internet, we now have the opportunity to make research results become part of public open science while cutting out expensive and inefficient journals. The National Library of Medicine’s ClinicalTrials.gov is an excellent example of an open science service that reports results directly to the public at no charge. PubMed Central, also part of the National Library of Medicine, makes available the full text of publicly funded research but only after one-year delay following the initial publication. Some groups are taking the bull by the horns. bioRxiv, for example, is a preprint server that allows scholars from diverse biology fields to post their work to a public server and receive feedback from their peers.

There’s no question that peer review is essential for evaluating and improving scientific manuscripts. But we don’t need profit-motivated publishers to take advantage of unpaid expert reviewers. The money universities would save on journal subscriptions with open science would go a long way toward a system that reduces costs to universities and students, compensates reviewers, and rates the quality of scientific contributions.

As taxpayers and consumers, we deserve access to the research we paid for. If it is in the public interest, it should be in the public domain.

Robert M. Kaplan is a faculty member at Stanford University’s Clinical Excellence Research Center, a former associate director of the National Institutes of Health, former chief science officer for the U.S. Agency for Healthcare Research and Quality, and author of “More than Medicine: The Broken Promise of American Health” (Harvard University Press, 2019).