On August 25, 2022, Dr. Alondra Nelson, then Deputy Assistant to the President and Deputy Director for Science and Society at the White House Office of Science and Technology Policy (OSTP), signed a three-page directive that has become the most consequential US federal research policy change since the 2013 Holdren Memorandum. The Nelson Memorandum applies to every federal department and agency with annual research and development expenditures above $100 million, and it sets out two hard dates: December 31, 2024 for updated public access policies, and December 31, 2025 for full implementation that makes peer-reviewed publications and supporting data freely available to the public without an embargo, unless an agency justifies a short delay in its written plan.
The second deadline has now passed. On campuses, the question that follows is less about the memo's intent than about its translation layer: which agency rule applies to which grant and whether the institutional repository can handle the new deposit volume. Who pays for the data curation work that no federal line item funds is the sharper question. Policy announcements do not deposit manuscripts; library staff and graduate students do, and they are already busy.
What the Memo Changes on Paper
The 2022 directive reverses the 2013 Holdren Memorandum, which allowed federal agencies to permit a 12-month embargo before papers had to appear in a public repository. The new default is immediate public access. The memo covers two categories of research output: peer-reviewed scholarly publications and the underlying scientific data needed to validate those findings. Agencies must also require persistent identifiers such as DOIs for publications and datasets, include machine-readable metadata, and ensure that funding recipients deposit outputs in agency-designated repositories at no cost to readers. An embargo longer than zero must be explicitly justified in the agency's public access plan.
For a researcher, the shift is concrete. A biomedical scientist holding an NIH grant has long been expected to deposit an accepted manuscript in PubMed Central, but the older policy allowed that manuscript to sit behind a 12-month delay. Under the agency's updated plan, that delay is supposed to shrink to zero or close to it. A geoscientist funded by the National Science Foundation (NSF) similarly must plan for early deposit in the NSF Public Access Repository. The memo does not require journals to flip to open access; it requires funded authors to make their work available through repositories regardless of the journal's own business model.
- Peer-reviewed publications must be deposited in an agency-designated repository at no cost to the reader.
- Supporting data must be shared in a format that allows validation and reuse where legally, ethically, and technically possible.
- Persistent identifiers and machine-readable metadata are required for both publications and datasets.
- Any embargo longer than zero must be explicitly justified in the agency's public access plan, and that justification must be public.
- The policy applies to new awards and new funding increments after the implementation date, with agencies expected to apply it consistently across all eligible research outputs.
Deadline Passed: What Agencies Have Done So Far
Large agencies met the December 31, 2024 policy deadline with revised public access pages and implementation plans. The National Institutes of Health published updated guidance that removes the 12-month embargo for publications arising from NIH funds, and the National Science Foundation's Public Access Repository now expects deposits at the time of publication or after a short delay. The Department of Energy and NASA have similar repositories and plans, though the wording of "acceptable repository" and "supporting data" varies by agency. A university research administrator trying to build one campus workflow must reconcile those differences across NSF, NIH, DOE, and a dozen smaller funders, each with its own forms and deposit systems.
The OSTP memorandum remains the authoritative source for the two deadlines and the justification requirement. Agency pages, such as the NIH public access site and the NSF public access plan, are where the operational details sit. That scattered implementation creates the first compliance problem: there is no single federal dashboard that tells a principal investigator which version of a paper must go where, by when. Research librarians at major universities have built internal matrices that map funder to repository to metadata standard, a task no one funded and every grant now depends on.
Campus Compliance Runs on Unfunded Work
The policy reaches every campus that receives federal research dollars, but it reaches first the campuses that already have institutional repository staff, data librarians, grant administrators, and the faculty who must submit the deposits. A large public research university with an established digital scholarship centre can add Nelson Memo compliance to existing tasks. A regional university with one data librarian and no dedicated repository support faces the same federal requirement with a fraction of the capacity. The gap between those two sentences is where the Nelson Memo's equity problem sits.
Early-career researchers feel the compliance burden personally. A postdoc applying for a first federal grant must now write a data management and sharing plan that meets the funder's standard, and that plan will be evaluated as part of the award. The memo did not create new money for data curation or repository fees, and it added no staff lines for compliance. Institutions that can afford to hire a research data specialist do so; institutions that cannot ask existing faculty to do more unpaid administrative work. The implementation gap is not a failure of intent, but it is a predictable feature of an unfunded mandate.
Publishers, Contracts, and the Cost Question
Publishers have responded unevenly. Some large commercial publishers have adjusted author workflows to support zero-embargo repository deposit, but read-and-publish agreements remain the main mechanism that universities use to manage article processing charges. The UK's standoff with Elsevier over contract terms, which AcademicJobs.com covered earlier in the UK and Elsevier open access contract disputes, shows how quickly a public access rule becomes a budget negotiation. When the US mandate requires immediate repository access, publishers lose the exclusive delay window that made subscription bundles defensible; university library consortia enter renewal talks with a stronger hand, but only if they are willing to walk away from big deal packages.
The memo does not set a funder-side cap on article processing charges, and it does not require journals to lower prices. For researchers choosing where to submit, the practical effect is that a journal's compliance with funder deposit rules has become a selection criterion. A journal that refuses to allow accepted manuscript deposit in PubMed Central or an institutional repository can force an author into a compliance exception request, creating friction that journals that cooperate with the mandate do not impose. The SPARC analysis of the OSTP memo usefully separates the publication obligation from the data obligation, which is the distinction most campus committees still blur.
The Global Mirror: Canada and Europe Move in the Same Direction
The Nelson Memorandum was not an isolated move. Canada's federal research funding agencies began enforcing their own public access requirements with a 2026 compliance target, as reported in AcademicJobs.com's earlier coverage of the Canada Tri-Agency open access policy. The European Research Council's grantee rules similarly demand immediate open access to publications funded by its schemes. For researchers weighing a postdoc in Boston against one in Toronto or Utrecht, public access compliance is converging across jurisdictions, which means the skills that make a researcher compliant with the Nelson Memo travel.
The convergence changes the job market in a specific way. Search committees and hiring panels increasingly ask candidates to describe their data management experience, not just their publication list. A postdoc who has deposited a dataset with a digital object identifier and written a data management plan, then worked through a funder's repository requirements, has a demonstrable administrative skill that a CV line about publications does not capture.
Data Sharing and the Year-Two Question
The data provision is the part of the Nelson Memorandum most likely to outlast the publication provision in campus meetings. Publications deposit is a known process with existing repositories; data sharing requires decisions about what counts as "supporting data," how long it must be preserved, who prepares the metadata, and what happens when a dataset contains personally identifiable information or commercially sensitive material. Agencies have issued data management plan guidance, but the operational work falls on individual principal investigators who are rarely trained in data curation.
The year-two question for implementers is not whether the memo will stay in force, but whether the universities that cannot fund data stewardship will quietly comply at minimum viable level while better-resourced institutions build robust research data services. That divergence is already visible in institutional hiring patterns: research data librarian postings cluster at R1 universities, while smaller institutions advertise for grant coordinators who can handle compliance as one of many duties. The policy on paper reaches every funded researcher. In practice, it reaches the researchers whose campuses have the staff to make deposit possible. That is the equity gap the next phase of implementation will have to answer.
Photo by Sven Piper on Unsplash
