January 1, 1970

Open Access Publishing in Academia: What Researchers Actually Need to Know

Diagram illustrating the four main open access publishing models: Gold, Green, Diamond, and Bronze

Open access publishing has had a remarkable decade. Gold OA jumped from covering 14% of global scholarly output in 2014 to 40% in 2024, quadrupling in volume while subscription-only content barely crept forward at 2% annual growth. That sounds like a victory. But the more you look at how the money flows, the more the victory looks like a redistribution rather than a reform.

What Open Access Actually Means (and Why the Flavors Matter)

"Open access" covers several distinct publishing models, and conflating them creates real confusion about what any given journal or policy is actually offering.

Gold open access means the final published version is freely available immediately on the publisher's platform. The cost is borne by the author through an Article Processing Charge (APC). Nature's flagship journals charge up to $11,690 per article. PLOS ONE, one of the largest gold OA journals in the world, sits at around $1,950.

Green open access means an author self-archives a version of their paper—a preprint or accepted manuscript—in a repository like arXiv, PubMed Central, or an institutional repository. It's free to deposit and free to read, though published versions often remain paywalled for 6-24 months under embargo.

Diamond open access is the model that deserves far more attention. No subscription fees for readers, no APCs for authors. Funding flows from institutions, academic societies, or governments instead. The mathematician Timothy Gowers launched Discrete Analysis this way in 2015: it publishes papers already hosted on arXiv and costs almost nothing to run.

Model	Who Pays	Free to Read?	Free to Publish?
Gold OA	Author (APC)	Yes	No
Green OA	Nobody directly	Often delayed	Yes
Hybrid	Author or library	Mixed	Optional
Diamond OA	Institution/Society	Yes	Yes
Bronze	Publisher's discretion	Sometimes	No

One more worth knowing: hybrid journals are traditional subscription journals that let individual authors pay for open access on a per-article basis. Critics call this "double dipping" because publishers collect both subscription income and APCs simultaneously. Transformative agreements are specifically designed to phase this out.

The APC Problem: Pay-to-Publish Has Real Consequences

Let me be direct: the APC model is broken in its current form.

The global average APC hit $1,626 in recent surveys—but that masks enormous variation. Elite journals at major commercial publishers charge $3,000 to $12,000 per article. Meanwhile, independent estimates put the actual cost of efficiently publishing a quality article at €200-€1,000. Major publishers consistently report profit margins exceeding 30%, on a product created almost entirely by unpaid academic labor (research, writing, peer review). Do the math.

The prestige premium is real and largely unjustified. Established journals charge more not because they provide better editorial value but because their brand appears on tenure applications and grant reports. It is the academic equivalent of charging airport prices for a sandwich.

This creates obvious equity problems. STEM researchers with active grant funding can absorb APCs as budget line items. A historian at a university in Bangladesh cannot. A postdoc without institutional affiliation definitely cannot. Waivers exist at most major publishers, but research consistently shows they're underutilized, inconsistently applied, and sometimes create uncomfortable power dynamics between authors and editors.

The US Office of Science and Technology Policy's Nelson Memo (2022) mandated immediate open access for all federally funded research by the end of 2025. A meaningful policy step. But it was written assuming APC coverage, and it left unfunded researchers to navigate a system that still expects them to pay.

Mandates Are Moving Fast—But Unevenly

Government and funder mandates have done more to push open access forward since 2018 than decades of advocacy combined. Plan S, launched by a coalition of European research funders called cOAlition S (including Wellcome Trust and the European Research Council), requires that all funded research be published open access at the moment of publication, with no embargo.

The mechanism most institutions use: transformative agreements. Rather than paying subscription fees, a library negotiates a bundled deal where its researchers can publish open access as part of the contract. The ESAC Transformative Agreement Registry surpassed 1,000 recorded deals as of early 2024.

The Royal Society's trajectory is worth pausing on. They committed to 100% open access in 2021, built up to 675 Read & Publish institutional agreements by 2025, converted eight subscription journals to Subscribe to Open (a model with no APCs for anyone), and hit their 100% target in April 2026. It took them 20 years from launching their first OA journal to complete the transition.

But progress stops being universal pretty quickly:

Transformative agreements require significant negotiating capacity—legal staff, multi-year budget commitments, dedicated library infrastructure
Smaller universities and institutions in low-income countries are largely absent from the ESAC registry
Plan S has updated its guidance multiple times after pushback from humanities and social science researchers, where APC funding is scarce or nonexistent

"Transformative agreements shift costs from libraries to authors and institutions—but they don't reduce costs. They redistribute who bears them."

The mandate wave is valuable. But it's accelerating a system that rewards institutions already at the top of the resource distribution.

Predatory Journals: The Dark Side of the Boom

Here is where the story gets genuinely alarming. The APC model accidentally created a financial incentive that predatory publishers have exploited with surprising sophistication.

Predatory journals charge APCs while providing little or no real peer review. They accept almost anything. They look legitimate: plausible names, fabricated impact factors, editorial boards populated with real scholars who were listed without their consent. Some operate from addresses that turn out to be empty lots.

The structural driver matters. When the publish-or-perish pressure is highest and funding is thinnest—early-career researchers, academics at under-resourced institutions, scholars in countries with weak research infrastructure—a guaranteed acceptance is a genuine temptation. Predatory publishers know exactly who to target and how.

AI is compounding this. Language models can generate papers that pass superficial review. Paper mills have scaled operations using AI assistance. A 2025 analysis in PubMed Central flagged that AI-generated content in predatory venues is now "diluting the informational value" of research databases—which are increasingly used to train future AI systems. That second-order problem is still largely unappreciated outside specialist circles.

How to identify a legitimate open access journal before submitting:

Check the DOAJ (Directory of Open Access Journals)—it vets journals against quality criteria before listing them
Search Cabell's Predatory Reports (formerly Beall's List)—maintained at cabells.com for known bad actors
Use Think. Check. Submit. at thinkchecksubmit.org—a structured checklist from major library associations
Verify indexing in Scopus, PubMed, or Web of Science before submitting
Treat unsolicited email solicitations as red flags—legitimate journals do not cold-email researchers about their "impressive profile" shortly after a preprint posts

Diamond Open Access: The Path Taken Too Slowly

The mainstream OA conversation keeps defaulting to APCs as if they're the only viable funding model at scale. They're not. Diamond OA proves it, and the gap between proof-of-concept and mainstream adoption is a funding decision, not a technical constraint.

The Open Library of Humanities, supported by over 300 universities worldwide, publishes 28 peer-reviewed journals in humanities and social sciences with no fees on either end. France's National Plan for Open Science includes direct funding for diamond OA infrastructure. Germany's Bibsam consortium has redirected subscription savings toward open publishing infrastructure. These are modest programs in absolute terms, but they represent real institutional commitment.

The objection to diamond is almost always scale. Wiley alone published 475,746 articles in 2023. Can diamond models absorb that volume? Not without far more infrastructure investment than currently exists. But that's asking the wrong question. Diamond does not need to replace commercial publishing overnight—it needs to absorb the portion that commercial publishers do not need to touch: society journals, university press output, specialist disciplinary journals with modest submission volumes.

For those segments, diamond is not only viable. It's already working.

What You Should Actually Do If You're a Researcher Today

If you have grant funding: Check your funder's OA policy before picking a journal, not at submission time. NIH, Wellcome Trust, and most European funders require OA and often have APC budget lines—but only if you planned for them at the proposal stage.

If you don't have APC funds: Green OA via preprint is your best move. arXiv, bioRxiv, medRxiv, and SSRN are free and indexed by Google Scholar. The SHERPA/RoMEO database tells you whether your target journal allows self-archiving of the accepted manuscript after an embargo—many do.

If your institution has a transformative agreement: Your library's website should list covered publishers. Publishing in a covered journal typically means the APC is handled at the library level, with no invoice reaching you.

One non-obvious benefit of preprint posting that often goes unmentioned: your work gets a timestamp and starts accumulating citations immediately, sometimes months before the final article appears. In physics and mathematics, this has been standard practice for 30-plus years. Medicine and the social sciences are catching up fast, and the reputational norms around preprints have shifted considerably since 2020.

Bottom Line

Open access is genuinely expanding—the data is clear. But the APC-dominated model has swapped one access barrier for another: instead of readers paying to read, underfunded researchers pay to publish. Neither outcome was the original point.

My position: the field should be investing more aggressively in diamond OA infrastructure rather than negotiating marginally better rates on a system that routes all value through commercial publishers. The Royal Society completing its 100% OA transition by 2026 is genuinely worth celebrating. Paying a major commercial publisher $9,500 to make your own publicly funded research available is not a sustainable solution.

Researchers: Use preprints and green OA wherever your funder and target journal allow. Check OA requirements at proposal time, not submission time.
Institutions: Prioritize transformative agreements for your highest-volume publication venues, and push for diamond OA funding at the library budget level.
Funders and policymakers: Mandates moved things forward. The next step is building alternative infrastructure so not every open access pathway runs through a for-profit publisher's accounts.

The writing was on the wall when academic publishers started reporting 30% profit margins on work produced almost entirely by unpaid researchers. How long academia keeps financing that arrangement is a choice, not an inevitability.

Frequently Asked Questions

What is open access publishing and how does it differ from traditional publishing?

Traditional publishing puts articles behind a paywall; readers or their institutions pay subscription fees. Open access makes articles freely available to anyone with internet access. Most OA publishing shifts costs to authors through APCs, but diamond OA uses institutional or government funding instead, with no fees on either side.

Are open access journals lower quality than traditional subscription journals?

This is one of the most persistent myths in the field. Journal quality depends on peer review rigor and editorial standards, not the access model. PLOS Biology, eLife, and many rigorous society journals are open access. The confusion arises because predatory journals are technically "open access"—but that association is coincidental, not causal. Predatory practices are a business model problem, not an access model problem.

Do I always have to pay an APC to publish open access?

No. Green OA (depositing a preprint or accepted manuscript in a repository) is free. Many publishers offer APC waivers for researchers from low-income countries or without institutional funding. If your institution holds a transformative agreement with the publisher, APCs are often covered automatically. Diamond OA journals charge nothing to authors at all.

What is Plan S and does it apply to my research?

Plan S is a mandate from cOAlition S—a group of funders including Wellcome Trust and the European Research Council—requiring that all funded research be published open access at the moment of publication with no embargo. If your funding comes from a cOAlition S member, it applies to you. The current list of participating funders is at the cOAlition S website, and specific requirements vary by funder.

How can I tell whether a journal is predatory before I submit?

Start with the Directory of Open Access Journals (DOAJ), which lists vetted legitimate OA journals. Cross-reference with Cabell's Predatory Reports for known bad actors. Confirm the journal is indexed in Scopus, PubMed, or Web of Science. If a journal contacted you by unsolicited email, treat that as a significant red flag regardless of how professional the message looks.

What is diamond open access and which journals actually use this model?

Diamond OA means free to read and free to publish—funded by institutions, academic societies, or governments rather than by author fees. The Open Library of Humanities runs 28 peer-reviewed journals this way, backed by 300-plus universities. Discrete Analysis in mathematics and many journals distributed through SciELO and Redalyc operate the same way. France and Germany have national programs specifically funding diamond OA infrastructure expansion.