Open Access and APCs: What Publishing Actually Costs

A registrar's paper is accepted. The email congratulating them also asks for an article processing charge of several thousand pounds, payable by the corresponding author. Nobody mentioned this at submission. There is no grant. The department has no budget line for it.

This is an entirely routine situation and it is almost entirely preventable — usually by an email sent before submission rather than after acceptance. This post covers how the models work, what they cost, and where the money actually comes from. It sits within our research integrity series.

The models

Diamond journals exist and are under-used. Society-funded and institutionally supported journals charge neither the reader nor the author. They are less visible because they have no commercial incentive to market to you.

What it actually costs

As of 2026, APCs at established publishers fall roughly between $500 and$3,500, with a commonly cited global average around $1,600–$2,000.

Medicine sits above that average, and varies considerably by specialty. Recent analysis found haematology and oncology journals with the highest median APC (about $4,690)** and **primary care the lowest (about$2,690).

The finding most worth knowing:

Hybrid journals charge more than fully open-access ones — a median of roughly $4,248 against$2,909.

That is counterintuitive and it matters. Paying a hybrid journal to open a single article in a publication your library is already subscribing to is the most expensive route available, and it is the one authors default to because it is offered at the point of acceptance in a journal they already wanted.

Accurate as of July 2026. APCs change annually and vary by journal — check the specific figure for your target journal before submitting.

Check your institutional agreement first

The single most valuable action in this post, and the one most often skipped.

Many universities and health systems hold transformative agreements — also called read-and-publish deals — where subscription spending is redirected to cover APCs for affiliated authors. Under such an agreement, publishing open access in a covered journal may cost you nothing.

These are extensive. In Australia and New Zealand they are negotiated collectively by CAUL, with individual institutions covered by many such agreements simultaneously; UTS, for example, facilitates 18 read-and-publish agreements in 2026. Comparable arrangements exist across the UK, Europe and North America.

Surgeons pay APCs their institution had already agreed to cover with reasonable frequency, because nobody told them the agreement existed. Before you submit:

1. Search your library's website for "transformative agreement" or "read and publish"
2. Check whether your target journal is covered
3. Email the library if it is unclear — they answer this daily
4. Note that eligibility usually depends on the corresponding author's affiliation

That last point catches people out on multi-centre papers: which author corresponds can determine whether the APC is waived.

Plan S has changed — and that matters

If your understanding of Plan S is the strict version, it needs updating.

Plan S, from cOAlition S, was launched as a firm mandate: research funded by member organisations had to be published open access in compliant venues, with hybrid journals explicitly excluded.

The Coalition S Strategy 2026–2030 takes a markedly softer approach. It steps back from the goal of supplanting paywalled journals, acknowledges that no single model meets all needs, and endorses multiple routes to open access rather than prescribing one. Notably, it contains no new mandate restricting the venues in which grantees may publish — a contrast with some individual funders, which have moved the other way.

The practical consequence: do not assume a blanket funder mandate, and do not assume its absence. Your specific funder's current policy is the thing to check, and it may differ from both the old Plan S and the new strategy.

Green open access is the forgotten option

If there is no money, green OA usually solves the problem and costs nothing.

You deposit the accepted manuscript — the peer-reviewed version before the publisher's typesetting — in an institutional or subject repository, after any embargo the publisher requires.

• Costs nothing
• Satisfies many funder requirements
• Permitted by most publishers, with conditions on version and embargo
• Requires you to actually do it, which is why it is under-used

Check the publisher's self-archiving policy before signing the copyright agreement, and keep a copy of your accepted manuscript. People routinely discover they cannot find the accepted version when they finally get round to depositing it.

Licences

If you are paying for open access, know what you are buying.

• CC BY — others may reuse, adapt and redistribute, including commercially, with attribution. The most permissive, and the one most funders require.
• CC BY-NC — non-commercial reuse only. More restrictive, and sometimes not funder-compliant.
• CC BY-NC-ND — non-commercial, no derivatives. The most restrictive.

Check what your funder requires before choosing, because a licence that does not comply may mean the payment did not achieve what it was for.

Where the money comes from

In rough order of preference:

1. Institutional transformative agreement — free to you; check first
2. Grant funding — budget APCs at application stage; retrofitting them is difficult
3. Institutional open-access fund — many universities hold central funds
4. Departmental research budget — ask early, not after acceptance
5. Waivers and discounts — most publishers waive or reduce APCs for authors in low- and middle-income countries, and many will consider hardship requests. These are under-claimed because people do not ask
6. Choose a journal without an APC — a subscription journal with green OA deposition costs nothing
7. Personal payment — last resort, and it should be genuinely rare

The predatory-journal connection

The APC model is exactly what predatory publishers exploit: charge the fee, skip the review. Two defences.

Know the APC before submitting. Legitimate journals publish theirs openly. A charge that only materialises after acceptance is a red flag in itself.

Verify the journal, not the price. A low APC is not evidence of legitimacy, and a high one is not evidence of quality. See How to Spot a Predatory Journal.

Practical sequence

1. Decide the target journal, on fit and readership rather than access model.
2. Check the institutional agreement. Free is better than cheap.
3. Check the APC on the journal's own site.
4. Check the funder's current policy, not what it was.
5. Decide who corresponds, since agreement eligibility usually follows that author.
6. Agree who pays, in writing, before submitting.
7. If there is no money, choose a subscription journal and deposit the accepted manuscript.
8. Keep the accepted manuscript, whatever route you take.

The summary

APCs in medicine run higher than the general average and vary widely by specialty. Hybrid journals cost more than fully open-access ones, which makes the default route — paying to open one article in a subscription journal — the most expensive option available.

Check your institution's transformative agreements before you submit; the charge may already be covered. Check your funder's current requirement rather than the version you remember, because Plan S has genuinely softened. And if there is no money, green open access costs nothing and satisfies most requirements.

Sort it out before submission. After acceptance, your options are worse and your negotiating position is gone.