Skip to main content

Free guide

What Publishing Actually Costs: the list price, the real price, and how to pay nothing

9 min read · verified August 21, 2026

Every price in this guide was read off the publisher or funder page that sets it, on the date above, and every one is linked at the end. No roundups, no remembered numbers.

The short version: the figure you were quoted is a list price, it is close to the worst case, and there are four separate ways to pay less than it or nothing at all. The expensive mistake is not choosing the wrong journal. It is missing a deadline that two major publishers enforce in almost the same words, and that neither of them puts in front of you.

Yours to keep, free

The whole guide is on this page. The PDF is the same thing, formatted to print or keep on your phone.

Download the PDF (PDF file)

01The deadline is submission, and after that there is no appeal

Start here, because this one is unrecoverable. Waiver and fee-assistance requests must be made when you submit. Not at revision, not at acceptance, not when the invoice lands.

  • PLOS: assistance "must be formally applied for at submission. Requests made during the review process or after acceptance will not be considered."
  • Springer Nature, on the Nature Communications fee page, says the same thing in its own words: requests made during the review process or after acceptance are unable to be considered.
  • Two independent publishers, one rule. Treat it as the industry default rather than a quirk of either, and assume your journal enforces it unless its own page says otherwise.
  • The reason it catches people is that the money only feels real at acceptance, which is exactly one step too late. When you are least sure the paper will be accepted is when the request has to go in.
  • It costs you nothing to ask early. A waiver request does not reach the editors handling your manuscript, and it has no bearing on the decision.

Four questions to answer before you hit submit

  • Is this journal hybrid or fully open access? Hybrid means the fee is optional and you can decline it. Fully open access means there is no free route and the fee is the price of publishing there.
  • Does my institution have a read-and-publish agreement covering it? If so the fee may already be paid, and you have to claim it at submission for that to work.
  • Do I qualify for a waiver on country, funding status, or field? Check before you assume you do not. The criteria are broader than most people expect.
  • If I am paying, which tier am I buying? Article type, licence, and immediate versus delayed access can move the price by thousands at the same journal.

Send this at submission, through whatever channel the journal specifies, which is often a dedicated waiver address rather than the editorial inbox. Keep it short and factual. You are not making a case for your paper, you are answering a funding question.

02At a hybrid journal the fee is optional, and Nature says so on its own page

This is the most widely misunderstood fact about publishing costs. An acceptance at a journal with a five-figure fee does not mean a five-figure bill. At most legacy journals, open access is one of two routes and the other one is free to you.

  • Nature charges £9,390 / $12,850 / €10,850 for gold open access. That is the real list price and it is the highest of any general journal.
  • Nature also offers, in its own words, the traditional subscription route: the article is assessed, peer reviewed, and published, and readers reach it through a subscription. The author pays nothing. Same review process, same journal, no fee.
  • So the fee buys free access for your readers. It does not buy publication, and it does not buy a better look from the editors. Papers at Nature are accepted or rejected before the question of which route you want comes up.
  • The fork matters: Nature Communications is fully open access, at £5,490 / $7,350 / €6,150, with no subscription fallback. There the fee is unavoidable. Knowing which of the two you are submitting to is the difference between an optional cost and a committed one.
  • One catch worth knowing at Nature specifically: reviews, comments, and other non-primary articles are not eligible for gold open access at all, and can only go the subscription route.
  • When you should pay anyway: if your funder mandates immediate open access, the subscription route will not satisfy it. Check the mandate before you choose, because changing the licence after publication is often impossible.

03Compliance is free, and any fee for it is a fee you should question

If you are funded by NIH, the thing you are legally required to do costs nothing. This gets conflated with open access fees constantly, and the conflation is expensive.

  • NIH policy is explicit that compliance itself is free. NIH does not require you to pay an open access fee to satisfy the public access policy. Depositing the accepted manuscript is what the policy asks for.
  • NIH allows publication costs including article processing charges to be paid from grant funds when they meet the usual cost principles, so paying is permitted. It is just not required.
  • Two things are specifically not allowable: fees charged for submitting a manuscript to PubMed Central, and charges that discriminate based on NIH funding. If a journal quotes you a PubMed Central deposit fee, that is the one line on the invoice to challenge.
  • The 2024 policy applies to manuscripts accepted on or after July 1, 2025.
  • Green open access is the general version of this and it predates the mandates. PNAS has offered it since 2004: you may deposit your accepted manuscript in your funder archive or institutional repository immediately on publication, at no cost. Most journals permit some version, with the embargo and the permitted version differing.
  • PNAS goes further and deposits the version of record in PubMed Central for you automatically, regardless of which licence you chose. All PNAS articles become free to read within six months whether you paid or not.

04Free is a real category, and the criteria are wider than people assume

Most authors never check because they assume waivers are for someone else. The published criteria include entire countries, entire fields, and simply not having funding.

  • PLOS publishes free for authors at institutions in Research4Life Group A countries who have no external funding for the study. Not reduced, free. Group B countries pay $940 at most journals, or nothing at selected titles.
  • Elsevier reports in its own annual report that it is a founding partner of Research4Life, with over 11,500 institutions across 125 countries participating. This is a large programme, not a token one.
  • PNAS grants waivers for authors without funding to cover the charge, and names mathematics and astronomy as examples alongside lower-income countries. Field is a criterion there, not only geography. If your discipline has no publication budget, that is a stated reason rather than an excuse.
  • Springer Nature waives or discounts automatically for authors in the lowest income countries, and considers requests from anyone else case by case on financial need.
  • The pattern across all four: financial need is a legitimate basis everywhere, and nowhere does asking count against your manuscript. The only thing that disqualifies you is asking late.

05Check whether it is already paid before you pay it

Thousands of institutions have agreements that cover the fee for their researchers. The money is sitting there and it goes unclaimed because nobody tells the author it exists.

  • These go by several names: read-and-publish, publish-and-read, transformative agreements. The effect is the same, which is that your institution has prepaid and your article publishes open access at no cost to you.
  • The ESAC registry collects these agreements worldwide and is searchable by publisher and institution. It is the fastest way to find out whether yours is covered, and it was current as of August 2026.
  • Your librarian is faster still, and this is squarely their job. A single email before submission is the highest-value thing in this guide after the waiver deadline.
  • Even partial coverage moves real money. At PNAS the immediate open access charge is $5,575 for corresponding authors at institutions holding a current-year site licence, against $6,075 otherwise, which is $500 for a status you either already have or do not.
  • You usually have to identify your affiliation correctly at submission for any of this to trigger. Getting your institution right in the submission system is not admin, it is the mechanism.

06The same journal has several prices, and the choice is yours

People compare journals on one headline number. Within a single journal the spread is often larger than the gap between journals, and it is driven by choices you control.

  • PNAS charges $3,025 for delayed open access and $6,075 for immediate open access on a regular research article. Same journal, same review, twice the price for the timing of free access.
  • Article type changes it again: a PNAS Brief Report is $2,600 and is immediately open access on publication.
  • Licence choice is bundled into these tiers. At PNAS the delayed option carries CC BY-NC-ND, while immediate open access allows CC BY or CC BY-NC-ND. If your funder requires CC BY specifically, that requirement is choosing your tier for you.
  • Across PLOS the range runs from $1,852 for some PLOS One article types up to $3,165 at the selective titles, so which journal in a family you target is itself a price decision.
  • Decide the tier before submission alongside the waiver question. Changes to licence or open access status after publication may not be possible, and the cheapest moment to think about any of this is before you are invested in an acceptance.

07What the price is not: an estimate of what it cost to publish you

Worth understanding, because it explains why the same service costs $2,600 at one journal and $12,850 at another, and why haggling does not work but asking does.

  • RELX, which owns Elsevier, reported 2024 revenue of £3,051m and adjusted operating profit of £1,172m in its scientific, technical and medical segment. That is a 38.4 per cent operating margin, and you can do the division yourself from their annual report.
  • PNAS publishes on its own charges page that it operates on a break-even basis, must cover its operating expenses, and is prohibited from retaining a surplus or maintaining cash reserves.
  • Same industry, same product, two entirely different financial models. Neither is hidden. Both are stated by the organisation itself.
  • The labour is the part that is not on any invoice. Aczel and colleagues estimated reviewers worked over 100 million hours globally in 2020, more than 15 thousand years of time, and valued the contribution of US-based reviewers alone at over 1.5 billion dollars. China-based reviewers came to over 600 million, UK-based to close to 400 million. The authors call these under-estimates.
  • The practical conclusion is not outrage, it is leverage: the price is set by position and strategy, not by a cost you have to respect. That is precisely why waivers and agreements exist and why asking works. You are negotiating with a pricing policy, not with a bill.

Where to check

The databases this guide sends you to, all free and open without an account.

  • ESAC agreement registry · Search whether your institution has a read-and-publish agreement with your target publisher. Do this before submission, because coverage usually has to be claimed at submission.
  • Open Policy Finder (formerly Sherpa Romeo) · Look up what a specific journal permits you to self-archive, which version, and after how long. This is how you find the free route at a journal you have already chosen.
  • Research4Life · Check whether your country is Group A or Group B. It decides whether several major publishers waive your fee automatically rather than on request.
  • NIH Public Access Policy · The requirement itself, in plain terms, including confirmation that compliance does not require paying anyone.

Where these numbers come from

  1. 01Nature gold open access is £9,390 / $12,850 / €10,850, and the subscription route costs the author nothing Nature, Publishing options
  2. 02Nature Communications is £5,490 / $7,350 / €6,150 with no subscription fallback, and waiver requests after acceptance cannot be considered Nature Communications, Open Access Fees and Funding
  3. 03PLOS fees by journal, Research4Life free publication, and assistance must be applied for at submission PLOS, Publication Fees
  4. 04PNAS tiers of $3,025 delayed, $5,575 site-licence and $6,075 regular immediate, $2,600 Brief Report, green open access since 2004, automatic PubMed Central deposit, and break-even operation PNAS, Publication Charges
  5. 05Compliance itself is free, APCs are allowable but not required, and PubMed Central submission fees are not allowable NIH, Supplemental Guidance on Publication Costs (NOT-OD-25-048)
  6. 06The 2024 NIH Public Access Policy applies to manuscripts accepted on or after July 1, 2025 NIH, Public Access Policy effective date (NOT-OD-25-101)
  7. 07STM revenue of £3,051m and adjusted operating profit of £1,172m in 2024, a 38.4 per cent margin, and Research4Life participation of over 11,500 institutions RELX, Annual Report 2024, Market segments
  8. 08Over 100 million reviewer hours globally in 2020, valued at over 1.5 billion dollars for US-based reviewers alone Aczel, Szaszi and Holcombe, Research Integrity and Peer Review, 2021

This guide will go out of date

I will tell you when it does

One email a week on what changed in academic publishing and whether it actually matters for your paper. Every number links its source.

One email a week, Mondays. Leave whenever you like.

The next step

The most expensive thing in publishing is not the fee.

A desk rejection costs nothing on paper and six months in practice, and then you pay the same fee at the next journal anyway. That is the cost worth attacking first, and it is decided by the manuscript before any of this pricing applies to you. The Academy works on that part, on your own draft, one task at a time. Everything in this guide still applies whichever journal you end up at.

Get published, one task at a time.

See what the Academy is