Upon acceptance, some journals offer open access through an optional fee. This model differs from the usual open access model in which all authors pay a fee and all accepted papers are accessible without restriction. Addendum: the latter is referred to as simply open access, while the former is referred to as hybrid open access.

My question is as follows: I publish (most of) my papers on the arXiv, and as such, the preprints are picked up by search engines. When the corresponding camera-ready journal version appears, anyone with an internet connection can still access my arXiv paper, and hence I always opt not to pay the optional fees to make my paper open access.

In fact, most respected researchers in mathematics/computer science/physics/etc. also publish to the arXiv. As such, I am confused as to why this model of open access exists. What demand is it satisfying? Is there some advantage to paying these optional fees that I am not seeing?

P.S. -- The journal I have in mind currently is a SIAM journal. Their open access policies are listed here. Their open access fee is $2,500 USD (this is not an unusual number for publishers following this approach). This (at least to me) is a substantial amount of money per paper.

  • 3
    Actually, SIAM even allows you to post their final typeset copy on your own website (though not on arXiv). Commented Feb 25, 2016 at 6:42
  • 6
    "This (at least to me) is a substantial amount of money per paper." This is of course in the eye of the observer. In Computer Science we often publish in conferences, which requires us to travel there, which in essence means that we commonly pay northwards of 2000 USD for a paper (in hotel fees, flights, and conference tickets). Of course there are added benefits and the "feel-good factor" is better, but the matter of the fact is still that each published paper incurs costs in a similar dimension to us - and then the paper isn't even open access.
    – xLeitix
    Commented Feb 25, 2016 at 8:32
  • 1
    I'm asking myself the same question, especially since in fields where Arxiv is not an option, the majority of good journals allow posting preprints online.
    – Cape Code
    Commented Feb 25, 2016 at 9:03
  • 1
    I think you meant "not an unusual number" instead of "not an usual number". In fact it's a typical, if not low, fee.
    – silvado
    Commented Feb 25, 2016 at 10:01
  • 2
    In most field peer reviewed papers a distinguished from open, non reviewed manuscripts
    – Greg
    Commented Feb 15, 2017 at 1:42

8 Answers 8


If you work in a field in which people use and trust the arXiv, then there's little reason to pay fees for hybrid open access. One reason would be if your funding agency requires open access publication and does not consider the arXiv to be an acceptable substitute. Another would be if you wish to encourage the journal to transition to fee-based open access publishing. In your case, I doubt either of these reasons would be compelling.

From the publisher's perspective, it's a no brainer: why not offer authors this option? It can't do any harm, since it pays for itself. If few authors choose to pay the fee, then the publisher can use this as evidence that there is little demand to change their business model. If many authors do, then it smoothens the transition to becoming a fully open-access journal.

  • 7
    "If few authors choose to pay the fee, then the publisher can use this as evidence that there is little demand to change their business model." I long feared that this is really the main reason behind the existence of this model, similarly to how some players in the music industry initially wanted to curb digital music sales with overtly terrible offers.
    – xLeitix
    Commented Feb 25, 2016 at 8:30
  • @xLeitix I don't see why would this evidence be needed. Publishers will naturally transition to author-pay OA if that becomes the fashion. It's much more profitable for them anyways since the number of authors wanting to publish their stuff largely outnumber people willing to pay to read it.
    – Cape Code
    Commented Feb 25, 2016 at 9:13
  • 1
    Another explanation for the hybrid model: if some of the authors choose to pay the fee, the publisher can get this money in addition to the subscription fees they are already collecting. So that's a win-win for them. (This is also known as "double dipping".)
    – a3nm
    Commented Jun 9, 2020 at 23:06
  • Is there some evidence for this: "If many authors do, then it smoothens the transition to becoming a fully open-access journal"? What I heard is the contrary, that hybrid journals' APC are higher than other OA journals, but I don't really know if many hybrid journals have transitioned to fully OA
    – llrs
    Commented Nov 5, 2021 at 10:09

In fact, most respected researchers in mathematics/computer science/physics/etc. also publish to the arXiv. As such, I am confused as to why this model of open access exists. What demand is it satisfying? Is there some advantage to paying these optional fees that I am not seeing?

Because not all respected researchers work in mathematics/computer science, etc. For example, posting on arXiv or an arXiv-alike is extremely uncommon in my field, and preprints are not readily accepted - nor necessarily searched for. My only two preprints are there, for example, because a co-author insisted. However, these authors may still value open access - especially if they don't have a funding mechanism like the NIH that will mandate it become open eventually. In that case, why not pay for it?

For why it exists for fields where arXiv is a substitute? Likely because it's a publisher wide program, and there's very little reason not to offer it on the off chance someone wants to pay.


Keep in mind that there's open access and there's open access. Let's compare for example the different self-archiving models (yellow, blue, and green open access according to the SHERPA RoMEO classification) with the hybrid open access you're asking about or the otherwise commonly used gold open access model.

With self-archiving, the copyright is typically transferred to the publisher, but they permit you to post preprints or maybe even post-prints to a publicly accessible repository. The license that these repositories typically require simply permits them to distribute the paper (e.g. the arXiv non-exclusive distribution license), without giving any rights to persons who download the paper from there.

In contrast, with hybrid or gold open access, typically the authors put their paper under an open license like the Creative Commons License. This is also the model that the SIAM Open Access policy proposes. In that case, everybody can use the paper under this license. That implies that other researchers and teachers may distribute the paper, and can legally reuse figures or excerpts in their works (presentations, posters, blog posts), with attribution, but without getting special permission from the publisher.

And then there's strange open access policies like the one from the IEEE, where authors pay a fee just to have their paper downloadable by everybody: essentially the same thing that arXiv or self-archiving provides, just with the official publisher's version...

In the end, the authors can (and have to) decide which open access model is best suited for their situation.


Our lab has a portion of its budget allocated to publication costs. If we don't spend it, it looks like we haven't been publishing enough. Having nothing else we can actually spend it on, we've been paying for open access recently (e.g. some of my recent open-access papers: 1, 2).

  • 9
    This the most depressing thing I have read on this website. Author-pay open access has negatively affected academia much more than I initially thought.
    – Cape Code
    Commented Feb 25, 2016 at 8:58
  • 2
    That money may be useful for the next project.
    – phenomenon
    Commented Feb 25, 2016 at 10:15
  • 4
    Who decides that a portion of its budget allocated to publication costs? Commented Feb 16, 2017 at 15:06
  • 3
    Are you really promoting your work here? Seems jarring, since there's little that it adds to the answer. Commented Nov 5, 2021 at 3:58

Prestige and branding. Publishers are well aware of the power of brands and many academics are obsessed with the prestige of having work accepted to a ‘high impact’ journal and the associated positive effects it has on their career and their ego. So much so they are willing to hand over tax payers (the majority, but not all research, is funded from taxes, at least in the UK) money as well as the copyright of the work to publishers in order to publish their work in these venues. RCUK (the body responsible for overseeing the different research councils in the UK) awarded £22.6 million in funding (2015/2016) for Universities to publish in open access journals. Or thought of another way, scientists have handed £22.6 million to publishers for the privilege of access and reviewing their own data. One can argue that the publishers operate a service, which they do, however with profit margins of some major publishers nearing 50% I am astonished that this practice continues given the potential ease of distribution using something called the internet. We can blame the publishers for charging high fees but we basically let them and it needs a large cultural shift of academics to stop this as the £22.6 million (and that is not counting the money from charities etc.) that is given to the publishers is lost from science and would be better spent on funding science rather than lining the pockets of publishers. For this we need to stop valuing the name of the journal in which the information is published more than the actual information in the scientific papers.

  • we need to stop valuing the name of the journal No, we need to stop paying article processing charges. OP is talking about journals where authors can publish without paying and upload a preprint.
    – Cape Code
    Commented Feb 25, 2016 at 14:50

I believe the short version is that some journals do not actually permit this. Many Elsevier journals, for instance, have embargo policies that appear to disallow uploading the article to your own website or to a site like arXiv/BioRxiv for a period of time following publication, usually 12-24 months. In "hybrid" journals such as you describe, I believe that paying for open access means that the article is available immediately.

  • Nope, if you read them, Elsevier's policies actually explicitly allow this ... See academia.stackexchange.com/a/177206/13044. Or just follow the link in Patrick's answer and click on "Article sharing guidelines"! They even write that "Authors can share their preprint anywhere at any time " and "updating a preprint in arXiv or RePEc with the accepted manuscript" is allowed ... Commented Nov 5, 2021 at 4:29
  • In 2017, when I provided this link, that page explicitly stated "Cell Press, The Lancet, and some society-owned titles have different preprint policies. Information on these is available on the journal homepage." Cell Press is an imprint of Elsevier. That page used to link directly to Cell Press, but if you investigate now, you will still find that for their hybrid OA journals, there is an embargo period of 6-12 months during which authors may not replace their preprints with the accepted manuscript: cell.com/rights-sharing-embargoes
    – Patrick B.
    Commented Nov 5, 2021 at 21:38
  • To be clear, I don't mean to dispute what they said in 2017; I don't know (though there are archives) and policies can change. They are quite friendly toward preprint servers now. I also note that the Cell page on embargoes you link to is extraordinarily noncommittal - "Upon official acceptance of a manuscript, authors will be asked to complete a "Journal Publishing Agreement."" and "Our support for posting of preprints only applies to ..." are the opposite of clear MUST or MUST NOT language. And the text of the "Journal Publishing Agreement" seems to be secret. Commented Nov 14, 2021 at 8:49
  • I agree that there is some waffle-y language about posting preprints in general, and Elsevier has definitely moved on this over time. But I think their policy for posting the accepted version of the paper post-review is still fairly unambiguous: "Authors of papers published by Cell Press can share their accepted manuscript (the post-peer-review version that does not incorporate copy editing and proofing) via non-commercial hosting platforms, such as their institutional repository, after a posting embargo period has elapsed."
    – Patrick B.
    Commented Nov 16, 2021 at 21:58
  • We agree on that. I am convinced the language you quote is intended to leave room for ambiguity not captured by your literally true short version, and we'll never know. Commented Nov 18, 2021 at 2:58

The major open source licenses are irrevocable˟. And arXiv's policies are in accord with this. (Thanks, to visionaries like RMS, LLL, and allies.)

"Articles that have been announced and made public cannot be completely removed." Source: arXiv's Withdrawal Policy.

So if you've submitted it to arXiv, the journals can require you to transfer your rights, but even if you do, they will be unable to get arXiv to take them down.

I'm reading Elsevier's Article Sharing Policy and they are -- shockingly -- VERY liberal with respect to sharing of both Preprints AND Accepted Manuscripts - at any time - but not the actual Published Journal Article.

The Article Sharing Policy and the Quick Definitions of those terms in bold, above, is at the following link, which brings you to the part of the page with the latter:

https://www.elsevier.com/about/policies/sharing#definitions *
Hmm! The above previews as a clickable link like this, but doesn't save as one.

So I think most of what's holding people back from changing with the times is inertia.

*For example, here's one definition:

  • An accepted manuscript is the manuscript of an article that has been accepted for publication and which typically includes author-incorporated changes suggested during submission, peer review, and editor-author communications. They do not include other publisher value-added contributions such as copy-editing, formatting, technical enhancements and (if relevant) pagination.

˟Yes, there's even been court precedent to further establish this.


It depends on the copyright transfer agreement.

Some journals require you to transfer the rights to the pre-print. Other than book chapters, in my humble opinion, this is a predatory behaviour and such journals should be boycotted, but as soon as you sign the copyright transfer agreement, you can't upload your manuscript to arXiv or personal repository. Some journals might even ask you to take down versions from arXiv or personal repository, even if you uploaded your work to those media before submitting to the journal. I am not well versed on the IP laws to tell you whether this is legal, but they can still ask you to take them down.

If the copyright transfer agreement gives you the pre-print or even the post-print, I guess, you can even upload them to scribd. Normally, since editors and reviewers are generally unpaid, the expenses of the journal starts after the post-print, so it makes sense that they retain the full rights to the final version.

What is the downside of not paying open access and uploading it to the repository?

  • The uploaded version will probably be poor in writing, as it did not go through a detailed script checking.
  • The visibility of your article might reduce. I am very sceptical of this, as Google scholar indexes personal repositories and arXiv. But I am not sure whether Web of Science indexes personal repositories too. Still, although it is largely depending on the field, I assume the reduction will be infinitesimal.


  • Since this question is asked on academia.SE rather than law.SE, any legal precedence carries little weight here. This is because suing an editor/journal might have unprecedented implications for an emerging academician. Winning such a lawsuit can easily turn into a pyrrhic victory, while you can easily resubmit it to another journal after the rejection.

  • Journals distinguish between for profit repositories and non-commercial ones. It is never a good idea to put your work in a for profit repository.

  • 1
    Please do not use edits to disagree with an answer. Please flag again if such edits are made again.
    – Bryan Krause
    Commented Dec 12, 2021 at 0:02

You must log in to answer this question.