Oh, and I forgot to mention an important feature, which I often take for granted as obvious but is worth spelling out. much of the work that goes into processing a manuscript goes into processing all the manuscripts that come in, not only the ones that get published. So when you have a journal that is intending to serve as a filter for importance and is as selective as possible (ie Nature, Cell, etc), you end up rejecting 90+% of articles. All those rejected articles take work to process and come to a rejection decision. So the name of the game then is trying to cover the cost of all the things you reject by monetizing all the things you accept. That's why megajournals that intentionally do not filter by importance (and thus can accept a higher proportion of submissions) are able to charge less (it's obviously not pure cost economics, top tier journals charge for their brand name as well of course).

So, how about giving the breakdown of the price that the parent comment asked? On a $1000 article, how much gets to the people doing this filtering? How much in the people organizing the peer review? How much ends up in the pockets of the rentiers?

I associate to jempekka's feeling:

> Before such breakdown is given, I keep on thinking you are leeching the work of scientists and taxpayers and making the world a worse place.

I would add that by not answering, you lose any right to complain about an "echo chamber" wrongly assuming that paywalls are unnecessary evils.

It's mostly people costs. There's also royalties to societies (the basic deal is academic societies own a lot of journals and license them to publishers for a royalty on sales). And there's tech/infrastructure costs for the systems to manage the manuscript workflow (and yes, for hosting the journal websites and serving PDFs). And even in this modern age some journals (most still maybe?) have some print component with costs. You can argue that's ridiculous (as I do) but parts of the academic library market is still in love with print for some reason.

But if you're trying to understand the real cost drivers it all comes back to the people involved. You're totally right in that peer reviewers are almost never paid (there are a few exceptions to this, like eLife). And often editorial boards of journals are not paid. Although there are also a significant amount of paid editors as well, either stipends provided to academics or in house editors. The internal staff manages the peer review process, which includes everything from finding reviewers to hounding reviewers to complete reviews, etc. That peer review management gets done by a combo of unpaid editors and paid editors.

So then there's society journals. More and more society journals have been scooped up by the major publishers, and the academic societies rely on the journal royalty to function. Now, you could argue (as I would) that academic societies should figure out ways to survive without journal subscription revenue royalties being an integral part of their finances (ie run conferences, purely via membership fees, etc) but the current reality is that a huge number of academic societies rely on the journal royalty to stay afloat.

Then there's marketing. A journal lives or dies by its reputation, and building a journal up requires a significant marketing effort to establish a journal to drive submissions. For new OA journals that marketing spend works very differently, and in fact requires a lot more spend on a consistent basis. OA journals live and die by steady submission flow (since the only revenue ever comes from accepted papers), and so you have to market the journal constantly to ensure you drive submissions.

There are also people employed that aren't direct costs to production but are responsible for new product ideas (R&D), and all the kinds of folks that are required to run a large organization, like HR, people in acquisitions trying to grow the business, people in sales, legal, etc etc.

I understand the visceral reaction against publishers. I have the same reaction to the subscription business model to be honest. My preference would be that the service publishers provide gets treated more like what it is - a service to academics who choose to use it - and gets charged accordingly. That's actually exactly what's happening with the move to Gold OA, where authors pay for the service of publishing. IMO that's actually working pretty well in STM, even though there's a long way to go. But unfortunately the non STM fields in the humanities and social sciences are having a lot of trouble in a transition like that. There are two main problems: in HSS there aren't tons of grants that academics can use to fund publication fees, and there's a huge, longstanding cultural assumption that publication should be free to academics, so it's hard to convince people that the publishing service should be bought (for any price). There are other people trying to solve this in different ways, like the Open Library of Humanities, but nothing has had significant success yet.

You can argue with profit levels (although I do think it's funny how much people are against profit in certain businesses and have no problem with profit in others, ie large tech cos). But what I object to is saying that publishers don't provide a valuable service. The use of terms like leech and rent-seeking imply a monopoly position and no value add whatsoever. First, nobody (even Elsevier) has a monopoly. There are 4-5 big publishers (Elsevier, Springer-Nature, Taylor & Francis, Wiley, SAGE) and then a huge number of smaller players (some of whom are substantial and highly influential, like Oxford University Press and Cambridge). So at best you can make the argument it's an oligopoly (although one that has about 2,000 total players).

Sorry, but I don't accept this as a breakdown. Surely as a board member you must know an approximate dollar amount/labor time going to different direct production phases per article published (or submitted). Eg, how much time it takes for the staff to manage the editors, check the manuscripts and lay them out.

The other stuff sounds like overhead, so feel free to lump it as a single figure.

PS. I'm not a fan of (obscene) profits in any industry. Even if one believes in capitalism, the usual basic tenet is that no company's, let alone industry's, profits should be able to consistently surpass the interest rate if the market is working.

I find it sort of ridiculous when publishers throw figures like $1000 a piece as a reasonable APC. That's over two weeks of research stipend in Finland at least. If it takes anywhere near two workweeks to process an article and handle the overhead, you guys really suck at what you do.

And yet nobody in the industry is doing the work for cheaper at scale. And the noble non-profit players like PLOS aren't able to do it cheaper either. So yes, you can take the stand that literally every single player in the industry really sucks at what we do, from the large commercial publishers all the way to the non-profits. Or you could acknowledge the alternate hypothesis, which is that it might simply cost that much to deliver the services performed.

PLOS's prices are ridiculous as well, and their surplus is quite staggering.

There are players like peerj that publish at the fraction of the price. Also many smaller non-profit journals have APCs in tye range of tens of dollars.

But whatever makes you sleep at night.

Edit: Also, no breakdown given, as expected.

PeerJ's APC pricing is $1,095. We're actually an investor in PeerJ and are really hoping they're a success. But they're not a counter-point to PLOS-level APCs (they are indeed still cheaper, but not substantially anymore).

And yes, there are examples of small-scale journals charging less, or society journals charging less/free because it's paid for by the society membership fees. There's Tim Gowers' arXiv overlay journal. There's a bunch of small-scale examples or anecdotes. But none of those are repeatable at the scale the academic community needs.

You're right, I didn't give you the financial breakdown of our operating costs that you wanted. First of all, I don't actually know the per-published-article costs and breakdown off the top of my head. And second, I wouldn't post that information publicly here. I guess that means you'll continue to think that we don't do any work and collect vast sums of money, sitting in dark rooms laughing about how we get to fuck everyone over. You clearly think every single player in the industry is doing it wrong. I'm not going to change your mind.

Ah, PeerJ did the bait and switch too. Thanks for the heads up.

And yes, that's what I'll keep on thinking. I wouldn't be much of a scientist if I'd take an industry shill's claim at face value, 'cuz secrets and all.

And I don't think you're very different from other corporate leeches. Probably the investors take the biggest cut of in the racket, but I'm quite certain many of the minions get compensated quite well too. Plenty of those in the academia as well.

You can call it bait and switch, but the reality is simple and not nearly as nefarious, it's just about trying to find a sustainable business model. It's either that or go out of business. There's no grand cabal running a conspiracy that forces players like PLOS and PeerJ to raise their prices. They're just trying to stay alive, and they price their product to do so, just like literally every other corporation in the world.

"Staying alive" meaning paying those yachts? If PLOS pays out hundreds of thousands to its execs, I can only imagine the gluttony going on in the for-profit scams. And they have to make those rich investors richer for being rich on top of it!

It seems to me like the status quo is working pretty well. Publishers do all the things you said, and university libraries pay the subscription fees. Anyone not affiliated with a university just goes to SciHub or the author's webpage and downloads the paper for free. Nobody cares about your copyright and they never have; before SciHub there was your college buddy with a roll of quarters and a photocopier.

You're lucky enough to have the revenue stream you do, so what makes you think there's another one in open access? Authors don't want to pay you, the university already takes half their grant money for such purposes. Unaffiliated readers definitely don't want to pay you. As you admit, you thrive on the prestige of your name, so why not stick to your business model of charging other prestige-focused institutions, and leave the rest of us out of it?

The OA revenue stream is meant to replace the subscription one, not be in addition to it (long term anyway). I can't tell how much of your comment is tongue-in-cheek or satire, but do you honestly not believe charging authors for the service they are buying (and then ensuring the results are freely available) is morally better than putting the results of science behind paywalls so the authors can publish for free? That seems like a no-brainer to me. Obviously nobody wants to pay for anything if they can not. But paying a transparent price for a service being performed seems like a huge win compared to the status quo of using copyright and paywalls.

I don't think you should charge authors, because academics should be encouraged to publish without having to worry about the cost. That's why society funds them with grants from taxpayer money, because we want to see the results of this research we have purchased. And now we can, with SciHub, and everyone is basically content with this situation except the lobbyists for your industry.

Besides, the authors are your golden geese. How much of the work of producing a journal article is performed by the authors and other unpaid academic volunteers (who are authors elsewhere), compared to the publishers? Is it 90%, or maybe even higher? Your journals would not exist in any form without the authors, so it behooves you not to piss them off. Currently, the most famous journals enjoy a prestige-driven advantage: everyone wants to publish there, because doing so (sadly) advances their careers more than publishing elsewhere. But this is all just an agreement among academics, and they could turn on you at any moment.

So I think you should stick with the subscription model. Authors shouldn't have to pay, and readers currently don't "have to" pay, regardless of whether you think they should. The only people who are really interested in paying you right now are the university libraries, because what sort of prestigious institution doesn't carry all the famous journals?

Yeah, so what you describe is basically what's happening now. Piracy is convenient enough for individuals to get around paywalls, but not so convenient or widespread (in terms of use) that universities start cancelling subscriptions. If piracy stays that way and doesn't kill subscriptions then it's sort of a win win, readers effectively get access and the work of publishers is compensated. Right now that's true, and piracy is a net positive (in my view anyway). But the obvious concern there is that that homeostasis won't last.

There are other business models that are being tried that mimic the current one a bit. For example, instead of a university library paying a subscription fee to gain access to content, they can pay a fee to allow all researchers at their university to be allowed to publish in all of a publisher's journals. Essentially moving the APC charges from individual researchers to universities. I know we've tried floating that idea and have gotten a lot of resistance from libraries, which I don't fully understand because it seems entirely reasonable to me.

One other note: Gold OA APC charges are not being paid out of the author's pocket. Nearly all the APC charges are coming from grant money that funds the research (or from central government funding, like China). All the major funders allow a PI to account for publication charges. So publication is treated just like any other research expense, like buying equipment. It all comes from the funders. That works great, and honestly seems like a perfectly reasonable setup, but it only works for funded research. That leaves entire swaths of the academic landscape out (as I've said elsewhere that means most of the humanities and social sciences).

>charging authors for the service they are buying

This description feels weird and vaguely misleading, if only because when I buy, say, auto repair or hair styling service, the service provider doesn't consult a panel of other customers to decide whether my service request is worthy of being fulfilled. "Charging authors for the service they are buying" seems a more apt description of straight-up vanity publishing.