Breaking even in scientific publishing

Or, how to (perhaps) cover the costs of an academic journal

The incidental journalist

By Albert Cardona, October 8th, 2016.

There ain't no such thing as a free lunch in human affairs, and academic publishing is no different. eLife, perhaps my favorite academic journal [1], had not been charging a manuscript publishing fee since its inception. Avoiding this charge lowered the activation energy towards publishing in a new journal, and, initially, the journal was able to absorb the expense as a sunk cost of launching from scratch.

In a move that did not surprise anybody, eLife announced that from January 1st, 2017, publishing a manuscript will cost the authors $2500. The interesting bit was what will the charge cover: not the fixed costs, which will be funded by eLife's sponsors, but mostly the payments to the editors, who are busy academics and shoulder most of the work of requesting and coordinating manuscript reviews. The rationale is that, if the journal is to fulfill its mandate of publishing all papers that make it past an absolute threshold of quality, then more editors will be needed if the number of submissions increases, which reportedly it is [2].

I argue that charging for submissions and not charging at all for published papers could be more appropriate at many levels.

First, charging for submissions keeps the authors honest. The submitting authors can put their money where their mouth is: if they evaluate their own manuscript to be good enough, place the bet by paying up front [3]. Given that there will be more submissions than accepted papers (we don't live in an ideal world), the cost per submission will be relatively low. This has the excellent side effect of truly paying for scaling up publishing: if the cost really is in processing article submissions and coordinating reviews as reported, payments will be linearly related to costs.

Second, not charging for published papers keeps the journal honest. The journal will have every incentive to encourage manuscript submissions (its source of revenue, in this model), and no incentives to accept any manuscript just to get paid. Only excellent manuscripts, on an absolute standard, ought to get published.

Third, when authors demonstrate that they cannot pony up $2500 [4], the waving of the considerably lower submission fee might be less of a financial burden for the journal (eLife announced that they will wave the publishing fee for authors that cannot pay it). And, in addition, given than those who cannot pay can submit (and publish, when accepted) for free, they will be encouraged to send their best work to eLife. A win-win all around.

The caveat, of course, is that I have never run, much less founded a journal. I wonder what I might be missing from the big picture.

PS: an intended consequence of the above proposal will be a reduction in the number of submissions, which, at the limit, might raise the cost of submission to that of publishing (when the number of submissions is as low as the number of acceptances). What might be the unintended consequences? Perhaps only high-confidence authors would submit. Perhaps the overall reduction in submissions, which is equivalent to less broadly sampling the pool of possible good papers, is to be avoided. Compensating forces at play might be the preprint archives like the bioRxiv: the public response (e.g. number of downloads and the content of comments) might build the author's confidence prior to submission to the journal. I wonder whether the above effects could be estimated and the approach modeled, and if anyone has already done it.


  1. The eLife journal's inception combined spectacularly open access, the adoption of the lens publishing format for academic manuscripts, the adoption and popularization of the discussion-oriented reviewing process with unified reviews (see my prior thoughts on this), the endorsement and adoption as normal of preprints, the online system for proofs, and the flexibility in fixing up papers post-publication without the awkward and mostly useless erratum format that is a legacy of the printed press.
  2. That submissions to eLife are rising is great news. I am very happy to see eLife rising, for many reasons. I've done my bit by, and been fortunate to, publishing three papers at eLife this year 2016.
  3. Paying for submitting an article also reminds scientists that what they are doing makes the recipient incur in actual economic costs. This, in contrast to the magical thinking that I've met every now and then in academia, that considers publishing as an activity that some else does volunteering their free time. Academics, though, are hardly at fault, since what we mostly experience are the unpaid reviews that we write for our peers' manuscripts, and the process of getting a paper published is mostly a painful, time consuming journey of which we are relieved to leave behind when finally receiving a gracious acceptance letter. eLife is also changing this too, partially, with their paid reviewing editors, who are active scientists with labs.
  4. It is hard to imagine now, but back in the years 2000 to 2005, the lab of Rafael Romero, my PhD advisor in Barcelona, had 4 PhD students and published several papers, all funded with what at the time were 48,000 euros for all 4 years, plus fellowships (that paid meager salaries) for two of the students. We could not have then paid for the papers, and actually, the publication fees of my PhD papers were paid by our close collaborator and most kind human being I've ever met, Volker Hartenstein, who later became my postdoc advisor.