Monday, October 31, 2011

PLoS Is Not As Lucrative As Elsevier

David Crotty of Oxford University Press made the headline-grabbing charge that PLoS will this year be more profitable than Elsevier. I responded skeptically in comments, and Kent Anderson, a society publisher, joined in to support David. Comments appear to have closed on this post, but I have more to say. Below the fold I present a more complete version of my analysis and respond to David's objections.


It has been clear from the start that high-volume open access publishers such as BioMed Central and PLoS are far more of a threat to smaller society and not-for-profit publishers than to the vastly more robust finances of the major commercial publishers. It is thus understandable that David and Kent are concerned by the competition from PLoS.

I agree with them that more transparency in financial reporting from journal publishers is desirable, but note that PLoS is no worse than other publishers in this respect. Because of this, it is necessary to make some rather heroic assumptions in order to extract meaning from their published numbers. The fact that I, and David and Kent, can draw opposite conclusions from the same numbers indicates both that we are making different assumptions, and that making dramatic charges based on these kind of projections is unsound.

The following analysis asks what would have to happen for David's charge that PLoS would be more profitable than Elsevier to come true. I am not making a projection myself, I am analyzing David's projection, and being explicit about the assumptions I am making to do so.

In 2010 PLoS ONE published 6800 papers at $1350 each. Thus the author fee income for PLoS ONE was $9.18M. This is 76% of the total net author fee income for PLoS. I assume that PLoS is not cross-subsidizing PLoS ONE from its other journals, which means that PLoS ONE should account for 76% of PLoS expenses, or $9.28M. Thus the cost per paper of PLoS ONE is $1365. The $15 per paper loss is more than made up by PLOS ONE's share of PLOS' 2010 $1M other income from advertising, membership and interest, which would be $760K, or $112 per paper.

Suppose PLoS ONE publishes 12,000 papers in 2011. What would it take for PLoS ONE to achieve a 35% operating margin? Some linear combination of these three possibilities would have to take place:

  • If the author fee remained the same and other income totaled the same $760K, there would be income of $16.96M, of which costs would have to represent $11.02M, so the cost per paper would have to be $919. This would represent a decrease of 36% in per-paper costs.
  • If the per-paper cost remained the same, total costs would be $16.38M. So total income would have to be $25.2M. If the author fee remained the same, it would contribute $16.2M, leaving $9M to be supplied by other income, or $750 per paper. This would represent a 670% increase.
  • If the per-paper costs remained the same, and other income remained at $760K, the author fees would have to contribute $24.44M, or $2037 per paper, an increase of 51%
Alternatively, if we are to treat grants as operating income, grants attributable to PLoS ONE would have to rise from $1.6M (76% of $2.1M) to $8.24M, an increase of 515%. I find it hard to describe any of these possibilities as "the trends continue".

In response, David writes:
But I have a hard time accepting your central premise, that if PLoS ONE brings in 76% of author fee revenue that it must automatically generate 76% of costs. I think this is a flawed assumption. PLoS has (I believe) 9 publications. Each must have its own staff, its own office space, electricity, etc. Given that PLoS ONE is handling a larger number of papers than the others, it likely generates some higher costs, but it’s unclear if there’s a linear 1:1 relationship between the two. Note also that because of PLoS ONE’s high acceptance rate, the papers published there have to pay for a much smaller percentage of rejected papers as compared with the higher end PLoS journals, reducing the likely per-article cost.
He misses the point that the other PLoS journals, precisely because their cost per paper is higher, have higher author charges. I agree that it is an assumption that PLoS is not cross-subsidizing its journals. But given that the other journal's charges are 67% or 115% higher than PLoS ONE's, their costs would also have to be massively higher than PLoS ONE's to need subsidy. David continues:
If your assumption holds true, then PLoS ONE is actually losing money on each paper and the massive increase in publication volume would be a financial disaster rather than a boon.
No, PLoS ONE generates income from advertising, memberships and interest just as the other journals do. Even if we unrealistically assume no increase in these income sources, PLoS ONE would have a 3.4% margin if it published 12,000 papers in 2011. David ends:
That to me doesn’t sound like a journal that must rely on grants to break even.
I am not suggesting that PLoS ONE relies on treating grants as recurring income. David was the one doing that to support his charge that it is more profitable than Elsevier. I am simply pointing out that David's charge requires him to make assumptions that appear to me completely implausible, even if we treat grants as recurring income.

It seems clear that PLoS is comfortably above the break-even point without grant funding. Some combination of reduced per-paper costs through economies of scale and increased income from advertising, memberships and interest should enable them to continue to be comfortably above break-even even if PLoS ONE continues to double its output every year. But that is very different from running Elsevier-like margins.

No comments: