Since commerce is inherently evil, and commerce connected to the academy even more so, then it must be an exceedingly dastardly deed for a publisher to engage in double-dipping, in which the same content finds one audience in institutions and another among the authors who create it, sometimes in violation of an author’s contract. What I would have characterized as a peccadillo at worst is for many a sin of staggering proportions. One imagines Satan smacking his lips at the very prospect. Indeed, were Dante writing today, he would have to go beyond the nine circles of hell–he would have to double them — to accommodate all the evils of the post-modern world. Yes, beyond the thieves, the murderers, the hypocrites and creeps of Dante’s first nine rings, there is an entirely new set of depravities: the supporters of the Trans-Pacific Partnership, the abusers of pronouns, the forces behind fracking, and, of course, at the very bottom, in circle eighteen, the scholarly publishers who dare to sell things twice. Their punishment for all eternity is to read and re-read their very own contracts.

Criticizing double-dipping displays moral outrage unmitigated by economic understanding. To be clear, approving of something is not a necessary consequence of understanding it, but understanding something before you condemn it is not a bad idea. Double-dipping is not a dark blemish on one’s character but an emergent property of current open access (OA) practices. It is, in other words, a structural aspect of OA. It will not be swept away and in fact, like so many things connected to OA, it will actually increase costs. As I have had occasion to remark on the Kitchen many times before, whatever the benefits of OA, reduced costs are not among them.

So how does double-dipping work? Double-dipping grows out of mandates and, occasionally (rarely), the desire of authors to make their work available in OA venues. (Why rarely for authors operating on their own steam? Because such authors are likely to publish in pure Gold OA journals and avoid hybrid journals entirely.) A typical process begins with a research grant that carries with it a stipulation that any publications that are based on the research must be made available in an OA manner. (Some traditional publishers think this is a terrible interference in the marketplace. I don’t agree: it makes perfect sense to me that he who pays the piper calls the tune.) A certain amount of money from the research grant (not unusually, around $2,000-$4,000) is earmarked for such publication. The author submits the article to a subscription-based journal, where it is accepted. The author then elects to make the article available as OA by paying the fee. Thus the first financial “dip” is the income from the subscription, typically paid by a library or other institution, and the second dip is the APC (article processing charge), typically paid out of the research grant. Yes, there are many exceptions to this and room for a host of clarifying remarks. But on with the story.

Although publishers were initially uncomfortable with making parts of their publications OA, most have come ‘round now and view the pockmarked journal (in which the OA articles represent holes in the smooth surface of the subscription publication) as not life-threatening. But the hybrid or pockmarked journal is also an irritant, as tracking all the rules for what should and should not be OA is a pain in the neck. It adds to administrative costs and can result in PR headaches when an article for which an APC was paid is not made OA in an exact or timely manner.

I have heard people ask, in what must be termed a rhetorical mode, “But how much could it cost?” This is the thing about overhead: it’s hard to quantify for any specific activity (which is why we call it overhead and not a direct cost). The managing editor who is stuck with handling hybrid payments takes time from other activities and may have to hire someone or several people simply to handle compliance. This does not include the bandwidth of the senior management that goes into formulating the policy, running it by the legal department, and training the staff for implementation. Nor does it include the complexity of accommodating different requirements from the various sources of funding, or the laborious task of explaining these rules to the authors, who mostly want these headaches simply to go away. After all, the primary reason to publish with an established journal is the brand aura: if the matter at issue were simple dissemination, why not submit the article to PLOS ONE and be done with it? Taken together, all this back and forth — the setting of policies, the compliance with policies, and the education concerning policies — are hateful activities because they are pure administration — that is, they add no value whatsoever. This incremental overhead is one of the reasons that publishers charge for hybrid OA.

But so far we are still double-dipping. To eliminate the second dip, and to be sure to be in full compliance with the stipulations in some authors’ contracts, publishers are demanded to reduce the cost of the subscription — the APC, in other words, is an offset to the subscription cost. There is a significant difference if the offset is applied locally (a specific institution subsidizes an APC and then wants its own subscription cost lowered) or globally (the APC offsets the subscription cost for all subscribers around the world, an offset that may come to mere pennies). Once again we hear the question, But how much could it cost? Heh. The academy is a place where everyone runs around spending a dollar to save a dime. The APC offset may have to track authors’ institutions or the global distribution pattern of the journal, map that information against a pricing database (especially complicated if publications are part of an aggregation), and take into account the special circumstances when an article has multiple authors from multiple institutions working from multiple grants. The invisible overhead continues to tick upward, increasing costs (while adding no value), prompting publishers to look for ways to offset that growing expense — or even, if they are shrewd, to turn it into a profit center.

The obvious way to do this is to increase the price of the journal, which is relatively easy to do since not many libraries pay retail prices for journals; rather, journals are mostly purchased as part of large packages in which the cost of an individual journal is obscured. The sleight of hand here is that the increase in retail price is immediately offset by a discount. This is such a common marketing tactic in our society that it surprises me when it has to be explained to anyone. Those men’s polo shirts at Brooks Brothers are “on sale” for $49, reduced from $75. But when were they ever $75? They were put “on sale” the day they arrived at the store. At $49 the store owner may be working with a 50% margin. What’s not to like? The retailer makes money and the consumer goes home wearing a new shirt, smiling smugly for having gotten such a good deal.

These little marketing tricks may seem harmless or simply annoying, but over time they can have big impacts. Take mortgages. In the U.S. interest on mortgages is tax-deductible. When this law was first introduced, it was an enticement to get more buyers into the housing market. But prices don’t remain constant; they vary with, among other things, the ability of customers to pay for something. So the short-term benefit of deductible mortgage interest evolves into a basis for raising the price of a house. There is much not to like here, particularly the hidden pressure for households to take on more debt to finance a house. And if you aren’t interested in the nutty U.S. real estate market, consider college tuition. With rare exceptions, it is as high as an institution can get away with. If money is increased for need-based scholarships, then the retail price — the tuition paid by the rich — will rise. Make cheap loans available to students and the tuition will rise again. A twenty-one-year-old goes out into the world with $40,000 in debt and meager job prospects, the victim of good intentions.

Once a publisher begins to see that the APC and the double-dip can be manipulated, then marketing will begin to influence the offerings. An author paying an APC may be presented with a different dashboard to track usage from what a library sees; the dashboard could be connected to other, augmented resources: now that the author is a customer and not simply the provider of raw materials, the publisher has an incentive to sell more and more services to him or her. In time the article paid for by the APC becomes a different thing from the article embedded in a subscription, and the APC-funded article becomes associated with many additional services.

This will make it easier to maintain double-dipping. It is here to stay, though its form will evolve, the better to ensure the revenue of the publishers, as long as journals’ brands carry weight with tenure and promotion committees (not to mention scholars’ sense of themselves). As I recall reading in Henry Fielding many years ago, when you lock Nature out at the door, She will come in through the window. But there would be less pressure for increasing prices if the door were not locked to begin with.