To Sally Morris's post on the SOAF list saying that she has "difficulty envisaging how the 'no-fee' OA model, dependent on (conscious or not) institutional or other subsidy, could possibly scale", Matt Cockerill responded:
"I think a reasonable analogy here would be to ask: can a road system scale without charging tolls? I think it is clear that road systems can scale without tolls. But on the other hand, tolls can certainly play a role, and play a bigger role in some countries than others. Non-toll roads can't be written off simply as 'unsustainable'. No one is arguing that building and maintaining roads doesn't have costs - just that there is more than one way in which they can be funded, and some forms of funding may have practical/convenience benefits (no one wants to have to pay 10 different tolls just to get to the supermarket)." End of quote.
We could take this road analogy further. Roads are not paid for by tolls at every turn, as that would disrupt the flow of traffic (though the technology to introduce just that via satellite tracking is advancing fast). So tolls are only used for 'premium' roads (and tunnels, bridges, et cetera). Instead, the vast majority of the road infrastructure is usually paid for by state subsidies, which in turn, we must assume, are funded by road taxes and fuel excise taxes. These excise taxes are interesting, because it means that there is already an element of 'user pays', as more road usage means more fuel consumption means more excise tax paid. But that user-related charge is just part of the road payment structure. Every potential road user also pays via road tax, levied on the owners of cars whether they use them or not. They pay for access.
Would something like that work in science publishing? And would it be desirable?
To a degree, and in a way, the road tax simile is already there. Institutions pay for subscriptions for potential users. It's a 'just-in-case' provision. They pay for access, not usage. It is often said that payment for usage would be fairer. But we have to be very clear as to what usage and who the user actually is. It's certainly not just the reader. It's definitely also the author, who uses publication in a journal to give his article the formal status he needs for career advancement and impact. And it's also the institution itself, depending for recognition and reputation on the formal publication record of its research population.
So it would be fair were they all to pay their share. Practically all of the money streams involved would come together on an institutional level. The purse is filled with overhead charges on research grants, and for the purpose of sustaining scientific journal literature the funds could be disbursed partially via the library (subscriptions, i.e. reader-usage charges) and partially via the authors (article processing charges, i.e. author-usage charges). But one might want to be pragmatic here. Disbursement via library subscriptions inherently limits access to the journal literature, because that is the basis on which the whole concept of subscriptions is built. Disbursement via article processing charges makes open access economically feasible. Could the reader-side charge and the author-side charge perhaps be rolled up into a single charge, on an institutional level?
Could that be a way forward? Would it be possible to come up with a charge that reflects the total usage of a journal, by its readers as well as its authors, in a given institute? A way to sustain the formal peer-reviewed journal literature that balances the need to publish (publish or perish) with the need to have access (read or rot)? Or would it be a road to nowhere?