Nature has a long and interesting article about the costs of open access scientific journals, exploring both the difference in publication costs between different journals, and the allocation of costs between funders/institutions employing scientists, researchers publishing papers, and subscribers. It’s well worth reading for the figures on costs and charges alone – I’d not seen them gathered together before. Coincidentally, I also met up this week with a friend who publishes journals for one of the big publishers (not Elsevier).

The open access case is intuitive: research is funded publicly by taxpayers, so it should be free to read. The Nature article shows that the charges open access science journals make to researchers vary widely, as do the cost bases and profit margins of the different journals. The publication fees are also the only revenue stream, in most cases. Not surprisingly, existing journal publishers see it differently, arguing that they provide other services – peer review, editorial judgement, curation, marketing and the delivery of attention to articles, hosting technology, and no doubt more. There are counter-arguments to some of these points. For example, peer review and editorial work is normally done by other researchers, who are not paid for these tasks – again, taxpayers fund the work via their financing of academic salaries.

It’s clear that journal publishers were, like other traditional media providers, slow to adapt their offer to the online world and one or two well-known ones have tried to hang on to excessive monopoly profits. But the average profit margins cited by Nature for the scientific journals don’t seem outlandish, and it is very likely that some traditional journals are hampered by a higher cost base than newer online journals, hence a partial reason for high subscription charges.

Still, there does seem to me to be a need for clarity about the bundle of services being discussed in this debate. It is not just a question of providing an online platform for researchers to report on their work, the bit that is taxpayer funded. If academics only needed to get their work out there, they could publish it on their own websites – as many do. But journals obviously do play an important role in organising peer review to give research a kite mark, in selecting and curating papers, and in marketing to draw attention to research. The importance of editorial objectivity in these services makes me tilt slightly towards preferring the old-fashioned subscription model, especially now non-academics like me can legitimately access research via JSTOR. If only the old model hadn’t been discredited by the greedy minority of publishers hiking subscription charges so much, or lumbered with its inherited high cost base.

It’s probably too late, even though there are, ironically, signs that the subscription model may be starting to function for other online publications. The new RCUK open access policy came into effect this month, & states: “Free and open access to publicly-funded research offers significant social and economic benefits. The Government, in line with its overarching commitment to transparency and open data, is committed to ensuring that such research should be freely accessible. As major bodies charged with investing public money in research, the Research Councils take very seriously their responsibilities in making the outputs from this research publicly available – not just to other researchers, but also to potential users in business, charitable and public sectors, and to the general public.” RCUK will in future fund eligible research institutes and universities to pay submission fees to open access journals.

My friend is bitter about the Government casually torpedoing the commercial business model, and predicts that academic journals of the future will increasingly be hosted and part-funded by universities, research funders and learned societies, as the business is becoming decreasingly attractive for commercial publishers. I’m not sure the future landscape for the publication of academic research is at all predictable.