A change of plan for UK open access?

Open Access to research is almost universally seen as a good thing – though we don’t half like to bicker about the best way to make it happen.

The benefits are enormous. Research can be analysed, critiqued, and built upon faster and more efficiently than ever before, automated tools can be use to manage and summarise huge areas of research, and the public is finally able to read research for itself.

But the costs are mounting. Almost uniquely, the UK plumped for “Gold” OA as our primary route – meaning that article processing charges (APCs) need to be paid to publishers by the author, institution, or research funders (there are numerous shades of meaning within OA terms, but this is the general position). The idea was that the cost of APCs would drop as publishers became more used to the idea and competition set in – but this has not happened. Instead, we’ve seen a rise of 16% – some of this can be linked to the weakness of the pound, but it’s still inflation-busting.

So for me the headline advice expected in a report to the minister from Adam Tickell is that the current ‘preference for Gold OA is expensive – and there is a need for clarity as to whether the UK should maintain this approach. We understand there will be a recommendation that calls for a clear, UK-wide policy ambition for Open Access for the next five years, with a focus on achieving open access as the default publication option.

The advice from Adam Tickell and the Open Research Data Task Force report have now been published, along with ministerial responses.

Where has all this come from?

In this smartphone age, isn’t it great that people still write each other letters? Back in February 2016 Jo Johnson sat down to drop a note to Adam Tickell about open access to research. This wasn’t completely out of the blue – Tickell had advised the government on open access earlier that year. JJ’s response committed the sector to deliver nearly all research outputs through open access routes by 2020.

The news so far on this is great – the UK hit 54% of outputs as OA in 2016, up from 15% in 2012. We are firmly on track to achieve the target. And there is substantial evidence that OA articles are downloaded more, cited more, and used more than their non-OA counterparts, both from journals and repositories.

Twenty twelve was the year of the Finch report, a landmark that has set the OA policy weather ever since. Finch expressed a preference for the Gold OA (APC driven) approach as the “main vehicle” for the publication of research. At the time some saw this as a safe, status quo, decision – buy even back then you had the Research Councils and the likes of Paul Aryis agitating for an alternative.

Beyond Gold

The main alternative to Gold is known as Green OA – authors publish wherever they like, but also make a copy of their article openly available on a research repository. There are numerous publisher approaches to this – some ask that this only happens after an embargo period, some place other restrictions (not allowing the final, as published, version to be shared is a common one) on sharing. Some publishers don’t allow it at all, others have gradually increased the number and type of restrictions they place on repository deposit – spend the day with Jisc’s SHERPA ROMEO for an idea of how much fun these variations can be.

The continued popularity of SHERPA ROMEO relates to differing requirements about open access publishing from research funders. Some of these publisher wrinkles about Green may be approved on the nod, others may be forbidden entirely. To get a handle on the funder end of the equation, take a look at the gloriously named SHERPA JULIET and then try matching the two with the disappointingly named SHERPA FACT (surely we were all hoping that in SHERPA VERONA we would set our scene, where two databases alike in dignity…).

This is all a bit of a mess, so recent years have seen a number of funders (notably Wellcome and the Gates Foundation) becoming publishers in their own right, and moves in Europe (most notably Plan S) to bring about a supranational solution to open access publishing. Recently, UCL have established an open access mega-journal. We’ve also seen the (completely illegal) Sci Hub host stolen papers in a searchable interface – bringing condemnation from many but throwing a lifeline to independent researchers.

For Gold OA “offsetting” deals have begun to emerge, where Jisc Collections negotiate a single payment that combines subscriptions (for non-OA materials) and APCs (for OAs) – this has lowered prices a bit, but costs to institutions are still rising. We expect Tickell to call for more sector engagement and more transparency about aims for these negotiations.

Overall there is a growing consensus that the current academic journal market is not working well – both within the UK and elsewhere in Europe.

Exploring DORA and finding funding

Open Access conversations are often couched interms of an academic publishing “market” – with recommendations set as interventions within it. As (arguably) with universities, there is a popular hierarchy of journals in various fields – largely maintained by nonsensical statistics like the much-derided Journal Impact Factor (JIF). A wider adoption of the San Francisco Declaration on Research Assessment (DORA) could bring pressure on publishers to stop this behaviour – as was recommended by the UK Forum for Responsible Metrics.

The JIF-centricism of the current journals market also stops high-quality new journals from becoming established. The legacy of Beall’s list is that we tend to perceive new and unknown journals as predatory – a state of affairs that established publishers may be quietly happy with.

The fact that Gold OA comes at a cost is a clear disincentive for many researchers despite being a REF requirement – especially where a grant may not cover publication costs or where interesting work is being carried out without a grant. The research councils currently provide block funding for APCs, but this is unlikely to be a permanent feature – we might see a short extension, but only if the Gold OA policy direction continues.

If it does, and if we assume that current price trends and publishing patterns continue – and that OA take-up in the UK were to reach 100% by 2025 – total expenditure would rise to £362m in 2020, and £818m in 2028 – over three times the 2016 figure in real terms. We understand there’ll be some economic modelling published alongside Tickell’s advice, which looks at these figures across a variety of scenarios.

Clearly, something needs to be done about this – the UK already leads the way in terms of infrastructure and expert advice, what’s now needed is a strong policy commitment to back this up. From what we know of the recommendations this could now be on the cards if the minister has the bravery and foresight to take the issue on.