The UK government’s Business, Innovation, and Skills (BIS) Committee issued a report this week that offers a harsh rebuke to the RCUK’s proposed plans to drive the adoption of open access (OA) publishing in scholarly journals.

The RCUK’s policy has been controversial from the start. It has resulted in multiple hearings in Parliament and a general consensus that the key stakeholders — researchers, research institutions, libraries and publishers — were not adequately consulted before the plans were issued. The policy has already seen significant revision, and this latest salvo from Parliament will likely mean an even more profound course change for the RCUK.

A reversal in strategy from the RCUK’s preferred Gold OA route to a Green OA dominant policy seemed inevitable. The RCUK’s policy was something of a bold gamble, an attempt to stake a leadership claim that would inspire other nations to follow suit. The RCUK was willing to significantly increase spending to pay article processing charges (APCs) for immediate Gold OA for articles under an assumption that this investment would pay off in eventual savings — that is, if other nations did the same, the UK would ultimately come out ahead and its outlay for OA articles would be returned in free access to articles funded by other countries, allowing savings by cutting subscription journal spending.

But things didn’t work out that way. Rather than increase spending like the UK proposed, most other governments instead chose a different route to increasing access to funded research articles. Policies like that of the Australian Research Council or the US White House Office of Science and Technology Policy essentially torpedoed the RCUK’s plans. The RCUK was left with a controversial and expensive policy that did not provide adequate return on investment.

That’s the major point made by the BIS report — that the RCUK’s policy is out of line with the rest of the world and that the focus on Gold rather than Green is “mistaken” and must be corrected. The report takes both the RCUK and the Finch Report to task for basing decisions on incomplete data and recommendations that lack both qualitative and quantitative evidence.

Despite recognizing the problems with setting policy based on incomplete (or nonexistent data), the BIS recommendations are frequently guilty of the same sin. Rather than basing policy on evidence or calling for collection of meaningful facts, numbers seem to be randomly pulled out of a hat. Some numbers within the report do not match from paragraph to paragraph, and urges for more international perspectives are followed by UK-centric observations.

The report regularly calls out the excessive profit margins of publishers, yet mentions only one such publisher, Elsevier, who due to its size likely represents an edge case. No attempt is made to get a broader view of margins of the publishing industry as a whole, and no thought seems given to understanding whether OA really does anything to help reduce those margins (Hindawi’s OA profit margins, after all, exceed those of Elsevier).

Throwing things against a wall and hoping you’ll be able to clean up the mess later on seems a poor substitute for evidence-based reasoning.

Analysis of increases in journal pricing is limited to one flawed comparison — total spend versus the rate of inflation. No attention is paid whatsoever to the increasing quantities of articles and journals that are being purchased, nor the obvious flaws in this sort of approach.

The BIS Report suggests that the Finch analysis of costs assumes far too high a price for APCs. This assumption seems to be based mostly on the rate charged by PLOS ONE — again, perhaps something of an outlier as no other journal has been able to come close to replicating their business model and success. Even worse, there’s no recognition that current APC levels are likely far lower than what could be expected in an all Gold OA world, because most rates are subsidized in some manner. Any rate charged by a hybrid journal is subsidized by the subscription revenues of that journal, and does not reflect the amount that would need to be charged should this subsidy disappear. Rates charged for fully OA journals from larger publishers are similarly subsidized by the publishers’ stables of subscription journals which are contributing to covering, if not entirely covering, editorial and production costs. PLOS suffers from these same issues–their high end journals are all priced at unsustainably low levels because they’re subsidized by the profits made from PLOS ONE authors. And that’s not even getting into the subsidy most journals receive from secondary licensing rights.

Embargo periods are another worrisome area. The BIS Report calls for a move back to 6 months for STEM journals and 12 months for everything else. The rationale for this is that no evidence of harm from short embargo periods has been presented. Such reasoning falls prey to the logical fallacy that absence of evidence is evidence of absence. No attempt is made to factually justify the 6 month figure. No attempt is made to create a logical framework for setting embargoes for different fields. The OSTP Policy at least started with some data, using the 12 month embargo that had already long been in effect for the NIH, and then offering a framework for using evidence-based arguments for changing embargo periods.

Setting an embargo period that is too short is the equivalent of an unfunded Gold OA requirement. As has happened for 6 month policies like that of the Wellcome Trust, many journals response may be that if a shorter embargo is required, then the author must pay an APC and make the article available immediately. There seems little incentive for journals to make a major shift here, given that much of the rest of the world seems to be setting at least a 12 month standard. And since BIS also recommends against allowing RCUK funds to be used to pay for APCs in hybrid journals, this may result in a reduction in academic freedom, as researchers would be limited to a subset of journals and may have to pay for compliance out of their own pockets. What alternative will a UK scholar have if his journal of choice refuses to play ball?

On the positive side, it’s nice to see someone listening to researchers and pulling back on demands for the use of CC-BY licenses. These blunt instruments create more problems than they solve. Researchers have made it clear that in a reputation-based economy, they want at least some level of control over the use and reuse of their names and work. The ultimate solution, which the BIS Report doesn’t get into, is to find better licensing terms that offer the advantages sought from CC-BY without all the concurrent negatives. The text- and data-mining friendly licensing terms of Rockefeller University Press seem a good starting point. One has to wonder though, if the guardians of the term “open access” will object to its use by the RCUK if a CC-BY license is not used, thus contradicting the sacrosanct definition that has been set in stone.

Despite the flaws listed above, the BIS Report is probably good news for OA progress. These are just recommendations, and it’s unclear how they will be implemented (or whether they will be implemented at all). But the UK government has spoken, and spoken quite clearly on the direction it wants to take.

Simplifying requirements and setting common global standards makes compliance easier for researchers. Ideal policies are those that find ways to expand access without unduly burdening the researcher, both in terms of effort and in terms of diverting funds away from conducting research. The Green route seems much more likely to succeed, and the growing consensus supporting it can only help move things forward. Common policies can lead to common tools and reduced development and support costs for all involved. It seems likely that if CHORUS can adequately meet the needs of US funding agencies, it could be fairly quickly expanded internationally, offering additional savings and allowing research funds to be spent where they should be — on research.

Update: Having spoken with my UK colleagues, some further information on what this report means, if anything, in terms of actual changes to RCUK policy: BIS is a committee of a department of the government, led by a member of the opposition (Adrian Bailey). It is not official government policy and directly contradicts the agenda of David Willets’ government. Presumably the next steps are that the government, and Willets will respond to this report. The report may be so damning that it causes the government’s OA policy to be re-opened and re-formulated, but then again, it may not. Stay tuned…