The Antaeus Column

1

The Antaeus Column:

The “author pays” model of open access and UK-wide information strategy.

Abstract

Purpose of this paper / To comment on recent trends in UK information strategy which aim to further the development of a coherent national “author pays” open access (APOA) research publication system.
Design/methodology/approach / A description of APOA national policy initiatives, which is put into a wider context by looking at some economic analyses of the principles underlying this form of open access. This is in turn followed by the author’s own conclusions, which synthesise these two perspectives.

Findings

/ It is not at all clear that the economics of ‘author pays’ open access are well enough understood to guarantee that the original aim of open access – to deal with unaffordable serials price inflation -will be achieved by a large scale move towards a national APOA system in the UK.An enhanced, nationally coordinated move geared towards establishing APOA on a proper footing would be a bold experiment, and, as such, it might not fully achieve its aims. In recognition of this possibility, it would be worth considering the establishment of a similar, enhanced, UK-wide programme for the development of purely repository-based open access materials, to be developed in parallel with an APOA system, as an insurance policy in case the author pays model does not realise its full potential.
Research limitations/
Implications / This paper does not give any clear description of the nature of an enhanced, UK-wide repository-based open access programme. Further investigation would be required to ascertain if this suggestion is feasible. It may be the case that existing national initiatives aimed at supporting the growth of open access subject and institutional repositories have exhausted the potential for coordinated UK-wide development of this strand of the open access movement.
Practical implications / The insights provided give some idea of the practical difficulties of taking forward APOA systems on a national basis.
What is original/of value in the paper? / This paper combines views taken from a variety of different sources in the hope that a fresh perspective on the issue of “author pays” open access is made available to the reader.

Paper type: Viewpoint

Keywords:Open systems,generation and dissemination of information, libraries, publishing, United Kingdom.

Introduction

Recently in the United Kingdom, there has been a spike of interest in improving the efficiency of the ‘author pays’ model of open access to published research. In 2008 the JISC (Joint Information Systems Committee) collaborated with a Universities UK/Research Information Network (UUK/RIN) working group on the payment of open access publication fees to conduct a couple of surveys –one aimed at authors, the other aimed at institutions – which elicited information from them on how well this form of open access was functioning1.

The surveys showed that there were problems. It appearedthat ‘author pays’ open access arrangements were highly variable from one institution to another and that levels of awareness of open access among research-active publishing academics also varied greatly. In fact only 10% of respondents worked in an institution with some kind of coordinated approach to the payment of the per-article open access fees charged by open access publishers, or publishers with an open access option.Overall, there seemed to be a need for a heightened degree of coordination and managementso that the full potential of this publication model could emerge.

In order to provide this greater degree of direction, in March 2009 UUK/RIN published guidance to research institutions, with recommendations on how to manage publishers’ open access fees at the local level (Universities UK/Research Information Network, 2009). It was suggested(for example) that the process be managed by a pro or deputy vice chancellor, and that central budgets be established to which researchers could apply for funds to meet the costs of publication fees.

And since then, the JISC has circulated another survey2 which explores the possibility of taking this vision of arrangements for author pays open access one step further, with the JISC itself taking a national management role (One question asked: “If JISC Collections were to create a centrally administered open access publication fee service for UK Higher Education Institutions, would this be of benefit to your institution and what might be the potential advantages/problems?”). There is no definite outcome from this consultation as yet, but the idea is in some ways reminiscent of JISC’s management of national consortial e-journal subscription arrangements for conventionally purchased ‘library pays’ journals.

The principle is that, where disparate local arrangements are not working optimally, the provision of a coherent national structure may create economies of scale and a greater consistency of approach. And certainly, as far as consortial national negotiations for e-journals are concerned, the JISC set-uphas proved good value for money, helping local institutionsget more ‘bang for their buck’ than they would have done by handling their e-journal subscriptions at local level, or in less efficient forms of multi-institutional consortia.

However, the interest in managing publisher payment fees at national level can be taken to imply one mighty big assumption. This is the assumption that the ‘author pays’ open access publication model is a straightforwardly good thing, in the same way that well managed e-journal purchasing consortia are an intrinsically good thing. If the national bodies approving specific macro-level forms of information management say that a certain form of open access should be managed in such a way, then it may be construed that they areimplicitly giving aseal of approval to what they are managing – one particular form of open access.

This paper examines whether this assumption is accurate, and makes some tentative, cautious conclusions about the best way forward on ‘author pays’ open access and national information strategies.

Two forms of open access

Firstly, we should say that ‘Author pays’ open access (APOA) has many positive features to recommend it, especially in comparison to the other main form of open access, deposit in an institutional or subject repository by self-archiving. APOA is an economically well-founded model of open access, because it clearly addresses the issue of resourcing the act of journal publication. The thinking behind the model is based on the idea that, if library subscriptions are not to be the source of funding for publishers to publish, then the author should pay instead.

Repository-dependent open access, by contrast, appears to be an economically naïve form of open access, which fudges the issue of resourcing the act of journal publication. It does so by making eccentric assertions about resourcing: it implies that it is not an issue because the act of self-deposit in an institutional or subject repository is so easy for researchers that it is effectively a resource-free activity. This theory essentially clings to the belief that ‘there is such a thing as a free lunch’, but does so in a world full of starving people whose experience shows that there isn’t. There has to be some sort of market for such a service to work, because markets show you what level of resourcing is necessary.

The economics of ‘Author pays’ open access

However, if we want to agree that APOA is open access for grown-ups with bank accounts, then the economics underpinning it need to be understood in some detail. And this is problematic.

Because open access encourages potentially infinite use of a published paper, then a ‘pay-per-use’ economic analysis of APOA does appear to prove that it is cheaper than traditional subscription-based publication, or the reader-pays-per-use journal publication model. The trouble is, before the benefit of greater reader access is enjoyed with APOA, someone has to pay the initial publication charges. And because there is no decrease in the demands on traditional library serials acquisitions budget, publication charges effectively have to come from a new source of funding.

So, when UUK/RIN makes the high-minded statement that ‘Higher Education Institutions (HEIs) should establish dedicated budgets to which researchers can apply for funds to meet the costs of publication fees,’ (Universities UK/Research Information Network, 2009) then HEIs will reply, ‘Where does the money come from?’

The hope has to be that all funders of research (the UK Research Councils, the Wellcome Foundation and the like) will provide specific funds within their awards to cover the costs of publication fees, and that these small specific amounts of funding can be aggregated into one large central budget at each HEI. And to a great extent this is true, research grants increasingly do appear to provide some degree of subsidy for publication charges, and these small subsidies show potential for centralisation.

But again, because the size of and call on traditional library serial subscription budgets is undiminished, how can we justify redirecting an enormous extra chunk of funding out of a pool of cash originally created purely to fund research in order to provide what is in many ways no more than a new revenue stream for commercial publishers?You can pay commercial publishers out of your front pocket or your back pocket, but you’re still paying commercial publishers to do something for you.

The bottom line

The answer is that it’s fine to pay commercial publishers under a new publishing model so long as it’s cheaper to do so.Thus, there are some important cash calculations that need to be made here, and it would be helpful to try and make these calculations with real data. Unfortunately, we can only make some tentative illustrations with suggested data.

For example, take the imaginary example of a journal costing £1000 per annum with a subscription base across Europe and the USA of 700 universities, in which 60 articles are published per year. The total annual cost to the research community is £700,000 to use this journal to disseminate research. If this journal were converted to an APOA model, then, if each article costs a fairly representative£1500 to publish, the total annual cost to the research community of using this publication outlet would be £90,000.

In these terms, the APOA model looks massively better value for money than the traditional model – some eight times cheaper. And the potential infinite increase in open access use by readers outside the traditional subscription base of purchasing HEIs does not even have to be built into the argument to make the sums add up.

Pessimistic voices

But, unfortunately, there are a number of voices who cast doubt on whether the publication fees commonly levied by APOA publishers at the moment are in fact commercially realistic. Butler (2006) produced some figures which were widely noted at the time and seem still highly relevant: he reported that the publication fees levied by the prestigious open access service Public Library of Science, PLoS3 receive a heavy degree of subsidy, but even then a hefty hike in article publication charges was necessary in 2006, from $1500 per article to $2500. This is a 66% annual rise (which sounds remarkably like a serials subscription inflation crisis, as predicted for APOA journals by Scaria, 2003).

Although Peters4speedily rejected the pessimistic report from Butler, his response was largely an assertion from someone in a well-informed position, and his rebuttal providedfew specifics and figures. Others agreed with the pessimism of Butler (e.g. Jumars of the American Society of Limnology and Oceanography: “I would challenge you to find an economic model that works for open-access publication, i.e., one that provides well reviewed, high-impact products on a sustainable economic basis.”5)

More sophisticated analysis

Most of these discussions are relatively accessible and journalistic in their treatment of the economics of ‘author pays’ open access. By contrast, there are some far more dense economic analyses of this topic, such as those by Litman (2006), McCabe and Snyder (2006),

Walters and Wilder (2007) and Willinksy (2009).These containsome impressive pieces of economic analysis, but they have considerable drawbacks.

Firstly, some of them are impenetrable to the average library and information management practitioner, and secondly, they do not give us the voice of the commercial publishers, who are themselves unlikely to use such recondite numerical analysis to make hard-nosed decisions about profit and loss. Indeed, publishers would probably view some of these papers with as much bemusement as working librarians.

But, even if most of us cannot challenge the sophisticated economic analysis of these intelligent authors, we can understand their conclusions. McCabe and Snyder say that‘The more market power the journal has, the freer it is to extract rent from both sides of the market, readers as well as authors, through high markups.’ So as long as an APOA journal is popular and can exert the monopoly power of its reputation through some sort of pricing mechanism, be it a library subscription or a publication fee,then ‘high markups’ are just as likely to be a feature of such author pays journals as in traditional, library subscription-based journals.

And Walters and Wilder make the common sense observation that APOA publication is wonderful for institutions that do not publish much, but horrible for the elite universities that do publish a lot: ‘the top universities would pay up to 10 times as much as they currently do.’ 1000% inflation? Now that really does sound like a serials pricing crisis.

Known unknowns

As we read further, our authors seem increasingly to be saying that the one thing they do know is that they do not know enough to assess the APOA model accurately. Litman states that ‘open access publishing seems unlikely to have significant impact on the cost of generating and disseminating research…Nor do we have enough experience with open access publishing to conclude with any confidence that it will reduce the overall costs of consuming scholarly research.’

Similarly, Willinsky’s analysis is highly complex, but it seems to bring neither the author nor his readers any further forward in making conclusions about the topic. On his last two pages he asks plaintively ‘How can a $3,000 USD article-processing fee be justified while thousands of open access journals are publishing peer-reviewed articles at no charge to authors or readers?’ He goes on to conclude, ‘What is missing from the current economics of open access is a more exact accounting for pricing differences and why they should be sustained.’ In other words, only commercial publishers have a grip on the real figures underlying their business (‘exact accounting for pricing differences’) and these data are commercially sensitive. On the other hand, academic institutions operate non-commercially with little clear idea of what it costs to publish in open access journal, so they are most probably operating with invisible subsidies. This leaves most commentators on this industry making back of the envelope calculations of varying degrees of sophistication about who does what for how much.

The cloud of unknowing in which we are operating is reflected in the UUK/RIN’s guidance on pushing forward the APOA model. As this form of open access is pushed forward with the admirable steps that the UUK/RIN propose, their recommendation is that ‘Publishers should be as open as possible about their business models…’ which accurately implies that otherwise we will probably not have enough financial information to assess how APOA fees are set, where they will rise to, and whether they willbe affordable or not.

However, quite rightly, no commercial company will readily disclose competitive intelligence about its own business processes. And if some commercial publishers were abusing their monopoly position, they would hardly furnish their customers with evidence to prove that they were acting inappropriately. So, there is a danger that this recommendation is simply rather naïve. Thus, if, as is likely, this information is not forthcoming, we really need to know what the next step is – should the APOA model be abandoned as too opaque to be assessed properly?

Conclusions

To sum up, let us return to one of the chief original purposes of the open access movement, which was to deal with the serials pricing crisis.I would suggest, based on the arguments above, that it is not at all clear that the economics of ‘author pays’ open access are well enough understood to guarantee that the original aim of open access will be achieved by a large scale move towards a national APOA system in the UK. Research institutions can be bankrupted by publication fee inflation just as much by serials subscription inflation. So UUK/RIN and the JISC should proceed cautiously.

Although the absence of any fee for use by the reader is the one indisputable advantage of APOA over the traditional library subscription serials purchase model, if the publication fee for putting an article on the web is prohibitive, then this advantage is irrelevant. And the fact that, as stated above, APOA is an otherwise economically well-founded model of open access which clearly addresses the issue of resourcing the act of journal publication, should not blind us to these realities.