Introduction

University libraries and academic publishers share a common mission: to foster access to the world’s knowledge across disciplinary, institutional and national boundaries for researchers, students and academics alike. Whatever the tensions experienced in the relationships between the two parties, our shared core mission should provide strong motivation to explore how we can work together in the most efficient and effective way. This article identifies three areas in which more effective collaboration could deliver substantive results. Firstly, it looks at how straightforward changes in publisher practice might yield significant rewards for their customers in the shape of improved workflows. Second, it considers long-term trends in academic libraries in relation to their core mission and the cost of content, and looks at the requirement that this places on publishers to reconsider their own business models. Finally, it looks at the scope for transformational change in academic publishing as a result of open access (OA), and argues that publishers that explore radically new business models will be in a position to flourish in the future.

Delivering efficiencies through changes in publisher practice

The development of OA within UK higher education has required institutions and publishers to adapt their workflows and systems to take account of their changed responsibilities. It is clear that we have not yet reached the end of that process of change, and there are myriad small ways in which publishers currently work which, down the line, have a negative impact on institutions, cause frustration and generate unnecessary costs for their customers. Jisc has been working with university libraries to identify what these practices are and is now sharing the findings with publishers to see how these issues can be addressed.1

One example is the author’s acceptance letter. Libraries would like to see this include some core basic information, for example at the very least an acceptance date, the article’s DOI and a clear statement on what the author may do with the accepted manuscript (AM), among others. Why does this matter? Without this information, staff at the author’s institution are required to search for, and interpret, any policy on the publisher’s website, which can be a time-consuming process. They may also need to investigate other details, such as the acceptance date, taking up valuable staff time.

Institutions would also like to see publishers supplying the AM as an attachment to the acceptance e-mail, in a form that permits deposit into an open repository. We would want this e-mail to go to co-authors as well as the lead author.

Another example is the use of ORCID IDs. If publishers were to adopt ORCID identifiers throughout their workflows from submission to publication, and expose them in published articles, this would help institutions to identify outputs from given individuals and institutions. It would help libraries to populate institutional repositories and monitor OA compliance.

Lastly, it would be entirely appropriate for publishers to ensure that a Creative Commons Attribution Licence (CC BY) is the only option available for authors where this is a funder requirement, to help authors and their institutions ensure that they are compliant.

It may well be the case that there are institutional work practices which publishers would like to see changed, and no doubt the library and research communities would be happy to engage in a dialogue on these issues. However, in the meantime, it is reasonable for universities to expect to see a step change on these issues over the coming year.

Reconsideration of publisher business models

Library trends and the big budget squeeze

SCONUL has been collating and publishing statistics on the work of academic libraries for the last 35 years, and this longitudinal data set provides a powerful tool by which to examine trends in the academic library environment. An examination of these trends gives us four key facts about the operation of university libraries:

  1. It is clear that the number of visits by students to their libraries is increasing rather than decreasing, despite the growing availability of digital resources. Why is this? A few years ago, opinion pieces predicting that the digital revolution would bring about the slow but certain death of the university library were a regular occurrence. These analyses misunderstood the importance of the library as space – in fact, the library has retained its central importance as a hub for learning on campus. The reason that individual students choose to visit and work in the library will vary – for some it is a place for study in companionable silence, for others it is a venue for group work. But footfall is going up, not down, and there is no sign that this trend is slowing.2 As a result, libraries across all mission groups are feeling the pressure on study spaces and on providing support to students. So, the first key fact is that library spaces are busier than ever.
  2. The library is working across a broader range of institutional priorities than ever before. Convergence, or at least collaboration, with IT and student services functions within the institution are long-standing aspects of library life. But libraries are now also deeply embedded in supporting institutions on issues such as research data management, where librarians’ knowledge of discovery and digital preservation is highly valued. The same is true of OA, the complexities of which have deepened working relationships with research offices and academic departments. So, key fact two is that libraries are being asked to do more to support a wider range of institutional priorities.
  3. Libraries have always played an important role in supporting student success through training and support services – whether that is information literacy, digital literacy or research skills.3 Again, there is more rather than less of this going on, with librarians reaching a greater number of students through an increase in the use of e-learning tools and massive open online courses (MOOCs) to deliver skills training. It seems to be the case that digital natives need as much support as their predecessor cohorts of students, although the exact nature of this support may be different. So, key fact three is that libraries have not done away with their more traditional roles in supporting students.
  4. The figures shown in Figure 1 are for all UK SCONUL institutions over the last seven years, from 2008 to 2015, and are for the amount of spend per FTE user (and so include academics and researchers as well as undergraduates). The top line of this graph shows that library budgets are essentially flat, and have been over the last six years. The lower line shows the amount that libraries are spending on content (including monographs, textbooks and databases as well as journals). These figures do not include spend on article processing charges (APCs) or institutional spend on textbooks where these deals are financed from sources other than the library budget.
Figure 1 

Library budgets and spend on content from 2008–09 to 2014–15

This graph illustrates very clearly that the proportion of their budgets that libraries are spending on content is increasing. So, key fact four is that paying for content is taking an increasingly large proportion of library expenditure.

And of course, as described above, other demands on libraries have not lessened at all, meaning that libraries are facing a challenging situation, and one which is fundamentally unsustainable.

How have libraries been responding to these pressures up until now? Some institutions have had no choice but to limit their take-up of the big deals. All libraries have examined and re-examined their working practices, to deliver their support and services as efficiently as possible. Many libraries have also cut staff. Figure 2 shows how staff numbers have fallen per FTE library user over the last six years.

Figure 2 

Staff per FTE user from 2008–09 to 2014–15

This is not where libraries need to be or wish to be. Staff numbers have now been cut to the very bone and there is nothing left to cut. Meanwhile, they are being asked to do more rather than less to support the learning and teaching and research missions of their institutions. Libraries will need to stem this fall in numbers in order to meet these demands. Other areas that have suffered are library spend on research monographs and spend with smaller publishers including learned societies, with all the concerns about the impact on academic freedom that this entails.

In other words, libraries are reaching a crunch point. There is very limited scope for more efficient working, so university libraries are facing unpalatable choices. Cutting access to content is not where libraries want to be, but it is where they are. The current situation is not in the interests of libraries, but it is not in the interests of publishers either. Maintaining income levels at the cost of diminishing access to the research published is not a ’win‘ for publishers, given the shared core mission described above.

Poor competitiveness in the legacy hybrid journal market

As a nation, we made the decision to embrace OA. But the transition is too slow, and too expensive, and so we need to be clear about why this is the case.

Table 1 is a summary of a table produced for a Jisc paper and sets out our collective joint position together with Research Libraries UK (RLUK) and the Association of Research Managers and Administrators (ARMA). It analyses the competitiveness of the OA market against the legacy hybrid journal market.10

Criterion New open access market Legacy subscription/hybrid market

Market concentration Market concentration is low. Market concentration is relatively high.
In 2015 just 16%4 of all OA journal titles were published by the ten publishers with the most titles. In 2009 35% of all journal titles were published by the largest ten publishers, but in just five years that figure had jumped to 45%,5 mainly because large commercial publishers continue to take over publishing operations from learned societies that operate close to the academic community.
Barriers to entry There are low barriers to entry. There are high barriers to entry.
Seven of the ten OA journal publishers with the largest range of titles were founded since 1994. Eight of the ten journal publishers with the largest range of titles were founded before 1900.
The success of diverse new publishers in the OA market, such as the Public Library of Science (PLOS)6 and Hindawi,7 is evidence that this market features high levels of innovation, new technologies and business models, that enable new entrants to operate at scale and to compete with incumbents. A variety of structural features limit the opportunities for new entrants and create significant barriers that very few new publishers have been able to overcome in recent years.
Strength of customer response Customer response in terms of price sensitivity is relatively strong. Customer response in terms of price sensitivity is weak.
‘…we believe that for full OA journals, author sensitivity to the levels of APCs has been working effectively in creating pressure to moderate the price of APCs.’8 ‘…hybrid OA articles are significantly more expensive than their full OA counterparts and the price level is an important factor in inhibiting uptake of the hybrid option.’9

Table 1

Analysis of the competitiveness of the OA market against the legacy hybrid journal market

This demonstrates that against the competition authorities’ measures of competitiveness,11 the OA journals market looks quite healthy. However, the hybrid journal market does not. Market concentration is high, there are significant barriers to entry, and customer responsiveness to price is relatively weak. For example, in 2009, 35% of all journal titles were published by the largest ten publishers, but that figure has now risen to 45%. Hybrid journals are significantly more expensive than their OA counterparts. This uncompetitive practice has a real impact on UK universities. We have seen the impact on library budgets and on the inclusion of research monographs in their collections.

But it has also had a really significant impact on the diversity of journal publishers as smaller publishers have been swallowed up. And perhaps most importantly, around 60% of UK research outputs are still hidden behind paywalls.

Next steps for publishers

It is clear that some publishers are working hard to shift their thinking and adapt their business models to the new realities, and this should be recognized in any discussion of the issues. Libraries are also exploring how to transition to a fully OA world, not least in moving away from an historic print spend model.

However, other publishers are not taking this approach and instead are adopting what might be called the ostrich posture – head in the ground, ignoring the fact that government and funder policies in the UK promote OA. However, OA is a reality in the UK, and change is coming whether or not these publishers are ready for it now.

Universities have a number of clear challenges for this group of publishers:

  1. Embrace offsetting now. There is no moral defence for charging institutions twice for the same content. Reputation matters, particularly when it comes to spending public money, and funders and government have finite patience. Wellcome’s recent analysis of OA data makes for interesting reading in this regard.12
  2. Understand the reality of library and institutional budgets. Arguments about increasing the number of journals included within big deals are not particularly relevant. Of course, additional content is welcome, but it does not change the fact that libraries cannot afford to pay above inflation increases.
  3. Adjust your business models in ways which aid the shift from hybrid to OA, not least as a way of reducing your own costs and administrative barriers. Publishers such as Springer and IOPP, which have already taken this step, will enjoy the benefits of first mover advantage.
  4. Improve the quality of the service that you are providing to authors and institutions. Does it warrant the APCs that you are charging as things stand? This is an area which is coming under increasingly close scrutiny.

There are a number of potential levers for forcing through change if publishers are unable or unwilling to make the necessary changes. Most obviously, funders could set out rules for the use of research funds which favour publication in fully OA journals.

Long-term transformational change

The changes discussed above are the changes that need to take place now. However, we are at the very early stages of the transition to a fully OA world. In many ways, the UK has taken the lead, which puts us in a powerful position to consider how we should re-engineer our scholarly communications ecosystem to take full advantage of its transformative power. Some of the critical questions libraries and publishers will need to consider together are as follows.

  1. For content that we cannot yet make OA, how should we manage questions of access? At the moment, we have created a system in which each institution is its own walled garden. Those inside the walls are able to share knowledge with each other but are very circumscribed in their capacity to share knowledge beyond it. If we still need walls for legacy content, where should we draw them?
  2. What sort of experience would we like those accessing books and journals to have? Libraries do very well at providing any time, anywhere, any device access, but what is the experience of access like? What conditions would need to be present for publishers to stop worrying about inappropriate usage and sharing? How might those conditions be created?
  3. Given new approaches to academic practice, in which collaboration and sharing feature so heavily, how should we rethink the formulation of scholarly outputs? How would publishers approach a world in which the journal article is no longer the primary mechanism for sharing research outputs? And how might libraries help capture and enable discovery of those new types of content?
  4. What options do we have for speeding up the very slow process of getting our metadata in order, and improving systems interoperability?
  5. And finally, we all know that journal impact factors and other measures of impact are often poor proxies for quality, yet we persist in using them for want of better tools. Should we continue the endless search for improved metrics, or move away from this approach altogether?

Conclusion

There are real challenges here for libraries and publishers. For libraries, we must avoid being wedded to traditional ways of doing things, and be willing to think radically. How much of what is done in each institution could in fact be done at an above campus level in a shared service or more informally through networked collaborative working?

And for publishers, how can you embrace, as businesses, digital transformation? Some companies and academic presses are willing to think radically about the future, but there are others whose business strategy seems to be to maintain hefty profit margins for as long as possible without accepting change, and to wring the last possible remaining funds from a system that has ceased to function in anyone’s interest but their own. They might benefit from reflecting on the experience of companies like Nokia and IBM, which found that comfortable profit margins cushioned them against the very necessary need for evolution, until they found that they had been left entirely behind. Perhaps it would be worth considering that, should funders and government lose patience, flipping the legacy journals market into a purely OA access model would not be too hard.