A University and Academic Imperatives for Research

The production and dissemination of research is a core purpose of universities. This necessarily involves the creation of copyright material in the form of journal articles, books, chapters, conference papers and other scholarly outputs. But more than that: universities operate in an environment where they compete for students and research funding They must be able to demonstrate to government and other funders, and to prospective students, that their research is of the highest quality, and hence that they are worthy of continued support and enrolment, respectively. Academic publishing disseminates research and provides a mechanism, however limited, for assurance regarding the quality of research. Journals manage peer review and confer general reputational benefits, especially where the journal is perceived to be prestigious. But they also provide access to quantitative indicators of quality, including the journal impact factor (a quantitative measure of the frequency with which the average article in a journal has been cited in a particular year) and measures of individual article citations. Publishing in journals is necessary to generate such quantitative evidence, because articles in recognised journals automatically feed into the systems managed by the large commercial organisations (Clarivate’s Web of Science and Elsevier’s Scopus) which track citations.

The Australian government regularly assesses and ranks universities’ research through the Excellence in Research Assessment (‘ERA’) exercise.^{Footnote 3} Within this framework, excellence is demonstrated in part by publishing in prestigious journals and with reputable publishers; for some disciplines citations tracked by the commercial providers are also relevant.^{Footnote 4} Other commercial systems such as the QS and Times Higher Education rankings also seek to rank the research excellence of universities, based in part on inchoate reputation and quantitative publication measures.^{Footnote 5} Both government and commercial rankings play an important role in university strategies for attracting students, including lucrative international students.^{Footnote 6}

Because publications in prestigious journals are important to the university, the imperative to attain them is impressed upon academics through hiring, annual performance reviews, and promotion policies and practices;^{Footnote 7} as well as the criteria applied by grant funding bodies that rate academic track records.^{Footnote 8} Publication in recognised high quality outlets is required to get hired in the first place, and to enjoy career success in the form of grants and ongoing promotions. Publication can be a matter of survival in an academic research career or the only way to avoid receiving burdensome teaching allocations. Institutions may have minimum research expectations that demand a certain number or quality of outputs; or a certain number of outputs in outlets of a particular ranking.

Prestigious publications are however not the only goal universities and scholars pursue for their research. Researchers and universities are also evaluated for their ability to produce high-impact research. This general imperative is also backed up by government requirements and periodic government assessment.^{Footnote 9} Universities — and by extension at least some academic researchers — must therefore take steps to ensure research is used beyond a scholarly context and has (one hopes, beneficial) effects in the world, whether economic, environmental, cultural or social. Wide dissemination of research — beyond the limited subscriber base of the high quality journals — can play an important role in achieving impact and uptake of research insights.^{Footnote 10} Increasingly, governments and research funders are demanding that research publications be made freely available, both to increase impact and based on the principle that the public should have access to research supported by public money.^{Footnote 11} University policies also reflect these demands, often requiring academics to make their publications available open access (‘OA’).^{Footnote 12} In Australia to date, funder policies in favour of open dissemination of research have not come with additional resources to cover any costs incurred as a result, at least in the case of government funded research.

There is also global pressure on researchers to consider the broader implications of their research practice. The United Nations Educational, Scientific and Cultural Organization (‘UNESCO’) Recommendation on Open Science (2021) acknowledges the transformative potential of Open Access, Open Science and Open Educational Resources in progressing the ambitions of UN Sustainable Development Goals and in view of:

The UNESCO agenda was quickly embraced by Australia’s Chief Scientist, Dr Cathy Foley.^{Footnote 14}

The various research objectives outlined above can be in tension, at least as commercial publisher, and government and university systems are currently configured. The goal of broad, open dissemination of research for impact and real-world applicability is in tension with the imperative to publish in prestigious journals which have historically made money from subscriptions.

B The Commercial Nature of Academic Publishing and Its Basis in Copyright Ownership

The industry of academic publishing and research metrics analysis is not simply an adjunct to the functioning of the academy. Academic publishing is a major and highly concentrated commercial industry, where four companies: Elsevier, Springer Nature Group, Wiley and Clarivate wield considerable global influence affecting the terms and conditions of publishing contracts, subscription licences and access to data. Feedback loops further concentrate the market power and maximise the brand value of certain journals. The use of research metrics in government research assessment and commercial rankings (journal impact factors, citation indices and journal rankings) compounds the importance of journal prestige for universities and, hence, for academics. These feedback loops limit the ability of academics to make any choice other than to play the prestige game. The resulting academic decisions about where to submit, whom to cite, what to send to review and whether to recommend publication further journal reputation, in turn produces goodwill and trade mark value owned by publishers.

The scientific publishing industry is based on control of copyright in journal articles, obtained from academic authors.^{Footnote 15} Academic authors, however, have little choice about the transfer of copyright. From the perspective of the individual academic author, publishing in a prestigious journal routinely means assigning copyright to the publisher via the publisher’s contract presented to them during submission or following acceptance of an article.^{Footnote 16} Publishers’ contracts set the terms under which research dissemination takes place, and the conditions under which access to the final product is offered, with little opportunity for researchers to affect these terms. Interviews we conducted with researchers, managers and DVC-Rs suggests that researchers feel disempowered and too overwhelmed to negotiate with publishers over access restrictions and fees, and do not ‘spend the effort’ negotiating when ‘the power imbalance is too big’.^{Footnote 17} As one DVC-R explained, ‘we have researchers who are just overwhelmed by, you know, computer says, sign here… [our library staff] keeps telling us to push back against all those things and negotiate stuff individually. But you know, that’s not the way researchers think or act’.^{Footnote 18} For many researchers, the ‘idea that’ they have ‘any ability… to, like, negotiate the terms on which a paper is accepted’ ‘is a bit … fanciful’.^{Footnote 19}

The bargaining disparity has been exacerbated by a recent shift away from emailed contractual documents to automated processes. Upon article acceptance, or even during submission, an academic author today may be sent a link to a portal that contains a standardised contract that must be accepted to progress in the publication process. Faced with the imperative described above to publish journal articles, and little or no opportunity for negotiation over terms, it is not surprising that academics, often the lead author or the research student who handles research administration on behalf of the team, may assent even if they have serious reservations about the contract terms because they perceive they have ‘no choice’ but to sign.^{Footnote 20} In our fieldwork, many researchers acknowledged that they do not routinely read the terms or keep formal records of the terms they have assented to.^{Footnote 21} Their primary focus is on securing the research output:

Other researchers noted a failure to read publisher terms because terms are difficult to understand, or time-consuming to engage with:

Publishers sell access to published journal articles back to universities, including those written by their own academics, and to others willing and able to pay high subscription or individual article fees. Subscription costs are substantial, especially for STEM publications. This is a problem that has been documented for over three decades and the challenges maintaining appropriate access to scholarly resources within institutions have been exacerbated by other pressures on Higher Education budgets more recently. Further, librarians report that the high cost of accessing essential STEM resources currently impacts disproportionately on acquisition of Humanities and Social Science publications.^{Footnote 24}

Academics in our interviews expressed resentment about a system in which they ‘do so much’^{Footnote 25} — undertake research, write it up, peer review it, act as editors for journals — only to have any financial rewards accrue to highly profitable scientific publishers, or be told that libraries cannot afford to pay rising subscription fees for these same journals. The requirement of paying Article Processing Charges (‘APCs’) in such circumstances characterised is also particularly galling:

Our fieldwork suggests that academics would prefer their work to be more widely available: ‘I think scientific work should be as widely disseminated as possible and as widely accessible as possible, and as low cost as possible’.^{Footnote 27} In addition to furthering the sharing of knowledge, openly accessible publications are perceived to have a positive impact on citation counts.^{Footnote 28} For many academics, however, regardless of their discipline, their need for prestigious publications as a prerequisite to career success overrides these concerns and the benefits of wide dissemination.^{Footnote 29} After noting a shift towards OA publications in their field of Education, one researcher nevertheless noted that ‘the traditional high-impact journals are still seen as very prestigious and you try to get into those if you can’.^{Footnote 30} A research administrator confirmed, ‘our engineers will always publish in the top journal that they possibly can within their field, no matter what [research funding or open access] policies are around them’.^{Footnote 31}

In summary, the imperative to publish in journals ranked as Q1 or high impact increases incentives for researchers to enter into onerous transfer or licensing arrangements and to pay APCs for the privilege of having an association with a prestigious journal and publisher brand.

These dynamics are well known and would in themselves justify a more strategic approach to copyright management. But we suggest there are three further contextual factors that are also important, but perhaps less immediately visible to researchers and university managers: (1) the rise of OA and its increasing cost; (2) the link between OA and costs incurred in educating students; and (3) the rising importance of data about research. Combined, these three factors make a compelling case for strategic university or even sector-level management of copyright in research publications.

C The Rise, and Cost, of Open Access

Although far from sufficient,^{Footnote 32} one way to promote the impact and application of research is to make the results of research available to broader non-scholarly audiences. Since audiences not based in universities generally have no reason to pay expensive subscriptions for scholarly publications, this means making work available outside publishers’ paywalls.

While the seeds for OA were planted in the early 1990s, it was not until the early 2000s that the concept of OA as it exists today began to take root in Australian research institutions. Open access to research outputs has primarily been achieved in one of two ways: (a) the gold model where the author publishes with a journal that makes all its articles freely available online, usually because authors pay an APC upfront to cover the publishing costs that would otherwise be paid through library subscriptions; or (b) the green model, where the author publishes in a traditional journal, but then deposits (‘self-archives’) their article (or the author-accepted manuscript (‘AAM’), ie the text prior to copy-editing) in an OA online repository, sometimes after an embargo period.^{Footnote 33}

Governments, universities and researchers have, through past investments, built platforms for green OA. Some research disciplines, such as physics and computer science, have active and well-established mechanisms for OA publication of preprints (eg, arXiv). In law, both AustLII and the Social Science Research Network facilitate sharing of pre-prints and workshop papers. Self-archiving and institutional archiving have also been widely used. Australian universities have institutional repositories designed to store, and provide online access to, research publications, built through investment by the Commonwealth from 2007 onwards under the Australian Scheme for Higher Education Repositories (‘ASHER’) initiative to ensure every university developed an institutional repository for the purposes of government research assessment processes (such as the ERA). In practice, some universities (and some schools and faculties within universities) promote researcher use of the institutional repository more actively than others. In our fieldwork, we encountered institutions where the repository is in practice (if not formally) an optional fallback; others in which deposit is required and checked; some where it has been streamlined into systems academics use to report their publications to their employer; others where active researchers we spoke to had no idea how to use it.^{Footnote 34} These green OA platforms, in other words, are under-used in some institutions. In terms of attracting academics to use and value green OA more, one of their limitations is that while papers are accessible, there is limited cultural cache attached to this. Inclusion does not contribute to an academic’s reputation and goodwill in the same way as having a paper included a proprietary system does, so the value of inclusion is much harder to sell to academics. This could be addressed were Learned Societies and other organisations with recognised disciplinary expertise supported to encourage alternatives to proprietary journals and promote them to researchers.

Open repositories are also increasingly under some threat. Publishers are wary of self-archiving and academic sharing of publications, and many discourage or limit it via contractual restrictions, imposing embargos or requiring authors to seek permission. Previously not-for-profit author sharing portals have been acquired by publishers, licensing agreements negotiated, or legal action threatened.^{Footnote 35} This impacts on researcher behaviour: researchers in our fieldwork also expressed uncertainty over whether and when they are allowed to self-archive, and may be unwilling to risk annoying publishers on whose services they depend for career advancement.

Publishers have also created ‘hybrid’ OA models: folding demands for OA publication into their commercial models by allowing journal articles in a traditional outlet to be made OA on payment of an APC. Article Processing Charges vary enormously: in our interviews university librarians cited figures from $200 up to US$9000 for the most prestigious outlets.^{Footnote 36} Our fieldwork also uncovered instances in some disciplines of publishers levying multiple charges for the same publication in addition to an APC: a fee to include a graph, table or picture (ie, increased if the image is to be published in colour), a fee to be featured on the cover of the journal, and a further fee for the journal’s cover art — and no right to further disseminate any of the above. As one researcher describes,

We found in our fieldwork that in at least some Australian universities, funding to pay APCs is not generally available or available only in some faculties, schools or research centres or to certain researchers; where available, such funds may be limited to a set of high-impact/high quality journals.^{Footnote 38} Even recipients of national competitive grants will often have to trade off payment for OA against other research costs in the context of budgets that rarely meet funding requests or the actual costs of research.^{Footnote 39} In some cases, academics have even personally paid APCs.^{Footnote 40} As these costs mount up, a clear risk is that only the best-funded researchers will have the funds to make work available OA: entrenching the advantage of a limited set of researchers (typically ARC/NHMRC-funded), who are then placed in a better position to achieve, and demonstrate, impact for their work.^{Footnote 41} If the future of OA becomes more reliant on APCs, universities and researchers face an expensive and unequal playing field.

D The Education Costs of Access to Research

A second contextual factor that makes copyright management more urgent is that publication and OA decisions have financial implications beyond universities’ research portfolios. When universities use journal articles as reading materials in teaching, and copy or make them available online in university systems, those uses are the subject of a compulsory (‘statutory’) licence under the Copyright Act 1968 (Cth). Permission from copyright owners is not required, but equitable remuneration must be paid.^{Footnote 42} Collecting societies receive university payments and distribute funds to copyright owners.^{Footnote 43} The licence fee is agreed between the collecting society and the universities, or determined by the Copyright Tribunal (a Federal Court judge sitting extra-judicially, sometimes accompanied by other expert panel members).^{Footnote 44} Past licence rates have been set at a per-EFTSU (equivalent full time student unit) cost for the right to copy and communicate any copyright material.^{Footnote 45} The per-EFTSU rate has not historically been based on an accurate calculation of the material copied and used by universities. Rather, the Tribunal conducts a more holistic assessment, informed by evidence about the amount and nature of copying. The universities and Copyright Agency did not reach agreement on the rates for a new statutory licence for 2019–2024, leaving the Copyright Tribunal to determine the value of the copyright licences.

Importantly for present purposes, OA material can be used without relying on the statutory licence. In the Tribunal, the university sector claimed that the amount paid for the statutory licence should be reduced to reflect, inter alia, rising OA publication rates.^{Footnote 46} In other words, if researchers make their articles OA, over time this should impact the cost of the statutory licence. Researchers however have no reason to take this into account when deciding when or how to publish, or whether to pay APCs. In formulating his decision on the statutory licence, Perram J wanted more evidence that OA policies were being adhered to by academics in practice, and that OA versions were being used in education.^{Footnote 47} A more holistic and strategic approach to copyright management in research has the potential to reduce costs in education. It is difficult however for universities to ‘join the dots’ to identify, let alone calculate or take into account the impact of research publication decisions on education costs, as these parts of the university are separately managed and led, often relying on information management systems that do not talk to each other.

E The Rising Importance of Research Publication Data

A final factor that should make management of copyright ownership more urgent is the increasing importance of data and analytics services about research performance. Multiple stakeholders have an interest in and use for data about research. Such data includes backward and forward academic and non-academic citations;^{Footnote 48} the size, nature and distribution of readership, including over time; and basic article metadata such as funding sources, academic authors, institutional affiliation and whether the publication is OA. Governments seek to evaluate the quality and impact of research in the university sector;^{Footnote 49} universities to ‘prove’ the worth of their research — and internally to assess the performance of research centres, faculties and schools, and individual researchers, or identify high performing units or individuals for further support. Analysis of patterns of citations and use of research can itself provide research leads or guide the work of researchers. Publishers have for some time been offering data services (especially in citations analysis), and forecasting these potential data and analytics services markets to be a significant revenue stream; the need for data explains a range of vertical acquisitions by the large publishers.^{Footnote 50} Potential services these firms can offer go deeper than many researchers would necessarily appreciate: for example, valuing the commercial worth of individual author employees and research teams and supplying this information to government and external parties.^{Footnote 51} The more researchers and universities hand over management of all storage and dissemination of research publications to publishers, the more dependent they will be on those same publishers for data analytics services, and the less alternative or independent sources of information about research dissemination there will be.

F Resolving the Resulting Tensions and Complicating Factors: Research Publishing’s Inflection Point

The Australian university sector is not currently addressing the tensions between competing objectives for research, how they play into decisions on where and how to publish research, and what it all costs. At present, individual researchers must seek to understand publishing contracts presented to them, manage copyright ‘negotiations’ and often find the money for OA from grant funds or, occasionally, central or school funds. The results are predictable. Individual researchers are mostly distant from the costs to teaching or global access to research to which their publication decisions give rise and have the least individual power to effect change even should they try. For most, their most immediate imperative is achieving prestigious publications, and even if they wish publications to be OA for long term impact, they may not have the motivation, opportunity or money, to press for it.^{Footnote 52} Putting the burden on individual researchers to reconcile inconsistent demands from universities and funders for both prestige publication and wide dissemination is unsustainable. Simply providing dedicated funds to enable all researchers to pay APCs would be a very expensive and likely unfair solution, especially if no cap is imposed — and would divert already-scarce research resources away from directly funding research activities.^{Footnote 53}

A number of solutions to remove this burden from researchers have been investigated and adopted overseas. The best known of these are the UK Scholarly Communications Licence (‘UK-SCL’) and Plan S.^{Footnote 54} At institutions which incorporate the licence into their OA policy, the UK-SCL states that researchers retain re-use rights in their own work and asserts that the institution retains a non-exclusive licence to make their author-approved manuscript^{Footnote 55} available without embargo through the university’s OA repository under the terms of a Creative Commons licence (CC BY-NC, allowing free non-commercial reproduction and communication). Where the UK-SCL has been incorporated and publications are uploaded to the institutional repository, UK researchers automatically comply with funder OA policies and ensure the output is eligible for the Research Excellence Framework (‘REF’) assessment.^{Footnote 56} Plan S is an initiative of cOAlition S, a group of research funding agencies including the European Commission and the European Research Council.^{Footnote 57} Plan S requires that all scholarly publications on the results from research funded by Plan S members must be published in OA Journals, on OA Platforms, or made immediately available through OA Repositories without embargo. Article Processing Charges are permitted only if commensurate with publication services delivered and if the fee structure is transparent. Both the UK-SCL and Plan S seek to support a transition to move publicly funded research from behind paywalls, free for all to access without restriction and to impact researcher practice beyond researchers at institutions that have mandated adoption of the policy.

A publisher solution has also entered the market. New generation agreements known as ‘Read and Publish’ agreements (sometimes called ‘Transformational’ Agreements) offer universities a journal subscription which, alongside reading access for students and academics, also enables affiliated academics’ journal articles to be automatically OA, without payment of any APC. The Council of Australian University Libraries (‘CAUL’) has negotiated a number of these agreements, with pilots occurring during 2021–2022: Australian researchers are receiving (doubtless welcome) notification that their articles are automatically OA.^{Footnote 58} Pilot deals negotiated by CAUL included the ‘publish’ element at no additional cost to previous ‘read only’ subscriptions. A variation of the publisher solution was also proposed by Australian Chief Scientist Cathy Foley.^{Footnote 59} Foley proposed a national-level deal involving centralised payment to publishers to make Australian research OA. Detailed analysis of either solution is beyond the scope of this article. It is sufficient, for now, to note that while Read and Publish agreements take a burden off researchers, they can also impact academic freedom to choose where to publish: grant-funded researchers may have to choose journals where a Read and Publish agreement is in place. Further, neither solution addresses universities’ increased dependence on publishers for dissemination and data about use. Perhaps most importantly, the longer-term economic cost and diversion of resources from paying for research activities to paying publishers is at best unclear. In our fieldwork, librarians who routinely negotiate with publishers expressed scepticism: ‘It won’t reduce the cost of access to publications. Many of our bigger publishers are commercial enterprises and you know, their role is to maximize profits for them’.^{Footnote 60}

These Australian developments should also be understood in the context of institutional initiatives elsewhere in the world: in Europe, for example, where Read and Publish Agreements have been available for longer, but where there are moves to reduce reliance on them, driven by concern about paying publishers, and paying them far too much, as well as to facilitate access to publicly funded research.^{Footnote 61} Across the EU, free access to publicly funded research publications without embargo is increasingly the norm, whether achieved by paying for OA or through deposit of the AAM in an institutional repository.^{Footnote 62} Many EU nations have also enacted a secondary publication right to ensure that the author retains rights to the AAM and can deposit it in a repository, regardless of publisher contracts.^{Footnote 63} There has also been some reconsideration of dependence on publisher metrics for research assessment, given the financial burden of OA publishing.^{Footnote 64}

In recent times too, university-level rights retention policies have been rolled out across many institutions across the globe.^{Footnote 65} In the UK, these policies allow researchers to comply with funder requirements of OA without embargo and ensure publications are eligible for inclusion in the REF research assessment exercise. Since 2021, United Kingdom Research and Innovation (‘UKRI’) has made OA publication of journal articles mandatory: either the Version of Record (final published article) must be made available, for free and unrestricted through a publisher journal or platform, or the AAM must be available in an institutional or subject repository. Books, chapters and edited collections (including images, tables and other supporting content) must similarly be made available for free and unrestricted from 1 January 2024, with a maximum 12-month embargo allowed.^{Footnote 66}

Against this background, it is important to canvas alternatives, and/or counterweights to publisher commercialisation of the drive for OA, research impact and research measurement in Australia. We set out such an alternative below: one that depends on universities asserting a right both to make research outputs of their academic employees available OA, and to use those outputs for teaching and educational purposes, and perhaps even to cross license other Australian universities to do the same. Such an alternative must find its foundation in a clear understanding of, and strategic management of, copyright in research publications by Australian universities. Publishers’ current dominance is built on the assertion of full, exclusive ownership of copyright in publications. As we will argue further below, better solutions may lie in taking advantage of the capacity of copyright to be divided in multiple ways, which depends in turn on clarifying, and leveraging, university ownership of publications.

A Preliminary Matters

The right of Australian universities to own and use the copyright generated by academic employees has long been confused.^{Footnote 67} Determining who owns copyright in an article^{Footnote 68} depends on general legal doctrines, copyright law and employment law, as impacted by many institutional policies, EBAs and the customary practices of the particular institution. Before we analyse how these sources of law intersect, certain basic legal principles should be noted.

First, copyright is a property right which arises immediately and automatically as soon as a literary work is created, and without any formalities.^{Footnote 69} Copyright law allows the assignment of copyright and of future copyright.^{Footnote 70} Thus, if an academic’s employment contract prospectively assigns copyright in scholarly articles, then as soon as any article is written, that assignment will take effect. Second, copyright, being a property right, is subject to the fundamental legal principle nemo dat quod non habet: ‘you can’t give away what you haven’t got’. This common law rule provides that a person cannot transfer better title than they themselves possess.^{Footnote 71} If an academic author of a scholarly work is not the copyright owner, they cannot transfer copyright title to a publisher,^{Footnote 72} and the publisher contract will not be the definitive statement of the rights and interests in the work.^{Footnote 73} Third, copyright licences bind successors in title.^{Footnote 74} Thus, if at the time that an author purports to assign copyright to a publisher, that author owned copyright subject to a licence in favour of the university, then that licence to the university remains binding on the publisher.^{Footnote 75}

A fourth fundamental principle is copyright’s divisibility. Ownership of copyright can be ‘sliced up’ in any number of creative ways; the owner can license or assign different rights order to maximise their benefits.^{Footnote 76} Thus, the author of a popular novel can divide up volume rights which may be split up as paperback and hardback rights; ebook rights; serial rights (to have extracts serialised in newspapers or magazines); translation rights; film rights; podcast rights; streaming rights; turning-the-book-into-a-musical rights and so on.^{Footnote 77} It is therefore legally feasible to assign partial copyright to enable publishers to exercise some subset of rights, while retaining other rights and thus protecting the interests of researchers, the university and wider public in having access to research outputs.

These basic principles allow copyright owners to deal with their rights in a number of ways. The next step in the legal analysis is to understand how copyright ownership initially arises.

B Copyright Ownership Under the Copyright Act

Authors are the first owners of copyright in their works,^{Footnote 78} unless made in pursuance of the terms of their employment, in which case the employer is the first owner of copyright.^{Footnote 79} Both these rules are default rules: they can be varied by agreement.^{Footnote 80} If there is more than one author of a work, each author holds an undivided share of the copyright as tenants in common; each author has the exclusive rights of copyright, meaning they can individually transfer their share to another person or entity, and the agreement of all authors is necessary to effect the transfer of the full copyright or to grant an effective licence.^{Footnote 81} To understand first ownership, then, it is necessary to understand both (a) whether research outputs are made in pursuance of an academic’s employment and (b) whether the default position is varied by agreement.

C Other University Policies That Complicate, or Intersect With, Copyright Ownership in Research Publications

So far we have discussed core documents determining copyright ownership: the Copyright Act; employment contracts and enterprise bargains, and university IP policies. But other intersecting university policies, including those pertaining to commercialisation, outside work rules, OA mandates and research metrics are also relevant to questions of ownership. University policies are typically developed in a piecemeal fashion, drafted by different people in different university portfolios, each with their own goals and agendas, and implemented within universities across departments and at different levels. The cross-institutional analysis revealed that these policies diverge or conflict, directly or indirectly.

D Summary of the Legal Analysis

This Part has revealed the shaky foundations of any assertion of ownership of copyright in university-affiliated research outputs, by any party in the research pipeline — whether that claim is made by the university, academic author or the publisher who receives an assignment from them.^{Footnote 154} Universities’ claim to ownership is theoretically strongly grounded in their status as employer (as recognised under Copyright Act s 35(6)) and by IP policy clauses, common across the sector that retain limited ownership or perpetual, irrevocable licences and which are incorporated into employment contracts. But that claim is weakened by the rise in casual employment; moves to remove research as part of the universal workload of academics; and unclear mechanisms to make IP policies binding on anyone other than contracted employees with research as part of their recognised workload. It is also weakened by universities’ failure to act on their claim. By acting as if all ownership and all responsibility for academic publications lie with academic authors, universities may make it true, via traditional legal principles such as estoppel. The picture is further confused by a morass of inconsistent university and funder policies which have turned academic authorship into a measure of credit, a proxy for research performance and a sine qua non of the academic career: further increasing the practical pressure to expand the list of authors, potentially in ways inconsistent with copyright law.

The copyright ownership claims of academic authors are weakened by IP policies that reserve certain rights to universities, and their status (where applicable) as employees. This in turn weakens the claims of publishers who purport to obtain ownership via assignments from researchers. Moreover, given universities’ default ownership under the Copyright Act, publishers’ only basis for copyright ownership lies in University IP policy clauses which assign rights back to academic authors. These are precisely the same clauses which confer perpetual, irrevocable rights on the university: publishers surely cannot take the benefit of the assignment without also accepting the university’s rights. Co-authorship rules also potentially weaken publishers’ claims in circumstances where publishers have received a written contractual assignment from only an executive or corresponding author.^{Footnote 155}

In short, the current situation is a mess. The parties benefiting from this mess are publishers, who have stepped in to assert, and to build automated and data systems that assert default ownership of the whole copyright, however shaky the foundation of that claim, largely without challenge from the university sector to date. So long as these assertions by publishers remain unchallenged by universities, universities could have difficulty seeking to enforce their rights. However, as default owners of copyright under the Act, universities are best placed to remedy, or at least mitigate, the current lack of clarity around copyright ownership: and use copyright ownership to achieve universities’ public goals. The final question is: how should this be done?