University IP policies and research commercialisation in Australia

Similar to many other countries, the Bayh-Dole Act of 1980 in the USA inspired and stimulated some similar policies in Australia in the late 1980s [16]. All Australian universities probably had an IP policy by the mid-1990s. This is because in the early 1990s, there were emerging claims by university administration to IP of staff and students [17] and there were calls for clarification and minimum standards of protection for academic IP rights [18]. Moreover, the Australian Research Council (ARC) in the ARC/HEC Advice on Intellectual Property in 1995 recommended that all universities should have an IP policy in place by November 1996 and all universities adopted the recommendation [19]. Having an IP policy that covered certain issues was also a requirement of receiving national research funding and all universities certainly had an IP policy by 2001 [20].

Having an IP policy is not sufficient for successful research commercialisation. While the pressure on Australian universities to exploit IP due to decreasing public funding started at least two decades ago when all universities had IP policy [21], Australia still lags behind in research commercialisation [22–24]. In the last five years, Australia’s rank has been declining in the Global Innovation Index and the number of its innovation outputs is small relative to its level of innovation investment [25]. This is while historically Australia’s expenditure on university R&D is proportionally more than most other advanced economies, particularly when compared with business expenditure on R&D [26]. Some experts suggest that the solution might be to remove barriers and increase incentives for academics to innovate and commercialise [27]. Barriers (e.g., complicated commercialisation procedures) and incentives (e.g., creator’s share of revenue) might be rooted in university policies and procedures. Two old surveys of academics and technology transfer officers in Australia showed that about a third of both groups considered weakness in university IP policy as a major barrier or a barrier to the success of research commercialisation [28,29]. Therefore, effective IP policy is paramount for Australian universities in the context of the existing criticism about Australia’s failure in research commercialisation and innovation [23,24] and the government’s initiative to improve the situation. Examples of initiatives are the Medical Research Commercialisation initiative that provides $450 million over ten years from 2023 to support innovative early-stage health and medical research in Australia. It will help researchers transform ideas into life-saving medicines, devices and treatments to help future patients [30]. Another example is a $2.2 billion investment through the University Research Commercialisation Action Plan which is to strengthen the role of university innovation and industry collaboration in Australia’s economic recovery [31].

However, the only proper review of Australian universities’ IP policies was done a quarter of a century ago by Monotti [32]. She examined the IP policies of 30 Australian universities in terms of revenue distribution and IP ownership and found that universities had different formulas for sharing commercialisation profit, with a common approach that usually divided net profits among the inventor, the faculty, and the university. Originators’ shares in policies ranged from 30% to as high as 60%. More recently, Ross [22] as part of her doctoral thesis examined policies and procedures to find out who was the responsible decision maker for certain commercialisation activities such as identification of IP, investigation of IP viability, and finding industry partners.

Research aim

While developing an effective IP policy is not enough, it is necessary for successful research commercialisation. Effective IP management not only safeguards the intellectual assets of academic institutions but also plays a crucial role in facilitating the transfer of knowledge and technology from academia to industry. In Australia, there is a need to dissect and understand the nuances of universities’ IP policies. Consequently, this paper aims to analyse Australian universities’ IP policies to find out about some of their key features, similarities and variations. The study specifically looks at what intent Australian public universities pursue in their IP policies, how they define IP, how they approach the ownership of IP created by various groups (e.g., staff and students) and finally how they distribute any net revenue generated through research commercialisation. The study aims to provide insights that could inform future policy enhancements, fostering a more effective and equitable ecosystem for research commercialisation in Australia.

Policy identification

To find IP policies, the Australian Government’s official list of universities [35] was consulted to identify thirty-seven public universities in Australia. The list of these 37 universities can be found in Table 3 in the findings section. Five private universities—Bond University, Carnegie Mellon University (Australian branch), Torrens University, University of Divinity, and University of Notre Dame—were excluded from the study due to their different nature and administration.

Multiple strategies were employed to identify and retrieve IP policies from the selected universities:

• Policy Libraries: For universities with accessible policy libraries, the policy libraries were explored, utilising their search and browse functions to pinpoint relevant IP policies.
• Web-Based Searches: search was conducted on Google, employing primary keywords such as "intellectual property." Furthermore, complementary terms such as "research commercialisation" were examined to uncover potential policies related to these domains. This dual-pronged approach aimed to ensure all relevant policies could be identified.
• World Intellectual Property Organization (WIPO) Database: the "Database of Intellectual Property Policies from Universities and Research Institutions" provided by the World Intellectual Property Organization (WIPO) was also consulted. This resource offered links to the IP policies of various universities.

Policy collection and snapshot

All of the thirty-seven Australian public universities studied were found to have IP policies or similar documents. All of the identified documents contained ’intellectual property’ in their titles. However, within this dataset, 33 were explicitly named as policies. Two of these were named “policy and procedure”. One document bore the title ’Intellectual Property’ without further specification of document type, but the document’s content confirmed its status as a policy. Additionally, three documents were designated as ’regulations.’ One university (Federation Uni) did not have a policy but only a procedure, however, the content of the procedure was similar to the policies of other universities and therefore, it was included in the analysis. To ensure a consistent dataset for analysis, all identified policies were collected and saved in November 2023. This particular time point was selected to provide a static snapshot for analysis, as policies may undergo revisions and updates over time.

Analytical framework

For the systematic analysis of these policies, a hybrid approach encompassing both checklist-based and content analysis was employed. The following key information was extracted from each policy document:

• Date of current policy version
• Date of next scheduled policy review
• Responsible office/policy owner
• Applicability recipients of the policy
• IP definition
• Purpose of policy
• IP ownership guidelines
• Revenue distribution

These categories were developed inductively following a review of a sample of IP policies, ensuring that the analytical framework was aligned with the nuances and intricacies of the policies under examination. This methodological approach was designed to facilitate a robust comparative assessment of IP policies.

Findings

Although most of the 37 policies were generally up-to-date with median being two and mean being three years old, some policies had not been updated for a significant time with the oldest being nine years old and the newest being updated in 2023. Twenty-six of the policies had a date for the next review and in the case of 10 policies, those dates were in the past, the longest being six years overdue, indicating the slow/delayed process of policy revision. The date of first version of the policy was available for 27 policies and the oldest was in 1983 (UNSW). For two policies this date was in 1996–7, for nine policies the first version was in the 2000s and for the remaining 14 policies the first version was in the 2010s.

In terms of the ownership of policies, this information was available for 35 of them. In the case of most universities, the policy was owned by the vice chancellor leadership team, mostly the research portfolio (DVCR or PVCR). Table 1 shows the terms used in the policies for ‘policy owner’.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Policy owner.

https://doi.org/10.1371/journal.pone.0304647.t001

Groups to whom IP policies apply

One important aspect of policies is their scope of applicability. Most policies had a direct statement about this. Four policies did not explicitly specify, however, it was implicitly clear in the documents. Students and staff/employees were the most common groups mentioned (35). See Table 2. Two policies that did not use these terms used broad (perhaps vague) terms of ’all university members’ and ’entire university community’. Policies mentioned other groups such as affiliates, visitors, honorary appointments, titleholders, and contractors. Some defined these terms specifically and some included such groups in their definition of staff.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Groups mentioned in the policies as people to whom the policy applies.

https://doi.org/10.1371/journal.pone.0304647.t002

Definition of IP

Policy documents usually have a specific section for the definitions of key terms used in the policy to avoid any ambiguity or misinterpretation. Given the core concept in an IP policy is the IP itself, one would expect it to be defined in any such policy. Surprisingly two policies did not explicitly define IP (University of Newcastle and Charles Darwin University), and five universities defined ‘IP rights’ and not IP itself. The analysis of the definitions provided showed a range of similarities and a few notable differences.

Broad and Inclusive Definitions: Many policies adopted a broad (e.g., CSU) and inclusive definition of IP. They typically encompassed both registered and unregistered rights. Registered rights like "patents" and "trademarks" were frequently mentioned, with patents appearing in 29 policies and trademarks in 32. These definitions often aligned with international standards, such as those outlined by the World Intellectual Property Organization (WIPO). This indicates a common understanding that IP encompasses a range of protection mechanisms, from formal registrations to more intrinsic copyright protections.

Detailed Enumerations: Some policies provided detailed enumerations of IP types (e.g., Deakin), explicitly listing various forms of IP types, such as "inventions" (28 mentions), "designs" (32 mentions), "trade secrets" (23 mentions), and "know-how" (17 mentions), indicating recognition of various forms of intellectual contributions. Categories like "discoveries" (6) and "creations" (3) are less frequently mentioned but their presence in a few policies showed the full spectrum of intellectual endeavours that university policies may protect.

Distinctions and Clarifications: A few universities made distinctions between IP and related concepts such as moral rights, providing explanations of what is considered IP and what is not. In fact, the name of the policy for one of the universities (CQU) was the Intellectual Property and Moral Rights Policy. These policies clarified the scope of IP and its relationship to other rights, emphasising the protection of moral rights as closely aligned but distinct from IP. "Rights" was one of the most frequently occurring terms with 33 mentions, highlighting that universities focus on the legal rights associated with IP. The term "legal" however, showed up only in two policies, suggesting that two universities explicitly reference the legal frameworks that underpinned IP definitions and rights.

Research Data as IP: There was an acknowledgment that IP may exist in research data (e.g., Monash), especially where intellectual effort has been applied to organise or describe the data or where the data is confidential.

Purpose of policy

University policies usually have sections that outline their purpose, intent or objectives. For this part of the analysis, text related to either the purpose, intent or objectives of policies was used. An analysis of statements in the 37 policies revealed a multifaceted approach to the management and commercialisation of IP arising from research. The overarching themes identified through these statements highlight the varied priorities and strategies of these institutions.

Ownership of IP created by staff

The policies’ content regarding the ownership of intellectual property (IP) created by staff showed all universities asserted ownership of IP generated by staff during the course of employment as part of their duties. Universities also claimed ownership to IP created by staff in a number of other ways. IP generated using university funding, facilities, resources, or IP derived from sponsored or contract research, or university commissioned projects, was also claimed by the university. Moreover, universities own any IP that is created using existing university-owned IP. While the policies generally supported university ownership over staff outputs, they had scope for exceptions which mostly included literary, scholarly (e.g., book, journal articles) and creative works. Additionally, there may be scope within the agreements for special arrangements to be made with staff members.

The overarching theme was that universities seek to retain ownership of IP to facilitate research sponsorship agreements, collaborations, and their educational mission. However, they also recognise the rights of creators through exceptions and the potential for individualised agreements, balancing institutional interests with those of the staff members.

Ownership of IP created by students

When it comes to the ownership of IP created by students, there were more variations in policies. Thirty-two policies included a blanket statement indicating that the university in principle does not assert ownership of IP created by students and then go on to list conditions under which the university would assert ownership of student-created IP. Five policies did not have blanket statements, instead only listed situations under which they would assert ownership (e.g., a student is also a staff and created IP during their employment).

While only one policy (ACU) explicitly named undergraduate students to say that the university laid no claim on IP created by them, others incorporated conditions applicable to postgraduates and research students. For example, postgraduate students might need to consent to non-exclusive IP rights for the university (e.g. Swinburn). In cases where a student is also employed by the university, IP ownership typically aligns with staff ownership policies, indicating the role of students (as students or employees) in determining ownership rights. Moreover, policies addressed IP ownership in student-involved projects, particularly externally funded or university research activities which often necessitate IP assignment or licensing. Furthermore, universities may assert ownership over student-created IP under specific agreements, joint creations with staff, third-party agreements, or when existing university-owned IP is used. In the case of the commercialisation of student-generated IP, the student is entitled to a share of revenue. Additionally, students may be required to notify university authorities of valuable IP to mitigate legal, commercial, or reputational risks. Students generally retain the right over their thesis (unless subject to a third-party agreement).

Ownership of IP created by visitors

Visiting academics are typically hired by one institution but spend time at a second university in a research role. Since they may not be employed by the host university, ownership of IP from their work can be complex. Policies were checked whether they specify ownership in such cases. In 10 policies, visitors were not mentioned at all. The analysis of the information on visitors in the other 27 policies showed some variation in how institutions address the intellectual property rights of visiting creators. The universities that included visitors in their policy can be divided into four groups in terms of their approach to visitors’ IP rights.

1. Six universities (e.g., DU) stated that the ownership of IP created by visitors is determined by the agreement they sign before their visitation, and five (e.g., CDU) said the IP rights are determined through case-by-case negotiation.
2. Four universities (e.g., ANU) took a presumptive exclusion (non-ownership) approach by stating that the university does not assert ownership of IP created by visitors, and then they go on to add conditions under which they would assert ownership of IP (e.g., when visitors use the university’s funding or facility and resources).
3. Eleven universities (e.g., Melbourne) took a presumptive inclusion (ownership) approach and stated that the university owns the IP created by visitors. Some (4) of these universities (e.g., SCU) considered visitors as a member of staff and therefore, the same rule applicable to staff (default university ownership) would apply to them. This ownership usually occurs under certain conditions such as the use of university resources, terms of employment and so on. Given the similarity of the conditions in non-ownership (category B above) and this ownership approach, in practice, the outcome of the two approaches might not be very different.
4. One university (VU) stated the IP of the work commissioned by the university (including computer programs, learning and teaching material etc will be owned by the university and visitors will own the copyright of any other work they might do.

Ownership of IP created by affiliates

Another category of researchers who might engage in IP creation is affiliates. Policies were searched for the words affiliates, adjuncts, honorary appointment/appointees, titleholders, and associates to find out how policies approached the ownership of IP created by this creator group. The approach to affiliates mirrored that of visitors. Seven policies (e.g., CDU) made no mention of affiliates, six (e.g., CSU) included them within the definition of staff, and five (e.g., WSY) grouped them with visitors. Ten policies took a presumptive ownership approach and stated that they assert ownership of IP created by this group, but of course, they had conditions under which such ownership would happen. Two policies took the opposite approach, i.e., presumptive exclusion (non-ownership) stating that they do not assert ownership on IP created by affiliates except in certain conditions such as the use of the university’s resources, or the use of university’s IP, teamwork with the university staff and so on. Four other university policies stated that the terms in the agreements signed by affiliates will govern the ownership of their IP. Finally, one university stated that they would negotiate with affiliates about IP cases, and one policy was not clear in this regard.

Ownership of Indigenous cultural and intellectual property

Indigenous matters are increasingly important in the Australian context and when it comes to Indigenous Cultural and Intellectual Property (ICIP), it is expected that universities have clear guidelines that respect the rights of First Nations people. Eleven policies had no information about ICIP. One of them (Griffith) mentioned Indigenous people in the context of Benefit Sharing Arrangements. The rest of the policies had a section or a statement concerning Indigenous people, some brief and some in more detail. Analysis of these statements revealed a consistent and respectful approach to managing Indigenous Knowledge and IP. The key trends and principles in these statements were:

Recognition and respect: Universities (16) explicitly acknowledged the significance and importance of Indigenous Knowledge, with a commitment to respect and protect it. This was often reflected in statements that emphasised the dynamic and living nature of Indigenous Knowledge.

Ownership and consent: There was a clear stance (13) against asserting ownership over IP relating to works predominantly created by Aboriginal or Torres Strait Islander people. Policies stipulated that the use of Indigenous Knowledge must have the prior approval of the appropriate knowledge holders and that informed consent is crucial.

Compliance with standards and ethics: Universities (12) committed to complying with legislation, national and international standards, and protocols, including ethical guidelines concerning the use of Indigenous cultural heritage and property.

Benefit sharing: The policies (10) often highlighted the need for equitable sharing of benefits arising from the commercial use of Indigenous Knowledge. They called for pre-agreed terms to ensure that any commercialisation activities are fair to the Indigenous Knowledge holders.

Consultation and negotiation: In some policies (14), before undertaking commercial development involving Indigenous IP, there was a process of consultation and negotiation with Indigenous Knowledge Holders to agree on benefit-sharing arrangements.

Culturally appropriate dispute resolution: In the event of disputes regarding ICIP, one policy suggested that they should be managed respectfully and in culturally appropriate ways, often guided by specific offices or bodies within the university.

Creator’s share of net revenue

One important aspect of research commercialisation is how the net revenue/proceeds will be distributed between the creators and the university. This was specified in the policies of 24 universities. Nine universities did not have such information in their policy, but they specified this in a ‘procedure’ document. In the case of four universities, no information was found. These universities may have such information in a document that is not publicly available, but this could also be a sign of a lack of transparency or an absence of a standardised policy in those cases. Upon reviewing this information (see Table 3), some trends can be observed.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Distribution of the net revenue generated by commercialisation of research.

https://doi.org/10.1371/journal.pone.0304647.t003

Notably, the distribution is not uniform; a few universities had a tiered system that adjusts the percentage based on the amount of revenue generated, while others had a flat rate regardless of the revenue amount. If we ignore the arrangement of a few universities for net revenues below $100k, the creators’ share ranges from a third to a half, with an average of approximately 42%. This suggests that a reasonable portion of the net revenue is commonly awarded to creators, acknowledging their contribution and incentivising further innovation. Universities had different approaches to the percentage of share that was not allocated to creators, however, generally, they used the rest of the net revenue to support the university’s infrastructure, including the academic units and central administration that facilitate research. Most policies (22) defined how this share is divided between different parts of the university and some others did not specify this and broadly indicated that it would go to the university. Those that specified usually gave some part of it to the academic unit to which the creator belongs (in some cases this is the school or department and in some other cases it is the faculty). Another part usually goes to the research portfolio perhaps to reinvest some of the revenue back into their research systems.

While the analysis suggests that there is no standard model for revenue sharing in Australian universities, the share of creators doesn’t vary greatly at different universities (varies between 1/3 and ½). However, different arrangements of the universities in how they divide the revenue and how they invest some of it back at the universities show a diverse landscape of policies that cater to different priorities and institutional values. This diversity could be due to various factors such as the size of the university, the focus of their research programs, and their overall strategic goals related to innovation and commercialisation. Further investigation into the impact of these policies on research output and commercialisation success would be valuable for a comprehensive understanding of their effectiveness.