Research and analysis

From public research spend to innovation: the role of registered IP

Published 16 August 2023

Executive Summary

1. Intellectual property (IP) protection is a policy instrument that aims to encourage innovation and creativity. By granting legal protection for inventions and creations, this incentivises individuals and organisations to invest time and resources into developing new ideas and products. In doing so, this helps to ensure that inventors and creators can reap the economic rewards of their innovations. IP rights also provide a framework for licensing and collaboration between different parties, thereby promoting knowledge dissemination and sharing. Overall, the aim of the IP system is to strike a balance between encouraging innovation and creativity, whilst promoting knowledge dissemination and sharing.
2. This paper examines the role registered IP protection, as a policy instrument, plays in the innovation process. This process starts with research and concludes when the invention or creation is put to use or commercialised, resulting in an innovation.^{[footnote 1]} Whilst other channels to innovation exist, commercialisation is the focus of this report, due to the availability of suitable data.
3. The role of IP in the innovation process is explored using UK Research and Innovation (UKRI) self-reported data on research council grants and their outcomes, including registered IP (patents and trade marks)^{[footnote 2]}and a range of commercialisation activities^{[footnote 3]}. Research projects in receipt of grant funding from the seven research councils^{[footnote 4]} between 2010 and 2020 are analysed.
4. IP and commercialisation outcomes in UKRI data are self-reported by grant recipients. Patents and registered trade marks are cross-validated against various IP data sources^{[footnote 5]}. Grant holders that do not complete the online reporting system risk withdrawal of future funding, however it is not possible to assess the completeness of outcomes reported, and under-reporting is possible.
5. The analysis does not extend to research funded by Innovate UK, Research England, Scottish Funding Council, Invest Northern Ireland or Higher Education Funding Council for Wales, due to differences in data collection by these funding bodies.
6. Because IP registration is a flawed indicator of innovation success, it cannot be used to measure or compare performance of research projects. Inventions that become innovations by being used are not always protected by registered IP. Registration of a patent or trade mark is not always appropriate or even possible, as evidenced by previous IPO research^{[footnote 6]}. Some inventors prioritise publishing their invention in academic journals, making it ineligible for patenting. Some avoid registering IP due to perceived difficulty enforcing their rights against infringers, or low risk of infringement. Some may protect their research outputs through other IPRs (copyright, design rights or trade secrets), which are not analysed in this paper due to data limitations. Others may prioritise other research outcomes; such as upskilling (particularly studentships and training grants). Furthermore, analysis in this paper shows that even if a patent or trade mark is filed for an invention, this does not necessarily mean the invention will be used.
7. Research grants awarded by the seven UK research councils^{[footnote 7]}) to projects that started between 2010-2020 are analysed, using UKRI survey data.
   • 70,152 research grants were analysed, awarded to projects that started between 2010-2020, but did not necessarily complete within this date range. Studentships, training grants, fellowships and intramurals are included alongside traditional research grants
   • of the 37,852 research grants awarded to projects that had completed by the end of 2020, 1 in 38 chose to protect their resulting IP with a published patent application, granted patent or trade mark registration. As highlighted above, IP registration is not a good measure of innovation success. As a result, this ratio should not be interpreted in terms of innovation performance as no benchmark exists to frame this finding in a comparative context
   • of the patent and registered trade mark outcomes reported by grant recipients, most were published patent applications (over 80%), followed by granted patents (over 10%) and registered trade marks (4%). As published patent applications progress, the proportion of granted patents may increase in future. More trade marks may also be registered by grant recipients with projects currently at early-stage, if they progress to commercial trading in future
   • the median research council grant sizes associated with granted patents, published patent applications and registered trade marks are £1.1 million (m), £0.9m and £0.7m respectively. This does not consider further funding from other sources, such as private investment or university funds, or differences in expense across research specialisms associated with these types of IP protection. The median grant size among funded projects that did not result in a patent or registered trade mark outcome was £0.2m
   • scientific research projects in the fields of medicine, engineering, and biosciences (funded by MRC, EPSRC and BBSRC) were more likely to produce outputs suited to patent protection. Other types of IP protection not explored in this paper, such as copyright to protect software, literature, and other creative works, may be better suited to project outputs in the humanities and social science fields (funded by AHRC and ESRC)
   • from the data provided, projects led by the four universities receiving the highest research council grant funding and largest average grants also led to the most patent and registered trade mark outcomes. These universities all have mature technology transfer offices (TTOs) that are at least 20 years old and have large net current assets. These universities locate in the South East, London and East of England^{[footnote 8]}, impacting the regional distribution of registered IP outcomes from research council grants
8. IP rights do not imply innovation has taken place, which requires the underlying inventions or creations to be used or commercialised. This paper focuses on commercialisation outcomes self-reported by the grant recipients. Direct commercialisation by researchers is measured through the creation of spinouts, product interventions and clinical trials in the medical field, while indirect commercialisation activity is measured through IP licensing and collaboration with the private sector. It is found that registered IP in the form of patents or trade marks arising from research council grant funding is frequently commercialised (taken to market):
   • 27% of UKRI research projects that reported a registered patent or trade mark outcome reported creating a spinout company to take forward these IP assets
   • 40% of registered patents and trade marks associated with research council grants over the period were reported as licensed^{[footnote 9]}. Data is not available on licensing across the whole population of patents for comparison as rights owners are not required to provide licensing information to the IPO.
   • almost half (49%) of UKRI research projects that reported a registered patent or trade mark outcome were associated with private sector collaboration. Projects that reported a registered trade mark were most likely to collaborate with the private sector
   • 17% of UKRI research projects leading to clinical trials and medical product interventions were associated with a registered patent or trade mark outcome

Introduction

Intellectual property (IP) protection is a policy instrument that aims to encourage innovation and creativity. By granting legal protection for inventions and creations, this incentivises individuals and organisations to invest time and resources into developing new ideas and products. In doing so, this helps to ensure that inventors and creators can reap the economic rewards of their innovations. IP rights also provide a framework for licensing and collaboration between different parties, thereby promoting knowledge dissemination and sharing. Overall, the aim of the IP system is to strike a balance between encouraging innovation and creativity, whilst promoting knowledge dissemination and sharing.

This paper examines the role registered IP protection, as a policy instrument, plays in the innovation process. This process starts with research and concludes when the invention or creation is put to use or commercialised, resulting in an innovation.

Innovation is a major potential route for impact of publicly funded research. The UK Innovation Strategy^{[footnote 10]} defines innovation as “the creation and application of new knowledge to improve the world”. The requirement for application differentiates innovation from other concepts such as invention or creation^{[footnote 11]}. Inventions and creations may only become innovations if they are put into use or made available for others to use^{[footnote 12]}. Their diffusion and uptake may then deliver value, prosperity, productivity and wellbeing^{[footnote 13]}. The UK Government has set the target of being the most innovative country in the world by 2030^{[footnote 14]}.

More specifically, this paper investigates how registered patents and trade marks^{[footnote 15]}, arising from research council grant funding, are translated to innovation through commercialisation channels^{[footnote 16]}. Whilst other channels to innovation exist, commercialisation is the focus of this report, due to the availability of suitable data. This analysis relies on the Gateway to Research (GtR) data from the UK Research and Innovation (UKRI). UKRI is the national funding agency investing in science and research in the UK^{[footnote 17]}. Its mission is to convene, catalyse and invest in close collaboration with others to build a thriving, inclusive research and innovation system that connects discovery to prosperity and public good^{[footnote 18]}. The GtR dataset provides a record of IP and commercialisation outcomes self-reported by grant recipients to Researchfish, an online reporting system. It brings together seven interdisciplinary research councils that invest in research across biotechnology and biological sciences (BBSRC), engineering and physical sciences (EPSRC), medicine (MRC), science and technology facilities (STFC), natural environment (NERC), economic and social sciences (ESRC), and arts and humanities (AHRC). Grant holders that do not complete the online reporting system risk withdrawal of future funding, however it is not possible to assess the completeness of outcomes reported, and under-reporting is possible.

The sample analysed includes grants allocated to projects that started between 2010 and 2020, and may not have completed by the end of 2020. Patents and registered trade marks are cross-validated against various IP data sources^{[footnote 19]}.

Because IP registration is a flawed indicator of innovation success, on its own it cannot assess or compare performance of research projects. Registration of a patent or trade mark is not appropriate or even possible for all inventions. This is evidenced by results of the 2015 Survey on Innovation and Patent Use (SIPU) ^{[footnote 20]}, which found most firms (68%) not using patent protection have inventions that are not eligible for patenting. In the UK, patent eligibility requires that an invention: (i) is new (not made public prior to application); (ii) involves an inventive step (not just an obvious modification to something that already exists); and (iii) can be made or used. Almost a third of firms in the SIPU with unpatented inventions did not fulfil the first criteria, for example because they had already published their invention in an academic journal. Furthermore, some types of inventions are simply restricted from patenting, such as medical treatments or diagnoses, scientific theories, mathematical methods, genetic processes and software with non-technical purpose. Patents can also be difficult to enforce, as reported by almost a third of SIPU respondents as a reason for their inventions being unprotected.

The SIPU found the most common reasons reported for not filing a trade mark were the presence of a pre-existing trade mark (27%), no danger of infringement (24%), and a trade mark not being perceived as important or possible (22%). A trade mark may be seen as unimportant if, for example, the owner does not intend to use their invention in a commercial setting. Some inventors prioritise publishing their invention in academic journals, making it ineligible for patenting. Some avoid registering IP due to perceived difficulty enforcing their rights against infringers, or low risk of infringement. Some may protect their research outputs through other IPRs (copyright, design rights or trade secrets), which are not analysed in this paper due to data limitations. Others may prioritise other research outcomes; such as upskilling (particularly studentships and training grants). Furthermore, analysis in this paper shows that even if a patent or trade mark is filed for an invention, this does not necessarily mean the invention will be used.

As a result, the ability of institutions to protect all possible research outcomes or IP will depend heavily on their IP protection budget and rules around filing. These parameters will change from institution to institution. Some research outputs may be better suited to protection through other IP rights not analysed in this paper, such as copyright, trade secrets, design rights or unregistered trade marks. The reader should bear this in mind when interpreting the results of this paper.

The channels through which inventions and creations are used, and therefore translated into innovation are diverse, and some are easier to monitor and quantify than others. This paper focuses on commercialisation outcomes self-reported by the grant recipients, on account of data made available by UKRI. Commercialisation is defined as “the process by which new or improved technologies, products, processes and services (arising through research) are brought to market”^{[footnote 21]}. For the purposes of this paper and based on outcomes reported by grant recipients in UKRI data, direct commercialisation by researchers is measured through the creation of spinouts, product interventions and clinical trials in the medical field, and indirect commercialisation activity of the research is measured through IP licensing and collaboration with the private sector.

This paper investigates four main questions to understand the role of registered patents and trade marks in translating research council grants into innovation:

1. How do research grant recipients protect their inventions and creations?
2. What types of research grant recipients protect their inventions and creations using registered patents and trade marks?
3. How long does it take research council grant recipients to apply for a patent once their grant is initiated?
4. What is the role of IP in the pursuit of a commercial route for the inventions and creations of research grant recipients?

Literature review

Several government reports have explored the role of IP in the journey from public research spend to innovation. Limited comparisons can be drawn in the results of these reports due to lack of standardisation on data collection methods, types of IP outcomes reviewed and time between research spend and data collection. The main findings are outlined below:

• a high proportion of surveyed participants of public research grant programmes report IP outcomes, ranging from 13-42% (based on programmes that have published impact assessments)

  • in the Space Science Programme (SSP), 42% of 26 projects funded between 2000 and 2018 reported that IP had been achieved^{[footnote 22]}, including 3 unpatented ‘industrial secrets’. Grants were relatively large in size, totalling over £520m (averaging £20m per project)
    
  • in MRC’s Confidence in Concept programme, 13% of 516 projects funded between 2012 and 2016 reported a new patent application^{[footnote 23]}. Grants were relatively small in size, averaging £59,000
    
  • the proportion of projects associated with IP in these studies is higher than found in this report, likely due to sampling differences. The type of research funded, research objectives, grant sizes, timing of the survey, and type of IP outcomes investigated differ, and may all influence findings. This report looks only at patents and registered trade marks, a wide range of grant sizes, research specialisms (including humanities and social sciences), and grant types (including studentships and training grants)
• mixed success of grant-holders in commercialising their IP through licensing and creation of spin-out companies:
  
  • projects supported by Research England’s HEI fund generated an estimated £7.9 for every £1 of funding 2015-2019, including through IP licensing income and spin-out creation^{[footnote 24]}
  • 18% of UK scaleups^{[footnote 25]} that received Innovate UK grants before 2018 went on to achieve a company sale or public listing (vs 10% average for UK scaleups)^{[footnote 26]}
    
  • royalty-free licenses increased from 39% to 77%, 2016-2020, as a proportion of all licenses sold by UKRI-funded universities^{[footnote 27]}
    
  • on average, BBSRC grant holders that incorporated a spinout did so 3.6 years after receipt of the associated BBSRC grant. Most that had filed patents did so 0-2 years prior to incorporation^{[footnote 28]}, a finding replicated in this report for a larger dataset of grants^{[footnote 29]}
• income-related measures may underestimate the impact of publicly funded research, for example:
  • per £1 HEIF grant 2015-2019, an estimated £2.2 was generated in non-monetised impact (e.g., royalty-free licensing and knowledge exchange through public networks)^{[footnote 30]}
  • the University of Oxford generated an estimated £4.0bn in productivity spillovers from £317m research funding allocated by UK Research Councils and UK charities 2018-2019^{[footnote 31]}
• differences in IP commercialisation success are observed across countries and regions:
  • english and Scottish universities, recipients of the largest Innovate UK grants^{[footnote 32]}, had the highest average patenting activity, licensing and spinout creation 2013-2017 as reported by the National Centre for Universities and Business (NCUB)^{[footnote 33]}
  • correlation between patents granted and public R&D support is stronger in Scotland, the North and the Midlands, based on a sample of 14,500 UKRI grant holders 2004-2016^{[footnote 34]}. This could be related to firms in peripheral areas facing higher financial constraints, as shown by evidence of an “urban premium” in accessing credit markets^{[footnote 35]}
• this paper contributes to the literature by analysing a wider sample of public research grants, namely all grants allocated by the seven research councils 2010-2020. Analysis is not limited to translational research programmes, scientific research, or certain grant types. For example, humanities research and student-led research funded by studentships and training grants are kept in the sample. This allows a more holistic picture of the types of grant holders that protect their research outputs using registered IPRs. A range of direct and indirect routes to commercialisation are investigated among the grant holders, based on their own reports

Data

UKRI Gateway to Research (GtR) data provides the main data source for this analysis, containing information on research grants allocated by the seven research councils (Figure 1). The sample analysed includes 70,152 research grants awarded to projects that started between 2010-2020, but did not necessarily complete within this date range. Studentships, training grants, fellowships and intramurals are included alongside traditional research grants.

IP and innovation outcomes are self-reported by grant recipients through the online platform, Researchfish. These include registered IP rights, including published patent applications, granted patents and trade marks, and commercialisation of research through IP licensing, spinout creation, clinical trials, medical product interventions and collaborations with the private sector.

As data is self-reported, it is incomplete and contains inconsistencies. These are mitigated where possible by cross-validation with IP data sources, including PATSTAT^{[footnote 36]}, the European Patent Office’s (EPO) worldwide patent statistical database, and trade mark data from the IPO and TM-Link. These sources also provide complementary data, such as patent family size.

78% of patent outcomes reported by grant recipients are found in PATSTAT, based on patent application number, the remainder are removed from the analysis. Those removed include patent applications that were withdrawn prior to publication or did not reach publication stage, for confidentiality reasons^{[footnote 37]}.

Just under half of registered trade marks reported by grant recipients are found in IPO data or TM-Link^{[footnote 38]}, the remainder are kept on account of insufficient information being provided by grant recipients for cross-validation.

Further information on the data cleaning and validation process is included in the appendix.

Figure 1: Research Council Acronyms

Research Council	Definitions
EPSRC	The Engineering and Physical Sciences Research Council
MRC	Medical Research Council
BBSRC	Biotechnology and Biological Sciences Research Council
NERC	Natural Environment Research Council
STFC	Science and Technology Facilities Council
AHRC	Arts and Humanities Research Council
ESRC	Economic and Social Research Council

Results

How do research grant recipients protect their inventions and creations?

This section outlines research council grant recipients’ commercialisation use of their registered patents and trade marks^{[footnote 39]}. The volume and type of projects that led to these outcomes are investigated, as well as whether the registered IP was a patent or a trade mark, with the following results:

• 1 in 38 completed research projects in receipt of research council grants during the period 2010-2020, led to at least one registered IP outcome in the form of a patent or trade mark. Out of the 37,852 completed research projects, 983 led to at least one registered patent or trade mark outcome^{[footnote 40]} after receipt of funding (2.6%), generating a total of 1,357 registered patent or trade mark outcomes. When ongoing projects are included, 1 in 60 projects in receipt of research council grants led to a registered patent or trade mark outcome^{[footnote 41]}, generating a total of 1,543 registered patent and trade mark outcomes. These low figures reflect that the two types of registered IP monitored in this study are not appropriate, or even possible outcomes, for all research outputs^{[footnote 42]}. The sample of projects analysed is broad in research areas, grant types, and is not limited to research with translational objectives^{[footnote 43]} that lend themselves to patent or trade mark registration. Significant sampling differences are present compared to other studies in the literature where higher translation of public research funding to IP registration is found^{[footnote 44]}. Underreporting may also be present in the self-reported data.
• published patent applications^{[footnote 45]} make up 80% of the IP outcomes (published patent applications, granted patents and registered trade marks) reported by 2010-2020 research council grant recipients. (Figure 2). Published patent applications associated with more recent grants are less likely to have yet translated to granted patents
• granted patents make up 12% of the IP outcomes reported by 2010-2020 research council grant recipients. The percentage of granted patents could increase in future as more recent patent filings progress to the stage of becoming granted
• registered trade marks make up 4% of the IP outcomes reported by 2010-2020 research council grant recipients. Only 59 registered trade marks were reported. As trade marks are most often used to protect a commercial trading advantage through branding^{[footnote 46]}, they may be inapplicable to research that is non-commercial or early-stage. Furthermore, there may be under-reporting of trade marks that are filed by parties other than the lead research organisation, such as a licensed entity or private company acting in collaboration. The lead research organisation is solely responsible for reporting outcomes to Researchfish
• research council grants associated with patent outcomes range more greatly in size than those associated with trade mark outcomes (Figure 2). The size of research council grants associated with patent outcomes is also positively skewed in distribution, meaning that a few very large grants push the average grant size above the median
• the median grant amount associated with a granted patent is highest (£1.1m), followed by a published patent application (£0.9m). The median grant amount associated with a registered trade mark is smaller (£0.7m). Projects that did not report a patent or registered trade mark outcome had the smallest median grant amount (£0.2m). The analysis does not consider funding that follows a research council grant, or differences in expense of different research specialisms associated with these forms of registered IP protection

Figure 2: Research council grant size (£m) allocated to projects funded by research councils 2010-2020, by registered patent or trade mark outcome^{[footnote 47]}, shown as a boxplot^{[footnote 48]}

• over three quarters (78%) of projects that report a registered patent or trade mark outcome also report receiving further funding from a research council, other public funding body, or private organisation. On average, projects that report a registered patent or trade mark outcome receive £8.2m further funding before the end of 2020. Most (59%) of these follow-on grants are allocated after the project’s first registered patent or trade mark outcome. By comparison, projects that do not report a registered patent or trade mark outcome receive just £1.0m further funding before the end of 2020, on average. The finding that patent holdings increase the likelihood of further research council funding is also made by Vanino et al (2019)^{[footnote 49]}, in their analysis of GtR data. Further funding is not reflected in Figure 2

• 87% of published patent applications and 90% of granted patents are protected in more than one jurisdiction^{[footnote 50]}. The most common number of jurisdictions for patent protection sought by research council grant recipients is four (Figure 3). Patent family size refers to the number of jurisdictional patent offices where an invention is protected by a patent. It is sometimes used as a proxy for patent value, based on the assumption that rights holders reserve their more valuable inventions for international patenting given the costs associated with filing patents in multiple jurisdictions^{[footnote 51]}. The number of jurisdictions in which a patent is protected could also be influenced by the institution’s patent budget, licensing income, and the technology readiness level (TRL) and field of their invention.

• more than half (55%) of grant recipients’ patent filings are Patent Cooperation Treaty (PCT) applications, providing international protection

Figure 3: Number of patent outcomes (published patent applications and granted patents) from research projects funded by research councils 2010-2020, by patent family size.

• pharmaceutical and biotechnology inventions were most commonly patented by research council grant recipients (Figure 4). 10% and 9% of patent outcomes were filed by research council grant recipients in the pharmaceutical and biotechnology fields, respectively. Previous UKIPO research^{[footnote 52]} has found that these technology fields rely heavily on patenting, relative to other UK industries. Research grant recipients less frequently patent research outputs in other fields, such as data analysis, service, textiles, consumer goods, furniture, and games

Figure 4: Percentage of patent outcomes (published patent applications and granted patents) from research projects funded by research councils 2010-2020, by technology field. Patents are assigned to technology fields by patent examiners when they are published^{[footnote 53]}.

What types of research grant recipients protect their inventions and creations using registered patents and trade marks?

• some grant types were more likely than others to lead to a reported patent or registered trade mark outcome (see appendix for GtR grant type definitions). Intramurals, research grants and fellowships were the grant types found most likely to be associated with registered patent or trade mark outcomes. These grants support research carried out by employees of universities and research institutes. Studentships and training grants were found least likely to be associated with registered patent or trade mark outcomes^{[footnote 54]} (Figure 5). These support research carried out by students with the objective of leading to a postgraduate qualification. It is notable that the statistical validity of findings may be reduced by the relatively small number of projects that received intramural grants and third party grants.

Figure 5: Percentage of projects funded by research councils 2010-2020 that led to at least one registered patent or trade mark outcome (published patent application, granted patent or registered trade mark), by grant type. In brackets, n = number of projects funded by research councils 2010-2020, by grant type (grant type definitions included in the appendix).

• grants administered by some research councils were more likely than others to lead to a reported patent or registered trade mark outcome. Research projects funded by science-focused councils (MRC, EPSRC and BBSRC) are more likely to lead to a patent or registered trade mark outcome (Figure 6). Whereas, research projects funded by councils in the social sciences and humanities fields (AHRC and ESRC) may be protected by other IP rights not explored in this paper, such as copyright to protect software, literature, and other creative works ^{[footnote 55]}.

Figure 6: Percentage of projects funded by research councils 2010-2020 that led to at least one registered patent or trade mark outcome (published patent application, granted patent or registered trade mark), by research council. In brackets, n = number of projects funded by research councils 2010-2020, by research council.

• The South East, East of England and London have the highest percentage of research projects with at least one registered IP outcome (Figure 7), and the highest number of patent and registered trade mark outcomes overall. The top 5 universities by number of registered patent or trade mark outcomes all locate in these regions, and these regions received the most funding: almost half of total research council grant spend, 2010-2020. Other studies have also found regional variation in IP outcomes of public-funded research to correlate with regional variation in funding size^{[footnote 56]}. The analysis does not control for regional variation in access to other funding sources, such as from regional funding bodies: Research England; Scottish Funding Council; Invest Northern Ireland; and Higher Education Funding Council for Wales^{[footnote 57]}, or private investors.

Figure 7: Percentage of projects funded by research councils, 2010-2020, that led to at least one registered patent or trade mark outcome (published patent application, granted patent or registered trade mark), by region of lead grant recipient^{[footnote 58]}. In brackets, n = number of projects funded by research councils 2010-2020, by region of lead grant recipient.

Of research council grants across the UK, 2.2% in the South East, 2.0% in the East, 1.9% in London and Northern Ireland, 1.6% in the North West, 1.5% in Scotland and East Midlands, and 1.2% in all other regions led to a registered patent or trade mark.

• projects led by hospitals, academic/universities and public organisations were most likely to result in at least one registered patent or trade mark outcome^{[footnote 59]} This does not take into account business as usual research conducted by those organisations, specialist equipment, or other sources of funding. The largest number of registered patent and trade mark outcomes came from academic/university-led projects, which constitute 94% of projects in the dataset. Projects led by charity/non-profit organisations, learned society and private organisations (mostly social enterprises) were less likely to associate with at least one registered patent or trade mark outcome. This may reflect that charitable status limits the commercial value that can be gained from IP, in accordance with the Charity Act 2011 and the UK’s international subsidy control commitments. The relatively small number of projects led by private companies may reduce the statistical validity of findings

Figure 8: Percentage of projects funded by research councils, 2010-2020, that led to at least one registered IP outcome (published patent application, granted patent or registered trade mark), by sector ^{[footnote 60]} of lead grant recipient. In brackets, n = number of projects funded by research councils 2010-2020, by sector of grant recipient (see figure 25 for definitions).

• a quarter of projects reported receiving further funding from a research council, other public funding body, or private organisation before the end of 2020. Of projects in receipt of further funding, just over 5% reported a registered patent or trade mark outcome (compared to 1.6% of all projects). Most (58%) of these had registered their first patent or trade mark outcome before receipt of further funding. Other than the research councils, the European Commission, Wellcome Trust, the National Institute for Health Research and the Royal Society were common providers of further funding
• just over a quarter (28%) of projects involved collaboration between the lead research organisation and another entity. These projects were more likely to have a patent or registered trade mark outcome than projects led by only one organisation (Figure 9). 16,085 projects involved a university collaboration, 6,571 projects involved a public sector organisation collaboration, 5,653 projects involved a charity/non-profit collaboration, and 1,466 projects involved a hospital collaboration. Of these, 4.1%, 4.6%, 4.3% and 6.9% led to a patent or registered trade mark outcome, respectively
• generally, as the number of collaborating entities increase, the likelihood the project will report a patent or registered trade mark outcome increases (Figure 9).However, this effect is not observed for charity/non-profit collaborations
• hospital collaborations are most associated with patent or registered trade mark outcomes (Figure 9). This includes collaboration between universities and their affiliated teaching hospitals, where universities can draw upon extensive medical resources and training. Not only is medicine a particularly patentable research area (see Figure 4), but it is also an area with a wide range of alternative funding sources, including from the National Institute for Health Research and topic/disease specific charities. This alternative funding is not considered in this analysis

Figure 9: Percentage of projects involving collaborations with universities, charity / non-profit organisations, public organisations and hospitals, that led to a patent or trade mark outcome.

• trade marks are the most common registered IP outcome among projects funded by the social science and humanities research councils trade marks are the most common registered IP outcome among projects funded by the social science and humanities research councils (ESRC and AHRC). Published patent applications are the most common registered IP outcome among projects funded by all of the other research councils (Figure 10). Different types of research generate outputs that are suited to different types of IP protection

Figure 10: Breakdown of type of registered IP outcomes associated with research projects funded by the different research councils, 2010-2020. In brackets, n = total number of registered IP outcomes associated with projects funded by that research council, 2010-2020.

• projects led by the four universities receiving the most grant funding, and largest research council grants on average, led to the most registered IP outcomes (Figure 11). These universities all have mature technology transfer offices (TTOs) that are at least 20 years old, with large net current assets (see Figure 32). They do not differ significantly from each other in the percentage of their projects involving collaboration (Figure 11).
• the University of Birmingham and University of Southampton have the highest share of research grants with a trade mark outcome (9% and 5% respectively), all funded by AHRC and EPSRC, mostly protecting brand names associated with video software, cleaning and medical technology
• some universities required less funding from research councils per registered patent or trade mark outcome. Brunel University, the University of Southampton and the University of Liverpool required the least research council funding per registered patent or trade mark outcome: £3.7m, £8.4m and £8.7m respectively (Figure 11). However, this does not reflect university performance as this analysis does not control for other sources of funding received by the universities, prior to registering the patent or trade mark

Figure 11: Top 20 universities by number of registered IP outcomes associated with research council funded projects for which they acted as the lead research organisation^{[footnote 61]}, 2010-2020.

How long does it take research council grant recipients to apply for a patent once their grant is initiated?

The time taken to apply for a patent after receipt of a research council grant is investigated among the sample of 2010-2020 grant recipients. Timing to patent is analysed across different research grant years, research councils, and grant recipients, with the following results:

• the most common time taken to apply for a patent after receipt of a research council grant is within 2 years. The percentage of research council funded projects that applied for a patent within 1 year of grant receipt was highest in the 2016 and 2017 cohorts of grant recipients. Patent applications typically take 18 months to reach publication, and only published patent applications are analysed. Therefore, more recent cohorts of grant recipients have had less time to apply for a patent, and for their applications to become published
• overall, a quarter of patent outcomes were applied for less than a year after receipt of a research council grant. Further analysis is needed to see if short times taken to patent following a research grant reflects funded research being late-stage, or at high Technology Readiness Levels^{[footnote 62]}.
• there is little significant time difference between grant receipt and patent application across projects funded by the different research councils (Figure 12), although projects funded by MRC tend to apply for patent protection slightly faster. Over 40% of MRC grants go to translational research programmes operating closer to market, including the Developmental Pathway Funding Scheme (DPFS) and the MRC Confidence in Concept (CiC) scheme (now the Impact Acceleration Account, IAA)^{[footnote 63]}.

Figure 12: Cumulative frequency^{[footnote 64]} of patent outcomes with a given number of years between receipt of a research grant ^{[footnote 65]} and patent application date of published patents, by funding organisation^{[footnote 66]}. Includes all published patent publications and granted patents applied for by grant recipients, following receipt of grant funding by one of the seven research councils 2010-2020.

• there is more variation in time difference between research grant and published patent application across universities (Figure 13). University College London, University of Edinburgh and University of Leeds tend to apply for patent protection faster than other universities. Universities’ IP filing strategies and technology transfer offices differ (see Figure 32), and they specialise in different research areas at different stages of development

Figure 13: Cumulative frequency^{[footnote 67]} of published patent outcomes with a given number of years between receipt of a research grant and patent application date, by university applicant. Includes all patent publications and granted patents applied for by the top 10 universities^{[footnote 68]}, following receipt of grant funding by one of the seven research councils 2010-2020.

What is the role of IP in the pursuit of a commercial route for the inventions and creations of research grant recipients?

To become innovations, inventions and creations must be used, or made available for others to use^{[footnote 69]}. Their use may then deliver economic and social value. The relationship between registered IPRs and use of IP is complex, with the former not necessarily guaranteeing the latter. Commercialisation is a channel through which inventions and creations can be made available for use, through bringing to market new or improved products and processes arising from research^{[footnote 70]}. The relationship between registered patent and trade mark IP outcomes and commercialisation outcomes reported by research council grant recipients is explored in this section.

Commercialisation can take many forms. Direct commercialisation by an inventor may involve spinout creation or clinical trials with the aim of introducing a product on the market. Commercialisation can also be indirect, if the invention is made available for others to bring to market through IP licensing or collaboration with the private sector, where the inventor and research team may still play an active consultancy role. The following section investigates some of these different routes to innovation among recipients of research council grants and the role of IP protection.

Direct commercialisation

Spinouts

A spinout is defined by GtR as any new business resulting directly from a research project that has received research council funding^{[footnote 71]}. These businesses may take forward projects’ inventions or creations by creating an entity to develop and market products, or license IP.

The incorporation of a spinout may be assisted by a university’s Technology Transfer Office (TTO)^{[footnote 72]} (see Figure 32). Universities also have access to other funding sources, including university seed spinout funds, external venture capital (VC) funds allied to universities or active in their spinouts, and general seed investor funds (see Figure 33), which are not considered in the analysis.

The association between registered patents and trade marks and translation of research council grants into spinouts is investigated, with the following results:

• over a quarter (27%) of projects that reported a registered patent or trade mark outcome also reported the creation of a spinout company. Granted patents were most frequently associated with spinout outcomes

• a total of 710 spinouts were reported by recipients of research council grants between 2010 and 2020. Based on information reported by grant recipients, of these spinouts, 279 (39%) have at least one published patent application, 86 (12%) have at least one patent granted, and 17 (2%) have a registered trade mark. This mirrors findings elsewhere in the literature: a third of spinout companies building on BBSRC investments were found to have a published patent application in a 2019 study^{[footnote 73]}. Under-reporting is possible, particularly where IP registration is carried out by the spinout or another third party that is not the grant recipient

• universities and their technology transfer offices (TTOs) are the most common applicants for patents and trade marks associated with spinouts that arise from research council funding^{[footnote 74]} (Figure 14, Figure 15). Where an application is made by the university or its TTO, the right would need to be licensed or transferred to the spinout to allow it to use the IP^{[footnote 75]} . A third of patents reported by grant recipients, associated with at least one spinout, were applied for by a company.

Figure 14: Applicants of patents associated with spinout companies launched from 2010-2020 research council funding, by type (as found in PATSTAT).

Figure 15: Applicants of trade marks associated with spinout companies launched from 2010-2020 research council funding, by type (as found in IPO data and TM-Link).

• of the spinouts associated with patents, 74% incorporated after the first patent application had been filed (positive values in Figure 16)^{[footnote 76]}. In these cases, the patent would be licensed or transferred to the spinout upon incorporation. The remaining 26% of spinouts incorporated before the first patent application had been filed (negative values in Figure 16). These early-stage companies are often referred to as “shell companies”, set up while their IP is being further developed by the researcher. These findings are in line with previous analysis of BBSRC grant data^{[footnote 77]}.

Figure 16: Years between patent application and spinout incorporation. Negative figures on the x-axis show patents being applied for before spinout incorporation, positive figures on the x-axis show patents being applied for after spinout incorporation. Includes all spinouts associated with grant funding by one of the seven research councils 2010-2020, that are also associated with at least one patent application.

• spinouts incorporate sooner after receipt of research council funding if they do so before the associated project’s first patent application filing. Applying for a patent is a timely process that may come before or after the incorporation of the spinout company. Spinouts incorporated after a patent application incorporated 3.7 years after receipt of research council funding, on average. Spinouts incorporated before a patent application incorporated 1.2 years after receipt of research council funding, on average

Clinical trials and product interventions in the medical field

Clinical trials study the effectiveness of a medical invention by evaluating its effect on health outcomes. This is an important stage in the commercialisation of some medical products, as gaining approval to market a medicine (“market authorisation”) by the MHRA^{[footnote 77]} or FDA^{[footnote 78]} requires assessment of evidence obtained at clinical trial. The association between reported registration of a patent or trade mark and translation of research council grants into medical inventions adopted by the market is investigated, with the following results:

• a total of 2,381 clinical trials and medical product interventions were reported by recipients of research council grants between 2010 and 2020, as outcomes of funding. Of these, 396 (17%) had at least one published patent application, granted patent or registered trade mark associated with the research council grant, based on self-reported data from grant recipients^{[footnote 79]}

• a third of clinical trials and medical product interventions associated with EPSRC grant funding had an associated patent or registered trade mark (Figure 20). AHRC, BBSRC and MRC grants were next most likely to have patent or registered trade marks associated with clinical trials and medical product interventions reported by grant recipients^{[footnote 80]}

• research council funded projects involved in clinical trials are most likely to report a published patent application if they reach the non-clinical refinement stage after initial development (Figure 18). Published applications may become granted or be unsuccessful/withdrawn. Granted patents are most often reported by projects that progress past the market authorisation stage. Trade marks are most often reported by projects that see their product adopted by the market, at least to a small scale.

Figure 17: Percentage of clinical trials associated with research council grant funding associated with a patent or registered trade mark outcome, by research council. In brackets, n = number of clinical trials associated with research council funding.

Figure 18: Percentage of research council funded projects that have reached or passed a given stage of clinical trial, and have a published patent application, granted patent or trade mark. In brackets, n = number of research council funded projects that have reached or passed that clinical trial stage.

Indirect commercialisation

Licensing

Entering a licence agreement enables an IP right owner (the licensor) to authorise another party (the licensee) to exercise some of those rights to use the IP, while the owner retains ownership and control of the IP. A patent licence allows the licensee to use the licensor’s patented invention, for example to make and sell products. A trade mark licence allows the licensee to use the licensor’s trade mark to sell products, sometimes through a franchise. Often, the licensor receives an ongoing fee from the licensee, or a ‘royalty’. Licensing of registered IP associated with research council grants is investigated, with the following results:

• 40% of all registered patent or trade mark outcomes associated with research council funding are licensed, based on information reported by grant recipients^{[footnote 81]}. All research councils achieved at least a quarter of registered patent or trade mark outcomes arising from their funding being licensed. The finding of high IP licensing activity among grant recipients mirrors findings of a 2021 UKRI study^{[footnote 82]}, where higher education providers in receipt of UKRI grants were found to generate £170m in non-software licensing revenue in 2019/20
• of licenses reported, just over half are commercial agreements in confidence. These involve the signing of a non-disclosure agreement (NDA), limiting public disclosure to protect commercially sensitive information^{[footnote 83]}. Due to greater sensitivity around patents, they are more often associated with commercial agreements in confidence than registered trade marks (Figure 19)

• trade marks are most likely to be reported by grant holders as licensed, of the IP outcome types analysed. Trade marks are reported as licensed in over half (56%) of cases. Granted patents are reported as licensed in 45% of cases (Figure 19).

• published patent applications are least frequently reported as licensed, of the IP outcome types analysed (Figure19). It is possible to license a published patent application before it is granted. Once a patent application is granted, the applicant has the right to seek damages in respect of any infringement that took place while the application was pending, after it had been published^{[footnote 84]}. A party wishing to use the invention is protected from such claims by agreeing a licence before the patent application is granted.

Figure 19: Percentage of registered patent or trade mark outcomes associated with research council funded projects (2010-2020) that are licensed, by IP right. In brackets, n = number of that type of registered IP outcome observed.

• the top 10 universities (by number of registered IP outcomes from research council grant funding) differ greatly in licensing activity (Figure 20). The University of Cambridge and University of Birmingham have the highest share of licensed patents associated with research council grants. This result does not reflect universities’ innovation performance. Different universities have different IP strategies and objectives for use of their IP, which may or may not involve licensing^{[footnote 85]}. Their licensing behaviour may also depend on the type of IP, the focus of the research, and whether the university has a sectoral specialism

Figure 20: Percentage of registered patent or trade mark outcomes from research projects with a grant date starting between 2010-2020 that are licensed, by university. Universities are ordered by number of registered IP outcomes arising from their research council grants, 2010-2020 (n).

Collaboration with the private sector

Any cooperative activity on a project between its lead research organisation and a private sector entity may constitute private sector collaboration. This can involve the provision of expertise or guidance, access to the private organisation’s data, software, facilities or funding, help launching a good or service, and provision of PhD partnerships or internships. The Lambert model^{[footnote 86]} outlines the various types of agreements between universities and business collaborators, including which party has rights to any resulting IPRs.

Research councils assess collaborative research projects to ensure grant funding is compliant with UK subsidy control legislation when it is provided directly or indirectly to enterprises that engage in economic activity^{[footnote 87]}.

The translation of registered IP outcomes that arise from 2010-2020 research council grants into innovation through private sector collaboration is investigated with the following results:

• almost half (49%) of research projects that reported patent or registered trade mark outcomes also reported private sector collaboration. Research projects associated with granted patents were most likely to report a private sector collaboration. Of projects that did not report any registered patent or trade mark outcomes, 10% reported private sector collaboration

• a total of 16,494 collaborations with the private sector were reported by recipients of research council grants between 2010 and 2020. Of these, 2,895 (18%) had at least one published patent application, granted patent or registered trade mark associated with the research council grant, based on self-reported data from grant recipients

• recipients of research council grants between 2010 and 2020 most frequently reported collaborations with AztraZeneca (374 collaborations), GlaxoSmithKline (320 collaborations), Rolls Royce Group (151 collaborations), National Biofilms Innovation Centre (149 collaborations) and Unilever (112 collaborations)

• sharing of expertise, experience, knowledge and other types of guidance was the most commonly reported contribution from private companies in collaborations (Figure 21)

Figure 21: Partner contributions as reported by research council grant recipients (2010-2020), by number of times reported within a private collaboration.

• projects that report a registered trade mark or granted patent are most likely to report collaboration with the private sector. 66% of projects that report a registered trade mark also reported private sector collaboration, compared to 57% of projects that report a granted patent, and 49% of projects that report a published patent application (Figure 22)

Figure 22: Percentage of research council funded projects that report private sector collaboration, by whether the project reports a published patent application, granted patent or registered trade mark outcome. Includes all projects funded by the research councils, 2010-2020 (n).

• most collaborations with the private sector took place before a published patent application or trade mark registration, as was the case for 59% of research projects (Figure 23). This could be because private collaboration provides researchers with knowledge (research know-how) of industrial relevance, access to proprietary property (data, equipment, molecules, etc.) or further funding needed to develop their invention to a stage where it may be patented, or protected by a trade mark. The private organisation may also have provided a letter of support^{[footnote 88]} as part of the grant application, as an industrial endorsement^{[footnote 89]}. A third of the patents reported by grant recipients as associated with private sector collaboration were found to have been applied for by companies in PATSTAT.

Figure 23: Percentage of research projects with a given time lag between first registered patent or trade mark filing (application year of published patent application or granted patent, or year of trade mark registration) and private collaboration^{[footnote 90]}. Includes all research council funded projects, 2010-2020, that led to at least one registered IP outcome and private collaboration.

• researchers’ collaborations outside the private sector may also be key to developing uses for inventions, for example in the non-profit charity sector^{[footnote 91]} or in hospitals. The association between registered IP and these types of collaborations is not assessed here

Findings and interpretation

The following broad conclusions arise from the results outlined:

Most projects receiving research council grants between 2010 and 2020 do not report a patent or registered trade mark outcome, though differences exist across grant and recipient types.

Universities with mature and well-resourced technology transfer offices are more likely to report registering patent or trade mark protection for outputs of their research council funded projects. Grant holders funded by science-focused research councils are more likely to report registering for patent protection, which may reflect differences in patentability across different specialisms of research. Grant holders that report receiving further funding from a research council or other public funding body or private organisation were more likely to report a registered patent or trade mark outcome. Regional differences in patent and trade mark registration from research council grants are observed. Research projects that completed by 2020 were more likely to report patent or trade mark outcomes; further outcomes may be expected from projects that are not yet complete. Differences in universities’ IP policies, IP portfolio budgets, access to research facilities and equipment, and access to alternative funding sources likely also play a role.

This paper examines the role registered IP protection, as a policy instrument, plays in the innovation process. It does not consider IP rights protection as a measure of performance. Not all outputs of publicly funded research are compatible with patent or trade mark protection, nor is this an objective of all research council funded projects. Some researchers may prioritise publishing their findings, invalidating any subsequent patent application. Some research outputs are non-patentable, including medical treatments and diagnoses, scientific theories, mathematical methods and genetic processes. Some research outputs are not intended for commercial use, so are unlikely to be protected by a trade mark, which protects a name, slogan, logo, symbol, or other replicable feature that indicates trade origin.

The dataset is incomplete and does not take into account patents that were filed and withdrawn before publication or have not published yet. Progressing a patent application to grant stage can be costly, and some organisations may instead sell their rights to a business, leading to under-reporting in data reported to Researchfish. It also cannot be ruled out that research council funded projects not associated with a patent or trade mark outcome could have led to IPRs not analysed here, such as copyright, trade secret, design right or unregistered trade marks.

Most patents associated with research council grants are applied for within 2 years of grant receipt, though differences are observed across grant years and grant recipients.

Other factors, not analysed, could influence timing to patent following receipt of a funding grant. Grants fund research at different stages of the innovation process. Some fund early-stage research, while others fund later-stage research that is closer to being translated into innovation, such as prototyping. Consequently, the relative timing of a patent filing could reflect maturity of the research project at the time of funding^{[footnote 92]}. Future analysis could investigate this relationship by looking at technology readiness levels associated with patent filings.

The relative timing of a patent filing could also reflect funding received for other research projects, as research is often cumulative.

Registered IP arising from research council funding is frequently commercialised.

Of registered patent and trade mark outcomes arising from 2010-2020 research council grant funding, 49% are reported by projects that also report private sector collaboration, 40% are reported as having been licensed, 36% are reported by projects that also report the creation of a spinout, and 14% are reported by projects that also report a clinical trial or medical product intervention^{[footnote 93]}. Therefore, registered patents and trade marks arising from research council funding are frequently made available for use in the market through commercialisation.

Not all patent and trade mark outcomes of research council funding are commercialised, which may reflect that:

• some research council grant recipients, including universities and other higher education institutions^{[footnote 94]}, have charitable status. This means that any private benefit of their research must be incidental to the achievement of their charitable aims, for the public benefit^{[footnote 95]}. Furthermore, Research Councils adhere to the UK’s international subsidy control commitments, limiting funding of business enterprises (any entity that puts goods or services on a market)^{[footnote 96]}. As a result, findings of commercialisation activity from research council funding may be limited

• not all research council grants are intended to facilitate inventions and creations for industry use. Funding of early-stage, basic and fundamental research may facilitate knowledge exchange, including sharing of ideas, research evidence, experience and skills, which may later contribute to innovation. Furthermore, some grants may be allocated to solve particular issues, for monitoring and evaluation, or for use of existing data. Not all of these activities will be appropriate for commercialisation

• several uses for inventions and creations are not considered in this paper. For example, innovation may take place through collaboration with the non-profit sector, or after the outright sale of IP, by its acquiror or another third party

• not all research projects that seek it will be successful in attracting industry interest (prospective licensees, private collaborators, or investor support for a spinout or clinical trial). Evidence has shown that inventions that have not reached prototype stage or demonstrated manufacturability and practicality in the market may be of limited interest to industry^{[footnote 97] [footnote 98]}

• developing an idea to commercialisation stage can be timely and cost-intensive, and further commercialisation outcomes may arise from ongoing projects and projects that receive further funding, that are not yet reflected in the data

A causal relationship between registration of published patent and trade mark IP rights and commercialisation is not established through this analysis, though correlation is observed.

Despite a positive correlation being observed between the IP registration and commercialisation of publicly funded research, one cannot conclude that patent or trade mark registration plays a causal role in the innovation process. It is possible that this correlation could reflect:

• filing requirements. According to the Patents Act 1977, patentable inventions must be capable of industrial application. Commercial use is also a precondition to trade mark registration; if a registered EU or UK trade mark is not used commercially within 5 years of registration, third parties can seek to cancel it^{[footnote 99]}. Outputs of research not intended for industrial application, such as fundamental, basic, and theoretical research, will not fulfil patent filing requirements, and are also unlikely to be commercialised

• selection effects. Grant recipients that register for patent or trade mark protection may differ from those that don’t register in ways that correlate with the commercialisation decision. For example, they may be supported by more mature or better-resourced technology transfer offices, their projects may be later stage, and the type of research may be better suited to application in industry

• response bias. Grant recipients that take the time to report registered patent or trade mark outcomes through the online reporting system may also be more likely to take the time to report their commercialisation outcomes

This analysis is also unable to comment on whether the registered patent or trade mark rights reported among research council grant recipients, and their commercialisation, would have occurred in the absence of research council funding. Projects awarded research council funding may have been selected based on characteristics that correlate with their likelihood of going on to register or commercialise IP. For example, Vanino et al (2019, ^{[footnote 100]} find that existing patent holdings increase the probability of receiving a research council grant.

Appendix

Further description of data

Gateway to Research (GtR) is a database that provides information on research council grants and their outcomes, including IP^{[footnote 101]} and commercialisation. The information is extracted from Researchfish, which is an online reporting system used by research councils to collect information on the outcomes and impact of the research they fund^{[footnote 102]}. The data in the system is live, and grant recipients are required to provide up-to-date information throughout the period of their grant and five years after the end of their grant^{[footnote 103]}. Grant recipients are only required to complete sections relevant to their research outcomes^{[footnote 104]}.

Though Researchfish went live in December 2009, full use by all research councils only began in September 2014. Previously, only MRC and STFC used Researchfish, and other research councils used ROS (Research Outcomes System), a separate platform for reporting their research outcomes. The two systems were harmonised in 2014^{[footnote 105]}, when ROS was frozen, and its data migrated to Researchfish. In September 2014, holders of grants from research councils that used the previous system were sent emails with a link to register with Researchfish^{[footnote 106]}. Underreporting of project outcomes may have been a greater risk among projects funded by these Councils, on account of their grant holders being relatively inexperienced using the system. Guidance on using Researchfish also differs across research councils, with EPSRC and ESRC providing additional guidance^{[footnote 107]} on the UKRI website, which may affect quality of reporting.

Failure by a holder of any of the seven research councils’ grants to report research outcomes through Researchfish results in sanctions according to the UKRI Compliance and Sanctions Policy^{[footnote 108]}. Sanctions include suspension of future payments of an awarded grant, and suspension of the principle and co-investigator from taking these roles in any future grant programmes. Sanctions are not applied to students supported through block grants, and UKRI units or institutes. While sanctions encourage project leads to use the online reporting system, it cannot be assessed whether project leads report all relevant outcomes, meaning under-reporting is still possible. Researchers are only required to provide information on one type of outcome, one of which can be IP^{[footnote 109]}, meaning data on outcomes is likely non-comprehensive.

The coverage of GtR grant data was calculated by comparing to research expenditure disclosed in research councils’ annual reports. Over the 2010-2020 period, annual reports show that the seven research councils spent £26.5bn on research funding. An estimate of grant size is provided for 67% of projects in UKRI GtR data. Grant sizes are not provided for most studentships in the data, as their allocation is decided by the research organisation receiving funding. GtR data records £26.3bn of research council spend 2010-2020^{[footnote 110]}. However GtR recorded spend exceeded spend disclosed in EPSRC, ESRC and NERC’s annual accounts over the period. Among the other research councils, data coverage ranged from 72% (STFC) to 99% (MRC), according to comparison with annual accounts.

Data cleaning and validation

PATSTAT (Autumn 2021 edition) is used to validate patent outcomes reported by grant recipients, and IPO and TM-Link data^{[footnote 111]} is used to validate reported trade marks. Other variables such as licenses are more difficult to cross-validate as reliable data on these are not available, and rights owners do not provide licensing information to the IPO. It is therefore not possible to check quality of coverage of licensing data self-reported by grant recipients, and it is possible that some licenses are not reported.

UKRI data was merged with PATSTAT based on IP reference, to:

• correct inconsistencies, including the status of patent outcomes (published/granted) and year of application;
• complement UKRI data with PATSTAT data to have a better coverage (e.g. patent in all jurisdictions);
• include additional useful information (e.g. patent family size)

78% of patent outcomes reported by projects receiving grants from research councils 2010-2020 could be matched with PATSTAT (1,883 out of 2,400 IP references in UKRI data). Of those not found, a minority were reported without a patent application number^{[footnote 112]}), and the rest were reported with a patent application number that could not be found in PATSTAT due to being entered incorrectly or pertaining to a patent application that has not been published. These were removed from analysis to maintain confidentiality^{[footnote 113]}.

Several projects only indicated having a trade mark or “protection not required” but do have patents according to PATSTAT (based on IP reference number provided in GtR data). In this case, non-patent outcomes were kept and patent outcomes were added along with corresponding PATSTAT variables.

Of the 1,292 published patent applications reported by researchers that were found in PATSTAT, 270 (21%) were identified as announcements of applications in the Patents Journal^{[footnote 114]}that have not progressed to full publication^{[footnote 115]}. These announcements all relate to UK patent applications, as the UKIPO is one of a select few IP offices that reports these outcomes to the EPO for inclusion in PATSTAT.

Several projects only indicated having a trade mark or “protection not required” but do have patents according to PATSTAT (based on IP reference number provided in GtR data). In this case, non-patent outcomes were kept and patent outcomes were added along with corresponding PATSTAT variables.

Trade mark outcomes self-reported by grant recipients in UKRI data were searched in IPO and TM-Link data. Some grant recipients provided registration numbers for their trade marks, whilst others only provided descriptive data, such as trade mark name, name of the organisation conducting the research, the lead researcher’s full name^{[footnote 116]} and the year of application. Based on the information provided, 46% of trade marks reported in UKRI data were found in IPO or TM-Link^{[footnote 117]}. This low number reflects that many researchers did not provide sufficient information for their trade marks to be cross-checked. Two trade marks were removed, one that had been withdrawn, and another that was a design right misreported as a trade mark.

Figure 24: Grant type definitions (source: GtR data dictionary^{[footnote 118]}.

Grant type	GtR definition
Research grant	A contribution to the costs of a stated research project which has been assessed as suitable for funding through the procedures established by the relevant Research Council.
Studentship	A research project leading to the award of a higher degree, funded from one or more Research Council Training Grants, possibly supplemented by funding from business or charitable organisations.
Training grant	A block grant to a Higher Education Institute to cover the cost of PhD and (previously) MSc studentships, including Doctoral Training Grants and Partnerships.
Fellowship	A type of project covering award made through a fellowship competition providing a contribution to the support of a named individual. It covers the cost of the time dedicated by the fellow to their personal research programme, and may or may not include research support costs.
Intramural	A project category covering research contracts and internal transfers made to Research Council Centres, Units, Institutes and Facilities.
Other grant	NA

Figure 25: Sector definitions.

Sector	Includes
Academic/university	Private and public universities, schools, academies and colleges.
Public	Public bodies and ministerial departments. Public bodies are organisations that are (at least in part) publicly funded to deliver a public or government service.
Hospitals	Hospitals, NHS foundation trusts and university teaching hospitals.
Charity/Non-profit	An organisation that is established for charitable purposes only, principally governed by the Charities Act 2011, with any private benefit incidental to the achievement of their charitable aims, for the public benefit.
Learned society	Organisations dedicated to research and scholarship, often focused on a particular academic discipline, profession or cause.
Private	Private limited companies, mostly consisting of social enterprises, established partly or wholly for to create profit that may be reinvested into the business or local community to address social or environmental challenges.

Distribution of funding across grant recipients

Figure 26: % of research projects receiving grant funding from the seven research councils 2010-2020, by research council.

Research council	Number of grants	Percentage of grants	Size of funding (£m)
EPSRC	22,549	32.1%	10,938
BBSRC	12,461	17.8%	3,822
MRC	8,014	11.4%	5,950
ESRC	7,918	11.3%	2,670
NERC	7,518	10.7%	2,583
AHRC	6,282	9.0%	916
STFC	5,410	7.7%	1,562
Total	70,152		28,441

Figure 27: Research projects receiving grant funding from the seven research councils 2010-2020, by sector of project lead.

Lead Research Organisation sector	Number of projects	Percentage of projects	Size of funding (£m)
Academic / university	65,811	94%	25,127
Private	1,549	2%	1,490
Public	1,848	3%	668
Charity/Non-profit	524	1%	995
Hospital	348	0%	143
Learned society	72	0%	19
Total	70,152		28,441
Collaborator Organisation sector [footnote 119]	Number of projects	Percentage of projects	Size of funding (£m)
Academic / university	16,085	23%	12,122
Private	7,651	11%	7,828
Public	6,571	9%	6,690
Charity/Non-profit	5,653	8%	5,839
Hospital	1,466	2%	1,924
Learned society	428	1%	683
Total	70,152		28,441

Figure 28: Percentage (%) of research projects receiving grant funding from the seven research councils 2010-2020, by region of project lead.

Region	Number of projects	Percentage of projects	Size of funding (£m)
London	12,545	21.8%	5,968
South East	10,476	18.2%	4,636
Scotland	8,356	14.5%	3,257
East of England	6,887	12.0%	3,365
North West	6,075	10.5%	2,366
South West	5,165	9.0%	1,742
West Midlands	4,072	7.1%	1,331
East Midlands	3,684	6.4%	1,145
Yorkshire and The Humber	3,669	6.4%	1,243
North East	2,779	4.8%	898
Wales	2,367	4.1%	804
Northern Ireland	834	1.4%	265
Outside UK	324	0.6%	208
Unknown	2,919	5.1%	1,212
Total	70,152		28,441

Figure 29: % of research projects receiving grant funding from the seven research councils 2010-2020, by grant type.

Grant type	Number of projects	Percentage of projects	Size of funding (£m)
Research Grant	35,066	50.0%	19,394
Studentship	23,116	33.0%	NA[footnote 120]
Training Grant	5,333	7.6%	4,129
Fellowship	4,247	6.1%	1,604
Intramural	2,258	3.2%	3,130
Third Party Grant or other	132	0.2%	184
Total	70,152		28,441

Applicant types (from IP data sources)

By cross-checking with IP data, including PATSTAT, UKIPO data and TM-Link, the type of applicants that register for registered IP outcomes reported in GtR data are identified.

Figure 30: Type of organisations filing for patents linked to public research grants allocated by the seven research councils 2010-2020.

Patent applicant	Number of patent outcomes	% of total
University	616	42%
Company	497	34%
Individual	288	19%
Government non-profit organisation	72	5%
Unknown	9	1%
Total	1,482

Figure 31: Type of organisations filing for trade marks linked to public research grants allocated by the seven research councils 2010-2020.

Trade mark applicant	Number of trade mark outcomes	% of total
University	18	31%
Company	3	5%
Individual	2	3%
Government non-profit	1	2%
Unknown	35	59%
Total	59

Technology transfer offices (TTOs) of top 10 universities using research council grants to register IP^{[footnote 121]}

All of the top 10 universities in the data by number of patent and trade mark outcomes from research council grants have technology transfer offices. These are responsible for identifying, protecting, and commercialising the creations and inventions that form from university research to deliver maximum positive impact^{[footnote 122]}. They differ in age and resource. Net current assets (also termed “working capital”) indicates the TTO’s ability to increase investment in the short term ^{[footnote 123]}. Universities that report the largest number of patent and registered trade mark outcomes are observed to have more mature TTOs with larger net current assets.

Figure 32 Technology transfer offices of the top 10 universities (ranked by total number of patent and registered trade mark outcomes reported by the university associated with research council funding, 2010-2020).

University	Technology Transfer Office	Incorporation date[footnote 124]	Net current assets 2019 (£m)[footnote 125]	Website
University of Oxford	Oxford University Innovation Ltd [footnote 126]	1987	£5.710m	https://innovation.ox.ac.uk/
Imperial College London	Imperial Innovations Ltd	2000	£5.148m	https://imperial.tech
University College London	UCLB Investments Ltd	1993	£2.312m	https://www.uclb.com/
University of Cambridge	Cambridge Enterprise Ltd [footnote 127]	1972	£2.419m	https://www.enterprise.cam.ac.uk/
University of Southampton	Research and Innovation Services (R&IS); SETsquared Partnership	2019; 2002 [footnote 128]	NA	https://www.southampton.ac.uk/research/ris.page
University of Liverpool	Virtual Engineering Centre (VEC); University of Liverpool Knowledge Exchange and Impact	2010[footnote 129] ; NA	NA	https://www.virtualengineeringcentre.com/ https://www.liverpool.ac.uk/collaborate/commercialising-research/
University of Leeds	Leeds Innovation Centre Ltd (Nexus)	1999	£2.556m	https://nexusleeds.co.uk/
University of Manchester	University of Manchester Innovation Factory Ltd[footnote 130]	2004	£0.910m	https://www.uominnovationfactory.com/
University of Edinburgh	Edinburgh Innovations Ltd[footnote 131]	1993	£0.545m	https://edinburgh-innovations.ed.ac.uk/
University of Birmingham	University of Birmingham Enterprise Ltd[footnote 132]	2008	£0.762m	https://www.birmingham.ac.uk/partners/enterprise/index.aspx

Funding available to UK spinouts, by university

Access to alternative funding sources (besides research council grants) for spinouts differs by university. These include university seed spinout funds, external venture capital (VC) funds allied to universities or active in their spinouts, and general seed investor funds. Cambridge Innovation Capital is the largest investor in UK university spinouts, with £500m funds under management. In total, UK universities currently have £1.4bn ready to be invested in their spinouts, as of 2022. Due to non-availability of earlier data, 2022 data is used to proxy earlier investment made available for UK university spinouts.

Figure 33: Total UK investment available for UK university spinouts, 2022.^{[footnote 133]}

University seed spinout funds	£m available	No. of investments	Average % equity stake	Average initial commitment (£)	Universities aligned with
University of Cambridge Enterprise Funds	£2m-4m per year	5 to 10	5%-35%	£0.25 - £1m	University of Cambridge
Imperial College Enterprise Funds	£2m-4m per year	5 to 10	5%-35%	£0.25 - £1m	Imperial College London
University of Oxford Innovation Funds	£2m-4m per year	5 to 10	5%-35%	£0.25 - £1m	University of Oxford
Old College Capital	£15m	30	10%-30%	£150,000	University of Edinburgh
UCL Technology Fund	£70m	66			University College London (UCL)
University of Bristol Enterprise Funds	£2m-4m per year	5 to 10	5%-35%	£0.25 - £1m	University of Bristol
Subtotal	£101m
External VC funds allied to universities or active in spinouts	£m available	No. of investments	Average % equity stake	Average initial commitment (£)	Universities aligned with
Cambridge Innovation Capital	£500m (£275m committed)	36	n/a	£5m	University of Cambridge
Northern Gritstone	£300m	n/a	n/a	tbc	University of Leeds; University of Manchester; Univeristy of Sheffield
Oxford Sciences Enterprise	£200m (£447m committed)	100+	0.25	£250,000 to £25m	University of Oxford
Parkwalk EIS funds	£50m-£60m per year	10 to 15	5%-35%	£1m-£10m	Across UK universities
BGF	£95m [footnote 134]	22	10%-15% approx		University of Cambridge; University of Oxford; Imperial College London; University College London (UCL); University of Bristol; Queen Mary, University of London; University of Southampton; University of Birmingham; University of Nottingham; Royal Holloway, University of London
Subtotal	£1,115m
General seed investor	£m available	No. of investments	% equity stake	Average initial commitment (£)	Universities aligned with
Scottish Enterprise	£78m [footnote 135] (£20m committed [footnote 136] )	340 overall, including 177 in 2020/21 [footnote 137]	n/a		Abertay University; Edinburgh Napier University; Glasgow School of Art; Glasgow Caledonian University; Heriot-Watt University; James Hutton Institute; Scottish Crop Research Institute (SCRI); Moredun Research Institute; Queen Margaret University (Edinburgh); Robert Gordon University; Scotland’s Rural College (SRUC); Scottish Association for Marine Science; The University of the Highlands and Islands; University of Aberdeen; University of Dundee; University of Edinburgh; University of St Andrews; University of Strathclyde; University of the West of Scotland; University of Stirling; University of Glasgow
Midven (Midlands Engine Investment Fund)	£16m (£2.323m committed)	3	n/a		Nottingham Trent and Cranfield Universities; University of Birmingham; Aston University
UKI2S (UK Science and Innovation Fund 2)	£47m (£4.8m committed)	22	10-15% at formation, diluted to 1-10% over time		University of East Anglia; University of Glasgow
Subtotal	£141m
Total	£1,357m

Trade marks and commercialisation

1. A trade mark registration grants the owner the exclusive right to use a name, slogan, logo, symbol, or other replicable feature that indicates trade origin, allowing consumers to distinguish the product offering of one company with that of another.

2. A trade mark registration may be interpreted as a sign of commercialisation.
   i. If a registered EUTM or UK trade mark is not used commercially within 5 years of re gistration, third parties can seek to cancel it on the grounds of non-use^{[footnote 138]}. In the US, commercial use of a trade mark is a precondition to registration according to the Trade mark Act (15 U.S.C. §1127), which requires “the bona fide use of a mark in the ordinary course of trade”. Holders are required to periodically prove use to maintain their trade mark registration.
   ii. Commercial requirements separate trade marks from other IP rights. The prevalence of large numbers of unused or ‘sleeping’ patents, including those filed tactically to block competitors, is widely recognised in the literature, as commercial use is not a precondition of patent filing^{[footnote 139]}.
   iii. The European Commission noted in its Trade Marks Directive 2015^{[footnote 140]}) that a trade mark can “attract and retain customer loyalty”, increasing the likelihood of innovations being successfully commercialised. Furthermore, it points to a “continuous process of product improvement and development” incentivised by trade marks, as their owners seek to maintain brand reputation^{[footnote 141]}. Trade mark-intensive industries generated almost 34% of economic activity (GDP) in the EU 2008-2010.
   iiiv. A US study of trade mark registrations by universities 1997-2007 found that achieving better market prospects to appropriate the output of innovation is a key motivation for academic institutions to file a trade mark^{[footnote 142]}.
   iv. Trade marks may also be filed to expand commercial activity into a new geographic market, as multiple trade mark applications are often required to achieve protection across multiple jurisdictions^{[footnote 143]}.
3. However, the registration of a trade mark does not always reflect commercial use of innovation, as evidenced by the phenomenon of “trade mark cluttering”:
   i. “Trade mark cluttering” occurs when firms hold trade marks that are overly broad (registered in several classes at the same time) or unused, as observed in IPO and European trade mark office (OHIM) data^{[footnote 144]}.
   ii. Firms may register several trade marks simultaneously to develop fallback options. This is observed especially in the pharmaceuticals sector^{[footnote 145]}, where trade marks must be submitted to medical regulators for inspection due to the increased danger for consumers arising from confusion of trade marks. As a result, pharmaceutical firms routinely apply for several trade marks for a single drug well before the drug has passed all regulatory hurdles^{[footnote 146]}.
   iii.Unused trade marks may also result from holders wanting to reserve space for possible future expansion of a product or service offering ^{[footnote 147]}, making use of the five year grace period permitted by the IPO and OHIM.

1. This follows definitions of innovation that require inventions and creations to be used/applied, to become innovations. The UK Innovation Strategy defines innovation as “the creation and application of new knowledge to improve the world”. The OECD Oslo Manual states “The requirement for implementation differentiates innovation from other concepts such as invention, as an innovation must be implemented, i.e. put into use or made available for others to use”. ↩

2. The analysis includes patents and trade marks applied for at IP offices worldwide. It does not extend to design rights or unregistered IP rights, such as copyright and trade secrets, due to limitations in available data. ↩

3. Commercialisation is defined by UKRI as “the process by which new or improved technologies, products, processes and services (arising through research) are brought to market”. See UKRI’s commercialisation plan (Nov 2021), available here: UKRI’s commercialisation plan: from ambition to action – UKRI%20are%20brought%20to%20market.). Commercialisation outcomes included in UKRI Gateway to Research (GtR) data and analysed in this paper include spinouts, licensing, clinical trials and medical product interventions, and collaborations with the private sector. ↩

4. AHRC, BBSRC, EPSRC, ESRC, MRC, NERC and STFC. ↩

5. The EPO Worldwide Patent Statistical Database (PATSTAT), TM-Link, and IPO trade mark data are used for cross-validation. ↩

6. IPO, 2018, Survey of innovation and patent use. ↩

7. The Engineering and Physical Sciences Research Council (EPSRC), Medical Research Council (MRC), Biotechnology and Biological Sciences Research Council (BBSRC), Natural Environment Research Council (NERC), Science and Technology Facilities Council (STFC), Arts and Humanities Research Council (AHRC) and Economic and Social Research Council (ESRC). ↩

8. These regions are also home to over a third of the UK’s Russell Group universities. ↩

9. Including commercial in confidence licensing agreements. ↩

10. UK Innovation Strategy, Department for Business, Energy & Industrial Strategy (July 2021). Accessible here: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1009577/uk-innovation-strategy.pdf ↩

11. Invention is the act of bringing ideas or objects together in a novel way to create something that did not exist before, such as a device, method, composition, or process. Creation is the generation of something new – a unique concept such as a logo, design or piece of art. ↩

12. The OECD’s Oslo manual states: “Innovation is more than a new idea or an invention. An innovation requires implementation, either by being put into active use or by being made available for use by other parties, firms, individuals or organisations”. See chapter 2.2.4 of the Oslo Manual. OECD/Eurostat (2018), Oslo Manual 2018: Guidelines for Collecting, Reporting and Using Data on Innovation, 4th Edition, The Measurement of Scientific, Technological and Innovation Activities, OECD Publishing, Paris/Eurostat, Luxembourg. ↩

13. UK Innovation Strategy, Department for Business, Energy & Industrial Strategy (July 2021). Accessible here: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1009577/uk-innovation-strategy.pdf. ↩

14. Department for Business, Energy & Industrial Strategy. (2017). “Industrial Strategy: Building a Britain fit for the future”. Pg.63 Retrieved from Industrial Strategy: building a Britain fit for the future - GOV.UK (www.gov.uk). ↩

15. The analysis includes patents and trade marks applied for at IP offices worldwide. It does not extend to design rights or unregistered IP rights, such as copyright and trade secrets, due to limitations in available data. ↩

16. Commercialisation is defined by UKRI as “the process by which new or improved technologies, products, processes and services (arising through research) are brought to market”. See UKRI’s commercialisation plan (Nov 2021), available here: UKRI’s commercialisation plan: from ambition to action – UKRI%20are%20brought%20to%20market.). Commercialisation outcomes included in UKRI Gateway to Research (GtR) data and analysed in this paper include spinouts, licensing, clinical trials and medical product interventions, and collaborations with the private sector. ↩

17. See https://www.gov.uk/government/organisations/uk-research-and-innovation. ↩

18. See https://www.ukri.org/about-us/strategy-plans-and-data/. ↩

19. The EPO Worldwide Patent Statistical Database (PATSTAT), TM-Link, and IPO trade mark data are used for cross-validation. ↩

20. IPO, 2018, Survey of innovation and patent use. ↩

21. See UKRI’s commercialisation plan (Nov 2021), available here: UKRI’s commercialisation plan: from ambition to action – UKRI%20are%20brought%20to%20market.). ↩

22. Winning Moves (January 2020), Report prepared for UK Space Agency. “An assessment of the industrial impacts of UK funding through the ESA Space Science Programme”. ↩

23. Ipsos MORI (June 2019). “Biomedical Catalyst Impact Evaluation Final Report”. The Confidence in Concept (CiC) programme is now called the Impact Acceleration Account (IAA). ↩

24. Research England (October 2020). “Assessing the Gross Additional Impacts of the Higher Education Innovation Fund (HEIF): An update for the period 2015/16 – 2018/19”. Estimated by dividing the sum of income attributable to HEIF funding (£4,542m, estimated by grant holders, including income from licensing IP, contract research, consultancy, facilities and equipment-related services and regeneration and development programmes) and value of spinouts created by HEIF-funded projects (estimated using private investment into spinouts) by total HEIF funding allocated (£745m). ↩

25. OECD definition of a scaleup company used: average annualised growth greater than 20% per annum, over a three year period, measured by the number of employees or by turnover. ↩

26. Beauhurst (15 August 2018). The relationship between equity and innovate UK grants. https://www.beauhurst.com/blog/innovate-uk-grants-equity-accelerators/ ↩

27. UK Research and Innovation (UKRI) (November 2021). “An Update on IP-Related and Commercialisation Activities in England in 2019/20”. Note: excludes software IP licenses (15% of IP licenses sold by UKRI-funded universities in 2020). ↩

28. BBSRC, 29 January 2019. “The making of bioscience spinout companies in the UK”. ↩

29. See pages 14-15. ↩

30. Research England (October 2020). “Assessing the Gross Additional Impacts of the Higher Education Innovation Fund (HEIF): An update for the period 2015/16 – 2018/19”. ↩

31. London Economics (August 2021). “The economic impact of the University of Oxford, Final Report for the University of Oxford”. ↩

32. On average, 2013-2017, grants to English, Scottish, Northern Irish and Welsh universities were £385,950, £282,370, £162,194 and £139,092, according to National Centre for Universities and Business (NCUB) data. A recent review found that higher spending in areas with “clusters” of universities, or research in one area, is associated with this distribution. It found no evidence of bias to English and Scottish universities in funding decisions. See https://www.bbc.co.uk/news/uk-northern-ireland-62243998. ↩

33. National Centre for Universities and Business (2019). “State of the Relationship”. ↩

34. BEIS (October 2021). “From ideas to growth: Understanding the drivers of innovation and productivity across firms, regions and industries in the UK”. ↩

35. Vavino (2019). “Knowledge to money: Assessing the business performance effects of publicly-funded R&D grants”. Lee and Luca (2018). “The big-city bias in access to finance”. ↩

36. https://www.epo.org/searching-for-patents/business/patstat.html. ↩

37. Information on unpublished patent applications is confidential, and data on unpublished patent applications is only available to the IPO for patents applied for at its office, not other worldwide IP offices, inclusion of these would skew findings. ↩

38. This low figure reflects that many researchers did not provide sufficient information for their trade marks to be cross-checked, and that TM-Link data is only updated to July 2019, and therefore does not include more recent trade marks. Of trade marks self-reported by grant recipients before 2019, 66% were found in IPO or TM-Link data. ↩

39. See the IPO’s Patent Guide https://www.gov.uk/government/publications/the-patent-guide and Trade Mark Guide https://www.gov.uk/government/publications/the-trade-mark-guide. ↩

40. A registered IP outcome is classified as a published patent application, granted patent or a registered trade mark application. ↩

41. 159 out of 32,300 research projects that were still ongoing in 2020 reported at least one patent or registered trade mark outcome. ↩

42. Non-patentable research outputs include scientific theories, discoveries and mathematical methods, literary, dramatic, musical or artistic works, methods of medical treatment or diagnosis, ‘essentially biological’ processes like cross-breeding animals or varieties of plants, software that has a ‘non-technical’ purpose (e.g. recreational apps), a way of doing business, playing a game or thinking, and the way information is presented. See https://www.gov.uk/patent-your-invention for more information. ↩

43. Some researchers may prioritise publishing their research, invalidating any subsequent patent application. For more information see https://www.gov.uk/patent-your-invention. ↩

44. Whilst the literature has focused on scientific research (particularly in biosciences), the sample of research projects analysed here also includes social sciences and humanities, research that is found in this paper to be less likely to translate to registered IP. Furthermore, studentships, training grants, fellowships and intramurals are included in the analysis in addition to research grants, which have been the focus of previous studies. ↩

45. 21% of published patent applications reported by grant recipients were found in PATSTAT as provisional patents, published in the Patent Journal (PDJ). More information on provisional patents can be found here: https://ipo.blog.gov.uk/2015/06/05/patent-pending/). ↩

46. See appendix for a short literature review on the association between trade marks and commercialisation. ↩

47. Where some registered IP outcomes associate with multiple research council funded projects, their respective grant sizes are summed. ↩

48. The box shows the interquartile range: where the middle 50% of grant size data lies. The line in the middle of the box shows the median. The distance between the two whiskers shows the full range of grant size data, excluding outliers. Grant sizes that are larger than the upper quartile (top of the box) by 1.5 times the interquartile range are considered outliers and excluded from the graph. 116 grants associated with published patent applications, 21 grants associated with granted patents, and 7 grants associated with trade marks fall within this definition of an outlier, and so are excluded from the graph. The largest outlier excluded from the graph is a £129m EPSRC grant allocated to the Faraday Institution in 2018 for a project researching energy storage in electric vehicles, associated with 3 patent applications and 2 trade marks. ↩

49. Enrico Vanino, Stephen Roper, Bettina Becker, Knowledge to money: Assessing the business performance effects of publicly-funded R&D grants, Research Policy, Volume 48, Issue 7, 2019, Pages 1714-1737. Analysis performed on 2004-2016 GtR data. ↩

50. Based on matching patents reported by grant recipients in GtR data to PATSTAT data on patent families. ↩

51. See Harhoff, Scherer, and Vopel 2003; Johnstone et al. 2012; Lanjouw and Schankerman 2004; Putnam 1996. ↩

52. “Use of Intellectual Property rights across UK industries”, UKIPO (2022). See https://www.gov.uk/government/publications/use-of-intellectual-property-rights-across-uk-industries/use-of-intellectual-property-rights-across-uk-industries. ↩

53. This classification system is called “International Patent Classification (IPC)”, involving the allocation of published patent applications to 35 technology fields, according to World Intellectual Property Organization (WIPO) concordance. See IPC and Technology Concordance Table (wipo.int). ↩

54. 23 studentships and 1 training grant led to a registered IP outcome ↩

55. WIPO define ‘core’ copyright industries as wholly engaged in the creation, production and manufacturing, performance, broadcast, communication and exhibition, or distribution and sales of works or other protected subject matter. WIPO defines ‘Inter-dependent’ copyright industries, which deal with products jointly consumed with the core industries, or with facilitation equipment. ↩

56. See Department for Business, Energy & Industrial Strategy (2021). “From ideas to growth: Understanding the drivers of innovation and productivity across firms, regions and industries in the UK”; and National Centre for Universities and Business (2019). “State of the Relationship”. ↩

57. In the financial year 2019-2020, Research England, Scottish Funding Council, Invest Northern Ireland and Higher Education Funding Council for Wales spent £2,383m, £1,867m, £303m and £180m respectively on research grant expenditure. ↩

58. Analysis excludes 2,919 projects out of 70,152 of unknown region in GtR data. Of these projects, 0.9% led to a patent or registered trade mark outcome. ↩

59. Whilst it is not possible to investigate underreporting among different grant recipient sectors, it may be expected that public organisations (government bodies and research council institutes) are more likely to adhere to reporting requirements, which may influence this finding. ↩

60. Sector categories taken from GtR data, see appendix for definitions.https://gtr.ukri.org/resources/GtRDataDictionary.pdf. ↩

61. The lead research organisation is defined in GtR data as “An organisation receiving project funding which is accountable for ensuring that the planned outcomes for the project are achieved and that the funds are spent in accordance with the terms and conditions of the funder”, see GtRDataDictionary.pdf (ukri.org). ↩

62. Technology Readiness Levels (TRLs) measure the maturity of evolving technology. There are nine, starting from the lowest level of maturity: observing and reporting basic principles, to the highest level of maturity: technology qualified through successful mission operations. See: https://publications.parliament.uk/pa/cm201011/cmselect/cmsctech/619/61913.htm#:~:text=Technology%20Readiness%20Levels%20(TRLs)%20are,Basic%20principles%20observed%20and%20reported. ↩

63. MRC Translational Research 2008-2018, available here: https://www.ukri.org/publications/mrc-translational-research-evaluation-report/, see page 17. ↩

64. For example, the second dark blue plot shows that 58% of patent outcomes were applied for within 1 year of receiving a research grant allocated by MRC over the sample period. ↩

65. The first research grant linked to that patent outcome, in the 2010-2020 period of analysis. ↩

66. AHRC and ESRC are excluded, as their grants were associated with 5 and 0 patent outcomes respectively. ↩

67. For example, the second purple plot shows 65% of patent outcomes were applied for by University College London within 1 year of receiving a research grant by one of the 7 research councils over the sample period. ↩

68. The top 10 universities are defined according to the highest number of patent outcomes. ↩

69. This follows definitions of innovation that require inventions and creations to be used/applied, to become innovations. For example, the UK Innovation Strategy defines innovation as “the creation and application of new knowledge to improve the world” and the OECD Oslo Manual states “The requirement for implementation differentiates innovation from other concepts such as invention, as an innovation must be implemented, i.e. put into use or made available for others to use”. ↩

70. See UKRI’s commercialisation plan (Nov 2021), available here: UKRI’s commercialisation plan: from ambition to action – UKRI%20are%20brought%20to%20market.). ↩

71. GtR Data Dictionary, available here: https://gtr.ukri.org/resources/GtRDataDictionary.pdf. ↩

72. See appendix for further detail on university TTOs. ↩

73. BBSRC, 29 January 2019. “The making of bioscience spinout companies in the UK”. ↩

74. Based on IP applicant data (patent applicants found in PATSTAT, and trade mark applicants found in IPO data and TM-Link). ↩

75. A previous study found that some universities choose to transfer their IP rights to a spinout company in return for an equity stake in the company, see IP Pragmatics (2020), “Best Practice in Equity Stakes for University Spin-Outs”, a report commissioned by Research England. Whether IP is licensed or transferred to the spinout companies by the university is not investigated in this paper, due to limitations of available data. ↩

76. Filing a qualifying patent allows a company to benefit from reduced corporation tax through the Patent Box scheme. However, as most new spinouts take years to enter profit, this is unlikely to be a key driver for early filing. ↩

77. BBSRC, 29 January 2019, “The making of bioscience spinout companies in the UK”. ↩ ↩²

78. Medicines and Healthcare Regulatory Agency. ↩

79. Food and Drug Administration. ↩

80. Of the 2,381 clinical trials and medical product interventions associated with 2010-2020 research council funding, 372 have a published patent application, 80 have a granted patent and 45 have a trade mark. ↩

81. Licenses self-reported by grant recipients could not be cross-validated, as reliable data on IP licensing is not available. Rights owners are not required to provide licensing information to the IPO. It is therefore not possible to check quality of coverage of licensing data self-reported by grant recipients, and it is possible that some licenses are not reported. ↩

82. UK Research and Innovation (UKRI) (November 2021). “An Update on IP-Related and Commercialisation Activities in England in 2019/20”. ↩

83. For example, the licensee’s intended use of the licensed IP may be kept private, if disclosure of this information would cause significant harm to the interests of the licensee, for example financial loss or loss of profitability. Different grant holders may interpret “commercial in confidence” differently when they submit information to Researchfish. ↩

84. Subject to conditions detailed in the Patent Act 1977, Section 69. ↩

85. IP licensing is usually carried out by the university’s technology transfer office (TTO). Information on university TTOs is included in the appendix. ↩

86. See University and business collaboration agreements: Lambert Toolkit - GOV.UK (www.gov.uk). ↩

87. See Subsidy control regime - GOV.UK (www.gov.uk). ↩

88. See Project partners letter of support – UKRI. ↩

89. The letter of support may include details of the industrial relevance of the project, including the projected market size, customers and sales, and how the organisation will commercialise the technology beyond the project. See Project partners letter of support – UKRI. ↩

90. Negative number of years means private collaboration took place prior to first patent application being published, or first trade mark being registered. ↩

91. For example, the C4 Rice Project involves researchers from 7 institutions collaborating to apply innovative scientific approaches to the development of high yielding rice varieties for smallholder farmers. https://c4rice.com. ↩

92. Information on the maturity of research projects at the time of funding is not included in GtR data, so this cannot be investigated. ↩

93. Of registered IP outcomes arising from grants funded by scientific research councils (BBSRC, EPSRC, MRC and STFC). ↩

94. Universities and other Higher Education Institutions (HEIs) have always been recognised as charities under the law of England and Wales. ↩

95. See Research by Education Institutions (publishing.service.gov.uk) and Charities Act 2011. ↩

96. See Funding rules – UKRI. ↩

97. Pressman, Lori & Guterman, Sonia & Abrams, Irene & Geist, David & Nelsen, Lita. (1995). PreProduction Investment and Jobs Induced by MIT Exclusive Patent Licenses: A Preliminary Model to Measure the Economic Impact of University Licensing. Journal of the Association of University Technology Managers. ↩

98. Shane S. (2004). Academic Entrepreneurship: University Spinoffs and Wealth Creation, Edward Elgar Publishing Ltd; UNICO, 2006a. Practical Guides: Commercialisation Agreements: Spin-out Transactions, UNICO; McMillan T, 2016. University Knowledge Exchange (KE) Framework: good practice in technology transfer, by the McMillan group, HEFCE. ↩

99. Also, in the US, commercial use of a trade mark is a precondition to registration according to the Trade mark Act (15 U.S.C. §1127), which requires “the bona fide use of a mark in the ordinary course of trade”. ↩

100. Enrico Vanino, Stephen Roper, Bettina Becker, Knowledge to money: Assessing the business performance effects of publicly-funded R&D grants, Research Policy, Volume 48, Issue 7, 2019, Pages 1714-1737. ↩

101. GtR data includes information on grant holders’ granted patents, published patent applications, trade marks and copyrights, reported as outcomes of grant funding. Copyright outcomes are excluded from analysis due to lack of reliable data available for cross-validation. Furthermore, the new 2022 questionnaire excludes copyright as an option. ↩

102. Researchfish questionnaire can be accessed here (“Intellectual Property & Licensing” questions are on page 26). ↩

103. However, researchers that received Medical Research Council (MRC) grants are required to submit data on a grant which ended more than 5 years ago. ↩

104. For example, the “Intellectual Property & Licensing” section of the online reporting system is only required to be completed by grant recipients that achieved these outcomes. ↩

105. Research Outcomes Harmonisation Project Research Operations Office (cam.ac.uk) ↩

106. Research Outcomes Harmonisation Project Staffordshire University Research and Funding Blog (staffs.ac.uk) ↩

107. See Reporting outcomes for EPSRC-funded projects – UKRI and ESRC additional guidance on Researchfish – UKRI. ↩

108. UKRI Compliance and Sanctions Policy: https://www.ukri.org/wp-content/uploads/2020/12/UKRI-081220-SanctionsPolicy.pdf. ↩

109. This also depends on the funding organisation. In addition, Researchfish was originally mainly used for collecting data on research grants by the Medical Research Council (MRC) and was later used by other funding organisations. This has likely influenced the design of the questions. ↩

110. Using the “award pounds” and “expenditure pounds” variables in the GTR dataset. ↩

111. Includes trade mark data from Australia, Canada, the European Union, New Zealand, and the United States. ↩

112. Patent application numbers had not been provided for 71 reported published patent applications and 17 reported granted patents. Together, these account for just under 5% of total patent outcomes reported. ↩

113. Public patent databases like PATSTAT only contain information on published patent applications and granted patents. Information from the UK patent register on unpublished patent applications is confidential, and only available for patents applied for at the IPO, not other worldwide IP offices ↩

114. Intellectual Property Office - Patents Journal : UK applications filed (ipo.gov.uk). ↩

115. Patent timeline. ↩

116. A search of researcher names yielded few matches as most universities and research institutions state in their IP policies that any IP discovered by an employee during the course of their employment is the property of their employer (the university or institution). ↩

117. Of registered trade marks reported by grant recipients before 2019, 66% were cross-validated. TM-Link data is updated to July 2019. ↩

118. Source: GtR Data Dictionary, accessible here: https://gtr.ukri.org/resources/GtRDataDictionary.pdf. ↩

119. Categories provided within GtR data. ↩

120. Grant size data unavailable in GtR for studentships. ↩

121. Top 10 universities by number of IP outcomes linked to research grants by the seven research councils, 2010-2020. ↩

122. Definition taken from IP Pragmatics (2020), “Best Practice in Equity Stakes for University Spin-Outs”, a report commissioned by Research England. ↩

123. Net current assets, or “working capital”, describes the value of the TTO’s current assets minus its current liabilities. Larger working capital indicates that the TTO can increase investment whilst continuing to fund its current operations. ↩

124. According to Companies House (except University of Liverpool). ↩

125. Information taken from annual reports published on Companies House. ↩

126. Previously ISIS Innovation Ltd. ↩

127. Previously Lynxvale Ltd. ↩

128. Leeds University Annual Report and Accounts 2019/20, page 21. ↩

129. https://www.virtualengineeringcentre.com/about-us/, “investing in protection and development of intellectual property”. ↩

130. Previously The University of Manchester I3 Ltd, and The University of Manchester Intellectual Property Ltd. ↩

131. Previously Edinburgh Research and Innovation (ERI). ↩

132. Previously Alta Innovations Ltd. ↩

133. Source: GrowthBusiness, https://growthbusiness.co.uk/british-investors-ready-to-invest-1-4bn-in-university-spinouts-20117/. Queens University Belfast and the University of Warwick were unable to provide figures to GrowthBusiness. ↩

134. In 2021 BGF invested £50m into 18 new early-stage businesses of which 11 were either spin-outs or had very close coupling to research previously undertaken in university. It also did a further c.£45m in follow-on investments into existing early-stage portfolio companies, many of which are university spin-outs. BGF expects to maintain the current level of investment in this area. ↩

135. Total amount Scottish Enterprise invested in companies in 2020/21, including spinouts and non-spinouts, through Scottish Co-Investment Fund (SCG) and Scottish VEnture Fund (SVF). ↩

136. Total invested via High Growth Spinout Programme (HGSP) over past 10 years, which has led to the creation of 28 companies. ↩

137. Total number of companies invested in, including 177 deals made in 2020/21 in, covering all equity investments, not just university spinouts. ↩

138. https://www.mewburn.com/law-practice-library/trade-marks-use-it-or-lose-it. ↩

139. See S Torrisi, et al (2016), ‘Used, Blocking and Sleeping Patents: Empirical Evidence from a Large‐scale Inventor Survey’. ↩

140. Directive (EU) 2015/ 2436/ EU of the European Parliament and of the Council of 16 December 2015 to approximate the laws of the Member States relating to trade marks (Recast) [2015] OJ L 336/ 1 ↩

141. INTA Presidential Task Force on Brands and Innovation, Final Report to INTA Board of Directors (9 March 2015). ↩

142. Squicciarini (2011), “Universities’ trade mark patterns and possible determinants”. ↩

143. However, international agreements such as the Madrid System for the International Registration of Marks reduce the number of applications required to achieve trade mark protection across multiple jurisdictions. ↩

144. Intellectual Property Office (2011), “Trade Mark Cluttering: An Exploratory Report Commissioned by IPO” ↩

145. Ibid. ↩

146. G von Graevenitz, C Greenhalgh, C Helmers, and P Schautschick, Trade Mark Cluttering: An Exploratory Report (UKIPO, 2012), 42. ↩

147. Ibid. ↩