Understanding the evolution of the entrepreneurial university. The case of English Higher Education institutions

2.1 Isomorphic pressures versus heterogeneous evolution paths: Towards an entrepreneurial university

The entrepreneurial university literature has developed over the years with different disciplinary approaches (e.g., Clark, 1998; Guerrero, Cunningham, & Urbano, 2015). The model has been portrayed as a step in the natural evolution of the university system (e.g., Rothaermel, Agung, & Jiang, 2007; Uslu, Calikoglu, Seggie, & Seggie, 2019; c.f. Martin, 2012). According to Etzkowitz (1998), universities have undergone two academic revolutions: a first revolution characterised by the institutionalisation of the research university, followed by a second academic revolution incorporating economic development as part of their mission¹ (Etzkowitz, 1998). In this context, universities are seen to be placing a higher priority on being relevant and responsive to broader stakeholder needs, and these efforts have resulted in a progressive institutionalisation of third mission activities alongside teaching and research (see Charles, Kitagawa, & Uyarra, 2014), including the objective of ‘improving regional or national economic performance as well as the university's financial advantage and that of its faculty’ (Etzkowitz & Leydesdorff, 2000, p. 313). The more recent concept of the engaged university (Goddard, 2009) goes even further and advocates that the third mission of economic development should be a guiding and integral principle of the organisation and practice of universities and not just a separate strand of activities.

DiMaggio and Powell (1983) developed the idea of institutional isomorphism. This perspective helps us understand the recent transformation of universities towards becoming more entrepreneurial, driven by pressures to access additional funding sources, and by government third mission policies and incentives encouraging collaboration between universities and external partners (Abreu & Grinevich, 2013; Guerrero et al., 2015; Kitagawa, Sánchez-Barrioluengo, & Uyarra, 2016; Laredo, 2007). Greater competition for funding as well as government incentives for the entrepreneurial turn could therefore be seen as top-down coercive, normative and mimetic isomorphic forces (DiMaggio & Powell, 1983) acting upon universities. This leads to our first proposition: universities have been pushed towards the entrepreneurial university model to meet external macro environmental demands by adopting similar practices and internal changes despite their institutional diversity and organisational differences.

The implicit portrayal of the entrepreneurial university model as an inevitable, homogeneous and isomorphic development path (Etzkowitz & Leydesdorff, 2000) has recently been put into question (Philpott et al., 2011; Tuunainen, 2005). Scholars have highlighted the tensions and contradictions that are likely to emerge between different university missions and activities and argued that the degree and form of this entrepreneurial transformation is likely to vary across countries and types of universities (see, e.g., Huyghe & Knockaert, 2014; Jacob, Lundqvist, & Hellsmark, 2003; Martinelli, Meyer, & Tunzelmann, 2008). In a study of Spanish universities, Sánchez-Barrioluengo (2014) identified strong differences in the performance and capabilities of universities to balance teaching with the new third mission. In the case of United Kingdom (UK), Degl'Innocenti et al. (2019) found a positive and non-linear effect between efficiency in generating income from industry engagement and research performance. However they found that for old universities this effect did not go beyond a certain threshold, whereas new universities appear to be in a better position to improve research by incentivising industry engagement. Case studies have also shown the lack of a unified approach regarding the appropriateness of the third mission, as well as clear tensions and differences across the sector on the meaning and type of entrepreneurial engagement (Benneworth & Jongbloed, 2010; Marginson & Considine, 2000; Philpott et al., 2011). Arguably, universities with different organisational heritage play different roles, reflecting institutional priorities, cultures and governance structures, as well as specialisation and research intensity (Abreu et al., 2016; Abreu & Grinevich, 2013; Hewitt-Dundas, 2012; Perkmann, King, & Pavelin, 2011).

Paradoxically, government policies to promote third mission activities² have been applied rather uniformly with little account for these differences (see, e.g., Hewitt-Dundas, 2012). This naturally generates a number of tensions, not least the ability and capacity of universities to balance a broad range of new tasks seemingly added on to their traditional core missions of teaching and research. As De la Torre, Rossi, and Sagarra (2018, p. 11) argue, ‘not all HEIs are equally equipped and prepared to interact with industry, and initiatives designed to get HEIs to work with business […] might not work with all of them’. It could therefore be argued that not all universities would respond to external pressures and incentives in the same way or interact with stakeholders in a similar fashion. Using indicators of research, teaching and engagement activities, De la Torre et al. (2018) found that institutional profiles influence the types of activities universities engage in and the stakeholders they respond to. In turn, activities influence resources and stakeholders have preferences for different activities, in a mutually interdependent and recursive way. Universities also respond to pressures and perceived challenges in their environments by adapting and developing their own third mission strategies. Based on documentary analysis of third mission strategies, Kitagawa et al. (2016) for instance showed how English HEIs strategically and deliberately target different areas of activities, partners and geographical areas based on their perceived strength. As suggested by Martin (2012), different types of universities' functions evolve differently over time as they respond to changing social, political and economic environments. Jacob et al. (2003) similarly argue that the transition towards an entrepreneurial university is an evolutionary process that takes several years because both infrastructural and cultural changes are necessary. By combining different set of missions, capabilities and resources universities build relationships and legitimacy over time, which would ultimately lead to third mission configurations that are highly differentiated rather than isomorphic.

In the light of this, we can argue that any analysis of the evolution of the entrepreneurial university needs to start from an acknowledgement of the heterogeneous nature of HEIs. Diverse pathways can co-exist within the ‘entrepreneurial university’, and hence no one-size-fits-all entrepreneurial model exists (Benneworth, Pinheiro, & Sánchez-Barrioluengo, 2016; Philpott et al., 2011; Sánchez-Barrioluengo, 2014). This leads to our second proposition: the choice of KE activities reflects individual heterogeneous evolution paths in terms of their positioning within the specific Higher Education sector. However, there is still limited understanding of the ways in which universities reorient their third mission over time (Charles et al., 2014; Hewitt-Dundas, 2012; Kitagawa et al., 2016; Klofsten et al., 2019). Evidence tends to be limited to particular types of universities (e.g., research intensive), narrow sets of activities (e.g., around commercialisation), or involve cross-sectional rather than temporal analysis of third mission engagement. In the next sections we aim to provide a broader analysis of the evolution and reconfiguraiton of KE activities of HEIs.

2.2 Unpacking third mission activities and the stakeholders involved in KE

The types of third mission activities within individual institutional contexts are wide ranging (Guerrero & Urbano, 2012; Huyghe & Knockaert, 2014; Kenney & Goe, 2004; Philpott et al., 2011; Rothaermel et al., 2007). These interactions have been variously referred to as entrepreneurial activities, academic entrepreneurship, knowledge transfer, academic engagement and KE activities. In the remainder of this paper we use the term knowledge exchange (KE) as it better captures the broad-encompassing and diverse nature of the third mission activities (Hayter, Rasmussen, & Rooksby, 2018; Perkmann & Walsh, 2007), reducing the linear conceptualisation of the highly used technology transfer denomination (Bradley, Hyter, & Link, 2013) and highlighting a bi-directional exchange of knowledge between academic and non-academic actors (Roper & Hirth, 2005). Unlike the narrower term of academic entrepreneurship, it acknowledges interactions that go beyond commercial benefit, including engagement with the public sector and non-governmental organisations. Scholars have demonstrated that KE mainly occurs through softer or open channels such as publications and consultancy activities (Cohen, Nelson, & Walsh, 2002; Perkmann & Walsh, 2007), student placements and generally the production of graduates as human capital development (Faggian & McCann, 2009; Iammarino & Marinelli, 2011) rather than ‘hard’ commercial activities such as patenting, licensing and spin-off activities (Philpott et al., 2011).

Despite this breadth of activities, there is a tendency to give privilege to the commercialisation of research results and the protection of intellectual property (IP) emanating from universities (such as patents and licences), neglecting other types of KE activities which can be less visible (or less easily quantifiable), but equally or even more important (Breznitz & Feldman, 2012; D'Este & Patel, 2007; Hughes & Kitson, 2012). Concerns have also been raised about the appropriateness of expecting all universities to effectively undertake the same narrow set of commercialisation activities (Hewitt-Dundas, 2012; Rosli & Rossi, 2016), particularly considering that most of the income from patenting is heavily skewed towards a few patents in specific sectors in a small number of universities (Mowery & Nelson, 2004), and that a too heavy focus on IP may even hinder collaboration by imposing additional transaction costs on firms (Nelson, 2001).

University–society interactions often entail the use of several activities simultaneously (Levy, Roux, & Wolff, 2009) and there is an increased tendency for universities and businesses to forge longer-term, strategic alliances encompassing a range of links closely tailored to the needs of the companies, rather than single, ad-hoc links (Geiger & Sá, 2008). Activities such as training, internships and consultancy tend to go hand in hand and generally enable the development of capacities to initiate other harder or more formal KE activities (Laredo, 2007).

Therefore the first dimension that we seek to explore is concerned with the breadth, or mix, of KE activities that universities are involved in. In the UK, Hewitt-Dundas (2012) found strong variation in the degrees and types of knowledge transfer activities across the sector. While highly research-intensive universities tend to focus on the exploitation of IP and maximising returns from research (see also Guerrero et al., 2015), low research-intensive ones tend to focus mainly on activities related to human capital development. Hussler, Picard, and Tang (2010) compared the approaches to academic valorisation activities in three different regions in Italy, Germany and China. They observed strong differences in terms of the key actors, with a strong focus on small and medium enterprises (SMEs) in Milan vis à vis medium and large firms in Chongqing in China, and the types of instruments used to support KE, with greater use of direct mechanisms for the adoption/exploitation of academic research like spin-offs, mobility of human capital or training programmes in Chongqing compared to the European regions. Another study in China (Eun, Lee, & Wu, 2006) found that the nature of academic entrepreneurship was affected by the low absorptive capacity of industrial firms and the underdeveloped intermediary institutions.

The second dimension of KE activities recognises the diversity of partners involved in these interactions, including with industrial partners but also with the public and third (not for profit) sectors (Hughes & Kitson, 2012), and their different inclinations and incentives to engage with (regional) universities. Different types of stakeholders may have different preferences in terms of activities and types of universities. For instance, large companies tend to be more attracted to work with a university because of its research reputation, while small firms tend to demand more routine services and consultancy, which are more likely to be sourced from their local universities (see also Pinto, Fernández-Esquinas, & Uyarra, 2015; Siegel, Wright, & Lockett, 2007).

Although universities are increasingly encouraged to facilitate KE within their regions, the extent to which universities engage with regional partners has been found to be contingent on the mode of interaction, the type of university (e.g., age and research intensity) and the characteristics of the region, including the presence of innovation support structures and structural characteristics of firms (D'Este & Iammarino, 2010; Hewitt-Dundas, 2012; Lawton Smith & Bagchi-Sen, 2010). Newer universities tend to give engagement with regional stakeholders a greater priority compared to more established ones (Uyarra, 2010). Huggins, Johnston, and Stride (2012) also found that more established universities in the UK were more likely to interact with a more diverse range of organisations and with organisations located outside their own region. The latter was also true for universities located in more competitive regions.

The studies reviewed above emphasise the complex and multi-layered nature of universities' third mission, evidenced by the diverse KE activities they undertake, and the many relations they forge over time with different partners. This study aims to provide a contextualised understanding of the above propositions by analysing the nature of the third mission, in terms of the breadth of KE activities and stakeholders involved, across different types of universities in England over the years. England, and the UK as a whole, is characterised by a large and diverse Highter Education sector, and a long-standing policy interest in incentivising the third mission of universities.

Since the late 1990s the UK government has sought to encourage university–society interactions by funding third mission activities at universities. The main vehicle for KE policy in England has been a series of funding allocated to individual Higher Education institutions (HEIs) including the Higher Education Innovation Fund (HEIF). These policy efforts, and the more recent emphasis on generating research impact, has led universities to increase their focus on delivering benefits from research and brought about a considerable expansion of KE infrastructure and capabilities in HEIs (PACEC, 2009), including the setting up of intermediaries such as technology transfer offices in order to exploit academic outputs and facilitate the knowledge and technology transfer from universities to the private sector (Decter, Bennett, & Leseure, 2007).

Universities in the UK have also been encouraged to facilitate KE in their regions, supported by regional policies and institutions such as regional development agencies and other intermediaries from the late 1990s and throughout the 2000s (Goddard & Chatterton, 1999; Kitagawa, 2004; Uyarra, 2010). The regional development agenda in England lost momentum however, with the abolition of the English Regional Development Agencies (RDAs) in 2010 and their replacement with smaller and less resourced Local Enterprise Partnerships (LEPs). In addition, the financial and economic crisis (Goddard, Coombes, Kempton, & Vallance, 2014; Hutton & Lee, 2012), led to knock-on effects on universities through reduced investments in innovation of the private sector, and public sector budgetary constraints, reducing demand for services such as consultancy (Charles et al., 2014). Similarly, Higher Education in England underwent further reforms since 2012 with a drastic reduction in teaching-related public funding and the introduction of higher tuition fees for home and European Union (EU) students. These changes in the institutional and economic landscape led to a widely differentiated institutional vulnerability affecting both universities and places (Goddard et al., 2014).

3.1 UK Higher Education sector: Organisational context

The UK Higher Education system is diverse for historical reasons (Goddard et al., 2014; Scott, 2014). These differences can be traced back to the medieval origins of universities such as Oxford and Cambridge, the civic universities created in industrial cities during Victorian times, and subsequent waves of expansion and reform, including the creation of red brick universities during the inter-war years, and new universities in the 1960s and the subsequent incorporation of the Colleges of Advanced Technology in the university sector.

A binary divide (McCormack, Propper, & Smith, 2014) is therefore often made within the UK Higher Education sector between the old universities, founded before 1992, which are typically more research focused, and new universities which were granted university status after 1992 as a result of the Further and Higher Education Act (HMSO, 1992), and also former university colleges that have become universities in recent years. New universities tend to be more teaching focused, and their third mission activities are assumed to be locally oriented given their traditional focus on vocational education and training, and their relatively low engagement in basic research (Charles et al., 2014; Goddard et al., 2014; De la Torre et al., 2018). Boliver (2015) indeed found large differences between the old pre-1992 universities and the new post-1992 universities in terms of research activity, economic resources, academic selectivity and social mix.

However, there is also arguably a further divide between the 24 most research-intensive older universities (known as the Russell Group) and other older universities, and also between newer universities that were former polytechnics (which offered higher diplomas and degrees, often in more technical subjects, that were governed and administered at the national level) and those that were previously further education colleges (McCormack et al., 2014). Given the heterogeneity that exists within the Russell Group of universities (see, e.g., Boliver, 2015) and particularly the dominance of a small number of top universities, sometimes a more fine-grained distinction is made between the latter and the rest of the Russell Group.

In this paper we therefore consider five categories of HEIs based on the historical development and research intensity (see Appendix 1). Within the group of old universities we distinguish three categories: the Top Five, the rest of the Russell Group, and others.³ The Russell Group universities represent less than 15 per cent of the sector in terms of the number of universities but capture around 75 per cent of the total quality-related research (QR) funding granted by Higher Education Funding Council (HEFCE) to universities.⁴ We treat the Top Five universities (Imperial College, Universities of Cambridge, Oxford, Manchester and University College London) as a separate group from the rest of the Russell Group based on the distribution of research funding: these five universities receive a disproportionate share of QR funding by HEFCE (32% in 2014–15). Within the new universities, two groups are identified: former polytechnics consisting of HEIs which were originally established as polytechnics under local authority funding and control, and converted to university status since 1992, and ‘other new universities and HEIs’, which includes HEIs that were granted university status after 2004, primarily former Further and Higher Education colleges, specialist colleges and current Higher Education colleges.

While acknowledging the limitations of adopting such broad categories, the different groups of universities capture the diverse nature of Higher Education sector in England and are consistent with other studies highlighting the institutional differences across the sector. For example, Boliver (2015) concludes that a binary divide persists between old (pre-1992) universities, characterised by higher levels of research activity, greater wealth, more academically successful and socioeconomically advantaged student intakes, and newer (post-1992) institutions. McCormack et al. (2014) suggest that research-intensive universities see themselves competing in international and national markets (for staff and students) while newer universities focus more on local markets. This is consistent with De la Torre et al. (2018)'s findings of a long-standing division between traditional universities and former vocational education institutions, in terms of specialisation, teaching and research intensity, and stakeholder engagement.

3.2 Source of information and variables

Our analysis draws on Part B of the HEBCI Survey for the academic years 2003/4 to 2011/12. HEBCI is an annual survey carried out since 2001, now administered by the Higher Education Statistics Agency. The questionnaire collects data on a broad range of third mission activities encompassing the contributions of universities to both economy and society, covering all the HEIs in England, Scotland, Wales and Northern Ireland. More specifically, HEBCI collects information on a range of third mission or third stream activities, defined there as: a set of selected knowledge exchange (KE) activities in which a university/HEI strategically engages as an institution.

The key KE activities used in this paper are: collaborative research (collaborations), consultancy (consultancy), contract research (contracts), facilities and equipment related services (facilities), continuing professional development and continuing education (CPD), IP activities including shares, sales (patents and licences) and spin-offs (spin-offs). These activities are frequently used in the literature to capture the relationship between universities and other actors in the society (see for example Hewitt-Dundas, 2012 or Guerrero et al., 2015). Table 1 presents a detailed description of the selected variables for the analysis as well as descriptive statistics for the whole period. Due to differences in the nature of the variables, our analysis uses normalised variables by year.⁵

We examine 107 out of the 130 English HEIs (176 in the UK) covered in the HEBCI survey. We exclude from the analysis HEIs for which no information was available for the whole time-window as well as those HEIs solely specialised in Arts and Design. In terms of the institutional types described in previous section, the breakdown of our population is as follows: 45.8 per cent pre-1992 universities (5% Top Five, 14% Other Russell Group and 27.1% Other Old) and 54.2 per cent post-1992 universities (29% Former Polytechnics and 25.2% Other New HEIs). Within these types, and in order to contextualise our findings, we complement our results with some illustrative examples of third mission strategies produced by universities.⁶

3.3 Methodology for empirical analysis

In order to better understand the evolution of KE activities, we employ two quantitative techniques. First, we perform a factor analysis based on a principal components technique with Kaiser Normalisation (Hair, Anderson, Tatham, & Black, 1998) to extract different modes of KE interaction by universities (Appendix 2 includes results of the factor analysis). We calculate the factor scores based on a least squares regression approach in order to predict the location of each individual on the factor (Thurstone, 1935). Under this process, the computed factor scores are standardised to a mean of zero and the standard deviation equal to one because the principal components method is used (DiStefano, Zhu, & Mîndrilă, 2009; Tabachnick & Fidell, 2001). Conceptually, factor scores represent the degree to which each university scores high on the group of items with high loadings on a factor. Thus, higher values of the variable with high loadings on a factor will result in a higher factor score (Hair et al., 1998). Subsequently, the standardised factor scores obtained are represented in a line chart in order to characterise the evolution of third mission interactions across universities. This information is used also to understand how universities prioritise their third mission engagement, including the ways in which they have shifted their focus and strategic activities over the years (third dimension of the model).

Second, the evolution of KE activities of universities is captured by means of temporal graphs depicting (a) the annual growth rates in income from different types of actors involved and (b) the share of regional income out of all income from KE. With this information, we look at the actors and the regionally embedded activities. The average annual growth rate (AGR) is calculated by dividing the slope by the income. The slope is determined by the regression line formed by the matrix corresponding to the years of study 2003/04–2011/12 and income raised by the universities by the type of partners.

4.1 Different forms of KE

Our factor analysis⁷ of KE activities yields five different categories, namely: research-oriented activities, facilities, consultancy, training and spin-offs (Table 1, last column). The five factors explain almost 70 per cent of the total data variability, which is considered as satisfactory for social science studies (Hair et al., 1998). In order to check the factors' internal reliability, we calculate the Cronbach alpha index. The coefficients for all groups are around 0.6,⁸ which is also considered as a satisfactory value (Hair et al., 1998). This categorisation of university activities offers a more nuanced approach to the grouping of KE activities compared to broader categories used in the literature such as soft/hard activities (Philpott et al., 2011), commercialisation activities (Abreu & Grinevich, 2013; Huyghe & Knockaert, 2014; Jain, George, & Maltarich, 2009) or entrepreneurial activities (Guerrero & Urbano, 2012; Guerrero et al., 2015) and is in line with the approach from Hayter et al. (2018) suggesting the broad variety of well-established pathways for KE.

In order to understand the evolution of interactions by different types of universities, the obtained factor scores are graphically described in Figure 1(a)–(e) and described in the sections below. The results show how different types of HEIs engage in different mixes of KE activities over time. The figures present standardised values and positive (negative) values of factor scores can be interpreted as the number of standard deviations of each group of universities in specific KE activities (factors). The higher the standard deviation is, the higher the distance with the population average and the more focused universities are in this particular activity compared to others. These standard deviations illustrate how efforts of universities vary within the Higher Education  sector but also how universities have re-configured and changed their third mission activities over the years.

As shown in Figure 1, Russell Group universities are more focused on research-oriented KE activities such as collaborative research, contract research and generation of IPs, which clearly distinguish themselves from the rest of HEIs because they are the only group with positive standard deviations for these activities. However, results suggest that within the Russell Group there are some differences. While the Top Five  universities (Figure 1a) show an increasingly pronounced focus on research-oriented activities compared to the other groups, The Rest of the Russell Group universities are not so specialised (standard deviation is close to 1 for the full period), balancing instead different activities and using facilities and training increasingly as forms of engagement (Figure 1b). This specific research-oriented mission of some Top Five  universities is evident in their third mission strategy. For instance, the University of Manchester strategically integrates translational research and IP exploitation as part of research activity instead of separating them as third mission.⁹

‘Other Old’ universities do not stand out in any factor (Figure 1c) and they reflect a mix of entrepreneurial activities throughout the entire period as their factor scores are very close to the population average for all the activities (values close to 0). A possible explanation for this finding is that within the Other Old universities, historically two types of institutions can be distinguished, both of which were founded during the 1960s. One type comprises the ‘plate glass universities’ founded in the 1960s (e.g., Aston, Essex, Lancaster, Sussex, Warwick,¹⁰ York) and the former Colleges of Advanced Technology that were converted into universities in the 1960s (e.g., Brunel, Cranfield, Loughborough). This latter group of universities show certain entrepreneurial characteristics with their strong historical links with industry partners. These universities tend to be relatively small and, in their KE strategies, demonstrate a wide range of KE activities as strategic areas, including collaborative research and exploitation; CPD; delivering high levels of graduate employability, entrepreneurial behaviour and enterprise; and innovative supports for SMEs and new business. This feature supports the findings from the HEBCI data showing a broad range of KE activities with low specialisation.

Post-1992 universities exhibit quite different behaviour: in general, they focus on all activities, but less so on the more research-oriented ones (they have the lowest standard deviations for these activities). Within this group, Former Polytechnics have increased their efforts in consultancy since 2007/08 (Figure 1d), while Other New HEIs appear to have accelerated the development of university spinoffs particularly in the last two years (Figure 1e).¹¹ Institutional strategy presents clearly focused and defined areas of KE activities, for example, the University of Derby (Other New), which demonstrates a strong focus on work-based learning with business partners, and has a strategic target on student start-ups by linking the employability agenda and inclusivity as an institution.

We thus observe that English HEIs seem to be intensifying their selectivity and specialisation in particular types of KE activities. Universities learn, build relationships with the partners and gain legitimacy in engaging certain types of activities over time. Research-intensive universities have, in the period studied, increased their share of income from research-oriented KE activities, as well as harder commercial activities such as licensing. New and generally less research-intensive universities have increased their share of KE income from softer activities such as consultancy and facilities.

4.2 Stakeholders involved in KE activities

The picture of an increasingly differentiated Higher Education sector is reinforced when we analyse the evolution of KE activities with different types of non-academic partners. In order to do this we distinguish the income derived from different types of stakeholders in the studied period. Specifically, we analyse differences in interactions between universities and SMEs, non-SMEs and non-commercial organisations—such as government bodies and third sector organisations—using the information on income from collaborative research, contracts, consultancy, facilities and licences (see Table 2 for more details).¹² Income evolution is presented in Figure 2a–e. We also calculate the evolution of the total income coming from the interaction with each mentioned partner as well as the annual growth rate for the full period (AGR) and for two sub-periods: 2003/04–2007/08 (the period preceding the economic crisis, AGR1) and 2007/08–2011/12 (AGR2), see Table 2.

Within this overall trend, there are clear differences in engagement with different partner types. While income with all types grew during the period (6% overall), income from KE activities with non-commercial entities grew more on average than income with firms (8.5% and 2.5% respectively). In contrast, engagement with SMEs increased before 2007/08 and declined afterwards (−1.2%).

However, within this broad picture we can observe stark differences across institutional types. KE income increased in the period for all groups, particularly for Top 5 and the Rest of Russell Group universities (on average the annual growth has been 10.3% and 5.5% respectively). Russell group universities (Top Five and the Rest of Russell Group) exhibit the biggest growth in income from KE activities with non-commercial organisations (16% and 7.2% respectively) whilst Former Polytechnics experienced the greatest increase in income from SMEs (7.5% compared with less than 2.5% for other universities). Top Five universities benefited from the largest increase in income from large firms (5%), while Other New universities experienced the biggest drop in KE income from large firms (−4.2%). Consequently, some of the new universities started to take an explicit strategy to reach out to new private sector clientele in specific fields, like the case of Middlesex University, as a response to the decline in funding from large firms as well as public sector organisations.¹³

Although growth patterns in KE income remained positive for Russell group universities after 2007/08 (particularly for the Top Five), the rest display low (for Other Old universities) or even negative income growth (for post-1992 universities). This slow or negative growth is explained mainly by a drop in KE income from large firms (particularly in the case of Other New and Other Old universities) and SMEs (particularly for Former Polytechnics). Other New universities also experienced a significant reduction in KE income from non-commercial organisations (−4% compared with an 11% increase in the previous period considered).

Our analysis shows a significant drop in income growth from all actor types, namely, SMEs, large firms and non-commercial organisations, from 2007/08 onwards. The decline in income from KE interactions with firms (industry helix) and non-commercial organisations (the government helix) may reflect the effects of the economic crisis and the following public sector funding cuts. This reduction is particularly severe for newer universities, which tend to be reliant on KE income from SMEs and public sector organisations, particularly at the local level. The trend towards greater selectivity and differentiation has intensified since 2007/08, which coincides in time with the economic slowdown as well as with a series of policy changes influencing regional development and Higher Education landscapes in England as described in Section 2.

This once again shows that old and more research-intensive universities appear to exhibit greater resilience and adaptation to policy changes and economic conditions. They have been able to diversify their funding sources away from the public sector and towards private sector opportunities, for instance through strategic partnerships with large private firms.

This in turn has implications in terms of universities engagement with regional actors. In order to understand this, we conduct an analysis of the regional (share and total) KE income (from KE activities such as contracts, consultancy, facilities and licences). Results are presented in Figure 3. In addition, the AGR is estimated for the regional KE income as well as for its share of the total KE income (see also Table 2). The data shows a slight increase (4.1%) of income from regional interactions in the full period, however the regional share of the total income decreased by 2.3 per cent. Further, we can observe stark variations in the two periods analysed. The positive trend (9.7%) in engagement with regional stakeholders in the first period (2003/04–2007/08) contrasts with a reduction of the regional income for HEIs of around −0.3% after the crisis. All universities exhibited a positive growth of regional KE income in the first period, with Former Polytechnics being the ones with the higher growth (18.9%), followed by Top Five  universities having increased regional income by 17.1 per cent. The decline has been shaper in recent years with the exception of the Rest of Russell Group that is the only group with a positive growth in both periods (5.8% in 2003/04–2007/08 and 3.9% in 2008/09–2011/12). For the rest, after 2007/08 all groups reduced their interaction with regional actors, both in terms of their overall regional engagement and as a proportion of total KE income (see Figure 3). This decline in the total and the share of regional engagement has been particularly marked for Former Polytechnic universities (−7.1% and −4.8% respectively) and Top 5 (−3.1% and −11.8%).

We can therefore observe that while total income rose markedly in the period, this was dominated by non-regional interactions, particularly in the period after the crisis. In turn, regional relations differ markedly across university types. Russell Group universities perform a smaller proportion of their KE activities in their region compared to other universities such as former polytechnics. This is not to underestimate the contribution of the former to the regional economy via KE activities which amounted to a combined income of near £14 million. However, we found that the relative importance of the region in KE diminished in the period for all universities, particularly for Top 5 universities but also, more surprisingly, for new universities traditionally considered to be more locally oriented. These universities have been more affected by public spending cuts, which translate into diminished resources to fund KE (particularly with SMEs) and lower demand for KE by the public sector, particularly by local government, and by changes in the governance of economic development in England. Locally oriented Former Polytechnics faced greater vulnerability as a result of policy changes including the disappearance of RDAs, whilst research-intensive universities are able to rely on other sources within and beyond regional boundaries (Charles et al., 2014). Despite this landscape, in the university strategy documents, Former Polytechnics demonstrate a strong commitment to local economic development agenda and engagement with local SMEs, seizing this as a new opportunity, especially at the time other support mechanisms are disappearing. On the demand side, firms and the public sector may have also responded to the economic crisis by refocusing their knowledge sourcing strategies and being more selective in their interactions with a fewer number of universities, or reducing their KE activities altogether (see Coates-Ulrichsen, 2014).