Industry–academic links: A new phase in Ireland’s FDI-led industrialisation strategy

Abstract

We analyse industry–academic links in the context of a dual economy (or disarticulated industrial structure) in Ireland, as an example of a peripheral territory in the EU. The duality found in the Irish industrial structure is the result of a FDI-led industrialisation strategy which has resulted in two distinct economic sectors – foreign and indigenous, respectively – with weak interactions between the two. Through increased public funding of academic research, the Irish government aimed to attract and embed new waves of higher-value foreign direct investment and increase the dynamism of its indigenous enterprise base. Based on a combination of quantitative and qualitative data, the paper analyses a crucial aspect of Ireland’s recent emphasis on STI policy – industry-academic linkages – and finds that the measures introduced reproduce in the public research system the uneven development found in Ireland’s productive system between indigenous industry and the foreign-owned industrial base.

Keywords

Dual economy foreign direct investment industry-academic links Ireland science and technology policy

Introduction

The establishment of dynamic academic research systems with entrepreneurial activities and strong links to industry has been at the forefront of policy initiatives in the area of science, technology and innovation (STI) in a number of EU countries since the 1980s. This increased policy emphasis on the importance of scientific and academic research for economic growth and competitiveness is a consequence of a number of factors, amongst which are perceived changes in the sources of innovation and economic dynamism, as well as in our understanding of the way innovation takes place. The widespread adoption of concepts such as the ‘knowledge’ and ‘learning’ economy (Lundvall and Johnson, 1994), ‘Triple Helix’ (Etzkowitz, 1998; Etzkowitz and Leydesdorff, 2000), ‘Mode-2 knowledge production’ (Gibbons et al., 1994) and new ‘Open Innovation’ models (Chesbrough, 2003) are expressions of the way in which the growing importance of scientific knowledge in technological change, as well as the increasingly networked character of innovative activities, have informed discussions about the nature of academic research systems and their role in economic development. It is in this context that national and regional STI research systems and the potential knowledge transfer activities they offer industry have become a central concern of industrial policy, as university research institutions are increasingly seen as playing a strategic role in industrial development (Mowery and Sampat, 2005). Despite the importance of scientific research for technological development, the relationship between advances in scientific knowledge and industrial innovation is not a linear one but depends on the historical development of countries and regions, as well as a number of industry and firm-specific factors. As a result, industry–academic links may influence different sets of firms in differential ways, thereby influencing the industrial structure of regions and countries. Examples of this are countries and regions in which FDI is an important source of industrial upgrading, but significant differences exist between indigenous and foreign-owned firms in their ability to engage with and benefit from university–industry links.

This paper explores changes in the patterns of university–industry linkages in the Republic of Ireland following STI policy initiatives introduced from the late 1990s. Ireland is an interesting case of a small, late-industrialising country on the periphery of the European Union, with an active industrial policy based on attracting higher-value FDI activities. Although during the 1990s Ireland’s FDI-led industrialisation model was perceived as a successful strategy, there remained ongoing concerns about the existence of a ‘dual economy’ characterised by productive but weakly embedded foreign-owned firms and an indigenous sector dominated by SMEs in mature, low-technology industries (Culliton Report, 1992; Enterprise Strategy Group (ESG), 2004; Telesis Report, 1982). Since the end of the 1990s, in the context of an intensification of international competition for FDI, the development of an internationally competitive academic research system capable of generating knowledge relevant for industrial innovation, along with dynamic university–industry links, has become a central plank of Ireland’s policy efforts both to attract and embed new waves of higher-value FDI activities and to strengthen the R&D efforts of indigenous industry. Given the importance of university–industry links in Ireland’s STI and industrial development strategy, our empirical study of the dynamics of these linkages sheds an interesting light on the impact of this central element of STI and industrial policy on countries and regions with disarticulated industrial structures.

We use mixed methods research to analyse university–industry linkages, drawing on two unique data sources. Quantitative data come from the Irish Innovation Panel, which provides information on (inter alia) university linkages and innovation performance of firms in Ireland over the period 2000–2008. The qualitative data are drawn from interviews with industry representatives and senior policy-makers responsible for the design and implementation of industrial and STI policy. The use of mixed methods research allows us to analyse not only the extent and nature of linkages between different types of firms and the academic research base following the change in policy, but also the rationale and anticipated consequences of the change in STI policy.

We find that the new Irish STI system prioritises the needs of firms with relatively advanced scientific and technological capabilities. These firms tend to be subsidiaries of MNCs. The system is less relevant for existing indigenous firms which find it difficult to engage with research being undertaken in academic institutions. We argue that, as presently conceived, the measures introduced deepen the uneven development found in Ireland’s productive system, potentially increasing the knowledge gap between indigenous and foreign-owned industry. We maintain that the deepening of this mis-match is not inevitable and can be reversed by a STI policy that is designed to support the needs of different types of firms and which creates the conditions for deeper linkages between the two sectors of industry.

Importance of university–industry links in NIS

The importance of public sector research systems as a source of innovation and industrial development has received increasing attention as policy-makers from countries and regions at various stages of economic development have turned to academic research institutions in their search for industrial renewal and economic dynamism (Mazzoleni and Nelson, 2007; Mowery and Sampat, 2005). One of the factors which explains the greater attention given to academic research institutions is the rise of a number of ‘science-based industries’, such as biotechnology, ICT and nanotechnology, where advances in science are often a necessary condition for technological innovation (Nelson, 1993; Pavitt, 2001) and where the emergence of new enterprises has been closely associated with the rise of university spin-out firms (Kenney, 1986). This is one of the reasons why policy initiatives that prioritise the practical relevance of publicly funded research and its accessibility to industry via university–industry links, often at the expense of ‘open science’, remain contentious areas (Cohen et al., 2002; Pavitt, 2001).

At the same time, as interest in the role of science as a source of innovation has grown, important changes have taken place in the ways in which innovative activities are organised, as firms have increasingly moved away from isolated in-house efforts towards networked or ‘Open Innovation’ models (Chesbrough, 2003). Whilst for most industries, the main focus of interest is the nature of relationships between firms and their suppliers and buyers, in the case of science-based industries attention has centred on the efficiency with which the transfer of knowledge from university research institutions to the rest of the economy takes place. It is in this context that academic research systems have increasingly been seen as playing a strategic role in industrial development (Etzkowitz, 1998; Mazzoleni and Nelson, 2007; Mowery and Sampat, 2005).

National and regional science and technology systems and their potential for knowledge transfer have also become an important element in policy efforts to attract MNCs’ higher-value activities, above all as the limitations of ‘branch-plant’ investment have become clear (Phelps et al., 2003; Young et al., 1994). Two trends are important here. The first is the increasing tendency for MNCs to locate important elements of their R&D functions outside their home countries and the rise of internationally dispersed R&D networks (Cantwell, 1995; UNCTAD, 2005; Zanfei, 2000). These new patterns of location and organisation of international R&D activities have been largely driven by the desire of MNCs to access the internationally dispersed centres of scientific excellence, as well as their need to understand diverse markets (Cantwell, 1995; Florida, 1997). The second trend has been the intensification of competition amongst governments to attract FDI, as multinational corporations are seen as important vehicles for technology diffusion (UNCTAD, 2005). For countries and regions which look to FDI as a source of industrial upgrading, the need to develop internationally competitive academic research systems as a means of embedding foreign investment in networks of collaboration with indigenous science and technology institutions has become an important element of FDI promotion policies. Nevertheless, the relationship between FDI and STI policies and their impact on national innovation systems (NISs) are not as yet well understood.

Despite the importance of scientific advances for technological innovation, research shows that the effect of scientific research on industrial development is not linear and will depend above all on the following three factors: type of firms, type of industry and the size and stage of development of countries. Next, each of these three factors are discussed in turn.

Type of firm

Empirical research indicates that firms vary in their propensity to establish direct linkages with academic research institutions according to firm characteristics and strategies. In the UK, the number of firms that draw directly from research carried out at universities is relatively modest (Laursen and Salter, 2004). However, firm size appears to be an important factor in both the US and the UK (Cohen et al., 2002; Laursen and Salter, 2004). The nature of firms’ search strategy, whether ‘open’ to external sources of knowledge or ‘closed’, was also found to be important, with firms adopting ‘open’ strategies more likely to draw from university research (Laursen and Salter, 2004; Tether and Tajar, 2008). Existing in-house R&D activities were also found to be a significant factor explaining the likelihood of links with academic research (Laursen and Salter, 2004; Tether and Tajar, 2008), suggesting that the development of a firm’s internal absorptive capacity (Cohen and Levinthal, 1990) is a necessary pre-condition to be able to benefit from the public-science and technology system.

One important distinction between firms and their ability to engage with and benefit from the academic research system that does not seem to have been studied is the difference between firms that are subsidiaries of MNCs, and therefore have a base in more than one regional or national innovation system, and indigenous firms, which are more closely rooted in (and may be limited to) one innovation system (Costa and Filippov, 2008). This distinction may be a significant one because although MNCs may be an important part of a host country or region’s innovation system, MNC subsidiaries are also part of global knowledge networks. As a result, the knowledge creation and innovative capabilities of these firms may be quite different from those of indigenous firms. They are also likely to have different needs and place different demands on national and regional research institutions.

Industry factors

Studies show that the degree and the ways in which academic research contributes to R&D also vary according to industry (Cohen et al., 2002; Laursen and Salter, 2004; Tether and Tajar, 2008). In an influential study of US firms, Cohen et al. (2002) found that only in pharmaceuticals and biotechnology did scientific research have a substantial direct impact on innovation. In most industries the impacts of scientific advances were indirect and the principal mechanisms used to access the results of scientific research were scientific publications, consultations and informal communication between industrial and academic scientists, training of researchers and development of new methods and instrumentations (Gibbons and Johnson, 1974). In the case of the UK, Tether and Tajar (2008) found that low-tech manufacturers and all service sector firms except for technical services were less likely to have links with the public-science base.

Regional and national factors

A number of scholars have argued that the impact on industry of investment in scientific research will differ depending on how countries’ and regions’ histories have shaped the nature of linkages between firms and knowledge institutions, as well as whether countries and regions are small or large, high-income or catching up (Lundvall and Borras, 2005; Pavitt, 2001; Tödtling and Trippl, 2005). As a consequence the role of university research in national and regional innovation systems and its impact on economic performance will vary according to the structure of national and regional economies (Cooke, 2005; Mowery and Sampat, 2005). On this basis, Lundvall and Borras (2005) and Mazzoleni and Nelson (2007) argue that in big, high-income countries and regions such as the US and parts of Europe, the focus will tend to be on establishing scientific and technological capacity at the technological frontier, whilst in less developed areas it might be a question of establishing science and technology institutions that support the capability to use, absorb and adapt modern technologies.

The point to stress is that competencies are unevenly distributed amongst firms, so practices for developing, absorbing and using new technology are not immediately diffused throughout regional and national economies. This has implications for which types of technologies and sectors thrive in different regional and national contexts. Although all firms need to undertake activities related to learning and upgrading, including activities related to R&D, the technological challenges facing nations, regions and firms at different stages of development are different. As a result, in the case of the EU, Liagouras (2010) points to the need for caution in the introduction of ‘high-tech’ or ‘best practice’ policies, as important differences exist in industrial structures and stages of development. So, for example, for EU laggard countries (which include ‘middle-income’ countries such as Ireland), the main challenge is not to make technological breakthroughs but to rapidly diffuse and adapt new technologies developed elsewhere. In this context, rather than the imitation of ‘best practice’ innovation policies developed in the ‘advanced economies’, which tend to be concerned with opportunities at the technological frontier, policies which focus on the assimilation and diffusion of existing technologies through the establishment of institutions which act as effective intermediaries between the public research base and firms may be most appropriate (Tödtling and Trippl, 2005). Our concern therefore is to gain a greater understanding of the impact of a central aspect of European STI policy, that of university–industry linkages, on an industrial structure characterised by significant disparities.

Science, technology and innovation policy in Ireland

The establishment of an internationally competitive academic research system to underpin the development of industry–academic links became a central element of Ireland’s industrial development strategy from the late 1990s. Although policies to encourage university–industry linkages were not new to Ireland, efforts to induce a radical step-change in the Irish NIS based on a major transformation of the academic research system can be seen in the substantial increases in funding commitments to this area. Whilst the National Development Plan (NDP) (1994–1999) committed €0.5 billion to STI, the NDP (2000–2006) increased this funding to €2.5 billion and the NDP (2007–2013) allocated €8.2 billion. Despite major reductions in public sector spending following the 2008 crisis, important elements of the STI programme have been protected from austerity measures. The significance of the new policies, which prioritise the strengthening of the indigenous STI system, has to be understood in the context of the need for a strategic shift in industrial policy. Until the late 1990s, policy efforts at industrial upgrading focused on attracting FDI in the higher-value functions of ‘high-technology’ industries based on Ireland’s location within the European market, the existence of a low-cost but relatively well-educated English-speaking workforce, low tax rates and a number of investment and training subsidies (MacSharry and White, 2000). This policy of ‘industrialisation by invitation’ underpinned the period of rapid economic growth of the 1990s during which Ireland became known as the ‘Celtic Tiger’. Towards the end of the 1990s, however, it became increasingly clear that in the context of increasing competition for FDI from low-cost regions in Easter Europe and Asia, increasing labour costs in Ireland and changes in the EU state-aid regime which meant an end to generous state grants to MNCs, new measures to attract FDI and industrial upgrading were necessary. The search for a new development strategy coincided with a growing awareness of the importance of strong national and regional research systems for attracting FDI at a time when technologically developed MNCs were increasingly collaborating with national and regional science and technology institutions.

The policy initiatives introduced from the late 1990s revolve around a number of programmes designed to create a ‘world-class research system’ through the funding of individual researchers, capital expenditure on state-of-the-art research facilities, the establishment of academic research centres with industrial partners (the CSETs) and the operation of research networks such as the Strategic Research Clusters. A new state agency, Science Foundation Ireland (SFI), was specifically created to administer the new STI funds, whilst Enterprise Ireland (EI), the industrial development agency responsible for the development of indigenous industry, was given the task of establishing new structures within the academic research system to foment the commercialisation of intellectual property and the spawning new firms. Initially much of the funding – above all, that disbursed by SFI – was directed to the two sectors identified by a Technology Foresight Exercise as of greatest strategic importance for Ireland, ICT and biotechnology, with a third sector – sustainable and energy efficient technologies – added in 2008. More recently EI, along with the Industrial Development Agency (IDA; responsible for FDI), have started a new programme of Competence Centres, where businesses come together to commission research from the academic system.

There is no doubt that the measures introduced since 1999 have fundamentally transformed the academic system, with the establishment – for the first time in the history of the Irish state – of a relatively well funded, peer-review research system. As a result of these measures, Ireland has been able to attract important groups of both Irish and international scientists to its universities. Central to many of these programmes, above all those associated with SFI, EI and the IDA, has been the objective to strengthen the knowledge base of industry through collaboration with the academic research system. Consequently the intensification of university–industry linkages has become an important element of Ireland’s policy efforts to attract and embed MNCs’ research activities, as well as to increase the R&D spending of existing indigenous firms. The creation of a new generation of start-up firms based around the commercialisation of the scientific and technological advances produced by the academic research system is another objective. Given the importance of university–industry links in Ireland’s STI strategy, we use both quantitative and qualitative data to analyse the character and dynamics of these linkages for the period 2000–2008 as an indicator of the success of this central element of STI policy. In the context of Ireland’s ‘dual economy’, we are particularly interested in examining whether a distinction exists in the extent and nature of linkages between the academic research system and multinational subsidiaries and indigenous firms, respectively.

Research methods and data

We base our study on mixed methods research, which refers to work that integrates quantitative and qualitative research approaches. There are a number of advantages associated with mixed methods research, such as its attempts to consider multiple viewpoints and perspectives and the fact that it allows for the triangulation of different data (Johnson et al., 2007). From our point of view, one of the strengths of mixed methods research is that it allows the elaboration and clarification of the results of one method with results from other methods. In this respect, mixed methods research can provide a more complete picture and a better understanding of the phenomenon under study. We use quantitative data to examine whether there is any evidence of differences between foreign-owned and indigenous firms in university linkages after 1999, and how these linkages have changed through time. Qualitative data are used to examine and interpret the nature of industry–academic links and the reasons why they differ across these two distinct populations of firms, and to interpret the policy intent behind the post-1999 STI changes.

The data come from two unique data sets. The quantitative data come from successive waves of the Irish Innovation Panel (IIP), which provides information on the innovation activities of manufacturing plants in Ireland and Northern Ireland over the period 1991–2008. This involves a series of plant-level surveys conducted every three years using similar survey methodologies and questionnaires with common questions. Like the EU Community Innovation Survey, each of the six IIP surveys covers the innovation activities of manufacturing business units over a three-year reference period. Here we make use of data from the last three waves of the IIP, covering the period following the STI policy introduction in Ireland. The survey covering the 2000 to 2002 period was undertaken between November 2002 and May 2003 and achieved an overall response rate of 34.1%. Subsequent surveys covering the 2003–2005 and 2006–2008 periods achieved response rates of 28.7% and 38%, respectively. The resulting panel is unbalanced, reflecting non-response in individual surveys but also the opening and closure of individual plants (for further details see Roper et al., 2008). The IIP covers both Ireland and Northern Ireland, but here we restrict the analysis to observations from the Republic of Ireland, employing a total of 1282 observations.

A common feature of the IIP is the following question: ‘Over the last three years did you have links with other companies or organisations as part of your product or process development?’ Plants responding positively to this were then asked to identify the types of external partners with which they were working. A total of eight potential partner types were identified in the questionnaire: customers, suppliers, competitors, joint ventures, consultants, universities, industry-operated laboratories and government-operated laboratories. In the present paper we are concerned solely with the data on linkages to universities, both in Ireland and elsewhere.

The qualitative data are based on semi-structured interviews with senior policy-makers, including the chief executives of Forfás (Ireland’s policy advisory board for enterprise and science) and the Office of Science and Technology, as well as representatives from the Irish Software Association and the Irish Bioindustry Association. All interviewees had participated in the strategic discussions and decision-making processes that have shaped Ireland’s industrial and STI policies and were therefore in a good position to explain both the motivation and the impact of initiatives in these areas. The data were collected during the period from 2002 to 2005 from Enterprise Ireland (EI), the Industrial Development Agency (IDA), the Office of Science and Technology (OST) and Forfás and Science Foundation Ireland (SFI), as well as from the Irish Research Council for Science, Engineering, and Technology (IRCSET) and the Irish Council for Science, Technology and Innovation (ICSTI), the body that advises government on science and technology policy. Further interviews were conducted with EI and IDA in 2011 in order to update the data after the introduction of major austerity measures. A total of 27 interviews were carried out.

Results

Based on an analysis of the quantitative data, the section below examines differences in industry–academic linkages between indigenous and foreign-owned firms, as well as the characteristics of firms more likely to develop relationships with universities. After this, qualitative data are used to address policy-related factors that account for the differences in linkages between the two sets of firms. Bringing these two datasets together allows us to analyse the actual effectiveness of policy measures on industry–university linkages and to compare the actual impact of these initiatives with the stated objectives of policy-makers when introducing measures to enhance industry–university links.

Industry–academic linkages: differences between indigenous and foreign-owned firms

The objective is first to determine whether there is any difference in the extent of university linkages between foreign subsidiaries located in Ireland and indigenous Irish plants, and whether the extent of linkages has changed since 2000. In addition, we wish to explore whether any identified differences are specifically driven by foreignness per se, or whether some other underlying features of foreign and indigenous plants explain the extent of university linkages. We carry out the analysis separately for linkages to any university and linkages specifically to Irish universities.

Table 1 shows the proportion of university linkages recorded for all three waves of the IIP, and performs a simple t-test of whether foreign plants were more or less likely to have university linkages than indigenous plants. Results suggest that (unsurprisingly) university linkages are carried out by a minority of plants in Ireland, both foreign and indigenous, but that the extent of this interaction has risen through time. More specifically, for all university linkages (the top panel of Table 1), the proportion of foreign plants having linkages is significantly greater than the proportion of indigenous plants for all three waves of the IIP. In addition, for both sets of plants the proportions with linkages rises steadily, with a substantial jump in 2006–2008. However, foreign plants’ proportion of linkages rises faster, so the gap between foreign and domestic plants’ linkages rises through time. For universities in Ireland (bottom panel of Table 1), the proportion of foreign plants having linkages is greater than the proportion of indigenous plants for all three waves, but the difference is significant only in the last period (2006–2008). The time pattern is slightly less consistent here, with a fall in the proportion of linkages in the middle period (2003–2005) and a large rise in 2006–2008 for both sets of plants.

Table 1.

Proportion of plants with knowledge linkages to universities.

	Period	Domestic-owned plants	Foreign-owned plants	t-test (H0, difference=0)	p-value of t-test
Linkages with universities in general	2000–2002	0.092	0.171	2.25	0.025
Linkages with universities in general	2003–2005	0.097	0.197	2.79	0.005
Linkages with universities in general	2006–2008	0.132	0.365	3.17	0.002
Linkages with universities in Ireland	2000-2002	0.074	0.106	1.11	0.266
Linkages with universities in Ireland	2003–2005	0.023	0.04	0.91	0.366
Linkages with universities in Ireland	2006–2008	0.122	0.296	2.54	0.012

Source: Irish Innovation Panel.

Overall the descriptive statistics appear to suggest that foreign plants are consistently more likely to have university linkages, and there is some evidence that the gap between the proportion of foreign and domestic plants with linkages has risen over the period 2000–2008. However, foreign subsidiaries in Ireland are very different from indigenous plants in terms of size, industrial sector, R&D intensity, and so on (Roper et al., 2008). Once we allow for these differences, is there still any evidence of differences in university linkages? Our objective here is not to come to a definitive account of the determinants of university linkages, but to allow for a reasonable set of control variables which shows the apparent differences between indigenous and foreign plants in a sharper light. We therefore run a panel probit estimation where the dependent variable is a dummy taking the value 1 if the plant has a relevant university linkage and zero otherwise. Following Roper et al. (2008), we allow for aspects of the plant’s internal resource base and aspects of absorptive capacity in addition to its location of ownership and industry controls; specifically, we include dummy variables indicating whether the plant conducted R&D in-house (an indicator of internal knowledge generation and absorptive capacity) and whether or not it engaged in exporting (a measure of international orientation which may permit ‘learning by exporting’ effects; see for example Salomon and Jin, 2008); employment and its square (logged); the proportion of workforce with a degree (a measure of labour skills) and the age of the plant. Industry and time-period controls are also included in the estimation.

Results are shown in Table 2. For both estimations (all universities and universities in Ireland), R&D, exporter status and proportion of workforce with a degree are significantly and positively associated with the likelihood of university linkages. Since the estimates are marginal effects we can say that, on average, having R&D in-house increases the probability of a plant having a university linkage of any kind by 15.6%, and increases the probability of a linkage with an Irish university by 7.8%. In the case of linkages to Irish universities, there is also evidence that larger and younger plants are more likely to have linkages. However, for both estimations, ‘foreign-owned plant’ is not significant. In other words, once the effects of in-house R&D, skill intensity and exporting (and possibly size) are allowed for, there is no longer a difference between the likelihoods of a foreign or a domestic plant having university linkages.

Table 2.

Probit model of linkages with universities (marginal effects).

Dep. var:	Model 1:	Model 2:
	Plant has linkages to universities (0,1)	Plant has linkages to universities in Ireland (0,1)
R&D conducted in-house	0.156^*** (0.023)	0.078^*** (0.019)
Exporter dummy	0.048^** (0.024)	0.031^* (0.019)
Foreign-owned plant	0.013 (0.026)	−0.002 (0.021)
Ln employment	−0.021 (0.056)	0.085^** (0.04)
Ln employment squared	0.008 (0.006)	−0.007 (0.005)
Workforce with degree (%)	0.001^* (0.0006)	0.0007^* (0.0004)
Establishment age (years)	0.0001 (0.0003)	−0.0004^* (0.0002)
Industry control variables	Yes	Yes
Time control variables	Yes	Yes
Observations	1120	1120
Pseudo R2	0.171	0.237

Notes and sources: Dependent variable is dummy indicating linkages to universities. Estimation by panel probit model. Marginal effects reported. Standard errors in parentheses. ^* p < 0.1, ^** p < 0.05, ^*** p < 0.01. Source: Irish Innovation Panel. Period: 2000–2008, three waves of the IIP.

Taking the results of the t-tests and probit estimations together suggests that foreign plants are more likely than indigenous plants to have university links and this gap is increasing through time; however, the reason is not necessarily that they are foreign but rather that they are more likely to have the characteristics that encourage university linkages. As shown in Table 3, foreign-owned plants in the sample tend to be larger, more likely to export and more skill and innovation-intensive than their Irish counterparts; foreign plants also spend more on R&D on average, although they are no more likely to carry out R&D. Our results suggest that the effect of these characteristics on university linkages has become stronger through time, and it is the effect of these resource and absorptive capacity indicators that underlies the apparent difference in the university linkage patterns of foreign and domestic plants.¹

Table 3.

Weighted averages of key variables by ownership status.

	Domestic-owned plants	Foreign-owned plants

R&D conducted in-house (0,1)	0.47	0.48
Expenditure on R&D (Euro)	82	504
Share of new or modified products in sales (%)	20.56	29.91
Exporter dummy	0.41	0.76
Ln employment	3.26	4.21
Workforce with degree (%)	10.91	18.88
Establishment age (years)	31.77	32.68
Observations	811	309

Source: Irish Innovation Panel. Period: 2000–2008, three waves of the IIP. Note: weighted averages.

Policy-related factors that account for differences in industry–academic linkages between indigenous and foreign-owned firms

Our interview data identified a number of factors that help interpret the distinct pattern of university–industry linkages between Irish-owned firms and MNCs found in the quantitative data analysis. It appears that the strategy adopted by SFI of prioritising research at the scientific and technological frontier in a limited number of research areas has made it difficult for indigenous firms to engage with the work undertaken in the academic research system. In particular, we find that three factors – the areas of research chosen for funding, the degree of complexity of the research projects selected and the amount of resources needed by firms to participate in research collaborations – account for the relatively weak linkages with universities amongst indigenous firms and the increasing gap between indigenous and foreign affiliates. The STI strategy adopted by SFI is explained by the need to create an internationally competitive academic research system capable of attracting MNCs’ research activities. We discuss these factors in more detail below.

SFI strategy

As the main funding body for STI, SFI has had a central role in shaping the new academic research system. The agency was established to bring about a radical change in Ireland’s research culture by creating ‘basic research capability’ in Irish universities; however, the concrete mandate that was given to SFI, and which influenced the criteria by which research projects were chosen, was far more focused (ICSTI, 1999). Initially only two research areas could be funded² (see NDP, 2000–2006) and emphasis was placed on research aimed at establishing capacity at the technological frontier rather than projects which enhanced technological diffusion and assimilation (Interview SFI-1).

Our interviews with various government bodies and industry representatives indicate that the measures introduced by SFI were driven by the need for a new strategy to attract high-value FDI through the development of a national research system (ESG, 2004; Inter-Departmental Committee on Science Technology and Innovation (IDCSTI), 2004). For the academic research system to play this role it had to be built on robust and globally recognised notions of scientific and technological ‘excellence’ established through an internationally recognised peer-review system. At the same time, the strategy was vague about the balance between ‘blue-sky’ research, which is usually distant from practical application, and applied research, which is closer to commercial use. In practice much of SFI’s effort has been directed at the selection of research projects close to the technological frontier which pass clearly established notions of quality and excellence. Examples of these initiatives include the Centres for Science, Engineering and Technology (CSETs), where SFI funds 80% of the research costs of a Centre whilst industrial partners are required to fund the remaining 20%. By 2011 nine CSETs had been funded, with industrial partners such as IBM, GlaxoSmithKline and Intel. Though the CSETs do not exclude the participation of indigenous SMEs, interviews with SFI indicate that they have found it hard to involve indigenous SMEs in these initiatives (SFI-2).

A relevant issue is who sets the research agenda for SFI-funded projects and the nature of the criteria used to select proposals. In the case of individual projects, it is academic scientists who set the research agenda and proposals are selected by SFI on the basis of scientific excellence through peer review. In the case of the CSETs, the research agenda is also set by academic researchers, although this has to be supported by the industrial partners. The selection of the CSETs to be funded is also based on scientific excellence, but includes a strategic review by the industrial development agencies in order to ensure the CSETs represent strategic value for Ireland. In the case of the Strategic Research Clusters (SRC) set up to strengthen research networks across academia but which have the explicit aim to develop linkages with industry, whilst it is academics who set the research agenda, this is meant to be ‘industry-influenced’ (interview EI-3-2011).

Impact of SFI-funded research on academic–MNC links

All interviewees – those involved in the policy-making system as well as industry representatives – highlighted the importance of establishing an internationally recognised research system within the context of Ireland’s new strategy for attracting higher-value FDI activities. Interviews with IDA representatives illustrate the importance of the new STI policy:

The work of SFI has certainly been very positive because it’s part of the infrastructure. For example, we bring senior managers from MNCs into Ireland and show them around and we bring them to an appropriate CSET to demonstrate the type of work that is going on there. So, even if they are not ready to engage with a CSET they are comfortable to know that it is there and it is doing that type of work. … Ten years ago this was not there. (IDA-2-2011)

From this point of view, SFI’s emphasis on the notion of ‘research excellence’ is considered crucial, as an increasing number of countries are also competing for FDI on the basis of collaborative research with national research institutions (Interviews OST, IDA and Forfás).

Though the IDA remains enthusiastic about the CSETs and the research collaborations between MNCs and academic researchers established under SFI, our interviews in 2011 revealed a more cautious attitude with respect to the effectiveness of these initiatives in comparison with our interviews in 2002–2005. The main concern was that the research carried out in universities was too far away from practical application to be useful to many of the MNCs investing in Ireland (IDA-2-2011). As a result, although there were important instances of deep collaboration between SFI-funded research and MNCs, the level of MNC engagement had not met the IDA’s initial expectations. As will be seen below, this has led to new initiatives regarding academic–industry links where it is the firm, rather than academics, that sets the research agenda.

Impact of SFI-funded research on linkages between academia and Irish-owned firms

Our interviews indicate that there are three main reasons why indigenous firms, which are mainly SMEs, have found it difficult to engage with and benefit from SFI-funded projects. One important reason is that most indigenous-owned firms are in traditional industries related to mechanical engineering and manufacturing, rather than the three industrial sectors prioritised for SFI funding, and firms in these sectors perceive that there is little in the academic research system of relevance to them (EI-3-2011). Figures from Forfás (2008) which analyse enterprise data according to ownership confirm this picture. For example in 2000, 46.4% of indigenous firms’ sales were in food/drink and tobacco, 34.6% were in traditional manufacturing and engineering industries and software accounted for 5% of sales. No mention is made of indigenous sales in the areas of pharmaceutical or biotechnology. On the other hand, foreign-owned firms dominate three sectors: chemicals and pharmaceuticals (where biotechnology is relevant), which accounted for 26.6% of sales; electrical and electronic equipment, accounting for 31.8% of sales and software, accounting for 17% of sales. The importance of these three sectors, and therefore of foreign-owned firms in the Irish economy, can be seen by the fact that in the 1990s they accounted for over 75% of Irish output and 90% of exports (Department of Enterprise, Trade and Employment [DETE], 2003).

A second reason explaining the weak linkages between Irish-owned industry and the academic research system is that the type of projects funded by SFI aimed to be close to the technological frontier and were too distant from the existing capabilities of Irish-owned firms (this is the case even for Irish-owned firms operating within the industrial sectors funded by SFI). Our interviews indicate that in the case of the ICT sector, Irish software firms do not at present have the absorptive capacity to engage with the type of research being conducted in the academic system.

I think Science Foundation Ireland doesn’t understand the issues that small companies have… This is self evident if you just look at the projects that have been funded… In terms of ICT, I have been very disappointed with the performance of SFI because in the first instance [the funding] went fundamentally into the basic sciences, physics… and in terms of the companies I deal with that is about as irrelevant as you can get. (EI-2)

According to the Irish Software Association and ICSTI, the long-term nature of research programmes and the requirement that industrial partners contribute to the funding is a challenge for indigenous SMEs and is a third reason why Irish-owned firms are not establishing links with the academic research system. Therefore although there are no conscious attempts to block Irish-owned SMEs’ participation in industrial partnerships such as the CSETs, the way in which the programmes are designed is too long-term focused and expensive for indigenous firms. Overall our interviews indicate that, although there are important exceptions, the perception exists that SFI-funded programmes have not been very effective in terms of knowledge transfer to indigenous SMEs.

The creation of a new generation of ‘high-potential-start-up’ (HPSU) firms founded on the basis of the output of the academic research system was initially identified as an area in which the new STI initiatives could result in a strengthening of indigenous industry (Interviews SFI, OST, EI-2). As a result, the activities of EI turned towards assisting the commercialisation of research output and the creation of academic spin-out firms (EI-1 and 2). In 2005, approximately 19% of the HPSUs created in Ireland came out of the academic research system. In 2011 this figure was approximately 15%, indicating that SFI-funded projects had not yet led to a renewal of Ireland’s indigenous industrial base, as was initially hoped (EI-3-2011).

The establishment of Competence Centres by EI and IDA

As awareness of the challenges involved in establishing academic–industry links and the long-term time frame of the economic impact of these collaborations has increased, the attention of IDA and EI has shifted towards the creation of nearer-market types of academic–industry collaborations in Competence Centres (CC). CCs, created as part of the government’s STI strategy for 2006–2012, are groups of companies that come together on a non-competitive basis to commission research from the academic system. What distinguishes CCs from other SFI initiatives is that it is the industrial partners who set the research agenda, with the academic system working as a provider of research services to industry. The funding for the research work is provided by EI in the case of indigenous industry and IDA funds a maximum of 40% of research projects involving MNCs. The research is relatively close to market (usually about 3–5 years from practical application); if it was longer than three years from market, firms – above all SMEs – would not participate (EI-3-2011; IDA 2011).

The establishment of CCs (and to a lesser extent the CSETs) explains the increase in IDA funding for R&D activities in Ireland. Whilst in 2000 approximately 10%–15% of projects funded by IDA were in the area of R&D, in 2011 this figure had risen to 35%–40% of projects. Most of this research activity is taking place in MNCs that were already established in Ireland, indicating that companies with production centres are now locating some of their R&D activities in Ireland (IDA-2011). In the case of indigenous industry, SME involvement in CCs was more limited (though there are examples of indigenous ICT and medical devices firms who are involved in CCs). According to our interview with EI (EI-2011), one reason for this is that the type of knowledge and expertise found in the academic research sector is not perceived as useful by existing indigenous SMEs. Moreover, attempts to make CCs instances of knowledge transfer between MNCs, Irish-owned SMEs and the academic research system have proved to be difficult because the problems which the two types of firms need to solve are very different.

Discussion and conclusion

Our results show that the STI policies introduced in Ireland since 2000 have resulted in an increase in academic–industry links on the part of both MNCs and indigenous firms, but that it is foreign-owned firms that have benefited most from the new academic research system, increasing the knowledge gap between indigenous and foreign-owned firms. This differential pattern of linkages with universities can be explained by the different characteristics of indigenous firms compared to foreign-owned subsidiaries and the particular STI strategy adopted by SFI, which has made it difficult for the indigenous sector to engage with the academic research system.

The quantitative data show that firms with higher R&D, skills and export intensity have created more linkages with universities. In Ireland these firms tend to be subsidiaries of MNCs, which are also deeply embedded in the global science and technology networks of their parent firms and have therefore developed substantial internal capabilities outside the Irish STI system. In contrast, indigenous firms, with the exception of those in the software and medical devices sectors, tend to be concentrated in less R&D, skills and export-intensive industries. This dual nature of the Irish production system is important, as the competencies and absorptive capacity of these two sets of firms are very different, and so is the nature of the support they require from the STI system. It also means that advanced knowledge and good practices will not diffuse easily between these two sectors – an important issue as the gap in university linkages between the foreign and indigenous sectors appears to be growing.

The particular STI strategy adopted by SFI is also central to the differential pattern of academic–industry linkages. The main objective behind the policy initiatives in STI was the creation of a national research system internationally recognised for excellence in work at the scientific and technological frontier. This choice of strategy was driven by the need to develop new ways of attracting and embedding high-value FDI in Ireland in the context of increasing international competition for this investment. The system established by SFI and the projects it chose to fund, however, made it difficult for indigenous firms to engage with the research activities performed in universities. This is because of the limited number of science and technology sectors being funded, the distance between the complexity of the knowledge base of the projects chosen and the absorptive capacity of indigenous enterprises and the high costs and long-term payback time of these partnerships. The evidence provided by the qualitative data discussed above suggests that this was an unintended outcome of the STI policy initiative; nevertheless, the lack of measures designed to enhance the diffusion and adoption of technological advances by less developed indigenous firms is an important factor in the failure of SFI’s effort to promote greater linkages between the academic research system and indigenous firms.

The danger posed by the particular set of measures designed to enhance industry–academic links in Ireland is that they are likely to reinforce the existing dual character of the country’s industrial structure and deepen the knowledge gap between an indigenous sector with low innovative and absorptive capabilities and foreign-owned firms embedded in global innovation networks. This duality has been identified as a factor that has historically undermined sustainable economic growth (Culliton, 1992; Telesis Report, 1982). The experience of the CCs, which have found it difficult to access expertise in the university research system that is of relevance to indigenous firms, suggests the need to establish bridging institutions between the academic research base and indigenous industry, in the form of research institutions able to support the learning efforts of indigenous SMEs. This points to the need for a twofold research system capable of catering to the distinct needs of the two different industrial sectors, but with the longer-term aim of building effective linkages between them. The establishment of a more complex research system however seems unlikely in the context of Ireland’s present economic difficulties. The fact that the budget for academic research has been relatively protected from austerity measures is an indication of the importance Irish policy-makers place on the academic research base in the process of industrial renewal. The protection of the science budget, however, has been accompanied by intensified demands for greater commercialisation of the results of the academic research system (Department of Jobs, Enterprise and Innovation, 2012), and is seen by sectors of the academic community as a potential threat to the objective of research excellence set out in initial policy initiatives.

The Irish experience is an interesting illustration of the tensions – identified by Power and Malmberg (2008) – that exist between STI policy driven by notions of ‘excellence’ which are determined by global standards and the need for national and regional development, above all in countries and regions with laggard or disarticulated industrial structures. Where a significant gap exists between the knowledge base and level of competency of different types of firms, an effective STI system needs to cater to the distinct requirements of these different sectors. The overall aim of such a system, however, has to be the creation of effective linkages that allow knowledge flows between different sectors of the productive structure, as well as with regional and national institutions for knowledge creation and diffusion. In the case of academic–industry links, the Irish experience shows that a variety of research institutions with different research objectives (e.g. some closer to the technological frontier and others prioritising more catch-up activities) may be required to ensure effective knowledge flows between the academic research system and the various parts of the industrial system. From a regional point of view, the study suggests that differentiated innovation policies are required not only between different regions, as suggested by Tödtling and Trippl (2005), but also within regions. The alternative is significant dislocation between parts of the productive and STI systems, hindering effective knowledge flows between different sectors of the economy, with negative implications for future sustainable growth.

Footnotes

Funding

The research on which this paper is based was partly funded by the ESRC under award RES-062-23-2767.

Notes

References

Cantwell

(1995) The globalisation of technology: what remains of the product cycle model? Cambridge Journal of Economics 19: 155–174.

Chesbrough

(2003) Open Innovation. Cambridge. MA: Harvard University Press.

Cohen

Levinthal

(1990) Absorptive capacity: a new perspective of learning and innovation. Administrative Science Quarterly 35: 128–152.

Cohen

Nelson

Walsh

(2002) Links and impacts: the influence of public research on industrial R&D. Management Science 48(1): 1–23.

Cooke

(2005) Regionally asymmetric knowledge capabilities and open innovation: Exploring ‘Globalisation 2’—A new model of industry organisation. Research Policy 34: 1128–1149.

Costa

Fillippov

(2008) Foreign-owned subsidiaries: a neglected nexus between foreign direct investment, industrial and innovation policies. Science and Public Policy 35(6): 379–390.

Culliton Report (1992) A Time for Change: Industrial Policy for the 1990s: Report for the Industrial Review Group. Dublin: Stationary Office.

Department of Enterprise Trade and Employment (DETE) (2003) Review of Industrial Performance and Policy. Dublin: DETE, Stationary Office.

Department of Jobs Enterprise and Innovation (2012) Putting public research to work for Ireland. Available at: http://www.enterprise-ireland.com/EI_Corporate/en/Research-Innovation/Companies/IPP-Putting-public-research-to-work-for-Ireland.pdf

10.

Enterprise Strategy Group (ESG) (2004) Ahead of the Curve: Ireland’s Place in the Global Economy. Dublin: Forfás.

11.

Etzkowitz

(1998) The norms of entrepreneurial science: cognitive effects of the new university-industry linkages. Research Policy 27: 823–833.

12.

Etzkowitz

Leydesdorff

(2000) The dynamics of innovation: from national systems and ‘Mode 2’ to a triple helix of university-industry-government relations. Research Policy 29(2): 109–123.

13.

Florida

(1997) The globalisation of R&D: results of a Survey of Foreign Affiliated R&D Laboratories in the USA. Research Policy 26: 86–103.

14.

Forfás (2008) Annual Business Survey of Economic Impact, 2007. The National Policy and Advisory Board, for Enterprise, Trade, Science, Technology and Innovation. Available at: www.forfas.ie

15.

Gibbons

Johnson

(1974) The role of science in technological innovation. Research Policy 3: 220–242.

16.

Gibbons

Limoges

Nowotny

Schwatzman

Scott

Trow

(1994) The New Production of Knowledge. London: SAGE.

17.

ICSTI (1999) Technology foresight Ireland: an ICSTI overview. Irish Council for Science, Technology and Innovation. Dublin: Forfás, ICSTI, April 1999.

18.

Inter-Departmental Committee on Science Technology and Innovation (IDCSTI) (2004) Building Ireland’s Knowledge Economy: the Irish Action Plan for Promoting Investment in R&D to 2010. Dublin: Forfás.

19.

Johnson

Onwuegbuzie

Turner

(2007) Toward a definition of mixed methods research. Journal of Mixed Methods Research 1(2): 112–133.

20.

Kenney

(1986) Bio-Technology: The University-Industrial Complex. New Haven, CT: Yale University Press.

21.

Laursen

Salter

(2004) Searching high and low: what types of firms use universities as a source of innovation. Research Policy 33: 1201–1215.

22.

Liagouras

(2010) What can we learn from the failures of technology and innovation policies in the European periphery? European Urban and Regional Studies 7(3): 331–349.

23.

Lundvall

B-A

Borras

(2005) Science, technology, and innovation policy. In: Fagerberg

Mowery

Nelson

(eds) The Oxford Handbook of Innovation. Oxford: Oxford University Press, pp. 599–631.

24.

Lundvall

B-A

Johnson

(1994) The learning economy. Journal of Industry Studies 1(2): 23–42.

25.

MacSharry

White

(2000) The Making of the Celtic Tiger: The Inside Story of Ireland’s Boom Economy. Dublin: Mercier Press.

26.

Mazzoleni

Nelson

(2007) Public research institutions and economic catch-up. Research Policy 36(10): 1512–1528.

27.

Mowery

Sampat

(2005) Universities in national innovation systems. In: Fagerberg

Mowery

Nelson

(eds) The Oxford Handbook of Innovation. Oxford: Oxford University Press, pp. 209–231.

28.

National Development Plan (NDP) (2000–2006) Government of Ireland, Stationary Office, Dublin.

29.

Nelson

(ed.) (1993) National Innovation Systems: A Comparative Analysis. New York: Oxford University Press.

30.

Pavitt

(2001) Public policies to support basic research: what can the rest of the world learn from US theory and practice? (and what they should not learn). Industrial and Corporate Change 10: 761–779.

31.

Phelps

MacKinnon

Stone

Braidford

(2003) Embedding the multinationals? Institutions and the development of overseas manufacturing affiliates in Wales and North East England. Regional Studies 37: 27–40.

32.

Power

Malmberg

(2008) The contribution of universities to innovation and economic development: in what sense a regional problem? Cambridge Journal of Regions, Economy and Society 1: 233–245.

33.

Roper

Love

(2008) Modelling the innovation value chain. Research Policy (37): 961–977.

34.

Salomon

Jin

(2008) Does knowledge spill to leaders or laggards? Exploring industry heterogeneity in learning by exporting. Journal of International Business Studies 39: 132–150.

35.

Telesis Report (1982) A review of industrial policy. National Economic and Social Research Council (NESR), Report no 66, February.

36.

Tether

Tajar

(2008) Beyond industry-university links: sourcing knowledge for innovation from consultants, private research organisations and the public science-base. Research Policy 37: 1079–1095.

37.

Tödtling

Trippl

(2005) One size fits all? Towards a differential regional innovation policy approach. Research Policy 34: 1203–1219.

38.

UNCTAD (2005) World Investment Report: Transnational Corporations and the Internationalization of R&D. Geneva: United Nations.

39.

Young

Hood

Peters

(1994) Multinational enterprises and regional economic development. Regional Studies 28(7): 657–677.

40.

Zanfei

(2000) Transnational firms and the changing organization of innovative activities. Cambridge Journal of Economics 24(5): 515–542.