This document is a Nesta Working Paper on the effects of innovation network policies. It includes an abstract, an executive summary, a detailed analysis of network policies, evaluation challenges, and a summary of findings and lessons. The document references several studies and reports.

The Effects of Innovation Network Policies

Paul Cunningham Manchester Institute of Innovation Research, University of Manchester

Ronnie Ramlogan Manchester Institute of Innovation Research, University of Manchester

Nesta Working Paper 12/04 March 2012

www.nesta.org.uk/wp12-04

Abstract

This paper is part of the Compendium of Evidence on the Effectiveness of Innovation Policy Intervention. It examines the evidence on the relationship between public support for networks and the impact on innovation. It provides a conceptual background for policy instruments that enhance innovation through the activities of networks, the rationale for their deployment and the main types of approach adopted. Drawing on a review of the available literature, it presents evidence on the impacts of policy interventions on network creation, management and behaviour and – where available – the effects of networks on innovation from a variety of forms of network support. This evidence is analysed against a framework that examines the rationale and goals of network policies and the extent to which expected outcomes have been realised with respect to such issues as collaboration/networking, partnerships, leading edge research, research training and the transfer and exploitation of knowledge and the cost effectiveness and design issues of networks.

From the analysis, two important issues emerge: first, the complexity of networks and the diversity of motivations, rationales, activities, outputs, outcomes and effects make the task of evaluation very difficult. Evaluations tend, therefore, to focus on specific aspects of network behaviour rather than covering the complete set of potential variables; second, the timing of evaluations is, in many cases, a critical issue. Finally, a number of key lessons are presented.

JEL Classification: O38

Keywords: Impact, Innovation policies, Network policies, R&D networks, innovation networks, networks

The Compendium of Evidence on the Effectiveness of Innovation Policy Intervention Project is led by the Manchester Institute of Innovation Research (MIoIR), University of Manchester, and funded by Nesta, an independent charity with the mission to make the UK more innovative. The compendium is organised around 20 innovation policy topics categorised primarily according to their policy objectives. Currently, some of these reports are available. All reports are available at http://www.innovation-policy.org.uk. Also at this location is an online strategic intelligence tool with an extensive list of references that present evidence for the effectiveness of each particular innovation policy objective. Summaries and download links are provided for key references. These can also be reached by clicking in the references in this document.. Corresponding Author: Dr Paul Cunningham, MIOIR, Harold Hankins Building, Manchester Business School, University of Manchester, M13 9PL. Email: [email protected]

The Nesta Working Paper Series is intended to make available early results of research undertaken or supported by Nesta and its partners in order to elicit comments and suggestions for revisions and to encourage discussion and further debate prior to publication (ISSN 2050-9820). © 2012 by the author(s). Short sections of text, tables and figures may be reproduced without explicit permission provided that full credit is given to the source. The views expressed in this working paper are those of the author(s) and do not necessarily represent those of Nesta.

This report is part of the Compendium of Evidence on the Effectiveness of Innovation Policy Intervention Project led by the Manchester Institute of Innovation Research (MIoIR), University of Manchester. The project is funded by the National Endowment for Science, Technology and the Arts (NESTA) – an independent body with the mission to make the UK more innovative.

The compendium is organised around 20 innovation policy topics categorised primarily according to their policy objectives. Currently, some of these reports are available.

All reports are available at http://www.innovation-policy.org.uk. Also at this location is an online strategic intelligence tool with an extensive list of references that present evidence for the effectiveness of each particular innovation policy objective. Summaries and download links are provided for key references. These can also be reached by clicking in the references in this document.

Executive Summary

Since the late 1970s, networks have become an important component of technology and innovation policy in many countries. Networks allow for rapid learning and facilitate the reconfiguration of relationships between suppliers (in the case of companies) and producers of knowledge (other companies or research institutions); they can also stimulate the development of additional cooperative activities including training, technological development, product design, marketing or facilitate knowledge pooling, skills sharing, the sharing of facilities, equipment or datasets and the co-development of programmes of joint research. However there are few reasons to believe that such beneficial cooperative relationships can emerge automatically and such failures provide a strong rationale for public intervention.

Governments may intervene in order to facilitate the establishment or continued development of a network due to the lack of, or insufficient awareness of, the opportunities they present. Intervention may be used to overcome barriers to network formation such as the fear that there may be unfair appropriation of the benefits accruing from collaborative undertakings. In such circumstances, governments are able to offer knowledge sharing frameworks which provide a level of security as a means of reinforcing the mutual trust upon which successful cooperative arrangements rely. Intervention can also be used to guide firms towards network membership (either with other firms or with the science base) in order to overcome technological "lock-ins", to enter a new area or to change management practices.

This report reviews and reflects on the evidence of the relationship between public support for networks and the impact on innovation. It provides a conceptual background for policy instruments that enhance innovation through the activities of networks, the rationale for their deployment and the main types of approach adopted. Drawing on a review of the available literature, it presents evidence on the impacts of policy interventions on network creation, management and behaviour and – where available - the effects of networks on innovation from a variety of forms of network support. This evidence is analysed against a framework that examines the rationale and goals of network policies and the extent to which expected outcomes have been realised with respect to such issues as collaboration/networking, partnerships, leading edge research, research training and the transfer and exploitation of knowledge and the cost effectiveness and design issues of networks.

Two important issues emerged from the analysis. First, the complexity of networks and the diversity of motivations, rationales, activities, outputs, outcomes and effects make the task of evaluation very difficult. Evaluations tend, therefore, to focus on specific aspects of network behaviour rather than covering the complete set of potential variables. Second, the timing of evaluations is, in many cases, a critical issue. Several of the evaluations and reviews considered found it difficult to make quantitative assessments of network effects, largely because many of the outcomes that could be used as proxies for this measure, such as patenting behaviour, had yet to materialise. Moreover, in many cases there was no baseline of existing capabilities and networking from which progress could be measured. Consequently the evaluation of network quality followed from activity analyses and interview responses.

Among the key lessons that have emerged from the analysis are the following:

1. Networks can have very positive effects on the stimulation of learning processes and the enhancement of skills levels.
2. Despite (or because of) the diversity and the complexity of various network forms, there is little evidence (especially quantitative evidence) to explain which forms of network most contribute to innovation or, indeed, whether networks do and, precisely how, lead to innovation.
3. Strong network management and leadership (such as through a board of directors), coupled with transparent and efficient administrative processes are overwhelmingly cited as essential contributory factors for network success.
4. Established (informal) networks, or pre-existing connections and relationships form the optimal basis for the establishment of more formal policy-led initiatives for the creation or development of networks.
5. Network participants need to actively manage their networking relationships; experience and network management competencies can strongly influence the gains to be made from network participation.
6. Policy instruments that facilitate network formation and development (such as support for network brokers or other intermediary organisations) are often successful in achieving these broad objectives. While all firms in a network benefit from being part of a network, the establishment and management costs are borne largely by the network organiser. Public intervention can therefore be used to mitigate this 'free-rider' effect.
7. Networks fail for a variety of reasons, but lack of demand, trust, commitment and excessive bureaucracy seem to be identified as major causes.
8. Government intervention can act as both a positive and negative force affecting the sustainability of particular networks and network infrastructures. It can be very difficult to predict the development path of a network since it can be influenced by unpredictable events or by the unintended effects of other policy actions or the regulatory context. In the absence of a bottom-up process of self-determination, top-down initiatives that select target industries, technologies or scientific fields, may not succeed.

1. Introduction

It is widely understood that innovation does not take place along simple linear lines from research, through invention to commercial product or process but is dependent on a variety of feedback loops "within a context of structured relationships, networks, infrastructures and in a wider social and economic context" (Perrin, 2002). Collaboration in research has become a widespread phenomenon, eliciting a number of studies (most recently in the context of 'open innovation'). Extensive use is now made of collaborative agreements as a mechanism for knowledge sharing and exchange (OECD, 2001). Paradoxically, one of the main driving forces behind innovation is competition among firms. Thus, there is inherent tension between sharing knowledge and protecting it and decisions must be made over which knowledge can be traded and of the benefits such exchange may generate.

Nevertheless, as noted by the European Commission (2003), networking now forms a key aspect of company strategy. Rather than single companies, key innovations and related global businesses have become increasingly developed and dominated by market oriented, value-chain based networks. Similarly, individual products and/or services have been superseded by more integrated systems - the nature of innovations is more systemic. At the same time, the market is seeking packaged solutions in contrast to single technologies or one-off services. Such "packaged solutions (or systemic innovations as they might be called) are developed and produced by networks". Hence, logic dictates that government intervention in support such network activities should represent another element in the innovation policy tool box.

As noted by Freeman (1991), the links by which networks are constructed may range from formal contractual agreements to loosely coupled informal networks. Although formal networks (such as multi-actor research co-operations, joint ventures, etc.) can operate within a (often codified) framework of standardised agreements and commitments between network partners, the 'real business' of knowledge exchange, dialogue and mutual cooperation often operates at the informal level largely through a process of incorporating tacit knowledge into their learning processes. Lundvall (1992) highlights the fact that the growth of distinct organisational networks comprising a variety of actors at all levels in the economy – what may now be termed innovation ecosystems, has accompanied the rapid growth in complex technologies.

Network membership need not be restricted to firms, indeed the encouragement of the formation of research networks presents a convenient way to address another policy goal adopted by many governments, namely that of increasing the exchange of knowledge between actors in the public and private sectors. Public sector actors typically comprise public sector research establishments ('government laboratories') and institutes of higher education (notably universities). Further significant value may be generated by the facilitation of research networks of public sector actors to develop critical research mass, or to encourage multi-disciplinary approaches to address scientific, technological and societal issues, for example.

The above features explain why networks have become an important component of technology and innovation policy in several countries (and indeed at the supranational level). Briefly, the use of policies to foster inter-firm networks originated in Italy in the 1970s. After the re-organisation of the country into 20 administrative regions, the regional government in Emilia Romagna in north-east Italy introduced a number of initiatives to stimulate collaboration among existing groups of companies. Eligibility for support was conditional on companies working in a collaborative network. Within 15 years, these policies had helped raise Emilia Romagna from the seventeenth to the second wealthiest region in Italy and to become the seventh most prosperous region of the EU. The Italian model was then adopted by Denmark, where it was applied through a top down initiative, the Danish Network Programme. Established in 1989, this had the objective of assisting small companies to compete in the Single European Market. A major feature of the programme was the use of brokers to facilitate the development of inter-firm networks. Within eighteen months, around 3,500 Danish companies had become involved in networks. Following the success of the Danish Network Programme, it was adopted in various forms by a number of further countries, including Norway, Australia, the United States, Canada, New Zealand and the United Kingdom. (Martin, et al., 2004) see Table 1 below.

Source: Liston, 1996

A study by the OECD (2001) notes that "by stimulating co-operation among the different actors in the innovation system, policy makers expect that the innovation potential can be better exploited in firms, both existing and new, in research, and in society as a whole". The report concludes that questions remain over the issue of appropriate policies is not clear-cut and that there is a need to clarify "the rationale and instruments for facilitating networking with the aim of generating optimal knowledge circulation and sharing in a context of intense competition". This conclusion remains valid and forms the underlying rationale for this report.

This report, one of a series produced under the NESTA Compendium of Evidence on the Effectiveness of Innovation Policy Intervention, will first focus on setting the broad conceptual background for policy instruments that focus on the enhancement of innovation through the activities of networks, the rationale for their deployment and the main types of approach adopted. We also provide a simplistic definition by which we distinguish networks from related innovation policy concepts such as clusters and R&D collaboration. This is followed by an overview of the available literature, both in the form of evaluation reports and in secondary academic and grey literature which explicitly present or reflect on the variety of forms of network support and on the evidence for its impact on innovation. Next, we will organise the available evidence according to the nature of the impacts that have been documented and the metrics (and their associated methodologies) that have been used to analyse such impacts. Finally, the report will present the main lessons learned, for example in terms of the main types of impact identified, the effect of contextual conditions on policy implementation (including interactions with other forms of innovation support), the implications for evaluation methodologies and more generally in terms of improving the understanding of innovation networks.

2. Background

2.1 Conceptual framework

2.1.1 Networks – a working definition

As a companion report in this series covers policies designed to support clusters, it is important to define precisely what is meant by networks in the context of this report. According to the European Commission (2003) networking measures comprise one of a set of direct measures which specifically "include support for clubs which exchange information and for activities such as foresight programmes which aim to develop common visions around which future oriented R&D networks can be formed".

As noted above, innovation occurs within the context of multiple forms of informal and formal collaboration. Recent policy attention has tended to focus on formal mechanisms of co-operation among groups of firms, or among firms and research institutions/centres of technical excellence (OECD, 2011). However, much of the policy literature addresses network and clusters policies in an almost interchangeable manner, with little distinction between them.

Cluster policies gained prominence with the work of Porter (1990) where they were associated with (national) competitive performance. Since then, governments have adopted a range of cluster approaches at the national and regional levels. Some of these be used to identify firm level networks and explain their competitiveness (micro level), typically with a strong emphasis on SMEs in the context of industrial, regional and/or innovation policy (European Commission, 2003). The implication here is that network policies operate at the micro-level between more restricted sets of innovation actors.

The European Commission (2003) report makes a further interesting distinction in that "industry-research clustering is basically about networking around specific knowledge bases or technologies and as such is closer to horizontal networking than vertical networking”. Here we stray into the area of science-industry collaboration – another topic reviewed by this series of reports, although we recognise that networks may encompass a range of actors from the public and private sectors (and, indeed, from the third sector).

According to the European Commission (2006), the term 'innovation cluster' refers to "groupings of independent undertakings innovative start-ups, small, medium and large undertakings as well as research organisations - operating in a particular sector and region and designed to stimulate innovative activity by promoting intensive interactions, sharing of facilities and exchange of knowledge and expertise and by contributing effectively to technology transfer, networking and information dissemination among the undertakings in the cluster". Clearly, this implies a much more intensive and sophisticated level of activity than might be expected within an innovation network.

In their study, Stahl-Rolf and colleagues (2002) define networks as the "usually formal collaboration of partners aiming at increasing the competences and innovativeness of the partners and to generate innovations".

In order to make this definition operational, they applied a number of criteria:

• Several projects may be conducted within the same network structure.
• While the network may be oriented towards the production of producing innovations, the role of networking policy is not to directly support innovation projects but to support co-operation and the building of competence which will result in innovations.
• For this reason, networking activities such as the exchange of experience, communication channels, etc. are part of the programme.
• Network management is institutionalised (often through some form of coordinating office, etc.).

However, as they further indicate, in many instances, the stimulation of networking activities forms an integral, if not the most important part, of broader and more encompassing cluster programmes. Thus, the characteristics of schemes supporting networks of innovation and those targeting innovation-oriented clusters can often be very similar. Accordingly, many of the conclusions arising from the evaluation of network schemes are equally valid as those for evaluations of cluster programmes.

In order to provide a pragmatic working definition of network policies (and to distinguish them in particular from cluster policies), in this work we refer to networks as measures aimed at promoting or sustaining the linkage of firms and/or knowledge producers where the activities concerned are centred on a specific technological or problem-oriented topic for the primary purpose of knowledge and information sharing. The basis of the network relationship is thus not based on specific individual projects or similar operational modalities such as personnel mobility or placement activities (which fall into area of schemes for R&D collaboration) but on broader notions of knowledge exchange between larger groups of actors. Whilst networks can be linked to the idea of innovation platforms, for our purposes they are not as developed as these (i.e. they are not necessarily focused on industry sectors).

Critically, from the perspective of this report, networks are not necessarily geographically co-located (which is the primary criterion used to differentiate them from clusters). Support for this argument comes from the literature[^1] which has increasingly emphasised that it is important to connect regional centres of activity to broader national and international networks rather than focusing exclusively on strengthening regional linkages. The inclusion of academic actors in regional and national networks of business actors is seen as important since the former are frequently embedded in international research networks and thus can function as bridges to a broader knowledge base (Bruno, et al., 2011). Nevertheless, it is inevitable that several of the findings of this report echo those reports within this Compendium covering innovation clusters and science-industry cooperation.

2.1.2 The rationale for intervention

A theoretical and conceptual rationale for the application of public innovation policies in support of networks can be traced through an extensive literature base. A seminal work in this area is that of Freeman (1982) in which the systems approach to innovation was first introduced. The resulting shift of focus to include the notion of systemic failures (rather than just market failures), highlighted the significance of actors, and the relationships between them, as a target for innovation policy. Initially focused on 'innovation bottlenecks' and still reflecting a rather linear view of systemic innovation - this led to a growth in the use of measures aimed at the support of collaboration and cooperation in R&D and innovation, at the expense of measures that directly supported R&D-related activities (which deal with isolated innovation 'events'). Further sophistication followed, shifting from the support of single projects carried out by, often limited numbers of, academic and industrial actors to wider ranging support for network development.

From the industry perspective, networking has become a key element of company strategy – instead of single companies, market-oriented, value-chain based networks now tend to dominate the development of key innovations and related global businesses. This provides government with the opportunity to act as facilitator with policies based on the innovation systems approach (European Commission, 2003). Again, a key target for support is the recognition (or assumption) that systemic failures are often due to sub-optimal knowledge flows arising from insufficient industry-science linkages - although, following the definition adopted above, this does not include single collaborative project or personnel mobility based activities but wider collaborations. In a sense, the rationale for network formation and, hence, for their support is the assumption that the whole (the network) is greater than the sum of its individual parts (the network members) in terms of the activities performed.

Thus, networks can allow for rapid learning and facilitate the reconfiguration of relationships – such as with suppliers (in the case of companies) or with producers of knowledge (who may be other companies or research institutions). As noted by the OECD (2011), networks can stimulate the development of additional cooperative activities around a diverse range of issues including training, technological development, product design, marketing, exporting and distribution. Similarly, in the field of scientific research, networks can develop activities based around knowledge pooling, skills sharing, the sharing of facilities, equipment and datasets, student and staff exchanges, the co-development of programmes of joint research, co-publication and others. Again, the objective may be the development of a critical mass in one or several activities, but one that is not necessarily geographically co-located and which may even be virtual.

In these contexts, there are reasons that provide justification for government involvement in supporting the development of networks. Governments may intervene in order to facilitate the establishment or continued development of a network due to the lack of, or insufficient awareness of, the opportunities that may be afforded by networks. Government intervention may also overcome barriers to network formation such as the fear that there may be unfair appropriation of the benefits accruing from collaborative undertakings (OECD, 2011). In this case, governments are able to offer knowledge sharing frameworks which provide a level of security as a means of reinforcing the mutual trust upon which successful cooperative arrangements rely. A typical goal for intervention is to guide firms towards network membership (either with other firms or with the science base) in order to overcome technological "lock-ins”, to enter a new technological area or to change management practice, etc. (European Commission, 2003).

Where the potential members of a network are geographically dispersed, intervention may be necessary to overcome inertia for their formation due to problems with coordination. In a similar fashion, government support may provide the central coordination and administration required to run the network, which may be beyond the resources or capacity of any single network member. Another opportunity for support is through the provision of channels of communication for the exchange of information about the network, such as members' details, information on activities and meetings and the dissemination of research outcomes, etc. Typically, this may be provided through the provision of a dedicated web resource (an illustrative example of which is the extensive web site that supports the UK Knowledge Transfer Networks and which offers both information resources and other services for both the public (open access) and for KTN members (restricted access)[^2].

Notwithstanding the barriers outlined above, networks can occur organically in a bottom-up fashion around issues of common interest to the network members. However, government support may aim to stimulate the general sharing of information between broader sets of network members (who may be drawn from the public, academic or private sectors, or further afield, such as NGOs and charitable foundations). Likewise, such support may focus on more specific goals - often aligned to scientific and technological themes or fields deemed to be of policy significance at regional, national or other levels and where the development of a critical mass of activity is desirable.

A further benefit of support for networking is that, by reducing the barriers or costs of network entry, network members may opt for a variety of levels of engagement. This may range from collaboration in specific research activities, for example, which involve the commitment of resources to participation in general level discussions over specific issues of mutual interest through to participation merely as an observer, incurring a minimal draw on resources. This flexibility makes networks attractive to actors across a range of scales, from SMEs (or even individual entrepreneurs) to large public research establishments or multi-national companies.

Drawing on the literature, Martin et al. (2004), offer the following benefits for the formation of networks:

• Increased scale and scope of activities
• Shared costs and risks
• Improved ability to deal with complexity
• Enhanced learning effects
• Positive welfare effect (increased R&D efficiency and overall R&D expenditure)
• Flexibility (in hierarchies)
• Efficiency (of knowledge transfer)
• Speed (of response to opportunities).

Citing O'Doherty (1998), they note that the benefits of networking can be summarised as follows:

• Material benefits: firms can increase sales and lower production costs by working together.
• Psychological benefits: as firms eliminate their isolation they learn that their problems are shared by others.
• Developmental benefits: By promoting interaction with other firms, networking increases learning and the ability to adapt to the changing economic environment.

Finally, Martin et al (2004) provide an overview of the major reasons for government intervention in the support of networks (note that the authors use the terms networks and clusters interchangeably) (see Table 2).

2.1.3 Overview of typical instruments, target groups, governance issues and practice

A useful 'evolutionary model' of the increasing complexity through which firm to firm relationships may be transformed into fully fledged innovation networks is provided by the OECD (2004). Around one-fifth of the networks identified across eight European countries were of the 'complete innovation network' type, comprising a range of industry actors, universities and government laboratories (see Figure 1). The figure contains no geographic dimension and thus serves equally well as a typology for networks or clusters (following our working definition).

Source: OECD, 2004

The same report also provides a broad typology of the types of support for networks, partnerships and clusters, populated with specific examples from a number of OECD countries.

Source: OECD, 2004.

Another useful overview of the various forms that networks may take is provided by Pittaway, et al. (2004) in their review of networking activities in the UK (see Table 3). Their categorisation is structured according to the specific characteristics, spatial location and composition of the various network forms.

Source: Pittaway, et al. (2004)

In a review of four Centres of Excellence (CoE) programmes (in Finland, the Netherlands, Norway and Sweden) Lemola and Lievonen (2008) note that all address the same broad failures in the innovation systems of their respective countries, namely:

• weak co-operation between innovating organisations
• low number of (growth-oriented) start-up companies
• the threat of multinationals relocating their R&D activities to low-cost countries.

According to Lemola and Lievonen (2008), all the programmes, which can be categorised as support for network development, share a set of common characteristics:

• they are explicitly driven by the strategic requirements of participating companies;
• they involve intensive collaboration between private enterprises and research groups from public knowledge organisations
• they are based on institutionalised long-term financing and collaboration commitments by private firms and other participants
• they emphasise oriented basic research rather than short-term applied research
• they tend to emerge in sectors of the economy in which technology progresses at a slow rather than revolutionary pace
• they are expected to perform well in acquiring RDI funding also from sources other than the CoE funding programmes
• they are designed to attract top international researchers and world class companies, and
• they take advantage of the principles of open innovation in partner-owners' mutual collaboration, although there may be formal or informal restrictions to collaboration with partner companies' competitors..

A further illustrative example is that of the Knowledge Transfer Networks (KTN) operated by the Technology Strategy Board in the United Kingdom. The objective of a KTN is “to improve the UK's innovation performance, by increasing the breadth and depth of the knowledge transfer of technology into UK-based businesses, and by accelerating the rate at which this occurs. Networks are aligned to, and actively contribute to, the goals of the Technology Strategy Board".

The KTN website sets out the following specific aims of a KTN:

• "To deliver improved industrial performance through innovation and new collaborations by driving the flow of people, knowledge and experience between business and the science-base, between businesses and across sectors;
• To drive knowledge transfer between the supply and demand sides of technology-enabled markets through a high quality, easy to use service;
• To facilitate innovation and knowledge transfer by providing UK businesses with the opportunity to meet and network with individuals and organisations, in the UK and internationally;
• To provide a forum for a coherent business voice to inform government of its technology needs and about issues, such as regulation, which are enhancing or inhibiting innovation in the UK".

Stahl-Rolf and colleagues (2002) provide the definition used by the German Federal Ministry of Education and Research for their "Kompetenznetze” (competence networks). These are characterized by the following criteria:

• They may have a thematic, strategic, and/or regional focus
  • They share common guidelines, targets
• They adopt an integrative approach based on:
  • scientific and technological know-how
  • educational offers
  • an innovation-friendly general framework
• They involve interdisciplinarity and co-operation, including:
  • close communication and interaction within the network
  • co-operation with external partners
• They promote international attractiveness, with:
  • products leading on international markets
  • international contacts

The role of government in supporting networks can take a number of forms. These can be categorised at three levels as shown in Table 4 below: