October 2011

Business Research, Development and Innovation in Denmark - policies and effects

by Thomas Alslev Christensen Head of Department for Innovation Policy in the Danish Agency for S cience, Technology and Innovation (DASTI),

Danish Ministry of Science, Innovation and Higher Education

Table of contents FOREWORD ..............................................................................................................3 PART ONE DANISH INNOVATION SYSTEM AND POLICY ..................................... 5 1 THE DANISH RESEARCH AND INNOVATION SYSTEM......................................7 1.1 The agencies for science, innovation, and higher education ..............8

1.2 Funding bodies for research and innovation in the Ministry of Science, Innovation and Higher Education...................................................10 1.3 1.4 2 The Danish universities and higher education institutions................14 Innovation Denmark 2010-2013 ........................................................15

DANISH COMPETENCE NETWORKS ..............................................................18 2.1 2.2 2.3 2.4 Description of competence and innovation networks ......................18 The impact of the Danish competence network scheme ..................22 Netmatch............................................................................................27 Overview over innovation networks and clusters .............................28

THE DANISH INNOVATION CONSORTIUM SCHEME .....................................33 3.1 3.2 Description of the IC scheme .............................................................33 The economic impact of the IC scheme.............................................34

THE KNOWLEDGE-PILOT SCHEME ................................................................38 4.1 4.2 Description of the knowledge-pilot scheme......................................38 The economic impact of the knowledge-pilot scheme......................39

THE DANISH INDUSTRIAL PHD PROGRAMME ..............................................41 5.1 5.2 Description of the PhD programme ...................................................41 The impact of the PhD programme ...................................................42

THE DANISH INNOVATION VOUCHER SCHEME ............................................48 6.1 Description of the voucher scheme ...................................................48

Side 3

6.2 6.3 7

The impact of the basic voucher........................................................50 The impact of the extended voucher.................................................51

THE DANISH TECHNOLOGICAL SERVICE SYSTEM (GTS-NET) ........................54 7.1 7.2 7.3 7.4 Description of the GTS-net.................................................................54 The performance of the GTS-net .......................................................56 The impact of the GTS-net .................................................................57 Description of the nine GTS-institutes ...............................................59

THE DANISH INNOVATION INCUBATOR SCHEME.........................................63 8.1 8.2 Description of the Danish Incubator scheme ....................................63 The economic impact of the incubator scheme ................................64

THE TECHNOLOGY TRANSFER OFFICES.........................................................67 9.1 9.2 The Technology Transfer Offices at the universities..........................67 Commercialization of public research - metrics ................................69

10

THE NATIONAL ADVANCED TECHNOLOGY FOUNDATION............................72 10.1 Description of the Advanced Technology Foundation.......................72 10.2 The economic impact of the Advanced Technology Foundation ......73

PART TWO THE ECONOMIC IMPACT OF R&D&I ............................................... 74 11 IMPACT OF BUSINESS R&D AND INNOVATION ON PRODUCTIVITY AND GROWTH .......................................................................................................75 11.1 Comparative descriptive analysis on average labour productivity....76 11.2 Impact of R&D&I on enterprise marginal productivity......................77 11.3 Overview of other empirical studies..................................................77 11.4. Data and method ...............................................................................80 11.5. Econometric results ...........................................................................82 12 THE SOCIO-ECONOMICAL IMPACT OF BUSINESS R&D INVESTMENT ..........85

Side 4

13

PRODUCTIVITY AND HIGHER EDUCATION....................................................88 13.1 The method and data.........................................................................88 13.2 The impact of higher education on productivity ...............................89

14

THE IMPACT OF R&D COLLABORATION ON PRODUCTIVITY, EMPLOYMENT AND GROWTH...............................................................................................91 14.1 International studies on the economic impact of national research collaboration.................................................................................................92 14.2 The impact of national collaboration projects...................................94 14.3 The impact of international collaboration projects ...........................96

REFERENCES .........................................................................................................100

Side 5

Foreword
Making the innovation system in Denmark more effective: The road to future productivity and growth
Denmark attaches great importance to supporting sound framework conditions conducive to collaboration between public research and the business sector and to improving the commercial exploitation of the innovation activities. Denmark provides a high level of public and private investments in an effective public research and innovation system. Denmark has in the last decade improved the governance of research institutions and institutions for technological development while at the same time reducing the level of state aid and direct subsidies for business activities too close to markets. Danish policies encourage the development of market-friendly approaches to support research and innovation which contributes to an improvement of the productivity and competitiveness of enterprises. In the past it has been difficult to document the microeconomic or socioeconomic effects of research and innovation policies. It has been difficult to obtain an accurate picture of the private economic returns for enterprises. However, the research and analyses presented in this report show that research, development and innovation of Danish enterprises are important for productivity, production, export and employment. A recent analysis of the return from private R&D investments in Denmark shows that R&D-active enterprises, which collaborate with universities or other research institutions experience an average 15 per cent higher productivity per employee compared to the average Danish R&D-active enterprises with no cooperation with research institutions. Furthermore, the productivity per employee increases 9 per cent for enterprises initiating collaboration projects with research and technology institutions compared to a control group of similar non-collaborating enterprises found by using the propensity score matching method among the 20,000 Danish R&D active enterprises. An additional analysis of the Danish innovation consortium programme which supports researchbusiness collaboration shows that EUR 400,000 invested by an average consortium enterprise in public-private research partnerships yield a EUR 2-3 million gross profits. Moreover, analyses of the return from private R&D investments in Denmark show that R&D-active enterprises have a 15 per cent higher average productivity per employee compared to non R&D-active enterprises. Further, innovative enterprises have a 6 per cent higher average labour productivity than non R&Dactive enterprises.

Page 3

The return of increasing private investments in R&D&I is in average between 30 per cent and 66 per cent for Danish enterprises. OECD analyses show that an effective diffusion of knowledge doubles the economic impact of private investments in research, development and innovation. In other words, it is beneficial to invest in research, development and innovation and to do so in cluster or project collaborations between research and business. Finally, the likelihood of enterprises to innovate is 3-4 times higher for enterprises participating in clusters and networks compared to similar enterprises not participating. R&D leads to the creation of new products, processes and services in businesses, increasing earnings and at the same raising their level of knowledge. They will thus be more competitive in the long run to the benefit of productivity R&D&I investments also provide society new solutions to the many challenges we face. Moreover, increased productivity and growth of enterprises yield higher exports, higher national income and thus more tax revenues. It thus contributes both to increased prosperity and welfare. This report describes the Danish innovation system and selected important innovation policy schemes and presents new analyses of the impact of enterprise R&D&I and impact assessment of selected innovation policy schemes. In terms of data, the report presents several new Danish research studies from 2010 and 2011 on the effects of policy schemes and corporate investment in research, development and innovation, providing a clearer picture of the productivity effects of innovation policy schemes. The studies presented have been undertaken by or in close collaboration with Centre for S tudies in Research and Research Policy (CFA) at Aarhus University, the consulting and analysis firm DAMVAD, and the Centre for Economic and Business Research (CEBR) at Copenhagen Business S chool (CBS ), The Danish Entrepreneurship Academy (DEA), The Danish Technological Institute (DTI) and with contributions from Statistics Denmark. To the extent that there are similarities between the analysed R&D enterprises across countries and between Danish innovation policy schemes and schemes in other European countries, the results and effects of the impact assessments presented in this report might be of general interest and could be integrated into cost-benefit analyses when evaluating the impact of private investments in R&D and innovation policies in other European countries. I hope you will enjoy reading the report! Thomas Alslev Christensen, PhD Head of Department for Innovation Policy, Danish Agency of S cience, Technology and Innovation in the Ministry of Science, Innovation and Higher Education

Page 4

The Danish Innovation S ystem and Policy

The Danish research and innovation system under the Ministry of Science, Innovation and Higher Education

Description of selected Danish innovation policy instruments

Impact assessments of innovation policies

Page 5

Page 6

1 The Danish Research and Innovation S ystem

The Ministry of Science, Innovation and Higher Education is responsible for the following policy areas: research, innovation, higher educations, including university educations and internationalization of education and training in Denmark. The ministry aims to make Denmark a leading entrepreneurial and knowledgebased society offering educations that rank among the best in the world, and to create the best possible opportunities for citizens and businesses to realize the vision about Denmark as a network society. The ministry consists of The Danish Agency for International Education The Danish Agency for Science, Technology and Innovation The Danish University and Higher Education Agency which, together with the Permanent Secretary's Department, are referred to as the Ministry of Science, Innovation and Higher Education. Also within the scope of the ministry are a number of funding bodies for research and innovation, research and advanced technological service institutions and the eight universities in Denmark. Figure 1.1. The Danish Ministry of Science, Innovation and Higher Education

The Ministry of Science, Innovation and Higher Education

Danish Agency for Science, Technology and Innovation (DASTI)

Danish University and Higher Education Agency

Danish Agency for International Education

Page 7

1.1 The agencies for science, innovation and higher education

1.1.1 The Danish Agency for Science, Technology and Innovation - DASTI
The Danish Agency for Science, Technology and Innovation (DASTI) has its main responsibilities in the following areas: Public research and innovation funding Researcher mobility Dialogue on priorities in research and technology initiatives Regionalization of research and innovation Commercialization of public and private research. Interaction between knowledge institutions and the business community. Innovation policy. EU research policy. International co-operation on research and innovation. Research and knowledge dissemination Technology transfer and knowledge transfer including legislation Legislation on funding bodies for research and innovation The Agency also functions as secretariat to the Danish Research Coordination Committee, the Danish Council for Independent Research, the Danish Council for Strategic Research, the Danish Council for Technology and Innovation, the Danish Research Policy Council and the Danish Committees on Scientific Dishonesty.

1.1.2 The Danish University and Higher Education Agency

The Danish University and Higher Education Agency is working towards ensuring that Denmark has world-class universities and higher education institutions. It is striving to create interesting and attractive learning and research environments within the best possible organisational, legal, physical and financial framework. The tasks of the agency are: analyses policy development, legislation and the administration of frameworks for the universities, higher education, including university, study programmes PhD programmes Government research institutions.

Page 8

1.1.3 Danish Agency for International Education

The Danish Agency for International Education is responsible for supporting the internationalisation of education and training in Denmark. The agency is the national agency for two EU education programmes, Lifelong Learning and Youth in Action, as well as for other similar programmes, including Nordic and Danish education programmes and initiatives. The Agency is also the central institution in Denmark where persons with foreign qualifications can get these assessed and recognised. In addition, the Agency is an information centre concerning internationalisation of all the educational sectors. The Agency has the following tasks: to help extend and strengthen internationalisation of education and training at all levels and to promote mobility to handle the national administration of international education programmes for school education, vocational education and training, higher education and adult learning as well as the youth sector to make authoritative decisions in the field of assessment and recognition of foreign qualifications to act as a national information centre in relation to internationalisation and recognition to contribute to the development of international cooperation between educational institutions and in relation to trade and industry to assist in launching new initiatives concerning the internationalisation of education and the development of competences in the global society.

1.2 Funding bodies for research and innovation in the Ministry of Science, Technology and Innovation
Figure 1.2 The innovation system in Denmark consist of several funding bodies for research and innovation that operate independently

Page 9

The funding bodies under the Ministry of Science, Technology and Innovation are The National Research Foundation, The Council for Independent Research, The Council for S trategic Research, The Council for Technology and Innovation and the Advanced Technology Foundation. Additional innovation funding bodies are found in the Ministry of Climate and Energy, The Ministry of Environment, The Ministry of Food and Agriculture, The Ministry of Finance, the Ministry of Foreign Affairs and the Ministry of Economy and Business.

1.2.1 The Danish National Research Foundation

The Danish National Research Foundation was established in 1991, and is an independent foundation, which works at strengthening Danish basic research within all research fields. The Foundation s main working method is to set up and fund research centres of the highest international standing Centres of Excellence for 1 -2 periods of funding. The Foundation annually distributes up to DKK 400 million (57 million euro) in 2011. This corresponds to approximately 2 percent of the annual public research expenditure. As a supplement to the Centres of Excellence, the Foundation experiments with various other programs, particularly with a view to strengthening the internationalisation of Danish research. Following this strategy the Foundation is active on collaborations with international foundations and organisations
Page 10

regarding establishment of joint research centres and initiatives on joint research programmes in general. Furthermore the Foundation has in the later years invested in six internationally recognised visiting professors and funded the recruitment of three outstanding professors to Danish research environments for a five-year period. The effect of these investments is clearly visible, e.g. in the exceptional quality of the research output, the high degree of international cooperation, the extensive PhD production, and in the ability to attract external funding from abroad.

1.2.2 The Danish Council for Independent Research

The Danish Council for Independent Research (DFF) funds specific research activities, within all scientific areas, that are based on the researchers' own initiatives and that improve the quality and internationalisation of Danish research. The council annually distributes up to DKK 1400 million (187 million euro) in 2011. The Danish Council for Independent Research also provides scientific advice in all scientific areas for the Danish Minister of S cience, Technology and Innovation, the Danish Government, and the Danish Parliament. The Danish Council for Independent Research is a key driver in the effort to create attractive career paths for researchers. Accordingly, the Council spends large parts of its funds on training young talented researchers, particularly at the postdoc-level. Furthermore, the Danish Council for Independent Research contributes to strengthening the dissemination and application of research findings as well as it participates in international research collaboration. The Danish Council for Independent Research is comprised of a Board of Directors and five scientific research councils.

1.2.3 The Danish Council for Strategic Research

The Danish Council for S trategic Research was established in 2003, and is an independent foundation, which works at strengthening Danish strategic research within all research fields. The council annually distributes up to DKK 1100 million (150 million euro) in 2011. The Danish Council for S trategic Research seeks to ensure that strategic research in Denmark is organised to meet the challenges facing Danish society. The aim is to ensure Denmark s position as a global frontrunner regarding welfare, wealth and science in the short and long term.

Page 11

The Danish Council for S trategic Research wishes to promote international cooperation in research, including cooperation with the new high-growth countries. The Council wants to strengthen the interaction between publicly financed research and the rest of society, such as privately financed research, private organisations, public institutions and the corporate world. The Danish Council for S trategic Research considers research education an important element in strategic research as it contributes to a positive development of the research environments in Denmark.

1.2.4 The Danish Council for Technology and Innovation

The Danish Council for Technology and Innovation was established in 2002, and is an independent council, which works at strengthening Danish private research, development and innovation and economic growth in Denmark. The council annually distributes up to DKK 1100 million (150 million euro) in 2011. The council's work consists of two parts. One is to advice the Minister of S cience, Technology and Innovation about technology and innovation policy. The other is to administer the initiatives given to the council by the Minister. The objectives of the council are to promote: Collaboration and dissemination of knowledge between researchers, research and educational institutions, advanced technology groups, knowledge institutions and enterprises. Innovation, development, diffusion, use and commercialization of new research and technology, and knowledge of organizations and markets. Flow and development of knowledge and technology based enterprises. Innovation and input of capital and expertise for knowledge and technology based enterprises. International collaboration on the utilization of knowledge and technology. The Danish Council for Technology and Innovation administers a number of initiatives the purposes of which are to promote private research, development, innovation and dissemination of knowledge between knowledge institutions and enterprises. The initiatives are: Cooperation and interaction between business and research: o Innovation Consortia scheme o Innovation voucher scheme o The scheme for new forms of collaboration o The competence and innovation network scheme Approved Technological Service (The Danish GTS-system)
Page 12

Industrial PhD scheme Knowledge Pilot Scheme Entrepreneurship and commercialization: o Technology transfer offices at the universities o Business Incubators (The Danish innovation incubator scheme ) o The proof-of-concept scheme

1.2.5 The Danish National Advanced Technology Foundation

The Danish National Advanced Technology Foundation was established in 2005, and is an independent foundation. The Foundation annually distributes up to DKK 600 million (80 million euro) in 2011. The aim of the Danish National Advanced Technology Foundation is to enhance growth and strengthen employment by supporting strategic and advanced technological priorities within the fields of research and innovation. Furthermore, the foundation shall make a special effort to promote research and innovation in small and medium-sized enterprises. The Foundation supports larger projects which are relevant to advanced technological research and/ or innovation. The foundation will pay special attention to applications which fall within the areas of nano-, bio-, and/ or information and communication technology, including the interface between these areas.

1.2.6 The Danish Council for Research Policy

The Danish Council for Research Policy (DCRP) gives the Minister for S cience, Technology and Innovation research policy advice. The Danish Parliament and any minister can also obtain research-related advice from the Council. This advice is given upon request or upon the initiative of the Council. The council does not distribute funds. The Council's responsibilities generally include advice on Danish and international research policy for the benefit of society, including advice on: Framework conditions for research. Funding for research. Major national and international research infrastructures. Development of national research strategies. Denmark's role and position in international research collaboration. Research training and recruitment of researchers.

Page 13

1.2.7 The Danish EuroCenter

EuroCenter in the Danish Agency for S cience, Technology and Innovation provides information and advice to all Danish potential and actual participants in European Framework programmes for Research and Innovation. Important tasks for EuroCenter are to enhance participation of Danish enterprises and research institutions in international research and innovation projects and thus strengthening Danish competitiveness and growth supporting the European Knowledge Society. EuroCenter are in current dialogue with Danish enterprises and research institutions on all aspects of participation in a Framework programme project i.e. clarifying if an idea fits a call, writing an application or advise on the legal and financial rules for participation. EuroCenter offers individual meetings with potential or actual project participants user defined workshops information meetings assistance in application writing advise on legal/financial rules courses on specific aspects of participation in international research and innovation projects telephone/e-mail consultation newsletters and publications.

Further EuroCenter provides information on the national support scheme for application writing, networking, contract negotiation and project coordination offered by the Agency for S cience, Technology and Innovation to all Danish enterprises and research institutions.

1.3 The Danish Universities and Higher Education Institutions

The system of higher education in Denmark includes the four types of higher education institutions: 8 universities conduct research and offer research-based undergraduate and post graduate programmes comprising Bachelor, Candidatus, and PhD degrees as well as the Diploma and Master degrees (continuing adult education) aimed at adult learners and in-service learning. 16 tertiary artistic education institutions and other institutions of higher education offer first, second and third cycle programmes comprising Bachelor, Candidatus, and PhD degrees and the Diploma and Master degrees (continuing adult education).
Page 14

8 university colleges and 2 colleges of engineering (both of which are to merge with the university colleges by the end of 2014) offering first cycle programmes comprising the Professional Bachelor and Diploma degrees (continuing adult education). 10 academies of professional higher education offering short cycle professionally programmes, comprising Academy Profession degrees and the VVU degree (continuing adult education).

Figur 1.3 The Eight Danish Universities

1.4 Innovation Denmark 2010-2013

The Danish Council for Technology and Innovation under the Danish Ministry of S cience, Technology & Innovation administers a number of initiatives for promoting innovation and dissemination of knowledge between knowledge
Page 15

institutions and enterprises. The council has in collaboration with the ministry and after a broad national consultation procedure with organisations, institutions and innovation actors established the second four year action plan called Innovation Denmark 2010-2013 which describes the main innovation policy initiatives under the Ministry of S cience, Technology and Innovation. The initiatives are divided across four broad priority areas: Figure 1.4.: Innovation Denmark Priorities and Measures

The objective of the Danish government is that private R&D should be minimum 2 per cent of GDP and Danish enterprises shall be among the most enterprises in the world. In addition, the overall goal of the Danish Council for Technology and Innovation is that Denmark should be in top-3 in the Innovation Union S coreboard of the European Commission. Other goals of Innovation Denmark 2010-2013 are:
Goal 1: Danish enterprises must be more and continuously innovative, the small and medium-sized enterprises included Knowledge dissemination and interact ion between research and industry shall be strengthened

Goal 2:

The most important tools of the Danish Council of Technology and Innovation: 1. Innovation Denmark project programme: a) Innovation consortia, b) innovation vouchers and c) new forms of research-business collaboration projects: i.e. large and small national and international innovation and research projects operated in collaboration between academic and research institutions and enterprises 2. Innovation Denmark Network Programme: 22 competence and innovation networks (cluster organisations)

Page 16

3. Highly educated staff and researchers in enterprises: a. The Industrial PhD Programme where the research student shares his or her time between an enterprise and a university b. The knowledge pilot scheme (an innovation assistant programme) which promotes employment of highly qualified staff in small and medium-sized enterprises. 4. The Danish GTS-net: The approved technological service institutes which are independent knowledge institutions delivering knowledge to enterprises 5. The Danish Innovation Incubator programme: 6 business incubators invest public capital in entirely new, high-tech enterprises. 6. The Danish Proof-of-Concept programme: Commercial exploitation of public research: In the form of support for maturation of inventions from public research institutions (proof-of-concept) and projects which promote technology transfer between national and international research institutions and enterprises.

Page 17

2 The Danish Competence Networks and Clusters Innovation Networks Denmark

This section presents the Danish competence and innovation network scheme ( Innovationsnetvrk Danmark ) which is a national cluster programme. The section presents the results of the scheme. These results are published annually, see Innovation Network Denmark Performance Account from the Danish Agency of Science, Technology and Innovation 2011.

2.1 Description of competence and innovation networks

A competence and innovation network is supported by the Danish Council for Technology and Innovation under the Ministry of S cience, Technology and Innovation. A network (or a cluster) is a flexible framework for collaboration between enterprises, research institutions and non-profit advisory/ knowledge dissemination parties. The annual budget of the total network programme of the ministry is approximately 10 million euro. The annual budget of an average network is approximately 0.9 million euro of which the 40 percent is financed by the network programme of the DCTI, at least 40 percent is financed by enterprises and the rest is financed by regional sources, universities, technological and research institutes and the European Union. From July 2010 22 national networks exist with support from the DCTI network programme. The average number of employee in each network was 5.4 man years.

Page 18

Table 2.1 Annual resources of the total network programme and of the average network

EUR 10 million are provided to the innovation networks in both 2010 and 2011. Some of these funds have been used for specific projects. The basic grants also include funds for network matchmaking and idea generation. This work can lead to partnerships that can be funded under other public sector schemes. One of the most important tasks of a competence and innovation network is to ensure that national innovation policy is not simply a matter for large research enterprises; both by ensuring that smaller enterprises participate in network projects, and by ensuring that the networks help this target group to make use of other innovation policy initiatives e.g. innovation consortia, innovation vouchers, the knowledge-pilot scheme and the Industrial PhD scheme. See chapter 3-11 for more detailed information about these various innovation policy schemes. The use of other innovation policy programmes is three times higher among enterprises that participate in the activities of the network than among similar enterprises not participating in innovation networks.1

The impact of cluster policies in Denmark, Danish Agency of Science, Technology and Innovation 2011

Page 19

The competence and innovation networks carry out a wide range of activities with the aim to share, develop and disseminate knowledge in a way which creates value for the enterprises. The activities can be divided into nine core activities briefly described in table 2.2.

Table 2.2. Core services for enterprises in the competence and innovation networks

It is the policy of the government and the DCTI that the various different networks should complement each other, and that enterprises and their advisors should be able, quickly and easily, to find out what each network is able to offer. The idea is that the innovation networks and clusters under the Danish Ministry of Science, Innovation and Higher Education should differ from other networks and clusters in the following ways, by: National wide. Many other networks are regional or local and their sole purpose is strengthening interaction with one particular knowledge environment. The innovation networks under the Danish Agency for S cience,

Page 20

Technology and Innovation should instead work as a portal to all knowledge institutions in Denmark in their particular subject area. Being open to new members/ participants. The goal of achieving national innovation networks also means that they must work continuously to attract new enterprises. Providing entry to the national innovation system. The innovation networks offer highly-qualified consultancy on financing options at both national and EU levels. At the same time, they work as a portal for the enterprises to a wide range of different knowledge environments - both in Denmark and abroad. Unlike most other networks the innovation networks under the Danish Agency for S cience, Technology and Innovation offer individual matchmaking to enterprises wishing to find e.g. a university researcher with knowledge in a specific area. Existing regional networks which aim to develop into national networks can do so by applying to the Danish Council for Technology and Innovation for innovation network status when future tenders are invited. There are tenders in 2010, 2014, etc. The DCTI finances national networks for a period of four years with the possibility to add additional 4-years periods after a tender. In summary there are nine core network services: Bridge-building activities and meeting places o Themed networks o Matchmaking o Idea generation o Conferences, seminars, etc. Partnership projects o Pre-projects o R&D&I projects o Business- to-business partnerships Knowledge information and communication o Consultation o Skills development Table 2.2 above provides a description of the core services of the various services for enterprises in the competence and innovation networks.

Page 21

2.2 The impact of the Danish competence network scheme

This section presents an analysis of the impact of the competence and innovation network scheme. It has been prepared by independent consultancy enterprises. The ministry publishes each year a performance statistics on the impact of the competence network. These statistics have been published since 2006. The statistics are based on a survey with the network coordinators, participating universities and technological institutes and enterprises having participated in the activities of the networks. The statistics follow the impact of the network in several success parameters in association with running and participating in the activities of the network. The success parameters are: R&D investments, partnership projects, collaboration between enterprises and knowledge institutions, innovation and idea creation. The 2011 performance statistics document progress in the ability of the networks to tackle the main tasks of the networks. A total of 3,155 enterprises participated in network bridge-building activities in 2010. Two out of three participants come from enterprises with fewer than 50 employees. In 2010, 906 businesses participated in partnership projects which were initiated as a result of network activities. 615 of these enterprises or 67 per cent had fewer than 50 employees.

On the other hand, there has been a minor reduction in the number of enterprises participating in partnership projects for the first time. Most
Page 22

enterprises had already worked with a knowledge institution. The reduction in first-time participants is due to the fact that some enterprises come back year after year. In 2006, many new enterprises joined the networks and subsequently participated in a number of projects. Many, however, do find new partners. The performance statistics from 2007 shows that 56 per cent of enterprises which in 2007 initiated a partnership project with a university or a GTS -institute had not previously worked with that specific knowledge institution. It is the aim of the Danish Council for Technology and Innovation to encourage a significant rise in the number of enterprises working with knowledge institutions, i.e. universities, GTS -institutes, other public research institutions etc. It is therefore an important future challenge for the networks to continue to attract new enterprises. When combined with other network activities, the partnership projects have produced significant results. Nearly 350 enterprises developed new products, services or concepts in 2010 - and 445 enterprises generated innovation ideas on which they expect to carry on working. Even more 896 enterprises benefited from the networks by building skills or obtaining specific methods or tools which significantly increase their ability to work with innovation. It is estimated by the enterprises themselves that the 408 new products and services which were created in 2008 by network activities will result in an increased annual company turnover of approximately DKK 1.6 billion. On top of this comes the value of the ideas and skills that the enterprises develop in their networks the value of which is not possible to estimate. Figure 2.4 provides an overview of the developments for various indicators.

Page 23

Networks carry out individual matchmaking for enterprises wanting to find the right researchers and scientists. They arrange idea generation e.g. in the form of special project camps or workshops whose purpose is to develop specific project ideas based on enterprise requirements and applicable knowledge held by the universities.

Many of the partnership projects resulting from the network activities are business to business projects. Other projects are public-private collaboration projects with the participation of enterprises as well as public research institutions. The projects are not the same as the Innovation Consortia S cheme
Page 24

(see next chapter). The collaboration projects of the Innovation Consortia Scheme also include a criterion regarding active knowledge transfer, including the participation of a knowledge transfer organisation in the project. Furthermore, a minimum of two enterprises must participate in an Innovation Consortium and enterprises cannot receive funding from the grant provided. The innovation network projects can take many forms. To the extent that there are similarities between the projects of the Innovation Consortium S cheme and the projects of the competence and innovation networks, however, the results of the impact assessment of the Innovation Consortium Scheme presented in the next section may be of general interest and could be integrated into cost-benefit analyses when evaluating the economic impact of the competence and innovation networks. In 2011 a separate econometric impact assessment of the Innovation Network Denmark programme was conducted; see The Impact of Cluster Policy in Denmark from The Danish Agency of S cience, Technology and Innovation 2011d. The study shows that participation in innovation networks and clusters increases the probability to innovate by more than 4,5 times year 1 after participation. Companies participating in different innovation networks have an increased probability of being innovative with the effects on innovation showing from the first year of participation. The probability of being innovative is 4.5 times higher for innovation network participating companies in innovation networks compared to a control group composite of other similar companies not participating in networks. This means that for every time 10 companies in the control group turns innovative 45 participating companies in innovation network will turn innovative. Among the participating companies in innovation networks we can identify 102 companies that become innovative of a treated population of 1,225 participating companies in innovation networks an increase of 8.3 percent. On the contrary for every 1,225 participating company in the control group 22 companies turn innovative an increase of 1.7 per cent. It is a difference of 6.5 percentage points. The control group is found by using the propensity score matching method. Increasing the number of innovative companies potentially has an economic impact. The most comprehensive Danish study of private return on investment in R&D and innovation has proven a significant return on investments in innovation of 30 percent, see chapter 102. Thus an increased ability to be innovative and

The Danish Agency for Science, Technology and Innovation, 2010

Page 25

increased probability to be innovative is expected to have a significant economic impact on the participating companies in innovation networks. The impact study also document that network participation increases the probability of R&D collaboration by 4 times year 1 after participation. Innovation networks assists companies in entering joint R&D and innovation projects by providing the companies with the competencies required for this complex task (and which S MEs, in particular, do not possess prior to participating). Additionally, innovation networks provide a platform within which participating companies in innovation networks identify potential collaboration partners. Already within the first year of participation, the probability of entering R&D collaboration increases by 95 per cent, and thus almost doubles their probability of entering R&D collaboration. Thus every time a company in the control group composited by other similar companies (found through propensity matching score) not participating in innovation networks enters R&D collaboration, two new participating companies in innovation networks enters R&D collaboration. The year after participating in an innovation network, the probability of entering R&D collaboration is almost 300 per cent higher than other similar companies not participating in networks. Chapter 11 and 14 presents new econometric studies on the impact of entering R&D collaboration3. The 2011-study shows that companies, who are entering R&D collaboration, have significant higher growth rates in productivity per employee compared to other similar and high productive companies. Entering R&D collaboration increases productivity with an average of 9 per cent a year over a 9 years time period. Thus, increased R&D collaboration is expected to have a significant economic impact on the participating companies in innovation networks.

The Danish Agency for Science, Technology and Innovation, 2010 and 2011

Page 26

2.3 Netmatch
In 2010, the Danish Agency of S cience, Technology and Innovation established Netmatch as a unifying platform and a support function for the innovation networks in Denmark. The objective of Netmatch is to increase the impact of the Innovation Network Denmark initiative through the development and supply of a number of services for the innovation networks in order to ensure optimum working conditions. By means of competence development and internationalization, and by assisting the innovation networks on operational and administrative tasks, Netmatch makes it easier for the innovation networks to focus on matchmaking and networking. Netmatch supports the innovation networks in three specific areas:
Being national - in contrast to many other Danish networks, which are regional or local, the innovation networks are national and should acts as national meeting points within their specialized field Branding - the Danish business and scientific communities should be familiar with the different innovation networks, their field of activities and how to interact with them. Being international the innovation networks are expected to give more focus on operating in international markets. Through the EU funded Enterprise Europe Network, Netmatch is able to support the innovation networks with international business and information services

Netmatch is in a close and continuous dialogue with the individual innovation networks and the Danish Agency of Science, Technology and Innovation. In close co-operation, the parties agree on which services Netmatch should develop and provide.

Page 27

Networks and clusters in the Innovation Networks Denmark programme AluCluster - Knowledge and technology centre for aluminium Contact: Michael Nedergaard ph. + 45 7473 3040, e-mail: mn@alucluster.com AluCluster, a centre within practical utilisation of aluminium, offers a wide range of competences within the aluminium area. AluCluster's objective is to create value for the customer through overall solutions. AluCluster's mission is to meet the need for highly qualified consultancy now and in the future within practical utilisation of aluminium and thereby become the preferred partner within consultancy and development of aluminium solutions. www.alucluster.com Animation Hub Contact: Viggo Johannes Jensen - ph. + 45 2850 9864, e-mail: vjj@animwork.dk Animation Hub is a cluster experimenting with animation in the context of development and communication. Based on the competences within animation, such as generating new ideas, dramaturgy, cinematography and simulation, the cluster wish to help companies find new ways of understanding, accepting and involving in regards of innovation processes, developing new concepts, marketing as well as learning, guiding and interaction design. www.animationhub.dk

Danish Sound Technology Network Contact: Jan Larsen ph +: 45 2243 0025, e-mail: jl@imm.dtu.dk Danish S ound Technology Network embraces individuals, organizations and businesses involved with sound technology. They will create a new space for innovation, collaboration and dissemination of knowledge. The vision of the network is that Denmark is a leading country with regards to sound technology in terms of knowledge, research and education. Danish S ound Technology will be the epitome of high quality in products and services, as well as in physical rooms and social contexts. www.lydteknologi.dk FoodNetwork Contact: Britt Sandvad ph: + 45 9612 7624, e-mail: bs@vifu.net FoodNetwork is an extensive network which includes a large number of Danish universities, research institutions, Approved Technological S ervice Institutions (GTS ), innovation- and development parks as well as technical- and vocational schools. The aim of the network is to create growth within the food industry through networks, projects and activities. It is also to be the link that ensures visibility of the relevant partners within the food industry and to support and facilitate existing and new clusters. www.foodnetwork.dk Biopeople Innovation Network for Biotech Contact: Per Spindler ph: +45 2875 6572, e-mail: per@biopeople.dk Biopeople embraces universities, research organizations, and hospitals, the Danish Medicines Agency, industry associations as well as pharma, medtech, medical device, food and biotech companies. www.biopeople.dk

Page 28

Infinit The ICT Innovation network Manager of Danish network Aalborg: Rikke Uhrenholt ph: + 45 9940 7220, e-mail: aalborg@infinit.dk Manager of network Copenhagen: Rikke Koch ph: + 45 2126 8724, e-mail: kbh@infinit.dk
Infinit is a Danish network for innovative utilization of IT. Our goal is to convert the infinite possibilities that technology offers into concrete collaborations between research and industry. www.infinit.dk

InnoBYG - Innovation Network for Energy efficient and Sustainable construction Contact: Henriette Hall-Andersen ph: +45 7220 2241, e-mail: hha@teknologisk.dk The construction industry's new network InnoBYG will facilitate sustainable and energy efficient development in the construction industry from 2010-2014. The focus of the network will be on development projects, knowledge sharing and dissemination and matchmaking across the industry and between companies and knowledge institutions/universities. www.innobyg.dk The Innovation network for Environmental Technology

Contact: Jrn Rasmussen - ph. + 45 4516 9200, e-mail: jar@dhigroup.com The Innovation Network for Environmental Technology, Inno-mt, was established in December 2010. The focus is to bring the sectors soil, water, air and waste together to enable innovative new products and services across the four sectors. www.inno-mt.dk InViO - Innovation network for knowledge-based experience economy Contact: Jens F. Jensen ph. + 45 9940 9028, e-mail: jensf@hum.aau.dk The objective of Innovation network for knowledge-based experience economy is to strengthen knowledge sharing, knowledge development, and cooperation between businesses and institutions of knowledge relating to innovation and research within the field of experience economy. In this way, the innovation capacity of the businesses is reinforced so that knowledge- and experiencebased growth within the industry will be generated. www.invio-net.dk Innovation Network for Biomass Contact: Michael Stckler ph +: 45 8999 2504, ms@agropark.dk The purpose of the Innovation Network for Biomass is to facilitate development within production, handling, and processing of biomass with the goal of better utilization for energy purposes. Our members consist of individuals and companies involved in the field of biomass, agricultural waste, and manure. The network has an international scope and welcomes both Danish and international members from private companies, research institutions, authorities, etc. www.inbiom.dk

Danish Lighting Innovation Network Contact: Lene Hartmeyer ph. +: 45 4717 1800, e-mail: info@dansklys.dk Page 29

The object of the Innovation Network for Danish Lighting is to promote the use of good and appropriate lighting and to advance knowledge and to disseminate information for the improvement of the lighted environment to the benefit of society. www.dansklys.dk The Innovation network for Market, Communication and Consumption Contact: Per stergaard ph. + 45 6550 3235, e-mail: poe@sam.sdu.dk The network embraces a broad range of core competences that are essential to understand future markets and consumers. Researchers from a classic marketing tradition collaborate with researchers from the humanities, art, and design. This combination is not common in a Danish context, but crucial in a market where symbolic and emotional dimensions of products are becoming more and more important. The participating researchers come from The University of Southern Denmark, Aalborg University, Aarhus School of Business Aarhus University, Copenhagen Business School, and Kolding School of Design. www.imkf.dk

Service Platform Service Cluster Denmark Contact: Mette Abrahamsen ph: + 45 2311 3719, e-mail: ma@dea.nu The vision for Service Cluster Denmark is to contribute to growth, innovation and competitiveness among service businesses in Denmark. S ervice Cluster Denmark aims to create new possibilities for cooperation between businesses and knowledge institutions, to strengthen research and innovation in businesses and to incorporate international knowledge and ideas by involving businesses, research institutions and networks outside Denmark. www.serviceplatform.dk Innonet Lifestyle Interior & Clothing Contact: Betina S imonsen ph: + 45 96 16 62 00, e-mail: betina@moebelcenter.dk The purpose of Innonet Lifestyle Interior & Clothing - is to promote growth and innovation into the industry of home and fashion by identifying, communicating and embedding new knowledge. The purpose is also to build bridges between companies and institutions of research and knowledge. www.innonetlifestyle.com Plastic and Polymer Network Contact: Dorte Blum ph: + 45 36973600, e-mail: dwb@plastcenter.dk The Plastcluster consists of a number of companies with in interest in plastic and polymer materials. The aim of the cluster is to increase the awareness of the materials, promote and innovate the use of the materials within and across sectors. www.plastnet.dk No Age innovative solutions for elderly people
Contact: Gunhild Garsdal, ph. + 45 3010 8080, e-mail: gg@vhhr.dk No Age aims to make to the older people of Denmark more resourceful by supporting their resources and making them capable of taking care of themselves. Companies, leading research institutions, municipalities, hospitals and organizations are part of No Age s work to develop innovative solutions within health, prevention, nursing, treatment, etc. www.lvvl.dk Page 30

OffshorePeter Center Denmark Contact: Blach ph: + 45 3697 3670, e-mail: pb@offshorecenter.dk Offshore Center Danmark is the official national competence and innovation center for the Danish offshore industry. On behalf of its +210 member companies and institutions Offshore Center Danmark push development with the aim of growth within the Danish offshore industry. www.offshorecenter.dk RoboCluster Contact: Bjarke Nielsen ph: + 45 2119 4797, e-mail: bjarke.nielsen@robocluster.dk RoboCluster is a Danish innovation network for robotics and automation. The object is to maintain and further expand the robotics sector in Denmark by generating and ensuring optimal conditions for innovation in new as well as in existing enterprises and set robotics into action in fields as hospitals, farming, industry, play and education. This is done by initiating technological projects between suppliers, producers, users, universities and knowledge institutions in the field of robotics and automation. www.robocluster.dk The Transport Innovation Network - TINV Contact: Steen Sabinsky ph: + 45 2966 2408, e-mail: ssa@maritimecenter.dk The Transport Innovation Network (TINV) is a national, cross disciplinary network aimed at the Danish Transport sector. The primary objectives of TINV are to create synergy, encourage match-making and generate research and development projects between stakeholders in the transport sector and research- and educational institutions, as well as related sectors such as energy and infrastructure. www.tinv.dk UNIC Use of New technologies in Innovative solutions for Chronic patients Contact: Dorthe Kjr Pedersen ph + 45 2498 4155, e-mail: dorthe.pedersen@robocluster.dk
The number of chronically sick people is increasing, and consequently so is the necessity of treatment and nursing for the chronically sick. Through development of technological and innovative solutions UNIC aims at a decrease in the number of hospitalizations, increase of chronically sick people in jobs, more people being able to care for themselves, etc. www.partnerskabetunik.dk

VE-Net - Renewable Energy Innovation Network Contact: Grete Bech Nielsen ph: + 45 7220 1113, e-mail: gbn@teknologisk.dk VE-Net (Renewable Energy Network) is an innovative network related to energy. The aim of the network is to create collaboration initiatives between companies and research institutes with the purpose of increasing the application of research-based expertise in the business community and to solve high technology matters. www.ve-net.eu Water in Urban Areas Contact: Ulrik Hindsberger, ph. + 45 7220 2285, e-mail: uhi@teknologisk.dk
Page 31

The partnership is directed towards the challenge of adapting cities to a changed climate, and thus it operates within the topic of energy, climate and environmental technologies. The partnership will contribute to realising the vision of Denmark as a climatically strong and green winner nation and establish Denmark as the global demonstratorium for viable water technologies, system solutions and integrated water resource administration. The goal is to develop, document and present technologies and planning tools for climatic adaptation of existing urban areas in Europe, US A and Australia, and for development of new, climatically strong cities in countries in financial and institutional transition, such as China. www.vandibyer.dk

Page 32

3 The Danish Innovation consortium (IC) scheme

This section presents the Danish Innovation Consortium scheme ( Innovationskonsortium ) and an analysis of the economic impact of the Innovation Consortium scheme on participating enterprises. The econometric analysis has been prepared by the Centre for Economic and Business Research (CEBR) at Copenhagen Business School (CBS). The Innovation Consortium scheme (IC scheme) is a Danish subsidy scheme granted by The Danish Council for Technology and Innovation under The Danish Ministry of Science, Innovation and Higher Education. The IC scheme has existed since 1995. The ministry invests annually approximately EUR 16 million in the scheme.

3.1

Description of the IC scheme

An innovation consortium supported by the Danish Council for Technology and Innovation (DCTI) is a flexible framework for collaboration between enterprises, research institutions and non-profit advisory/knowledge dissemination parties. The budget of an average innovation consortium is approximately between 3 million and 7.5 million euro. The average funding by the DCTI is 40 per cent of a consortium budget, i.e. between 1 and 3 million euro. An innovation consortium must consist of at least two enterprises which participate throughout the entire project, one research institution and one advisory and knowledge dissemination party. Additionally, an innovation consortium may involve or attach other types of partners which are considered to be relevant for the project. The collaboration should be based on a joint project aimed at developing and bringing research based knowledge to maturity, so that it can form the foundation for the innovation of Danish enterprises. The joint project should result in the completion of high-quality research relevant to Danish enterprises. Furthermore, the project should ensure that new knowledge is converted into competences and services specifically aimed at enterprises, and that the acquired knowledge is subsequently spread widely to the Danish business community including in particular small and medium-sized enterprises. Any project initiated by the consortiums must comply with the following: The project has a generic content and the results must be of relevance to a wide group of enterprises. The project is at a high level of innovation and research. The project cannot have the character of product development for individual enterprises. The project requires close collaboration between the consortium parties. The project duration is between two and four years.

Page 33

The role of the enterprises in a consortium is to ensure that the joint research and development project is based on relevant development needs of Danish enterprises. Consequently, the project theme should be of significance to the business development of the participating enterprises. However, it should not take the form of actual product development. Also, the enterprise participation ensures that the knowledge and competences of the business community are utilised in the project. Therefore, the participating enterprises should contribute with knowledge and competences at a high level within the project field. Easy application process and reporting procedures The University or the GTS -institute participating in the consortium takes care of the application process and the dialogue with the Danish Council for Technology and Innovation Project can start 3 month after application Light reporting procedures with the ministry every year if there are deviations from the project description Stop-go mid-term evaluation of the project

The enterprises may be Danish or foreign (or both). Between 1995 and 2003, 274 different enterprises have participated in an Innovation Consortium, and a number of enterprises have participated more than once. On average there were approximately 40 enterprises joining an Innovation Consortium per year, but there are large differences between years, with the years 1998-2000 being characterised by the highest activity with an average of almost 70 enterprises joining an Innovation Consortium. Approximately 50 per cent of all participating enterprises are in the manufacturing business, 25 per cent are in financial or business services and 15 per cent are in trade and retail.

3.2

The economic impact of the IC scheme

This analysis - based on The Analysis of Danish innovation policy - The Industrial PhD Programme and the Innovation Consortium S cheme , from The Danish Agency of S cience, Technology and Innovation (2011) - follows 220 enterprises which have participated in at least one Innovation Consortium using national enterprise-register dataset. Enterprise level developments are primarily studied in two success parameters: gross profit and employment. Note that while employment is defined as the number of employees in a given enterprise at a given point in time, gross profit is defined as annual net sales subtracted annual costs of variable inputs (raw materials, energy, intermediate goods purchases, etc.) except labour costs. Gross profit is the most precise measure of the enterprise s value creation, but one should, of course, keep in mind that part of the enterprise s total value creation may be passed on to
Page 34

consumers, may be retained in the firm and increase its value (of which there is no data available for this analysis), or may take the form of positive externalities, such as knowledge and/ or innovations, that benefits other enterprises or society as such. As a measure of knowledge creation, the analysis could in principle also have considered firm-level patenting activity. No actual data on patenting activities were, however, available for this analysis. While employment is simply the number of employees in a given firm at a given point in time, gross profit is a measure of the firm s value creation. The analysis considers (absolute and percentage wise) growth in gross profit and the number of employees both before and after programme participation and analyses the changes in the growth patterns in association with participating in the programme. The analysis identifies a control group of enterprises that do not participate (non-participants), but which are similar to the participants in terms of size, industry, and region. Under the assumption that gross profit and employment developments of participants and non-participants would be symmetric in the absence of programme participation, differences between the two groups of enterprises can be interpreted as the causal impact of the programme on participating enterprises. ICs subsidise and facilitate cooperation between private enterprises and research and knowledge institutions. Cooperating institutions can apply for financial grants at the RTI/ FI, and the grants subsequently finance the expenses incurred by the research and knowledge institutions whilst undertaking the cooperative project. Typically grants amount to DKK 7-19 million (approx. 1-2 million). Enterprises finance their own participation in the collaboration project and minimum 50 per cent of the total budget of the project. The results of the analysis can be summarized as follows: Of the enterprises that participated in the IC scheme small enterprises have experienced significant increases in the growth of gross profit and employment in association with programme participation. These results are robust to controlling for preparticipation growth and developments in the growth of enterprises in the control group. It is important to note that these potential effects depend on the size of the enterprises under consideration. The analysis finds positive potential gross profit effects (increase in growth) that are significant at the five per cent significance level for enterprises with a gross profit below 150 million DKK (approx. 20 million) the year before the programme. The analysis also finds potential employment effects for enterprises with less than 150 employees in the year before the programme. For enterprises with a gross profit less than DKK 150 million in the year before participation, estimates suggest a total potential gross profit effect of approx. 20 million DKK (approx. 2.5 million) over a ten year period, which should be related to own financing per firm averaging approx. 3 million DKK (approx. 370,000).

Page 35

This result is robust changing sampling conditions and using enterprises that applied for funding and got their application rejected as an alternative control group. Results for employment growth are not robust to using the alternative control group, and should thus be interpreted as being more tentative. It is important to note, that the size and statistical significance of these potential effects depend on the size of the enterprises under consideration. The analysis finds positive potential gross profit effects that are significant at the five per cent significance level for enterprises with a gross profit below 150 million DKK (approx. 20) the year before the programme. It also finds potential employment effects for enterprises with less than 150 employees in the year before the programme. The analysis looks at the survival rates of participant enterprises and compares these with the survival rates of enterprises in the control group. The analysis finds high survival rates (most likely due to IC participants and their control counterparts being relatively large) and no difference in the survival rates of participants and non-participants. Although this analysis is an evaluation of a specific research-business collaboration subsidy scheme, its results might be of general interest, as schemes similar to the IC scheme have been implemented in a number of countries. The results of the analysis with respect to gross profit developments of participating enterprises compared to an identical control group is illustrated in Figure 3.1. The figure is a graphical depiction of the absolute differences in gross profit. As illustrated in the figure the analysis finds similar increases in gross profit for participants and controls in the years before the base year. This suggests absence of any inherent differences in gross profit growth between the two groups of enterprises, which also indicates that the matching procedure succeeded in finding a group of controls of similar inherent growth compared to the group of participants. After the base year, the gaps between the graphs widen, with participants having larger increases in gross profit compared to the controls. Under the assumption that participants and controls would have continued their pre-participation (prebase-year) growth patterns in the absence of the programme or would have changed their growth patterns in the same fashion, the higher increase in the group of participants must measure positive effects of the programme on participants employment and gross profit. If pre-base-year trends are indeed equal, the graphs suggest participation in an ICs to have a gross profit effect of approx. DKK 7.3 million five years and approx. DKK 16 million ten years after the base year.

Page 36

FIGURE 1: Grossprofit. Mean differencescompared to base year. Firmswith grossprofit lessthan 150 million DKKin the base year. 3-year moving averages.
80000 60000 40000 20000 0 -20000 -40000 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 Participants C ontrols

Yearsafter the base year

Note that enterprises, to be observable long after the base year, need to have participated or have to be selected as controls at the time of the start of the programme in the mid-nineties, and must not have left the data before the end of the observation period. Also, to be observable long before the base year, enterprises need to have started to participate or been selected late in the observation period, and need to have existed long before the base year. As a result, there are only a limited number of observations long before and after the base year, implying that findings based on these observations get increasingly tentative at the left and the right sides of the figures. To establish evidence on whether or not the above differences in the two groups growth patterns are statistically significant, i.e., too large compared to the general variation in the data to be considered coincidental, a regression model is used. The findings agree with the presence of considerable effects of the IC scheme on participants increases on gross profit. Findings for both absolute and logarithmic differences are statistically significant at the 5 per cent significance level for enterprises followed over the first five years after the base year and significant at the five per cent level for percentage-wise increases for those enterprises which are able to follow for at least ten years after the base year. In a follow-up study Long run Effects of Public-private Research Joint Ventures: the Case of the Danish Innovation Consortia S upport S cheme , Ulrich Kaiser & Johan M. Kuhn (2011)- it is found, that program participation effects are instant for annual patent applications and last for three years and that employment effects materialize after one year. The study shows that these overall results are primarily driven by enterprises that were patent active prior to joining the shceme and that there are no statistically significant effect for large enterprises.

Page 37

4 The knowledge-pilot scheme

This section presents the Danish Videnpilotordning the Knowledge pilot scheme (the VP scheme). In some countries this scheme is also called the Innovation assistant scheme . The section describes the scheme and presents an analysis of the economic impact of the scheme on participating enterprises prepared by the Centre for Economic and Business Research (CEBR).

4.1 Description of the knowledge-pilot scheme

The VP scheme is a Danish subsidy scheme granted by The Danish Ministry of S cience, Innovation and Higher Education. The scheme aims at increasing knowledge dispersion throughout the economy by subsidizing the employment of university graduates in small and medium size enterprises which do not typically make use of the resources of these individuals. The VP scheme was initiated in 2005. Under the scheme, enterprises with less than 100 employees and having little or no experience in hiring employees with a higher education can apply for wage subsidies of up DKK 150,000 (approx. 20,000) over a twelve-month time period when hiring an employee with an academic degree recognised by the Danish Agency for S cience, Technology and Innovation. Until 2009, enterprises could apply for a knowledge voucher of an additional DKK 50,000 (approx. 6,500) for services purchased or education at a number of Danish research and education institutions, i.e. education and up qualification activities. Until early 2009, approx. 220 VP projects were completed, totalling approx. DKK 15 million kr. (EUR 2 millions). In the period between 2005 and 2009 425 Knowledge Pilot grants were approved including an additional 425 knowledge vouchers . However, knowledge vouchers were only used in 48 cases (11.3 per cent). In 2010 and 2011 additional 260 grants were approved. Very easy application process Application of maximum 1 page. Guaranteed answer with approval or rejection from the ministry within 20 working days. Project can start 1 month after application. Fast lane with approval in the ministry within 1 week if necessary. Max. page report to the ministry when the project is concluded

Page 38

4.2 The economic impact of the knowledge-pilot scheme

This section presents an analysis of the economic impact of the VP scheme on participating enterprises. It has been prepared by the Centre for Economic and Business Research (CEBR, 2010) at CBS in Copenhagen. The analysis follows the 127 enterprises that have participated in the VP scheme before 2007 using enterprise level register data. It studies these enterprises developments in three success parameters in association with participation in the VP scheme: changes in gross profit, changes in enterprise level employment and changes in average wage cost per employee. While employment is measured as the number of employees in a given enterprise at a given point in time, gross profit is a measure of the enterprise s value creation, and average wage costs per employee can be interpreted as a proxy for productivity per employee. The firm survival is also analysed as an additional success parameter. In terms of methodology, the analysis considers the increases in gross profit, the number of employees and wage costs between before and after running a VP project. Moreover, we apply a matching procedure and identify a control group of enterprises which do not participate in the scheme (non-participants), but which are similar to the participants in terms of size, industry, and region. Again, we can use enterprise-level data to calculate the changes in gross profit and employment for the non-participants, allowing us to address the question of whether participants have experienced higher increases in the success parameters than would be expected on basis of the trajectories of the matched non-participants. The results of the analysis can be summarised as follows: (a) Over a three year-period, starting in the year before participation in the VP scheme, participating enterprises on average increased annual gross profits by an additional DKK1.1 million (approx. 150,000) compared to nonparticipants. (b) Enterprises that participated in the scheme did not experience higher growth in the number of employees. (c) Average annual wage costs of enterprises participating in the scheme increased by an additional DKK 24,000 (approx. 3,200) compared to nonparticipants. (d) Enterprises that participated in the scheme had lower firm closure rates than non-participants. For these comparisons, any differences between the two groups of enterprises can be interpreted as the causal impact of the programme on participating enterprises under the assumption that the developments of participants and
Page 39

non-participants would be symmetric in the absence of programme participation. It should, however, be noted that the relatively small number of observations and large heterogeneity in the experiences of the different enterprises imply that results are not statistically significant, and should, consequently, be interpreted as being tentative. S till, the most qualified guess on the basis of the register data is that the participation in the VP scheme is generally associated with a higher growth in gross profit. Given the high similarity between participating and nonparticipating enterprises, the growth difference in gross profit can be interpreted as the scheme s causal effect. The finding of participating enterprises having larger increases in wage costs but not in the absolute number of employees compared to non-participants could be taken as an indication of participating enterprises increasing their shares of highly educated employees.

To address the potential problem that participating enterprises also would have hired highly educated employees in the absence of the scheme, there has been looked at the enterprises that found their application for participation in the scheme turned down by the ministry. As part of the evaluation the rejected applicants were contacted and asked whether or not they had hired a highly educated employee even though being denied financial support. The answers of this survey indicate that approximately one third of the rejected enterprises ended up hiring highly educated employees, which strengthens the interpretation of the VP-scheme having a genuine impact on participating enterprises employee compositions.

Page 40

5 The Danish Programme

Industrial

PhD

The Industrial PhD Programme was established in Denmark in 1970. It is internationally recognised for its combination of business experience and academic research. The programme has been evaluated several times and in 2007 an impact assessment was conducted.

5.1 Description of the PhD programme

An Industrial PhD is a three-year business-focused PhD project where the student is hired by an enterprise and enrolled at a university at the same time. The enterprise receives a monthly wage subsidy of DKK 14,500 while the university receives coverage for its supervising expenses. The PhD student works full time on the project and divides his or her time between the enterprise and the university. In 2010, DKK 135 million (EUR 18 million) has been allocated for new Industrial PhD projects. This is equivalent to app. 135 new PhD projects. Accordingly, it is assumed that all qualified applications from the private sector will receive funding. Last year, more than half of all applications were approved. Easy application process Guaranteed answer with approval or rejection from the ministry within 45 working days. Project can start 2 month after application. Max. page report to the ministry every year if there are deviations from the project description

The programme s structure allows the integration of an academic research project into a practical business environment by placing the project firmly within the enterprise organisation. It allows the PhD student to experience and learn from two distinctly different working environments while creating a personal network which spans them both. Thus, the programme has shown to be a very effective network promoter between the universities and the private sector. The programme is open to projects within all scientific fields, all types of enterprises, domestic and foreign, and candidates from all countries. However, the student must be employed in an enterprise branch located in Denmark. The enterprise receives a wage subsidy of about EUR 1,950 per month or EUR 70,000 for the entire three-year duration of the project, which usually corresponds to about half the PhD student s wages. The university enrols the student and receives a subsidy of between EUR 34,000 and EUR 48,000 to cover its various costs.
Page 41

Applications are assessed by the Industrial PhD Committee, an advisory interdisciplinary committee appointed by the Danish Ministry of S cience, Innovation and Higher Education. Table 5.1 shows the development in applications submitted and approved. The activity drop in 2009 is due to the economic crisis. Table 5.1 Applications submitted and approved, 2003-2010

2003 Applications submitted Applications approved

2004

2005

2006

2007

2008

2009

2010

88

89

115

130

174

192

184

179

64

70

83

85

109

119

98

116

The Industrial PhD Programme receives applications from all fields of science. The submitted projects are traditionally mainly from the fields of natural, health and technical sciences. The number of projects within the fields of economics, social sciences and humanities has risen throughout the past few years for a broader academic distribution of Industrial PhD projects, cf. table 5.2: Table 5.2: research field distribution of approved applications, 2003-2010
Agr. and veterinary sciences 13 2 4 2 5 5 5 12 Economics, managemen t and organisation 4 2 6 5 11 8 8 6

Year 2003 2004 2005 2006 2007 2008 2009 2010

Health sciences 23 19 15 18 16 9 11 19

Natural sciences

Technical sciences 15 30 36 34 34 55 40 43

Social sciences 6 6 7 9 11 10 13 15

Humanities 3 8 7 3 12 12 6 7

3 4 14 20 20 15 14

5.2 The impact of the Industrial PhD programme

This section presents an analysis of the impact of the Industrial PhD programme on participating researchers. It has been prepared by Oxford Research Company and Centre for Economic and Business Research (CEBR) at CBS in 2007 and 201011 respectively.

Page 42

5.2.1. The 2007-Oxford analysis impact on participating researchers

The Oxford analysis follows the persons involved in an Industrial PhD project in the period 1994-2004 using personal level register data. It studies these persons developments in several success parameters in association with participation in the PhD scheme: salaries, employment, employment place and mobility changes Employment is measured in terms of employment rates at a given point in time, employment place is a measure of the PhD candidates place of employment before and after project end. Wage is measured in terms of average wage costs which can be interpreted as a proxy for productivity per PhD employee. In terms of methodology, the analysis considers the level of employment, wage levels and places of employment for individuals participating in an Industrial PhD project compared to individuals with a PhD not participating in an Industrial PhD project. Moreover, we apply a matching procedure and identify a control group of PhD students that do not participate in the scheme (non-participants), but are similar to the participants in terms of education, sex, age and region. Again, we can use personal-level data to calculate the development in wages and employment for the non-participants, allowing us to address the question of whether participants have experienced higher increases in the success parameters than would be expected on basis of the matched non-participants. The results of the analysis with respect to employment is summarised as follows: (a) Over a nine year-period, starting in the year of end of the Industrial PhD project, participating individuals have a higher employment rate on average compared to non-participants with a PhD project. (b) One year after the end of the project the employment rate for participating individuals were 90.1 percent and 84.4 percent for non-participants. (c) Nine years after the end of the project the employment rate for participating individuals had increased to 96.1 percent and 91.5 percent for nonparticipants. The development over the period is to be found in table 1 below. For these comparisons, any differences between the two groups of PhD projects can be interpreted as the causal impact of the programme on participating persons under the assumption that the developments of participants and nonparticipants would be symmetric in the absence of programme participation.

Page 43

Table 5.3. Employment rate for participants and non-participants

The results of the analysis with respect to wages can be summarised as follows: (d) 22 per cent of Industrial PhDs have an annual income between 550,000 DKK and 1 million DKK. For ordinary PhDs only 18 per cent have an income in this interval. (e) 24 per cent of Industrial PhDs have an annual income between 450,000 DKK and 550,000 DKK. For ordinary PhDs only 18 per cent have an income in this interval. (f) For income levels below 450,000 DKK the share of ordinary PhDs having an annual income in this interval is higher than the share of Industrial PhDs having an annual income below 450,000 DKK. Figure 5.1. Wages for participants and non-participants

Page 44

The enterprise is the project owner. It ensures research education at a highly specialised level. The university is responsible for the quality of the research training. The evaluation process for applications is relatively fast, and there are three application rounds each year, for optimal flexibility. S ince enterprises employ the Industrial PhD-student and are project owners the impact on enterprises is also highly relevant. Therefore, in 2011 an impact assessment of the programme was concluded. The results of this analysis are presented in section 5.2.2.4

5.2.2 The 2011 CEBR-analysis: Impact of the Industrial PhD Programme on individuals and enterprises
A research-based impact analysis of the Industrial PhD Programme has been carried in 2011 by the Centre for Economic and Business Research (CEBR) at Copenhagen Business S chool at the request of DAS TI. The analysis follows approximately 430 individuals and approximately 270 companies that have participated in the programme. At the individual level, Industrial PhD wages and occupations are compared to regular PhDs and other university level graduates. At the company level, developments are analysed within four success parameters: the number of patents applications, gross profit growth, total factor productivity, and employment growth. The results of the 2007-Oxford analysis with respect to impact on individual level were confirmed. On the individual level, Industrial PhDs are compared with regular PhDs and other university graduates with regard to income and occupation. Groups compared are matched by field of study, gender and other similarly significant factors. The results of the 2011-analysis indicates that Industrial PhDs:

Based on The Analysis of Danish innovation policy - The Industrial PhD Programme and the Innovation Consortium Scheme , The Danish Agency of S cience, Technology and Innovation 2011

Page 45

earn approx. 7-10 pct. higher wages than both regular PhDs and comparable university graduates when adjusting for age, are more likely to be found at the top levels of their organisation than regular PhDs, are more likely to be found in positions requiring high-level specialist knowledge than regular university graduates, but less so than regular PhDs. With respect to the impact at enterprise level, the 2011-analysis identified a control group of enterprises not participating in the programme but with the same characteristic as the participating enterprises. The selection of the control group is based on the propensity score matching method. The basis for comparisons is the year before the first Industrial PhD project was initiated. Under identifying assumptions, these models isolate the causal impact of the programme on companies hosting Industrial PhD projects. The analysis shows that enterprises that host Industrial PhDs see on average increasing patenting activity, illustrated by Figure 5.2. The figure shows the average number of patent applications per enterprise in terms of change relative to the year before initiating an Industrial PhD project. Figure 5.2 Number of patent applications

The 2011-analysis also shows that participating enterprises are characterised by high growth in gross profit, and more positive developments in gross profit and employment growth than companies in the control group, see figure 5.3 and 5.4. However, the analysis is not able to identify robust relationships between hosting Industrial PhD projects and total factor productivity developments. In

Page 46

summary the impact assessment at enterprise shows that hosting Industrial PhD projects: Almost doubles the number of patents per year in the years after initiating the first project (significant to the 10 pct. level), Improves gross profit development by DKK 2 mil. a year for a cumulative effect of DKK 30 mil. in the first five years after initiating the first project (significant at the 5 pct. level), Has no significant effect on total factor productivity, Improves employment growth by more than two employees per year in the first five years after initiating the first project (significant at the 1 pct. level) Figure 5.3 Gross profits in participating and non-participating enterprises

Page 47

6 The Danish Innovation Voucher S cheme

The Danish Innovation Voucher Scheme (in Denmark called Knowledge vouchers ) was established in 2008 and has been a success which is documented by the number of applications and grants (approximately 1000 projects) since 2008. It is acclaimed for its combination of clear set-up, smooth administration as well as for its capacity to generate synergies between business experience and academic research. The scheme has been introduced to inspire small and medium sized enterprises to utilise the opportunities and potential of making use of the knowledge of the knowledge institutions. At the same time, the Innovation Voucher Scheme is expected to enhance the knowledge institutions awareness of the need for knowledge and thus secure the quality and societal relevance of public research. The scheme is open for projects within all scientific fields and the administrative structure of the scheme is designed in a slim manner to reduce bureaucratic measures as much as possible for the project participants. The Council of Technology and Innovation runs the scheme and has decided that state co-funding is channelled directly from the Danish Agency of Science, Technology and Innovation to the research or technological partners of the projects, relieving the SME from the burden of invoicing its project partner.

6.1 Description of the voucher scheme

An Innovation Voucher project is a research-business-partnership project designed along one of two different tracks: 1) A basic voucher for a research-based business development project with a state co-funding level of 40 percent, but max. DKK 100,000 (approximately 14,000). Main focus in the basic voucher is the successful transfer of knowledge from research to SME, in order to place the SME and its activities in a position ahead of its market 2) An extended voucher with similar characteristics as a larger scale R&D collaboration project with a state co-funding level of 25 percent, but maximum 500,000 (approximately 67,000). Main focus of the extended voucher is finding new solutions to current problems. It is a prerequisite for the extended voucher projects that the participating knowledge institution itself carries out research on the field in question. The enterprise is the formal applicant and experiences an evaluation process for applications which is relatively fast - 30 days from end of month of applying.
Page 48

There is one call every year with a first come first served basis and no fixed deadline, for optimal flexibility. This structure enables integration of academic research and practical business. The projects are effective network promoters between research institutions and the private sector. Two very different working environments are matched in the projects, making way for creating a personal network with a potential to spin off future collaboration. The applying SMV must be a private sector SME located in Denmark whereas the knowledge institutions may very well be placed abroad. Applications are assessed by the secretariat of the Danish Council for Technology and Innovation To meet a changed societal context and in close dialogue with key actors, the scheme has undergone a revision during the first quarter of 2010. Central aspects of this revision included a reduction of the co-funding share for the basic voucher from 50 per cent to 40 per cent of the total budget (max. DKK 100,000 / approximately EUR 14,000) and the maximum limit of funding for the extended vouchers was reduced from DKK 1.5 million / approximately EUR 200,000 to DKK 500,000 / approximately EUR 67,000. Thus far the voucher scheme has been received positively by the target group. Enterprises as well as knowledge institutions have been active partners in a wide range of projects. During the first three years of the scheme, the top five research fields of approved applications have been: 1. Industry 2. Retail 3. Knowledge service 4. Information and Communication Technology 5. Agriculture/forestry/fishery Table 6.1 Applications and grants approved in 2008-2010 Vouchers Applications submitted Applications approved Basic Voucher Extended Voucher

1059 699

53 35

Table 6.1 shows the accumulated number of applications submitted and approved in 2008-2010. The total budget of the basic vouchers was approximately EUR 24 million of which 40 per cent was financed by the Danish Council for Technology and Innovation. The total budget of the extended vouchers was EUR 13 million of which 25 per cent was financed by the Danish Council for Technology and Innovation.

Page 49

6.2 The impact of the basic voucher

The Innovation Voucher S cheme is up for a comprehensive econometric impact assessment in 2013 of the basic voucher. This up-coming analysis of the impact of the voucher scheme on participating enterprises will follow the same methods as the impact analyse of the Industrial PhD and Danish VP scheme (chapter 4-5). However, the basic voucher type was mid-term evaluated in the last half of 2009. The evaluation was based on a survey among participating enterprises, and the conclusions provided positive indications for the long-term effect of the scheme. Focus in the evaluated projects has been on the classical and functional innovation areas like product and process development. According to the participating enterprises and research institutions the basic voucher contributes via the adoption structured, long-term procedures and numerous different types of processes to promoting a professionalization of the collaboration between business partners and research and technology institutions. 77 per cent of the enterprises are expecting that they will take part in future development projects with a research and technology institution. 53 per cent of the enterprises and 54 per cent of the research and technology institutions declare that throughout the course of the innovation voucher projects they have developed increased motivation to establishing long-term cooperation relations with one another. 69 per cent of the knowledge institutions state that they are gaining a better understanding of the enterprise requirements. 53 per cent of enterprises state that they are getting better at viewing and assessing the advantages of collaborating with the knowledge institutions. This indicates that the innovation voucher scheme is functioning as an eye-opener with regard to gaining an effect of increased knowledge collaboration. The main lesson with regard to the knowledge at play in the projects is that this is new knowledge or a combination of existing knowledge. The evaluation shows that the Danish innovation vouchers help supporting and stimulating the development of small and medium sized enterprises in co-funded projects. Without the co-funding of the scheme, 43 per cent of the S MEs would not at all have initiated similar activities. The majority of the enterprises expect that as a direct result of their experiences from participating in an innovation voucher scheme project, they will enter into future collaboration projects with research and technological institutions.

Page 50

Likewise, a majority of the enterprises hold the view that the technology or knowledge they acquire from an innovation voucher project is very much leading market technology. The Danish Innovation Voucher scheme thus generates additional business research, development and innovation just as it contributes to stimulating research-, development- and innovation activities amongst the S MEs which are not used to collaborating with universities and technological service institutes. Easy application process and no administrative burdens on SMEs Application of maximum 1 page 90 per cent of S MEs do not make the application themselves but asked the R&D institution. Guaranteed answer with approval or rejection from the ministry within 1 month. Project can start 1 month after application. Fast lane with approval in the ministry within 1 week if necessary. Max. page report to the ministry when the project is concluded No administrative burden for the SME money are channel directly from the agency/ministry to the R&D institutions

6.3 The impact of the extended voucher

The Danish Council for S trategic Research and the Council of Technology and Innovation have in the period 2007-2011 run three various schemes for extended research vouchers. These schemes have been evaluated in 2011. This section presents an analysis of the impacts of the extended voucher. The schemes are relatively young and an econometric impact assessment will be conducted in 2014-15. The analysis includes participation of S MEs in research projects at the universities or the technological institutes in the period from 2007 to 2010. The three schemes included in the analysis serve different purposes, criteria and administration; however, common for them is that the central means to include S MEs lies in research. The purpose of the analysis is to illustrate: Which SMEs are participating? How and why are the SMEs participating? What is the potential for participation? What do the SMEs gain from the collaboration? How can the collaboration be expanded? The main conclusion is that the three schemes form an important framework for involving S MEs in the research collaboration with public research institutions. The collaboration creates value for the companies and contributes by making the research relevant and practice-oriented. At the same time there is a big potential to gain more of the collaboration and to involve more relevant companies.

Page 51

Overall, the analysis shows that it is a complex research and innovation policy job to support the knowledge and competence requirements of S MEs. The great variation in the S ME profiles, motivation and barriers demands a multi-pronged strategy which can include a plurality of schemes and means. The general scheme of the Council for S trategic Research provides for more long-term research collaboration with many actors and appeals to the more research-intensive S MEs. Additionally, large grants are provided with a smaller share for the participating SMEs. The Council for S trategic Research s special S ME-scheme is addressed to S MEs either without (or with limited) experience with public-private interaction, and small grants are provided for subprojects, but often with a large share of participating S MEs. The grants are relatively small and short-term and involve relatively few actors. The Council for Technology and Innovation s extended voucher scheme contemplates for short term R&D project and appeals more to S MEs with some experience with R&D activities and who have a specific idea about a development project. An extended voucher provides indirect grants to S MEs via universities (option to purchase) i.e. a kind of voucher for purchasing R&D services from knowledge institutions and thus contemplates for collaboration between an S ME and a knowledge institution. S MEs can receive money directly from SME-scheme of the Council of Strategic Research. The analysis shows that the three schemes, in many ways, together constitute just such a multi-pronged strategy. Although the three schemes appeal to many of the same types of S MEs, a clear division of labour seems to be present among the schemes. However, there may be a potential for development in clarifying how the schemes build upon each other as well as how the three schemes can be connected to other research and innovation schemes, e.g. the industrial Ph.D. initiative, Knowledge Pilot scheme or the GTS system. Over 90 per cent of the participating S MEs are R&D performing and are generally well placed to participate in research collaboration with research institutions. Thus, much suggests that the schemes attract S MEs with greater R&D activity than the general picture of Danish SMEs. The analysis shows that the S MEs participating generally are motivated by access to new knowledge, networks, strengthening of competences as well as the development of new products or services. However, a more varied picture emerges of the motives across the means, which reflect different strategies, conditions and readiness to collaborate with different research institutions while making something out of it. Furthermore, the analysis shows that there are great variations in the barriers for different S MEs. The great variation in motives and barriers among S MEs for participating in R&D activities illustrates that the idea of a straight path from idea to invoice is not straight at all, but rather complicated and very diverse. The analysis suggests that the participation of S MEs in the schemes have significant behavioural effects and effects on innovations for the participating S MEs. Participation in the research activities thus strengthens their ability to develop and apply new technologies and knowledge. Moreover, the participation strengthens their ability and willingness to continuously cooperate with the research institutions. Furthermore, a majority of the companies (approx. 70 per cent) expect concrete innovations from the collaboration. Especially, the R&D capacity and strategies of the enterprises are of

Page 52

importance to the effects and interviews suggest that project financing and size are of importance to the effects as well. At the same time the participating research institutions say that the projects improve their ability to apply research in practice e.g. in the business sector. The research institutions point out that S MEs contribute to strengthen research towards better exploitation of research results, increased practice orientation of the research as well as greater interdisciplinarity in the research, hence an increased ability to solve societal challenges. There are significant spill over-effects of the participation of S MEs in R&D activities, which go far beyond the actual project (e.g. increased ability to cooperate and exploit technology etc.) and which are relevant for learning and networks both for the companies themselves as well as their cooperating partners. For the identification of the long-term economic effects see chapter 11 and 14.

Page 53

7 The Danish technological service system (GTS -net)

GTS Advanced Technology Group is a network consisting of nine individual research and technology organisations in Denmark the GTS -institutes. The main function of the network is to disseminate new knowledge and technology to enterprises and public institutions in order to support innovation and development. The GTS -institutes constitute the core of the technological infrastructure in Denmark, e.g. testing facilities, certification and approval activities. The network has about 3,700 employees (2010) and a total turnover of 460 million EUR (3.4 billion DKK). Research and development represents 22 p.c. of the turnover. Approximately, 40 p.c. of the research and development turnover comes from performance contracts with the Danish Council for Technology and Innovation. The GTS -institutes each have their individual profile, which varies according to size, turnover, research intensity, sector vs. technological focus (broad or deep) and historical origin.

7.1 Description of the GTS-net

GTS is an abbreviation of Godkendt Teknologisk S ervice which is translated as approved technological service provider . Approved refers to the official procedure where the Minister of S cience, Technology and Innovation approves each institute for a period of three years. The approval is given on the basis of technological/ professional performance, financial performance and organisational solidity. The nine GTS -institutes are independent not-for-profit organisations with the purpose to transfer and disseminate technical know-how, new methods and knowledge to industry and society in order to create and increase development. All services are marketed on a commercial and competitive basis in Denmark and abroad. The government does not financially support these services. Therefore, the GTS -institutes must develop competences and services, which are attractive in their particular markets. In general the GTS-network has two main functions: 1. To develop and maintain the basic technological infrastructure in Denmark: Provide access to basic technological competences, which enterprises do not have internally and which are not available on a market basis, e.g. standardisation activities, access to equipment, facilities and laboratories, testing, educational courses and other technological services. 2. To create technological innovation and development within Danish industry: Development and transfer of new technological know-how, e.g. new methods, concepts and service, in order to stimulate the

Page 54

development of new knowledge-intensive products, service and processes in Danish enterprises and institutions.

Approved institutes have the opportunity to apply for government financed performance contracts to finance R&D activities with the purpose to develop new competences, knowledge and technology and to disseminate this to enterprises and institutions. The performance contracts can also focus on developing new technological infrastructure such as testing facilities and laboratories. In 2010 the performance contracts amounted to EUR 39,9 million (DKK 297 million). The GTS -institutes are also involved in research and development projects in cooperation with enterprises, universities and research institutions in Denmark and abroad. In total, the institutes generates approximately EUR 60,8 million (DKK 453 million DKK) from competitive programmes that are administrated by Danish ministries, agencies, regional authorities as well as the European research programme and other foreign programmes. The GTS -institutes themselves add another EUR 27,2 million (EUR 203 million) in research and development. The core function of the GTS -network is to deliver technological know-how to enterprises and public institutions in order to increase innovation and competitiveness of the Danish industry and society. This is being done in close collaboration with Danish and foreign universities. The GTS -network develops competencies, know-how, methods and technological services, which are strongly relevant for Danish industries and which is not available in the private sector. The GTS -network thereby ensures that the newest knowledge and knowhow is available to the Danish industries.

7.2 The performance of the GTS-net

GTS turnover, EUR (2010) Danish commercial turnover, private Danish commercial turnover, public 460 million EUR 146 million EUR 34 million EUR

Page 55

Commercial turnover outside Denmark Performance contracts (R&D) Other R&D funding, national and international Number of unique customers in Denmark Unique enterprise (customers) in Denmark Public customers in Denmark Number of employees

179 million EUR 40 million EUR 61 million EUR 20.664 17.342 1.679 3.681

As the table shows the GTS-network reaches a broad part of the private sector. The number of private customers in Denmark was 17.342 in 2010. The total number of active enterprises in Denmark was 153.000 in 2008. Due to the GTS network a larger part of Danish enterprises, primarily small and medium-sized enterprises are in contact with the Danish innovation system, ensuring that knowledge gets translated into value The GTS -institutes are increasingly internationalising its commercial activities all over the world. In 2010 nearly 40 per cent of the commercial turnover came from abroad. S ome institutes are already represented in a range of countries; some mostly rely on exporting their services.

GTS turnover, EUR 460 million (2010)

13% 32% 9% Danish commercial turnover, private Danish commercial turnover, public Commercial turnover outside Denmark Performance contracts (R&D) 7% 39% Other R&D funding, national and international

Despite international expansion the core of the knowledge base is still located in Denmark, though development activities outside Denmark is starting to flourish. The large majority of foreign offices are located in S weden, Norway and Western European countries whereas a few offices exist in North America, Russia and East Asia. Regarding research, the GTS -institutes play an important role in the Danish participation in the global knowledge exchange and import. The network has a responsibility in adapting and transferring global knowledge to the local context through participation in international research and development projects, e.g.

Page 56

the European research programs and through formal and informal relations with universities and research institutions abroad. Core activities of the GTS-institutes 1. AgroTech Institute for Food and Agriculture - the fields of expertise cover research in agriculture, food, biology and technology 2. The Alexandra Institute - research in research, user driven innovation, software, pervasive computing, ICT security, pervasive healthcare and interactive spaces 3. Bioneer focuses on biomedicine, biotechnology and medical technology, and facilitates commercialization of biotechnology research. 4. The Danish Institute of Fire and S ecurity Technology focuses on fire, protection and security through consultancy, knowledge and S ecurity Technology building, educational activities and accredited services. 5. DFM - Danish Fundamental Metrology has the primary task to support Danish businesses by offering technical measuring, consultancy and calibration at the highest international level. 6. DHI - Institute for Water, Environment and Health focuses on ecology and environmental chemistry, water resources, hydraulic engineering, flow technology, sediment transport and toxicology consultancy. 7. DELTA - Danish Electronics, Light and Acoustics carries out testing and certification as well as relaying knowledge in the fields and of electronics, microelectronics, sensor technology, light, optics, acoustics and vibration. 8. FORCE Technology focuses on production optimization, materials, inspection, testing, maritime and technology, energy and the environment, certification, sensor technology and management systems. 9. Danish Technological Institute (DTI) focuses among others on construction, energy, informatics, environment, institute materials, food, productivity, logistics and industry development.

7.3 The impact of the GTS-net

S ome of the main activities of the GTS -net are to promote R&D collaboration projects with enterprises. Below is presented a new analysis from the Danish Agency of S cience, Technology and Innovation (2011) of the impact of GTS business collaboration projects on labour productivity and profits in the participating enterprises. The analysis is conducted on panel data covering a total of more than 17,000 unique Danish enterprises and their productivity levels and R&D collaborations over the period 1997-2008. The analysis is based on three samples: Research and development statistics from 1997-2008. The annual sample varies from 2.097 4.523 covering a total of 17.825 unique companies. Corporate data on company performance from 1999-2008. The annual sample varies from 14.707-17.026 covering a total of 19.633 unique companies.
Page 57

Educational statistics data on employee educational levels from 1999-2008. The annual and total samples correspond to the Danish corporate data. Even though the analysis is based on Danish data the results should not be expected to be different in similar studies conducted on data of similar enterprises from Germany, S weden, Finland, Norway, Poland, Iceland or other countries with the same type of projects. It is shown that 332 enterprises having national collaboration with technological organizations (GTS institutes) have a significantly higher growth in labour productivity in each of the years after the collaboration in comparison to a control group of 332 statistical identical R&D-active enterprises with the same likelihood of having collaboration but without this type of collaboration projects. See chart 1. The positive productivity effect is noticeable until five years after the first instance of collaboration. After five years, other companies catch up on labour productivity growth. The control group is established using propensity score matching. The treatment group had collaboration with GTS -institutes once in the time period covered by the analysis. The matching group is determined by having the same probability of having this kind of collaboration given a variety of factors (education level, R&D department, R&D personnel, sector, size etc.), but ending up with no collaboration with GTS -institutes in the analyzed period. Each match is based on the nearest-neighbor principle comparing the two most equal companies. S ubsequently, development in value added per employee is compared over time among the treatment group and the control group. The analysis is based on various productivity functions to be implemented through OLS regressions for each of the years in the data period. The analysis is made for 664 enterprises in the data set and is based on data covering the period 1997 - 2007.

Page 58

Chart 1 Impact of collaboration with GTS-institutes

20 18 16 14
13,39 11,68 10,07 8,75 6,6 3,92 2,26 0 4,42 13,24 12,58 11,63

S amspil med GTS

16,76 15,56 13,91

18,88 17,9

Procent

12 10 8 6 4 2 0 1

Sign. 5% S ign.10% Treatment Kontrol 3 4 5 6 r efter samspil 7 8 9

Based on 1998-2008 data the impact of collaboration projects between enterprises and technological institutions is estimated. The analysis finds that over a period of 9 years the average increase in labour productivity is 7,3 percent. The 332 companies in this analysis have a total added value of just over 5.8 billion over a 9-year period. These results are significant. 10.

Page 59

Table 1 Brief description of the nine GTS-institutes for technological service AgroTech AgroTech was established in July 2007 based on a separation of the more research intensive activities from Danish Agricultural Advisory S ervice, National Centre (DAAS ). AgroTech was approved as GTS -institute in the middle of 2007. Attaching the agricultural industry to the technological service system symbolises the fact that modern agriculture is a technology-intensive industry with consequences for food and energy production. The activities in AgroTech merge knowledge within biology and technology in order to promote development and evolution in enterprises operating in the agricultural and food processing sectors. AgroTech s focus areas compose agricultural technology and food innovation, environment and energy technology, biomass processing, sensor technology and ICT, horticultural development, rural development and testing and documentation of new technologies. AgroTech s Danish market segment includes approximately 3-5,000 enterprises that deliver products and services to customers within agriculture, food processing and production. The Alexandra Institute The Alexandra Institute is a research-based limited organisation which bridges the gap between the IT corporate sector, research and education. The objective of the Institute is to participate in creating innovation in the corporate sector based on research results as well as providing researchers and students with an opportunity of testing their theories and ideas in practice and get inspiration through concrete collaboration projects with the enterprises. The Alexandra Institute represents the direct collaboration between research environments and the corporate sector. The services provided by the Institute are called research-based knowledge services . This comprises joint research and development projects, research networks and dissemination of information, e.g. courses, seminars, workshops and conferences. Furthermore, the Alexandra Institute provides consultancy services. All the services are based on the latest results within Danish and international IT research. The services cover a broad range of sectors and are provided to all kinds of enterprises and public institutions. Bioneer Bioneer A/ S is an innovative research-based organisation providing services within biomedicine and biotechnology. Bioneer contributes to the acceleration of enterprises pre-clinical development of new innovative pharmaceuticals and thus strengthening these enterprises competitive position. Bioneers commercial mission is to increase the competitiveness of their global customers by facilitating the costumer s early, explorative development of new therapeutics. Bioneer offers access to essential tools, technology and unique biological resources within cell and molecular biology, and Bioneer further aims to create new product opportunities at the interface between biology and other sciences.
Page 60

The Danish Institute of Fire and Security Technology The Danish Institute of Fire and S ecurity Technology (DBI) is responsible for improving the Danish knowledge base and services within fire technology. DBI is also responsible for developing and maintaining modern testing facilities within fire technology. The activities and business areas primarily target prevention and security/ safety in and around buildings, facilities and ships. In accordance with increasing societal needs for safety and security, DBI is expanding its knowledge-base and services into non-fire related security technologies. These activities compose i.a. risk management and specific services related to mechanical and electronic security systems. DBI is also the largest Danish provider of education services and short term courses in safety and security.,DBI recently built a Fire and S ecurity Centre where professionals and users of fire and security technologies have the opportunity to try out and learn about all types of fire and security equipment and systems. DBI plays an important role setting norms and developing national as well as international standards, through participation in around 70 national and international fire and security technological committees. DELTA Danish Electronics, Lights and Acoustics DELTA is an ICT (Information, Communication and Technology)- organisation with core technologies in electronics, microelectronics, software technology, light, optics, acoustics, vibration and sensor technology. DELTA s traditional business area concentrates on testing of new microelectronic ICT-products. Moreover, DELTA disseminates know-how concerning the application of new technologies. DELTA develops and produces AS ICs, micro sensors for application in medico technological products, application of optical filters and equipment for process surveillance within biotechnology and the pharmaceutical industry. DELTA specialise in the innovative yet intensely practical application of advanced technologies. DELTA seeks to develop competences to meet the customers within their product development processes. The purpose is to support product design and choice of technology as well as identify potential failures as early as possible. Thereby, DELTA is developing a one-stop-shopping concept. It offers the enterprises advisory services from idea to proto type, testing and certification as well as troubleshooting when the new products fail during the testing phases. Finally, DELTA has a strong focus on innovative deployment of existing technologies. The goal for DELTA is to develop a technology platform based on its sensor technology competences. Danish Fundamental Metrology DFM supports Danish industry with metrological know-how at an internationally competitive level in order to ensure confidence in Danish calibration aboratories and providers of measurement services. DFM develops and maintains selected national standards. In order to ensure international acceptance of Danish standards, DFM participates in European as well as international research projects and global cooperation within metrology. DFM is responsible for coordinating Danish performance within metrology through
Page 61

the network association DANIAmet. S trategy and policy development in metrology is rooted in the Ministry of Business Affairs and Growth. The ministry appoints and approves national metrology institutes within specific areas. DFM s customers are primarily enterprises that develop and produce measurement equipment for the industry as well as calibration enterprises. DFM is strongly research-intensive and has a smaller relative commercial turnover compared to the average GTS-institute. DHI DHI is one of the leading consultancy and research enterprises within in waterrelated technologies. The advisory services are based on research, development and innovation activities resulting in advanced technology within ecology and environmental chemistry, water resources, hydraulic structures and hydrodynamics. DHI performs advisory services, modelling and environmental analyses in its laboratories and it markets a range of software products all over the world. DHI has carried out projects and services in 140 different countries. DHI is the most internationalised GTS -institute with offices and customers in all parts of the world. Its customers include industrial enterprises, consultants, entrepreneurs, energy enterprises, harbours, shipping enterprises as well as public authorities, Danish and international NGOs and development banks. DHI hosts and services UNEP-DHI Centre for Water and Environment, WHO Centre for Water and Health and the Global Water Partnership. FORCE Technology FORCE Technology is among the leading technological consulting and service enterprises in Denmark and internationally. We make a targeted effort to sell highly specialised engineering knowledge for practical and cost-effective solutions to a wide range of industries. The solutions enhance customer competitiveness and are based on the customer and industrial insight that we have acquired over more than 65 years. FORCE Technology is divided into 9 business areas, each with its expert knowledge and its core competences, cooperating in cross-organisational project teams of experts. They ensure the road from knowledge to value, and that our customers are always provided with the most qualified solution. Danish Technological Institute The Danish Technological Institute (DTI) was founded as an independent institution in 1906. With more than 900 employees, DTI is one of the world s largest Institutes of its kind supplying approved technological services such as consultancy, tests, certification, applied research and training for enterprises and public sector organisations. The Institute adopts an interdisciplinary approach to innovation and to the task of improving the ability of small and medium-sized enterprises to exploit new technologies. Thus one of the Institute s most important functions is to ensure that new knowledge and technology can quickly be converted into value for the customers

Page 62

8 The Danish Innovation Incubator scheme

In 1998 The Danish government established an innovation incubator scheme with the general objective to promote commercialisation of new innovative ideas, inventions and research, particularly through the creation of new knowledge based start-ups.

8.1 Description of the Danish Incubator scheme

Under the scheme six regional incubators provide professional counselling and early stage capital for entrepreneurs and new innovative enterprises. The innovative incubators are co-located with regional science parks, also providing access to office-, ICT- and laboratory facilities vis-a-vis Danish universities. In legal terms, the innovative incubators are private limited enterprises appointed as operators by The Danish Ministry of S cience, Innovation and Higher Education following a European public procurement.
Figure 8.1 Location of the incubators in Denmark

Page 63

Financially, the incubator programme is supported by a national budget grant of approximately 200 million DKK annually (26.5 million annually). The innovative incubators operate at the earliest stage of the investment chain, where venture capitalists and other private investors are reluctant to engage. According to the scheme provisions the innovative incubators can engage financially in three successive stages: 1) Pre-investigation: A preliminary analysis and evaluation of the technological perspectives and commercial potential of the project (due diligence). In average an amount of DKK 80,000 can be allocated for this stage. 2) Primary project funding: Pre-seed funding for the initial capital injection and early stage development activities in the start-up. On behalf of the government the incubator in this stage can invest a maximum of DKK 2.5 million (EUR 330,000) in the form of loans or equity, provided that a supplementary private investment is raised equalling 18 per cent or more of the total primary investment. 3) Secondary project funding: S eed funding for further development activities. On behalf of the government the incubator in this stage can invest a maximum of DKK 3.5 million (EUR 470,000) in the form of loans or equity, provided that a supplementary private investment is raised equalling 60 or more per cent of the total secondary investment. The government investment in start-ups under the innovative incubator programme is managed in accordance with the EU block exemption on state aid for young innovative enterprises. At an annual basis the innovative incubators perform 250-300 pre-investigations and invest in 50-70 new primary projects. The investment portfolio of the innovative incubators includes a total of approximately 270 enterprises. Exit is required, when the start-ups are ready to rely exclusively on private market investors.

8.2 The economic impact of the incubator scheme

This section presents an analysis of the impact the Danish Innovation Incubators have had on the innovative entrepreneurs. The analysis has been prepared by the Centre for Economic and Business Research (CEBR). The analysis combines collected data from the Innovation Incubators and from S tatistics Denmark and will show if enterprises and entrepreneurs, that have received economic support from the Innovation Incubators have performed better, worse or at the same level as other enterprises in relation to specified success parameters.

Page 64

Focus for the analysis is survival rate, economic results and growth as well as the development in wages and employment situation for the individual entrepreneur, compared to a control group of fellow enterprises and entrepreneurs that have not received financial support from the Innovation Incubators. In terms of methodology, the analysis considers econometric/ statistical methods, where the effect of Innovation Incubators effort will be clarified by controlling for determining factors that can affect the variables. Approximately 66 per cent of the 245 enterprises which have received economic support from the Innovation Incubators belong to only one of the 27 main categories in the S tatistic Denmark classification of trades; Business service. Even within the category Business S ervice , the enterprises are mainly located under the sub-categories IT-service and R&D. Because of this unequal distribution, two datasets are presented. One dataset includes a control group consisting of all branches. A second control group consists of the two branches where approximately 80 per cent of the enterprises which have received economic support from the Innovation Incubator scheme belong. S ince the investment horizon is relatively long for enterprises which have received economic support from the Innovation Incubators, the impact measurement for the year 2005 is based on enterprises which have received economic support from the Innovation Incubators between 1998 and 2002. The impact assessment variables for enterprises are: 1) Probability for survival, 2) Average value increment, 3) Member of staff and 4) Turnover. The impact assessment variables are for individual entrepreneurs: 1) Income before tax, 2) Disposable income and 3) Unemployment rate a year. The results from the survival analysis as well as the three effect variables show a highly insignificant variation with a value not statistically different from zero.5

The same tendency goes for the entrepreneur related results.

Page 65

Table 1 Key figures

All Branches Innovator enterprises Incubator Control Group All Branches Two Branches Innovator Incubator enterprises Two Branches Control Group

Survival

- 11,9 percentage point

- 2,95 percentage point

Staff members

4,3

3,9

3,8

Turnover

3,97 M DDK

5,6 M DDK

3,28 M DDK

3,86 M DDK

Average increment

value

1,46 M DDK

1,62 M DDK

1,27 M DDK

1,37 M DDK

Enterprises and entrepreneurs that have received economic support from the Innovation Incubators are on par with enterprises and entrepreneurs that have not received economic support from Innovation Incubators. This applies especially to the data from the confined analysis, where the control group profile has the best possible match. Enterprises that have received economic support from the Innovation Incubators have a higher risk profile than other enterprises and have more R&D intensive products with a long way to potential marketing. Despite an amplified risk profile the Innovation Incubators seem to succeed in developing enterprises as viable and economical powerful as other enterprises. On the short range Innovation Incubators produce viable enterprises with staff, turnover and income on par with other enterprises and entrepreneurs. They contribute to the economic activity in society as such and create value that would not have been. On the long range one would anticipate that the effects of the investment within R&D&I will make them exceed the performance of other enterprises.

Page 66

9 The Technology Transfer Offices

9.1 The Technology Transfer Offices at the universities
At each university there is a technology transfer office and a science park. In January 2000, the new Act on Inventions at Public Research Institutions took effect with new regulations for commercial exploitation of the results of public research. From this date, researchers have been obliged to report their inventions to the Technology Transfer Office of their institution. The staffing profile 2010
FT E 70,0 9,5 60,0 9,0 50,0 10,9 40,0 9,3 30,0 20,0 10,0 0,0 2006 Law 2007 2008 Economy 2009 Science Other 2010 28,0 26,6 16,7 8,0 8,4 10,3 15,4 13,3 18,2 13,7 19,35 20,4 20,2 19,6 17,9 10,3

The total number of technology transfer staff within the institutions was 68. The staffing profile has an equal balance between lawyers and professionals with an educational background in economy or science .

In 2010, five of the 14 institutions surveyed reported a staffing level of more than five full-time equivalents. In average, the institutions occupied a technology transfer staff 5 full time equivalents.

Page 67

Technology transfer staffing education/ background 2010

Number of full-time equivalent employees in technology transfer at end of 2010 What is the primary educational background of staff in the institution s technology transfer unit (state as full-time equivalents)?

Legal, in total

Commercial, in total

Technical/ Scientific, in total

Other, in total

Universities (8) Aalborg University Aarhus University Technical Denmark University of 4,5 14,3 15,5 0,9 8 0,25 7 0 0,1 8 2 0,9 3 0,25 2 2 2 6 0 2,5 0 2 1,6 2,3 4 0 1,5 0 2 0,8 2 3,5 0 1 0 1 -

IT-University Copenhagen University Roskilde University University of Denmark Copenhagen School Government institutes (2) GEUS Research hospitals (4) Region of Northern Jutland Region of Denmark S outhern S outhern Business research

Statens Serum Institut

2 0,1

1 0,1

0,5 0

0,5 0

0 0

4 1 10 0,2

2 0 1 0

0 1 4 0,2

2 0 4 0

1 0 1 0

Capital Region of Denmark Region of Central Jutland

Total (13 institutions)

67,75

20,35

20,2

17,9

10,3

S everal of the employees occupied with technology transfer only spend part of their time on this task. Consequently, the total number of individuals engaged with technology transfer is substantially greater than the number of full-time equivalents stated in the survey. All Technology Transfer Offices are members of the National Network for Technology Transfer, a network of public research institutions in Denmark that commercializes research results and patents.

Page 68

9.2 Commercialization of public research metrics

The annual commercialisation survey illustrates the efforts and performance of public research institutions in technology transfer. The commercialisation survey has been conducted jointly by The Danish Agency for S cience, Technology and Innovation and The National Network for Technology Transfer. Data is collected through a questionnaire according to the international standards for such surveys. The presented data include all public research institutions with significant commercialisation activities. The survey for 2010 is based on data from a total of 14 public research institutions. The group of institutions surveyed includes eight universities, two government research institutes and four research hospitals (regional hospital administrations). All public research institutions with significant commercialisation activities are hereby covered by the survey. IP exploitation by public research institutions in Denmark 2000-2010

140 N u m b er o f co m m ercialisatio n s 120 100

14 16 11 9 12 7 115 7 8 4 35 10 35 45 81 89 90 74

80 60 40 20 0 2000 2001 2002 2003 2004 2005 2006 2007 2008

2 11

10 3

28

2009

2010

Licenses, options- and assignments

Spinout companies formed

Page 69

Key figures 2010

Universities Government research institutes Research hospitals Total 2010 Total 2009

Invention disclosures received

255

69

328

291

Patent applications filed

95

25

122

129

Patents issued

13

16

Licenses, options and assignments executed (incl. software) Total license portfolio
(excl. software)

96

103

74

107

17

127

107

Spin-out enterprises formed

10

11

Technology transfer staff

(full-time equivalent)

50,5

2,1

15,2

67,8

66,2

Commercialisation costs
(DKK million)

30,6

7,5

42,1

46,4

License income received

(DKK million)

42,8

64,7

2,8

110,4

83,1

Research agreements with private enterprises

2.020

710

2.737

Page 70

Institutional data
Invention Patent disclosures applications received filed Patents issued Licenses, Total options license and assign- portfolio ments (incl. software) (excl. software) Spin-out Technology enterprises transfer staff formed (full-time equivalents) Commercia License -lisation income costs (in DKK (in DKK 1,000) 1,000) Research agreements with private enterprises

Universities (8) Aalborg University Aarhus University Technical University of Denmark IT-University Copenhagen University Roskilde University University of Southern Denmark Copenhagen Business S chool Government research institutes (2) Statens Serum Institut GEUS Research hospitals (4) Region of Northern Jutland Region of Southern Denmark Capital Region of Denmark Region of Central Jutland Total (14 institutions)

53 49 87 6 40 4 16

13 11 46 2 11 1 11

0 0 6 0 2 0 0

35 17 22 0 18 1 3

9 21 10 0 58 1 8

4 4 2 0 0 0 0

4,5 14,3 15,5 0,9 8 0,25 7

1.801 7.808 14.010 406 5.009 0 1.604

2.347 3.663 31.208 0 5.063 350 214

391 331 802 7 306 72 76

35

4 0

2 0

3 0

0 0

3 0

0 0

2 0,1

3.875 102

64.770 0

7 -

10

400

68

25

34

13

16

10

6.300

2.617

466

22

0,2

808

189

151

328

122

13

103

127

11

67,8

42.123

110.401

2.737

Page 71

10 The National Advanced Technology Foundation

The Danish government has founded the Danish Advanced Technology Foundation, whose general objective is to enhance growth and strengthen employment by supporting strategic and advanced technological priorities within the fields of research and innovation. Furthermore, the foundation shall also make an effort to promote research and innovation in small and medium-sized enterprises.

10.1 Description of the National Advanced Technology Foundation

All projects which are relevant to advanced technological research and/ or innovation may apply for grants at the Advanced Technology Foundation. The foundation will pay special attention to applications which fall within the areas of nano-, bio-, and/ or information and communication technology, including the interface between these areas. The Danish National Advanced Technology Foundation will support selected fields and technologically advanced projects or consortiums which have a range of participants that will contribute financial means in varying degrees. The projects and consortiums that receive support must live up to three criteria: Market: Noticeable commercial potential Technology: Internationally high quality research and/or innovation Entrepreneurship:Public-private collaboration This means that at least one public research institution and one private enterprise must participate. Furthermore, public enterprises and Centres for higher Education may participate. The foundation attaches importance to whether or not the participating parties have the necessary means at their disposal in order to carry out a major project. According to the Act s definition of registered in Denmark , all enterprises and research institutions, Danish as well as foreign, may participate in the priorities and projects supported by the Advanced-Technology Foundation. Foreign enterprises and research institutions may also participate as sub-contractors in a priority or project. All grants must be awarded on the basis of an evaluation of scientific quality and industrial relevance. For this evaluation, the executive committee may choose to appoint its own advisory expert sub-committees, but scientific research evaluation is to be made by the Danish Research Advisory S ystem or the Danish National Research Foundation.

Page 72

All intellectual property rights, which can be obtained on the basis of projects financially supported by the foundation, will, according to the Act on the Danish National Advanced Technology Foundation, fall to the participating public research institution(s) and Centres for Advanced Education respectively. The enterprises that participate may obtain these rights on ordinary market terms. The types of expenses which may receive grants are listed below: Expenses for personnel, including PhDs, who contribute to the activities. Expenses for lasting material (equipment and instruments) which are used permanently for the activities. Expenses for consultants and services which concern the activities only, including the purchase of research, technical knowledge, patents etc. Other overheads which stem directly from the activities Other operating expenses (e.g. materials, supplies or the like which are directly connected to the activities).

10.2

The economic impact

In the period 2005-2010 the Advanced Technology Foundation has financed 143 projects with more than 200 enterprises. The economic impact of the Advanced Technology Foundation has not yet been evaluated by using the same methods as in the other econometric impact studies presented in this report. However, it is up for an econometric impact assessment when sufficient data and time series are available.

Page 73

The Economic Impact of R&D and Innovation

Impact of business R&D on productivity and growth

S ocio-economic impact of R&D

Impact of R&D personnel on productivity

Impact of research-business collaboration

Page 74

11 Impact of business R&D&I on productivity and growth

The competitive position of all countries is increasingly dependent on the development and application of new knowledge and technology. The rapid rate of technology development and growing global competition make it essential that enterprises conduct Research, Development and Innovation (R&D&I) in order to keep pace. Many European governments have the aim that their countries should continue to be among the leading countries in terms of R&D. This focus makes it important to have detailed quantitative evidence on the impact of investments in research and development. Increased knowledge on the effects of research and development among enterprises will also be valuable in informing the design of research and innovation policies to improve conditions for R&D&I. This chapter provides a presentation of the results of a comprehensive analysis of the impact of R&D&I on productivity for Danish enterprises, Produktivitetseffekter af erhvervslivets forskning, udvikling og innovation (Danish Agency for Science, Technology and Innovation, 2010). The analysis includes a broad range of all enterprises with 10 or more employees that were covered in the R&D&I statistics from 1997 to 2005 and where financial data was available from S tatistics Denmark. The analysis includes 2,694 enterprises with R&D activities over the period 1997-2005 and 9,558 enterprises without R&D for the same period. This chapter summarizes the results of: A descriptive analysis which compares the productivity of R&D active enterprises to enterprises without R&D. The analysis examines a number of groups of enterprises, by industry, exports and cooperation with public research. Firm level econometric analysis of the effect of R&D on productivity. An analysis of the aggregated effects of Danish business R&D using S tatistics Denmark s ADAM model. The results and effects of the impact assessments presented in this report might be of general interest when evaluating the impact of private investments in R&D&I in other European countries.

Page 75

11.1 Comparative descriptive analysis on average labour productivity

As a first step the analysis compares the labour productivity measured by value added per employee - of enterprises with and without R&D. A variety of different groups were compared in the analysis, for example by industry and enterprise size, exports and whether enterprises had collaborated with public research institutions. Table 11.1 shows basic results for industry groups and size classes. In all cases, average value added per employee is higher for enterprises with R&D; and in all cases this difference is statistically significant. The same pattern was found for exporting enterprises, where average value added per employee for R&D active enterprises was DKK 513,000 (EUR 69,000) compared to DKK 457,000 (EUR 61,400) for exporting enterprises without R&D. 1 EUR is equivalent to 7.44 DKK. Table 11.1. Value added per employee by R&D activity, industry and firm size
Value added per employee (R&D active) High tech manufacturing Other manufacturing Knowledge intensive service Other service < 100 employees 100+ employees Total (all enterprises) 492,100 471,000 512,600 518,600 482,600 504,500 493,900 Value added per employee (no R&D) 399,100 382,800 480,100 447,100 435,700 422,000 431,300 Difference in pct. 23 % 23 % 7% 16 % 11 % 20 % 15 % Number enterprises 1,871 3,806 2,608 3,620 9,336 2,916 12,252

Source: Calculations by CFA based on R&D statistics and financial data from Statistics Denmark. Amounts in DKK.

Among enterprises with R&D, those enterprises that have collaborated with public research have significantly higher value added than other enterprises with R&D. And, both groups of R&D active enterprises have significantly higher value added than enterprises without R&D.

Page 76

Figure 2

Average productivity per employee by R&D activity and cooperation with public research

Source: Calculations by CFA based on R&D statistics and financial data from Statistics Denmark.

Finally, R&D active enterprises are compared to innovative enterprises without R&D and with non-innovative enterprises. Average value added for both R&D active enterprises (DKK 493,000 equivalent to EUR 66,000) and innovative enterprises without R&D (DKK 455,000 equivalent to EUR 61,000) is significantly higher than for non-innovative enterprises (DKK 429,000 equivalent to EUR 57,000).

11.2 Impact of R&D and innovation on enterprise marginal productivity

The impact of R&D&I on enterprise productivity has been estimated both for individual years (cross-sectional analysis) and for the entire period from 19972005 (panel data analysis). Analyses were also conducted for individual industry groups and size classes. For the analyses, a measure of enterprises R&D capital was created using standard methods. Table 11.2 shows the main results of the analysis6. The results show a significant, positive relation between R&D and productivity. This positive impact was found both for the pooled analysis (analysis for entire period using OLS ) and for the panel data analysis. This result was found likewise for all regressions for

The quantitative results for the estimation of these two models can be found in Annex 11.A 1-2.

Page 77

individual years and industry groups. Additional details on the data, methods and results can be found below in the Annex. For the complete set of results, please see the full report. Table 11.2
Variable: Log (R&D capital/employee) Log (Physical capital/employee) Log (number employees) Public ooperation Business cooperation Export intensity Market concentration (Herfindahl index) Constant + *** Insignificant

Main results of the productivity analysis

OLS model 1 + *** + *** - *** Insignificant Insignificant + *** Insignificant + ** -* OLS model 2 + *** + *** - *** Fixed effects model 1 +* + *** - *** Insignificant Insignificant + ** -* Fixed effects model 2 + *** + *** - ***

+*

-*

+ ***

+ ***

R-sq: 0.389 0.389 0.231 0.232 Number obs 4143 4143 4143 4143 Dependent variable: Log of value added per employee. Signs of coefficient estimates shown with +/-. Level of statistical significance given by: *** = 1 pct.-level, ** = 5 pct.-level and * = 10 pct.level. Source: Calculations by CFA based on R&D statistics for 1997-2005 and financial data from Statistics Denmark for 1998-2006.

The results of the econometric analyses have also been used to estimate what the results imply in terms of increases in value added. In other words, for a firm with average R&D intensity, what is the return of an extra crown of R&D expenditures in terms of value added, based on the results from the econometric models? As table 11.3 shows7, based on the results for all enterprises, the return of R&D expenditures in terms of value added is 66 percent an extra crown invested in

More information on the method used for this calculation can be found below in Annex 11.A 1-2, and in the full report.

Page 78

R&D results in an increase of 0.66 crowns in value added. Returns vary across industry groups, ranging from 17 percent for the high tech sector to 112 percent for knowledge intensive service and 135 percent for low tech services. Table 11.3 Returns to value added of investments in R&D
Return of an extra crown invested in R&D High tech manufacturing Other manufacturing Knowledge intensive service Other service Less than 100 employees 100 or more employees All enterprises 17 % 98 % 112 % 135 % 52 % 76 % 66 %

Source: Calculations by CFA based on R&D statistics for 1997-2005 and financial data from Statistics Denmark for 1998-2006.

11.3

Overview of other empirical studies

In order to put these results into a broader context, Table 11.4 provides an overview of results from other analyses for comparison. Please consult the full report for a more detailed discussion of other Danish and international analyses. Table 11.4 Overview of empirical studies of R&D and productivity
Study Graversen and Mark (2005) Griliches (1980a) Schankerman (1981) R&D elasticity coefficient 0.02 - 0.11 0.03 0.07 0.10 0.16 Population 662 Danish enterprises; 19912001 39 US manufacturing industries; 1959 to 1977 110 US enterprises (Chemical and petroleum); 1963 crosssection 77 US enterprises (research intensive sectors); 1966 to 1977 525 US enterprises; 1973 to 1980 406 Japanese enterprises; 1973 to 1980 197 French firms; 1980 - 1987, cross-sectional estimation. time series-estimation

Griliches and Mairesse (1984) Griliches and Mairesse (1990)

0.19

0.25 0.41 Sample 1 0.20 0.56 Sample 2 0.05 0.25 Hall and Mairesse (1995) 0.00 0.07

Page 79

Minasian (1969) Griliches and Mairesse (1984) Cuneo and Mairesse (1984) Griliches (1986) Sample 1 Sample 2 Englander, Evenson, and Hanazaki (1988) Mansfield (1988) Wang and Tsai (2003) Danish Agency of Science, Technology and Innovation (2010)

0.08 0.09 0.20

17 US enterprises; 1948 to 1957 133 US enterprises; 1966 to 1977 182 French enterprises; 1972 to 1977 491 US enterprises;

0.11 0.20 -0.16 0.50 0.42 0.19

1972 cross section 1977 cross section 16 industries in 6 countries; 1970 to 1983 17 Japanese industries 136 Taiwanese manufacturing enterprises; 1994 to 2000 20,000 Danish manufacturing and service enterprises investing in R&D activities; 1997 to 2007

0.05-0.199

Source: Chapter 19, Congressional Budget Office 2005, Graversen and Mark (2005) and Hall and Mairesse (1995) and Productivity effects of business research, development and Innovatoin Danish Agency of Science, Technology and Innovation (2010).

It can be seen that our results are quite comparable to many of the results in the table, both concerning the cross-sectional estimates and the fixed effects estimates. This is particularly noteworthy given that the sample used in the study presented in this chapter is far more diverse and considerably larger than those used in earlier Danish and foreign analyses. In comparison with earlier studies, the analysis presented in this chapter is much more representative, including enterprises from both manufacturing and service sectors, and as small as 10 employees. Both the inclusion of service enterprises and small enterprises are fairly unique to R&D productivity analyses.

11.4

Data and method

The objective of the econometric analysis is to estimate the effects of R&D on productivity for Danish enterprises, and to make the results as representative as possible by including essentially all enterprises (with 10 or more employees) for which we have adequate data. This thus gives a sample which is much more comprehensive than in earlier studies (see the overview of earlier studies below). The analysis is based on data from R&D&I statistics and business data from S tatistics Denmark, over the period 1997-2005 for R&D and 1998-2006 for business data. This annex provides a brief description of methods used and main results of the econometric analysis. A more detailed description of the data, methodology and results can be found in the full report.

Page 80

The econometric model is based on a Cobb-Douglas function which includes R&D capital:
1 Qt Ae t K t 1 Lt 1 Rt 1e

where Q is output (value-added), A is TFP, K is physical capital (fixed assets), L is labour (number employees) and R is R&D capital. In the analyses, all variables are expressed per employee and in logs. To minimize potential simultaneity bias, we use input measures from the beginning of the year and output from the end of the year (in other words, all input measures are lagged one year). R&D capital is calculated using the perpetually inventory method. And, we correct double counting of labour by subtracting the number of R&D full time equivalents (FTE) from number of employees. Earlier analyses have typically considered either a cross-sectional analysis for a single year or a panel data analysis (cross-section over time), for example using a fixed effects model. This analysis has conducted estimations using both methods to ensure greater robustness of results. To a certain degree the two methods are also complementary in their interpretations; estimations using cross-sectional methods (OLS ) can be interpreted as examining the permanent effects of R&D for example the effects of a permanent increase in R&D while panel data methods (which focus more on variation for individual enterprises over time) can be seen as measuring the effects of a temporary increase in R&D. Over the sample period, the R&D statistics include in total 19,228 observations (12,252 enterprises), 6,647 with R&D and 12,581 without R&D. This full data set is used for the comparisons of value-added per employee. The panel used in the analysis was constructed with the intent to include as many enterprises as possible: Given that the model is in logs, all observations with values that are negative or equal zero are removed from the sample. Enterprises with 2 or more consecutive observations are included. For single period gaps in a time series of 4 or more observations, the single missing observation was estimated. For gaps of more than one period, the later part of the series was kept and observations before the gap were removed. Enterprises with shares of R&D/sales greater than 1 or less than 0.001 were removed from the sample. Extremely large growth rates (negative or positive - growth rates of value added, physical capital, employees or R&D capital were either less than -90 percent or greater than 300percent) in main variables have been removed.

Page 81

This yields a dataset for the econometric analyses with 4,143 observations (out of the total of 6,647 observations with R&D). The table below shows some descriptive statistics for the panel dataset. Table 11.5 Descriptive statistics for the panel dataset
All
Value added Value added/emp. R&D expenditure R&D/sales R&D capital R&D capital/emp. Employees Number obs. 61,200 470 2,896 2% 15,712 124 143 4,143

High tech manuf.

55,970 453 4,438 3% 24,282 184 138 1,599

Other manuf.
79,361 430 1,824 1% 9,757 55 196 1,224

Knowledge intensive services

31,184 556 3,464 7% 20,080 289 68 984

Other services
81,446 503 2,277 1% 12,390 101 162 324

Under 100 employees

18,210 458 1,251 3% 6,510 159 41 1,562

100+ employees
122,703 477 5,545 2% 31,980 108 263 2,531

S ource: Calculations by CFA based on data from R&D and innovation statistics for 1997-2005 and financial data from Statistics Denmark for 1998-2006. All nominal variables in 1000 DKK and are deflated using a GDP deflator (all variables in 2004 prices). The analysis includes enterprises with 10 employees or greater.

11.5

Econometric results

Table 11.6 and 11.7 show the main results of the econometric analysis. Table 11.6 shows the results of the pooled regression: by pooled is meant that all observations over time are pooled and estimated as a single cross section (using both year and industry dummies). Table 11.7 displays the results of the fixed effects analysis. The dependent variable is the log of value added per employee. The explanatory variables used in the analysis are: Log of R&D capital per employee Log of physical capital per employee Log of number employees (corrected for double counting) Dummy variables for cooperation with public research and businesses Interaction terms between cooperation and R&D capital Export intensity (exports as share of sales) Herfindahls index as a measure of market concentration and competition Model 1 in Table 11.6 shows a coefficient of 0.125 for the log of R&D capital, which means that one per cent increase in R&D capital will lead to a 0.125 per cent increase in productivity per employee. The table also shows different modeling of the relationship between corporate investment in R&D and productivity, in particular to examine whether cooperation affects the relation between R&D and productivity. The various models show positive, significant and robust impacts of R&D in all cases.

Page 82

Table 11.6 Results from the OLS-regression of the model

Variable: Log (R&D capital/emp.) Log (Physical capital/emp.) Log (Employees) Public cooperation Public cooperation * Log (R&D capital/emp.) Business cooperation Business cooperation * Log (R&D capital/emp.) Export intensity Market concentration (HHI*) Constant R-sq: Obs 0.087*** -0.014 0.113* 0,389 4143 0.088*** -0.014 -0.087* 0,389 4143 0.089*** 0.022 0.113* 0,392 4143 0.008 Model 1 0.125*** 0.172*** Model 2 0.126*** 0.172*** Model 3 0.109*** 0.171*** Model 4 0.102*** 0.169***

-0.019*** -0.017*** -0.021*** -0.018*** 0.022 0.102*** 0.043*** 0.107*** 0.052*** 0.089*** 0.043 -0.174*** 0,394 4143

Dependent variable: Log of value added per employee. Signs of coefficient estimates shown with +/-. Level of statistical significance given by: *** = 1 pct.-level, ** = 5 pct. and * = 10 pct.. Source: Calculations by CFA based on R&D statistics for 1997-2005 and financial data from Statistics Denmark for 1998-2006.

Moreover, the models show that there is a positive correlation between export intensity and productivity of enterprises. This is a result which was indicated in the comparative analysis, where R&D active enterprises with exports had an average productivity per employee, which was higher than non-R&D active enterprises. The competition which enterprises face is included in the models through a market concentration index. Corporate competition is not found to affect the productivity of the established models. These estimates are also used to calculate the marginal return of R&D in terms of value added (see Table 11.3), which make use of the fact that coefficients can be interpreted as an elasticity ; (e.g. using the coefficient for model 1) a 1 per cent increase in R&D capital per employee yields a 0.125 per cent increase in value added per employee. The calculations are based on the average firm in the sample, and size of the returns will depend on the elasticity, the ratio of R&D capital to R&D expenditures and the ratio of R&D to value added. Table 11.7 shows the results of fixed effects regressions on the same data as Table 11.6. The table shows the model 1, a correlation of 0.044, which means that one per cent increase in R&D capital will lead to a 0.044 per cent increase in productivity per employee. The table also shows different modelling of the
Page 83

relationship between corporate investment in R&D and productivity. The various models again show positive, significant and robust results for the impact of R&D in all cases. Table 11.7 Results of the fixed effects regression of the model
Variable: Log (R&D capital/emp.) Log (Physical capital/emp.) Log (Employees) Public cooperation Public cooperation * Log (R&D capital/emp.) Business cooperation Business cooperation * Log (R&D capital/emp.) Export intensity Market concentration (HHI*) Constant R-sq: Within between Overall Obs. 0.113** -0.614* 2.014*** 0.389 0.231 0.03 0.018 4143 0.113** -0.620* 2.026*** 0.389 0.232 0.03 0.019 4143 0.112** -0.609* 2.026*** 0.392 0.231 0.03 0.018 4143 -0.022 Model 1 0.044* 0.058*** Model 2 0.045* 0.059*** Model 3 0.047* 0.058*** Model 4 0.043* 0.058***

-0.523*** -0.524*** -0.525*** -0.522*** -0.008 -0.022 -0.008 -0.012 0,005 0.113** -0.619* 2.012*** 0.394 0.232 0.03 0.019 4143

Dependent variable: Log of value added per employee. Signs of coefficient estimates shown with +/-. Level of statistical significance given by: *** = 1 pct.-level, ** = 5 pct.-level and * = 10 pct.-level. Source: Calculations by CFA based on R&D statistics for 1997-2005 and financial data from Statistics Denmark for 1998-2006.

While significant, coefficient estimates are lower using the fixed effects model compared to OLS above. This result fits well with earlier analyses and provides a reasonable interpretation. The results indicate that both permanent and temporary increases in R&D have positive impacts on productivity, but increases that are maintained over time will have the greatest effect.

Page 84

12 The socio-economic impact of business R&D investment

This chapter provides a presentation of the results of an analysis of the socioeconomic impact of private investments in research and development on growth, exports, productivity and employment. The analysis is based on simulations in the macro econometric model for Denmark called ADAM. The full description of the analysis is found in the report Produktivitetseffekter af erhvervslivets forskning, udvikling og innovation (Danish Agency for S cience, Technology and Innovation, 2010). Businesses have positive productivity effects of their R&D investments, regardless of whether the R&D takes place in Denmark or abroad. The location of R&D, however, is not indifferently viewed with society's eyes. There are important medium and long term socio-economic productivity and welfare effects if enterprises choose to invest in Denmark. The Danish government has joined the EU objective (Barcelona target) to invest minimum 3 per cent of GDP per year in research. Of these, enterprises must invest 2 per cent of GDP in R&D. But what are the socio-economic effects of meeting the Barcelona target? This question is analysed by using a full-scale dynamic model for the Danish economy called ADAM. The model gives a simplified mathematical description of how the Danish economy functions and is used to predict economic trends and calculate the effects of such economic policy intervention. As input to the ADAM model is used the level of business R&D investments and the marginal return on their investment in R&D (elasticity coefficients from the econometric analysis presented in chapter 11). It is assumed in the model that the changes are permanent. ADAM analyses the effects of a 15-year period. The analysis of the socio-economic effects of increased investment in business R&D is based on business R&D investment levels in 2005. In 2005, business investment in R&D was 1.67 per cent of GDP. This means that business investment in R&D should increase by approx. 5.5 billion DKK to reach the Barcelona target of 2 per cent of GDP. That is equivalent to a permanent increase of 20 per cent. The 5.5 billion DKK is set at 2010 prices and used as input in the ADAM model. In relation to developments in GDP using the factual figures forward until 2008 - then used projections arising from price movements calculated by the Danish Statistics Office.

Page 85

Results from the ADAM model show that in the short term the effects are significant. Increasing R&D investments in the medium term, the effects show that employment has increased by almost 30,000 and GDP by 1.75 per cent. In the long term, effects on employment and exports are, however, declining. This is due to the fact that there will be pressure on employment, since the increased activity results in increased employment and that the workforce is declining in Denmark due to demographic development. This means that wages will eventually increase, making it more expensive to recruit to the same degree. Long-term employment increases slightly. The increased activities also increase prices, thus losing competitiveness and hence exports. The results thus imply that: There will be a significant growth in the Danish economy as a result of the increased R&D investments. GDP increases by 0.88 per cent already in 2011, and in 2015, GDP increases with 1.75 per cent equivalent to 27.7 billion DKK in 2010-prices. R&D investment will further improve competitiveness as Danish enterprises have improved productivity per employee by providing competitive products based on new knowledge and technology. The increased R&D investments thus strengthen Danish exports. Exports grow by 0.67 per cent already in 2011 and in 2015; exports will have increased by 1.01 per cent compared with the development if no changes are made in business R&D investment. At the same time Danish enterprises experience a more efficient production process, which will make production more valuable and thus lead to greater value per employed in Danish enterprises. According to the model productivity increases by 0.3 per cent in 2011 and 1.15 per cent in 2015. Tax revenues rise by 0.5 per cent in the year 2011, and in 2015 they increase by 2.3 per cent. Furthermore, the increased investment implies that more Danes come into work, thus helping to improve public finances. The increased R&D investment means that almost 8,000 more people will be employed in 2011. And in the year 2015 enhanced investment, means that employment will increase by almost 30,000. Table 12.1 summarizes the results after 1 year, 5 years and 15 years. Table 12.1 Effects of permanently increasing business R&D investments by 5.5 billion DKK per year, constant prices.
Year: 2011 2015 2025 Change in employment
7,700 employees 29,810 employees 12,375 employees

Change in productivity
0.3 pct. (7.6 bill DKK.) 1.15 pct. (33.0 bill DKK) 1.13 pct. (37.8 bill DKK)

Change in export
0.67 pct. (5.6 bill DKK) 1.01 pct. (9.1 bill DKK) 0.34 pct. (3.7 bill DKK)

Change in GDP
0.88 pct. (12.8 bill DKK) 1.75 pct. (27.7 bill DKK) 1.87 pct. (34.6 bill DKK)

Change in tax revenue

0.5 pct. (3.6 bill DKK) 2.3 pct. (13.3 bill DKK) 4.0 pct. (11.9 bill DKK)

S ource: Calculations in S tatistics Denmark s ADAM model based on R&D statistics for 1997-2005 and financial data from Statistics Denmark for 1998-2006.

Page 86

The results will underestimate the full socio-economic impact of business R&D. The full socio-economic impact implies that: Increased employment from increased investment will increase the disposable income, which in turn strengthens the domestic demand. Depending on the extent of increased imports resulting from increased domestic demand, productivity in Denmark will grow and provide increased prosperity through increased tax revenues and lead to even higher disposable income for the consumer. An increased transfer of knowledge helps to increase productivity, since new knowledge incorporated in products and processes contributes to a more effective production and makes it more efficient, e.g. knowledge transfer can be implemented more or less formally by interaction between enterprises and between business and public knowledge institutions. Knowledge is a "non-rival 'good, and therefore it does not lose value by being shared. Rather, it is argued from different sides that the value of knowledge increases the more it is used. An increased technology transfer which will have lasting positive effects for the Danish economy, e.g. investing in R&D means that there can be incorporated new technology into products and services that will provide increased productivity. The manufacturer will have higher marginal revenue of products either due to higher efficiency or higher prices. The effect applies not only for the producer, but for the entire value chain all the way to retail or wholesale.

Page 87

13 Productivity and higher education

The Report on Productivity and higher education has been conducted by the Centre for Economic and Business Research (CEBR) for the Danish Business Research Academy (DEA). The following is an executive summary of the report.

13.1

The method and data

The report analyses the effect of different types of highly-educated working capacities on the productivity in 138,372 Danish enterprises. A researcher has followed the development in the enterprises over 9 years, from 1999 to 2007, to obtain results on the added value for each enterprise. In contrast to most others, this analysis differentiates the individual effect of education and the joined effect. The individual effect is expressed by the highly educated employee obtaining a higher productivity which is often reflected in the fact that people who are educated receive higher wages than people without education. The joined effect is expressed by the effect on the productivity of the whole enterprise if more highly educated working capacity is applied. This means that the higher productivity is not only reflected by higher wages among employees with higher education, but also by the individual enterprise becoming more productive. With regard to the individual effect the analysis shows that the productivity for each person becomes increasingly higher, the longer the person s educational background is, regardless of the field of education. The analysis also shows that education within social sciences results in the highest individual productivity. Technical, health sciences and scientific educations result in a slightly lower productivity than the social sciences.

Page 88

With regard to the joined effects the analysis shows that these are at least as clear, and in some calculations almost twice as big as the individual effects. All higher education has a significantly positive effect regardless of the field of education. It is shown that technical, health scientific and scientific educations have the greatest effect within the manufacturing sector while the social sciences have the greatest effect within private services. Furthermore, it is a clear tendency that long-cycle higher education has a greater effect than shortand medium-cycle higher education.8

13.2 The impact of higher education on productivity

The results should be taken with caution. In the above and in the following it is implicitly assumed that there is causality where highly educated working capacity causes a higher productivity. However, this is not documented in the analysis. This means that it cannot be denied that highly productive enterprises hire highly educated working capacity and not the other way around that highly educated working capacity makes the enterprises highly productive. The impact of including the joined effects of education can be illustrated by calculating how much value a highly educated person contribute to the productivity compared with a person without education. If only the individual effect of education is taken into account, the contribution of an individual with a long-cycle higher education is equivalent to somewhere in-between 1.2 and 1.6 individual without education. If the joined effect of education is taken into account, the contribution of an individual with a longcycle higher education contribute is equivalent to approximately two individuals without education ranging from 1.7 and 3.2 persons. It is also examined if better results come from hiring people with higher education within different fields, instead of having one kind of employee profile.

Higher education in Denmark can be divided into three sectors, the names of which are used in this summary. Short-cycle higher education includes two-year academy profession programmes, mediumcycle higher education is used to denote professional bachelor s programmes, usually lasting 3-4 years, while long-cycle higher education includes the academic bachelor s and master s programmes offered at university level. (Source: The Agency of International Education)

Page 89

The gain of hiring highly educated individuals is decreasing, but the decrease only appears when the share of highly educated employees exceeds 40 percent. Hiring a different mix of educated people will make this decrease appear much later. On the basis of the results of the analysis it cannot be concluded that there is particular mixture of the highly educated which are more valuable than others. An important question is: to what extent can education contribute to an increase in productivity? Based on calculations in the report the following table (conclusive table) shows how a one percent increase in the share of people with a long-cycle higher education will affect the GNP in the economic part examined by the analysis. If the remaining part of the economy is similar to the part which is analyzed in the report with regard to productivity effects, the results will also reflect the effect on the GNP for all of the economy by increasing the share of employees with a long-cycle higher education by one percentage point. Table 13.1. Conclusive table: Effect on GNP by one percentage point more with long-cycle higher education Humanities Social Sciences Production 0.3 0.6 0.4 1.1 0.5 0.5 0.6 1.5 0.7 0.9 Private service 0.2 0.5 0.3 0.7 0.6 0.6 0.5 1.0 0.7 0.9 Technical 0.6 1.1 0.6 1.5 0.9 0.4 0.2 0.5 0.5 0.8

Individual effect Joined effect measured by TFP Joined effect measured by wage Aggregated effect measured by TFP Aggregated effect measured by wage Individual effect Joined effect measured by TFP Joined effect measured by wage Aggregated effect measured by TFP Aggregated effect measured by wage
Note: Source:

The aggregated effect is the sum of the individual effect, which is weighted with the wages share of the BVT, and the joined effect. Own calculations

As shown a one percentage point increase in the share of employees with a longcycle higher education, will cause an increase in the Gross National Product by approx 1 percent, when taking the individual and the joined effects into account. It should be noted that the effects shown in the table only refer to the increase in GNP that isolated can be connected to the increase in persons with a longcycle higher education. In reality the GNP effect could be a great deal bigger thus more highly educated employees can give the enterprises incentives to invest more. And more investments will naturally cause an increase in GNP.

Page 90

14 The impact of R&D collaboration on productivity, employment and growth

It is difficult to verify whether or not the R&D and innovation collaboration programmes produce significant economic additionality. The main reason is that appropriate data or long data series are not available in most countries. However, in some countries there exists historical data on participation in publicprivate partnerships, research-business collaboration projects and cluster and network activities with the participation of enterprises and research institutions. There also exist enterprise registers with time series information on exports, employment, turnover, profits, productivity per employee etc. Some collaboration projects are pure regional or national activities. Other projects involve enterprises and organizations from other countries. Collaboration projects imply enterprise investments in R&D and innovation activities and research activities in public institutions which in turn create new innovations, new competences and knew knowledge in the enterprises. With help from register data bases it is a difficult but not impossible task to verify some of the economic effects of the collaborative activities. This chapter presents new evidence on the impact of collaboration activities which could be based on the activities taking place in cluster and networks organizations or be the result of various innovations support programmes. The evidence focuses on the impact on employment, economic growth, productivity per employee and exports. The chapter is based on the international literature on innovation and R&D collaboration programmes and new Danish econometric impact studies9.

(1) Productivity effects of business R&D and innovation investments , Forsknings- og innovationsstyrelsen (DAS TI) (2010a), (2) Productivity of business R&D collaboration with public research , Forsknings- og Innovationsstyrelsen (DAS TI) (2011c) and (3) Economic Impact of International Research and Innovation Cooperation , Forsknings- og Innovationsstyrelsen (DAS TI) (2011b)

Page 91

These analyses sheet new light on the economic impact in enterprises of private investments in single R&D projects, national and international R&D collaboration projects, and cross-country business-to-business projects and knowledge transfer activities. The international part of the analyses has been conducted as a collaboration project with participation of Denmark, France, Israel, the Netherlands, S witzerland and Sweden with the support of the EUREKA secretariat.

14.1 International studies on the economic impact of national research collaboration

In the literature different indicators are used to measure the impact of research and innovation collaboration projects and cluster programmes. The most frequent used key performance indicators10 are innovations, sales of new products and services, price mark-up, turnover of the enterprise, patents and productivity per employee. Several studies and evaluations conclude that collaboration between research and business in clusters and innovation support programmes promotes collaboration projects between enterprises and external partners on research and innovation projects and activities. This in turn increases the innovation performance of enterprises, patent activities and sales of new or innovative products and services. Many international studies are less clear on the relationship between cluster and network activities and research and innovation collaboration projects on the one side and productivity per employee, profits and exports on the other side. There are only a few empirical studies on the economic impact of collaboration between enterprises and domestic or foreign enterprises or research institutions. The weakness of most international studies is the lack of access to national data on cluster and networks activities, R&D and innovation collaboration projects and registers on enterprises and employees with detailed information on profits, productivity per employee, education, exports, turnover etc. Most international

10

S ervice innovation: Impact analysis and assessment indicators Final report of Task Force 2 of the EPISIS-project of The European Commission, (2011)

Page 92

studies only have access to databases with few relevant statistics on enterprises and employees. This has made it difficult to conduct econometric analysis on the economic impact of clusters and collaborative activities and to obtain statistical significant results based on sufficient statistical information and a high number of historical observations. Table 14.1 Selected international studies of R&D collaboration and productivity Impact of collaborative activities in enterprise productivity
Data and methods Dutch analysis based on CIS 2 statistics of collaborative enterprises from the Netherlands in 1996. Collaborative projects (with costumers, universities, other enterprises etc.) and productivity development from 1996 to 1998. Analysis of collaborative enterprises from Korea. The analysis focuses on which factors are important for the collaborative behaviour of enterprises and what are the impacts of collaboration. Authors Belderbos, Carree and Lokshin (2004) Key performance indicator: productivity Key performance indicators are productivity per employee and sales of products new to the market. Results of impact study: Positive impact on sales of products new to the market. Eom and Lee (2010) Key performance indicators are productivity per employee, number of innovations, patent activities and turnover. Results of impact study: Significant impact on patent activities but no short term significant impact on productivity per employee, turnover in enterprises and number of innovations. Key performance indicator: Total factor productivity Result of impact study: Return from buying R&D is higher than from own R&D expenditure Key performance indicators: Labour productivity and price mark-up Result of impact study: EUREKAparticipation increases labour productivity and price mark-up

Paneldata from 1974-75 to 1981-82 for buy of R&D of indian enterprises

Basant and Fikkert (1996)

Participants in EUREKAprojects, EUREKA-clusters and EUREKA-networks and 3rd and 4th framework programme of the EU are compared to similar enterprises with respect to labour productivity and price mark-up Analysis with 455 manafacturing enterprises (chemical sector) from Japan

Benfratello and Sembenelli (2002)

Baba, Shichijo and Rita Sedita (2009)

Key performance indicator: Number of patents registered Result of impact study: R&D collaboration increases the number of registered patents

Page 93

14.2 The impact of national collaboration projects

This section presents an analysis of the impact of national research-business collaboration projects on labour productivity in the participating enterprises. It is based on the report Productivity of business R&D collaboration with public research from DAS TI (2011c). The analysis is conducted on panel data covering a total of more than 17,000 unique Danish enterprises and their productivity levels and R&D collaborations over the period 1997-2008. The analysis is based on three samples: Research and development statistics from 1997-2008. The annual sample varies from 2.097 4.523 covering a total of 17.825 unique companies. Corporate data on company performance from 1999-2008. The annual sample varies from 14.707-17.026 covering a total of 19.633 unique companies. Educational statistics data on employee educational levels from 1999-2008. The annual and total samples correspond to the Danish corporate data. Even though the analysis is based on Danish data the results should not be expected to be different in similar studies conducted on data of similar enterprises from UK, the Netherlands, France, Germany, Sweden, Finland, Norway, Poland, Iceland or other countries with the same type of projects. It is shown that 547 enterprises having national collaboration with universities and research have a significantly higher growth in labour productivity in each of the years after the collaboration in comparison to a control group of 547 statistical identical R&D-active enterprises with the same likelihood of having collaboration but without this type of collaboration projects. The effect is largest in the first two years subsequent to an instance of collaboration, as indicated by a significantly higher value added growth per employee compared to firms who have not engaged in collaboration but have the same probability of engaging in this kind of collaboration. See figure 14.1.The positive productivity effect is noticeable until five years after the first instance of collaboration. After five years, other companies catch up on labour productivity growth but not on employment growth and growth in turnover. In the analysis we have established a control group using propensity score matching. The treatment group had collaboration with universities or research institution once in the time period covered by the analysis. The matching group is determined by having the same probability of having this kind of collaboration given a variety of factors (education level, R&D department, R&D personnel, sector, size etc.), but ending up with no collaboration in the analyzed period. Each match is based on the nearest-neighbor principle comparing the two most equal companies. Subsequently, development in value added per employee is compared over time among the treatment group and the control group. For the
Page 94

analysis of effects we make an assumption of causality. The analysis is based on various productivity functions to be implemented through OLS regressions for each of the years in the data period. The functions on a Cobb-Douglas productivity function under the assumption of constant returns to scale for the exponents. Figure 14.1 Impact of collaboration with universities and research institutions
20 18 16 14
12,92 10,38 9,79 8,29 11,13

Collaboration with universities and research institutions

17,77 16,65

17,65

14,65 13,36 12,18 9,98 9,97

Percent

12 10 8 6 4 2
1 5,18

S ign. 10% S ign. 5% Treatment

3,68 2,51 0

Kontrol

0 1

4 5 6 Yearsafter initial collaboration

Based on 1998-2008 data the impact of collaboration projects between enterprises and public research and technological institutions is estimated. The analysis finds that over a period of 9 years the average increase in labour productivity is 9 percent. The 547 companies in this analysis have a total added value of just over 5.7 billion over a 9-year period. The largest enterprise specific gain is found in the manufacturing industry. For enterprises in knowledge services, other services and trade and transport, the effect is less clear. Both small and large enterprises are experiencing gains. However, large enterprises with more than 250 employees experience the largest returns. These results are significant. The effects the first 4 years after the initiation of the collaboration projects are shown in table 3. Table 3 Effect on labour productivity of collaboration with public research (EUR)
Year 1 High-tech manufacturing Low- and medium tech manufacturing Knowledge intensive business services Other enterprises Enterprises with more than 250 employees Enterprises with less than 250 employees 12439*** 9776*** 2341 9254** 13030*** 5321*** Year 2 7893 *** 10745*** 6495** 10509** 13986*** 6559*** Year 3 9088*** 12193*** 8957*** 7573 11016** 8172*** Year 4 7791*** 8122*** 3288 7950* 8879** 4404***

Page 95

Significance: *: p > 0,9; **: p > 0,95; *** p > = 0,99

14.3 The impact of international collaboration projects

This section is based on the Danish report Economic Impact of International Research and Innovation Cooperation from DASTI (2011b). The report analyses EUREKA-data. EUREKA supports international projects such as cross-border collaboration projects and cross-border clusters. As a funding instrument, EUREKA is set apart primarily through its efforts to coordinate national funds for R&D at the international level. EUREKA has no common pot, and relies on national funds. A EUREKA-project is a R&D collaboration project between minimum two enterprises from two countries. EUREKA Clusters are long-term, strategically significant industrial initiatives. They usually have a large number of participants, and aim to develop generic technologies of key importance for European competitiveness, primarily in ICT and, more recently, in energy and biotechnology. Umbrellas are thematic networks within the EUREKA framework which focus on a specific technology area or business sector. The main goal of a thematic network is to facilitate the generation of EUREKA projects in its own target area. Based on data from projects carried out in the period 1990-2000 the report concludes that: Participation in international projects significantly increased growth in labour productivity for participating enterprises compared to similar enterprises not participating. Likewise, these productivity gains were on par with similar businesses participating in national funding instruments. Export growth significantly increased in participating enterprises. Three years after participation, growth in exports for EUREKA-participants was higher than both similar businesses in general and similar businesses participating in national funding instruments. Growth in turnover significantly increased in participation enterprises compared to businesses in general. Likewise, growth in turnover is on par with similar businesses participating in national funding instruments. Participation significantly increased employment growth. This impact is significant from year-one after participation and persists for several years compared to both similar businesses in general and to businesses participating in national funding instruments. In the analysis there have been established two types of control groups. The first control group consists of similar enterprises participating in national collaboration projects. The second control group is found by using propensity score matching as explained in the previous section. The matching group is determined by having the same probability of having this kind of collaboration

Page 96

given a variety of factors (education level, R&D department, R&D personnel, sector, size etc.), but ending up with no international collaboration in the analyzed period. Each match is based on the nearest-neighbor principle comparing the two most equal companies. The treatment group consists of enterprises participating in EUREKA projects with participation of minimum two enterprises from minimum two countries. One third is business-to-business projects. Two third is projects with participation of universities and research or technological institutions.

14.3.1

The impact on labour productivity growth

Labour productivity is defined as the business turnover minus its expenses divided by the total number of full time equivalents in the business. It is used as a measure of the average value created by a business per labour year performed. Figure 14.2 and figure 14.3 compare growth in labour productivity of enterprises in cross-border projects to similar enterprises in general and with similar enterprises participating in national funding instruments, respectively. Participants in cross-border projects are labelled treatment , while the comparison group is labelled control . The shaded areas indicate statistically significant differences between the two groups. Both comparisons are made over a time period of six years since participation in EUREKA. Figure 14.2 shows that compared to similar enterprises, enterprises participating in cross-border projects experienced significantly higher growth in labour productivity three and four years after participation. Three years after participation, labour productivity grew more than twice as fast as nonparticipants. Four years after participation, the difference in growth rates between participants and non-participants further increased. Figure 14.2 Labour productivity growth of enterprises in cross-border collaboration (international projects) compared to similar businesses without cross-border collaboration

25

Growth in productivity (percent)

20

15

10

Sign. 5% S ign. 10%

T reatment C ontrol

0 1 2 3 4 5 Yearsafter partic ipation in E UR E K A 6

Page 97

Figure 14.3

Labour productivity growth of enterprises in international projects compared to businesses participating in national funding instruments

25

Growth in productivity (percent)

20 15 10 T reatment 5 C ontrol 0 1 2 3 4 5 Yearsafter partic ipation in E UR E K A 6

Figure 14.3 shows that growth in labour productivity of EUREKA-participants was not significantly different from that of similar businesses participating in national funding instruments. With some variation, EUREKA-participants and participants in national instruments share similar growth rates in labour productivity throughout the six year period after participation.

14.3.2

The impact on export growth

This section presents the impact on export growth of cross-border collaboration projects. In the analysis, export is defined as the business turnover abroad divided by the total number of full time equivalents in the business. Export per full time equivalent is a measure of the value of the goods and services that businesses sell outside of their national markets per unit of labour. Figure 14.4 shows that compared to similar enterprises participating in national funding schemes, participants in international projects experienced significantly higher growth in exports three years after participation. Three years after participation, exports of participants in international projects (EUREKA-projects, EUREKAclusters and EUREKA-networks) grew more than twice as fast as participants in national instruments. At this point the export growth for participants in international projects and participants in national instruments were 22.4% and 8.8%, respectively.

Page 98

Figure 14.4

Export growths of participants in international projects compared to similar businesses participating in national instruments

45 40

Growth in export (percent)

35 30 25 20 15 10 5 0 1 2 3 4 5 Yearsafter partic ipation in E UR E K A 6 S ign. 5% T reatment C ontrol

14.3.3

The impact on turnover growth

Turnover growth per full time equivalent (defined as the business turnover divided by the total number of full time equivalents in the business) is a measure of the income that businesses generate from the sale of goods and services. Growth in sales is an indication that businesses are growing their markets, selling more goods and services to existing or new customers. Figure 14.5 shows that compared to similar businesses, participants in international projects experienced significantly higher growth in turnover four years after participation. Turnover of participants grew more than twice as fast as non-participants (by 10.5% and 5.1%, respectively). The analysis shows that growth in turnover of participants in international projects is not significantly different from that of similar enterprises participating in national funding instruments. EUREKAparticipants and participants in national clusters, networks and projects share similar growth rates in turnover throughout the six year period after initiation of a collaboration project. Figure 14.5 Turnover growth of participants in international collaboration projects compared to similar businesses not participating

18 16
Growth in turnover (percent)

14 12 10 8 6 4 2 0 1 2 3 4 5 Yearsafter participation in E URE KA 6 S ign. 10% Treatment Control

Page 99

14.3.4

The impact on employment growth

Growth in employment is measured as the relative increase in the number of full time equivalents of labour carried out by a business. The analysis shows that growth in employment is significantly higher in enterprises participating in international collaboration projects compared to similar enterprises not participating and also compared to similar enterprises in national collaboration funding instruments (such as regional or national clusters, networks and projects). The impact of employment appeared from year one after participation. Figure 14.6 shows the picture of employment growth when participants in international EUREKA projects are compared to similar enterprises participating in national funding instruments. EUREKA-participants experienced significantly higher growth in employment throughout the six year period of the impact study. Figure 14.6 Employment growths of EUREKA participants compared to similar enterprises participating in national instruments

30

Growth in employees (percent)

25 20 15 10 5 0 1

Sign. 5% T reatment C ontrol

3 4 5 Yearsafter partic ipation in E UR E K A

Page 100

References
Baba, S hichijo og Rita S edita (2009), How do collaborations with universities affect firms innovative performance? The role of Pasteur scientists in the advanced materials field , Research Policy, Vol. 38, issue 5, side 756-764 Basant og Fikkert (1996), The Effects of R&D, Foreign Technology Purchase, and Domestic and International S pillovers on Productivity in Indian Firms , The Review of Economics and Statistics, Vol. 78, No.2, side 187-199 Bayona m.fl. (2001), firms motivations for cooperative R&D: an empirical analysis of S panish firms , Research Policy, Vol.30, side 1289-1307 Belderbos m.fl. (2004), Cooperative R&D and firm performance, Research Policy, Vol. 33, issue 10, Dec 2004, side 1477-1492. Benfratello og S embenelli (2002), Research joint ventures and firm level performance , Research Policy, Vol. 31, issue 4, side 493-507 Cassiman og Veugelers (1998), R&D cooperation and spillovers: S ome empirical evidence , Economics WP 328, Department of Economic and Business, Universitat Pompeu Fabra Crpon m.fl. (1998), Research, Innovation, and Productivity: An Econometric Analysis at the Firm Level , Economics of Innovation and New Technology, No.7, side 115-58 CBO, 2005, R&D and Productivity Growth , USCongressional Budget Office. Cuneo P. and J. Mairesse, 1984, Productivity and R&D at the firm level in French manufacturing. In Z. Griliches (ed.) R&D, patents, and productivity. University of Chicago Press. Czarnitzki, D. and O'Byrnes, N., 2007. The impact of R&D on productivity, Open Access publications from Katholieke Universiteit Leuven. D Aspremont og Jacquemin (1988), Cooperative and noncooperative R&D in duopoly with spillovers , AER, Vol. 78, issue 5, side 1133-37 Danish Agency for S cience, Technology and Innovation (DAS TI), Forsknings- og Innovationsstyrelsen (2008) Effektmling af forskningsog innovationssamarbejder fokus p innovationskonsortier, Forsknings- og Innovationsstyrelsen, Innovation: Analyse og evaluering 2008 Danish Agency for S cience, Technology and Innovation (DAS TI), Forsknings- og Innovationsstyrelsen (2009) Analyse af forsknings- og udviklingssamarbejde

Page 101

mellem virksomheder og videninstitutioner, Forsknings- og Innovationsstyrelsen, Innovation: Analyse og evaluering 03/2009 Danish Agency for S cience, Technology and Innovation (DAS TI), Forsknings- og Innovationsstyrelsen (2010a), Produktivitetseffekter af erhvervslivets forskning, udvikling og innovation Danish Agency for S cience, Technology and Innovation (DAS TI), Forsknings- og Innovationsstyrelsen (2010b) The Innovation Consortium Scheme an analysis of Firm Growth Effects, Forsknings- og Innovationsstyrelsen, Innovation: Analyse og evaluering 03/2010 Danish Agency for S cience, Technology and Innovation (DAS TI), Forsknings- og Innovationsstyrelsen (2011a) Analysis of the Industrial PhD Programme, Forsknings- og Innovationsstyrelsen, Innovation: Analyse og evaluering 01/2011. Danish Agency for S cience, Technology and Innovation (DAS TI), Forsknings- og Innovationsstyrelsen (2011b), Economic Impact of International Research and Innovation Cooperation , Innovation: Analyse og evaluering 12/2011 Danish Agency for S cience, Technology and Innovation (DAS TI), Forsknings- og Innovationsstyrelsen (2011c), konomiske effekter af erhvervslivets forskningssamarbejde med offentlige videninstitioner , Innovation: Analyse og evaluering 02/2011 Danish Agency for S cience, Technology and Innovation (DAS TI), Forsknings- og Innovationsstyrelsen (2011c), The Impact of Cluster Policy in Denmark (2011d), Innovation: Analyse og evaluering 21/2011 Eom og Lee (2010), Determinants of industry-academy linkages and, their impact on firm performanca: The case of Korea as a latecomer in knowledge industrialization , Research Policy, Vol. 39, side 625-639 EPIS IS (2011), S ervice innovation: Impact analysis and assessment indicators Final report of Task Force 2 of the EPIS IS -project of The European Commission, (2011) Englander, A.S ., Evenson, R. and Hanazaki, M. , 2008, R&D, Innovation and the Total Factor Productivity Slowdown. OECD Economic Studies 11, pp. 8 42. Graversen og Mark (2005), Forskning og Udviklingsarbejdes pvirkning af produktivitet og beskftigelse , Dansk Center for Forskningsanalyse, Aarhus Universitet, Rapport 2005/1; Graversen og Mark (2003), Bestemmende faktorer for danske virksomheders valg af forskningssamarbejdspartnere , Nationalkonomisk Tidsskrift 141, side 339-363

Page 102

Griliches, Z., 1980, Returns to Research and Development Expenditures in the Privat S ector New Developments in Producticity Measurement and Analysis, University of Chicago pp. 419-461. Griliches, Zvi, 1986, Productivity, R&D, and the Basic Research at the Firm Level in the 1970 s, American Economic Review, vol. 76(1), pages 141-54. Griliches, Z. og Mairesse, J., 1984, Productivity and R&D at the Firm level, in Z. Griliches (ed.) R&D, Parents and Productivity, University of Chicago Press, pp. 339-374. Griliches, Z. and Mairesse, J., 1990, R&D and Productivity Growth: Comparing Japanese and U.S . Manufacturing Firms , in: Hulten, C. (ed.), Productivity Growth in Japan and the United States, Chicago: University of Chicago Press. Hall, B.H. and Mairesse, J., 1995, Exploring the Relationship Between R&D and Productivity in French Manufacturing Firms , Journal of Econometrics 65, 263293. Hagedoorn (1993), Understanding the rationale of strategic technology partnering: interorganizational modes of cooperation and sectoral differences , Strategic management journal. Vol. 15, side 291-309

Kastelli m.fl. (2004) , Cooperative R&D as a means for knowledge creation , International Journal of Technology Management, Vol. 27, Number 8, side 71230 Loecker, Jan, (2007): Do exports generate higher productivity? Evidence from S lovenia , Journal of International Economics Vol. 73, side 69-98. Mairesse, Jacques og Hall, Bronwyn Hughes, (1995), Exploring the Relationship Between R&D and Productivity in French Manufacturing Firms Mansfield, E., 1988, Industrial R&D in Japan and the United S tates: A Comparative Study, American Economic Review 78, 223-8. Minasian, J.R., 1969, Research and Development, Production Functions, and Rates of Return , American Economic Review 59(2), 80-85. OECD (2008), OECD Information Technology Outlook 2008 OECD S cience (2008), Technology and Industry Outlook 2008 OECD (2010), Measuring Innovation A New Perspective . Polt m.fl. (2001), Benchmarking Industry-S cience Relations The Role of Framework Conditions , Financial Report. European Commission, Enterprise DG, Benchmarking Coordination office.

Page 103

Robertson og Gatingnon (1998), Technology development mode: a transaction cost conceptualization , S tategic management journal, Vol. 19, side 515-31 Rogers, Mark (2006), Estimating the Impact of R&D on Productivity using the BERD-ARD Data., Oxford University. S chankerman, M., 1981, The Effects of Double-Counting and Expensing on the Measured Returns to R&D , Review of Economics and Statistics 63(3), 454-458. Segarra-Blasco og Arauzo-Carod (2008), S ources of innovationn and industriuniversity interaction: Evidence from S panish firms, Research Policy, Vol. 37, issue 8, side 1283-1295 S olow, Robert M, (1957) Technical Change and the Aggregate Production Function , The Review of Economics and S tatistics, Vol. 39, No. 3. (Aug., 1957), side 312-320. Tether, Bruce S ., (2002), Who co-operates for innovation, and why: An empirical analysis , Research Policy, Vol. 31, issue 6, Aug 2002, side 947-967 Van Beers, Berghall og Poot (2008), R&D collaboration and public knowledge institutions in small economies: evidence from Finland and the Netherlands , Research Policy, Vol. 37 , issue 2 Veugelers, Reinhilde og Cassiman, Bruno, (2005), R&D cooperation between firms and universities. S ome empirical evidence from Belgian manufacturing , International Journal of Industrial Organization, Vol. 23, issues 5-6, side 355-379 Vinding (2002), Absorptive capacity and innovative performance: A human capital approach. Kapitel 7 i Interorganisational Diffusion and Transformation of Knowledge in the Process of Product Innovation., Institut for Erhvervsstudier, lborg Universitet. Wang, J-C. and Tsai, K-H, 2003, Productivity Growth and R&D Expenditure in Taiwan's Manufacturing Firms. NBER Working Paper No. W9724.

Page 104