Technology and Innovation Management: Theories, methods and practices from Germany and China 9783486593129, 9783486586336

Citation preview

Technology and Innovation Management Theories, methods and practices from Germany and China

Edited by

Reinhard Meckl, Mu Rongping and

Meng Fanchen 3., vollständig überarbeitete Auflage

OldenbourgVerlag MünchenWien

Bibliografische Information der Deutschen Nationalbibliothek Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über abrufbar.

© 2008 Oldenbourg Wissenschaftsverlag GmbH Rosenheimer Straße 145, D -81671 München Telefon: (089) 4 50 51- 0 oldenbourg.de Das Werk einschließlich aller Abbildungen ist urheberrechtlich geschützt. Jede Verwertung außerhalb der Grenzen des Urheberrechtsgesetzes ist ohne Zustimmung des Verlages unzulässig und strafbar. Das gilt insbesondere für Vervielfältigungen, Übersetzungen, Mikroverfilmungen und die Einspeicherung und Bearbeitung in elektronischen Systemen. Lektorat: Wirtschafts- und Sozialwissenschaften, [email protected] Herstellung: Anna Grosser Coverentwurf: Kochan & Partner, München Gedruckt auf säure- und chlorfreiem Papier Druck: Grafik + Druck, München Bindung: Thomas Buchbinderei GmbH, Augsburg ISBN 978-3-486-58633-6

Introduction The editors volume “Technical innovation and management in German and Chinese corporations” mainly contains presentations which were given during the “2nd Sino-German workshop on technology management and innovation” which took place at the University of Bayreuth from July 11 to July 14, 2006. The basic objective of the workshop, which was preceded by a workshop two years earlier in Beijing, was to further intensify the transfer of information and knowledge between Chinese and German researchers in the field of technology and innovation management. It was intended to reach and enhance an exchange of current state-of-the-art methods and topics in the field of technology management, and to establish and further deepen the personal dialogue between the respective researchers. We grouped the presentations into five chapters which cover different aspects of the research field. The chapter “Policy and innovation” takes a look at the macro framework of research mainly in China. The articles span from the impressive advances China has made in establishing a market-oriented research environment up to the entrepreneurial spirit evolving in China. The second chapter focuses on the classical field of “Corporate management of innovations”. The articles here reflect and give an impression of the wide domain of topics researchers from both countries currently deal with. The third chapter emphasises the importance of “Network and cooperation in innovation”. It becomes obvious that in both countries the formal and informal networks surrounding the progress in technological knowledge play an important part. A very operative approach is chosen in chapter four with “Tools for R&D management”. A theoretical instrument for market entry and decision making are presented in order to optimize R&D success. In a cross-country workshop, the “Intercultural aspects of technical innovation” must not be missing. In chapter five we deal with different topics in this field from the Chinese and German side. The workshop was financed by the “Sino-German Center for the Promotion of Sciences”, Beijing. We thank the center for the generous support and the efficient handling of the application and organisation process. The center also made a financial contribution to the realisation of this book which we strongly appreciate. We also thank the second sponsor of the book, the “Geheimrat Dr. Fritz Hornschuch’sche Allgemeine Wohlfahrtsstiftung” for supporting us financially and making the book possible.

Bayreuth, Beijing, November 2007

Reinhard Meckl, Mu Rongping, Meng Fanchen

Contents 1

Policy and innovation

1

Mu Rongping: The changing strategy and policy of innovation in China...............................1 1 Introduction .............................................................................................................1 2 Review of the changing innovation capacity in China since 1995 ..........................2 3 Key Factors for Building Innovation Capacity in China .........................................6 4 Changing Innovation Strategy and Policy in China.................................................9 5 Conclusions ...........................................................................................................14 References.....................................................................................................................15 Author ...........................................................................................................................16 Liu Xielin and Nannan Lundin: Toward a market-based open system of innovation in China ............................................................................................................................17 1 Introduction ...........................................................................................................17 2 From the dominance of government research institutes to an enterprise centered system .....................................................................................................18 3 Industry-science linkages ......................................................................................20 4 Company system ...................................................................................................22 5 Key determining factors for the Chinese innovation system .................................24 6 Conclusion.............................................................................................................37 References.....................................................................................................................38 Authors .........................................................................................................................40 Fan Chunliang: The role of the national R&D program in setting up China’s national innovation system..........................................................................................................41 1 Introduction ...........................................................................................................41 2 Theoretical framework ..........................................................................................42 3 The role of China’s national high-tech R&D program (863 Program) in setting up the national innovation system..............................................................44 4 Conclusion and suggestion ....................................................................................48 References: ...................................................................................................................50 Author ...........................................................................................................................50 Zhao Lanxiang: The Role of ERCs: Public and Private Aspects...........................................51 1 Introduction ...........................................................................................................51 2 Background for Establishing the ERCs .................................................................51 3 Characteristic of Chinese ERC Program ...............................................................52

VIII

Contents 4 Do the ERCs Work? A Case of CAS .................................................................... 53 5 Issues to be Discussed ........................................................................................... 55 6 Conclusions and Implications ............................................................................... 56 References .................................................................................................................... 56 Author........................................................................................................................... 57

Gao Jian and Shi Shude: The determinants of entrepreneurial activity in China – Empirical analysis by regional level data .................................................................... 58 Abstract ........................................................................................................................ 58 1 Introduction ........................................................................................................... 58 2 Measurements and CPEA...................................................................................... 59 3 Determinants analysis ........................................................................................... 61 4 Empirical results.................................................................................................... 64 5 Conclusion............................................................................................................. 68 References .................................................................................................................... 68 Authors ......................................................................................................................... 70 2

Corporate management of innovations

71

Hans Koller: Intercultural technology intelligence – a process and communication oriented approach......................................................................................................... 71 1 Introduction ........................................................................................................... 71 2 Technology intelligence – a short overview.......................................................... 72 3 A process and communication oriented approach of technology intelligence....... 74 4 Reflections from an intercultural perspective........................................................ 78 5 Conclusion............................................................................................................. 79 References .................................................................................................................... 80 Author........................................................................................................................... 83 Dirk Holtbrügge and Jonas F. Puck: The change of foreign investors’ ownership modes in the PRC – theoretical framework and empirical results ............................... 84 1 Introduction ........................................................................................................... 84 2 Research design and methodology ........................................................................ 85 3 Findings................................................................................................................. 86 4 Conclusion and implications ................................................................................. 90 References .................................................................................................................... 91 Authors ......................................................................................................................... 95 Chen Jin and Wang Fangrui: A research on relationships in the knowledge management system of biotechnology firms in China................................................... 96 Abstract ........................................................................................................................ 96 1 Introduction ........................................................................................................... 96 2 Definition of the key elements for the knowledge management system ............... 97 3 The concept model and hypotheses..................................................................... 101 4 Methods............................................................................................................... 103

Contents

IX

5 Results .................................................................................................................105 6 Managerial implications and conclusions............................................................114 Appendix A.................................................................................................................116 References...................................................................................................................118 Authors .......................................................................................................................120 Torsten Eymann, Christoph Niemann and Falk Zwicker: Innovating mindfully in healthcare IT using RFID technology.........................................................................121 1 Introduction .........................................................................................................121 2 The IT innovation phenomenon ..........................................................................121 3 Applying the framework to patient logistics in hospitals ....................................125 4 Conclusion...........................................................................................................128 References...................................................................................................................129 Authors .......................................................................................................................130 3

Network and cooperation in innovation

131

Hu Lijun and Shi Junwei: Market structure, social structure and technological innovation: An interpretation of firm behavior in the transitional stage of China .....131 Abstract.......................................................................................................................131 1 Introduction .........................................................................................................131 2 Theoretical development: The role of the social structure in the process of technological innovation......................................................................................133 3 Discussion and application: The technological innovation behavior of firms in transitional China ............................................................................................137 4 Concluding remarks.............................................................................................142 References...................................................................................................................142 Authors .......................................................................................................................143 Duan Yibing and Tang Le: Key Factors in the Successful Commercialization of Public Sector Research: Implications of the Case of CAS Spin-offs ......................................145 Abstract.......................................................................................................................145 1 Introduction .........................................................................................................145 2 Spin-offs policy and performance in the CAS.....................................................146 3 Key factors to create and operate spin-offs in the CAS.......................................147 4 Conclusion...........................................................................................................149 References...................................................................................................................149 Authors .......................................................................................................................150 Martin Kloyer and Roland Helm: Contractual design of contract R&D: state-of-the-art of empirical research ..................................................................................................151 1 Research question................................................................................................151 2 Contract R&D......................................................................................................152 3 Positive Institutional Economics and contract R&D ...........................................154 4 Conclusion...........................................................................................................158

X

Contents References .................................................................................................................. 159 Authors ....................................................................................................................... 165

Liu Yun, Cheng Guo-ping and Yang Yu: Research on Influencing Factors of the Performance of Enterprises’ R&D Cooperation ........................................................ 166 Abstract ...................................................................................................................... 166 1 Introduction ......................................................................................................... 166 2 Micro-mechanism and Influencing Factors of R&D Cooperation between Enterprises........................................................................................................... 167 3 Hypothesis and Variable Design ......................................................................... 168 4 Case Study........................................................................................................... 170 5 Conclusion........................................................................................................... 172 References .................................................................................................................. 173 Authors ....................................................................................................................... 175 Uwe Cantner and Holger Graf: Evolving innovation networks – some empirical results .. 176 1 Introduction ......................................................................................................... 176 2 Theoretical building blocs ................................................................................... 176 3 The case and the data .......................................................................................... 178 4 Results ................................................................................................................. 179 5 Conclusion........................................................................................................... 188 Authors ....................................................................................................................... 189 Ricarda Bouncken, Thorsten Teichert and Michael Koch: Modularity and synergistic specificity in inter-firm collaboration......................................................................... 190 1 Introduction ......................................................................................................... 190 2 Theory ................................................................................................................. 190 3 Empirical study ................................................................................................... 193 4 Results and conclusion ........................................................................................ 194 References .................................................................................................................. 195 Authors ....................................................................................................................... 197 4

Tools for R&D Management

199

Michael Dowling and Roland Helm: Licensing strategies in the biotech industry ............. 199 1 Introduction ......................................................................................................... 199 2 Theoretical background....................................................................................... 200 3 Biotechnology in Germany, Jena and Regensburg.............................................. 201 4 Conclusions ......................................................................................................... 207 Literature .................................................................................................................... 208 Authors ....................................................................................................................... 209 Liang Liang and Li Yong: A lexicographic maxmin approach for allocating the fixed cost based upon data envelopment analysis ............................................................... 210 1 Introduction ......................................................................................................... 210 2 Efficiency evaluation........................................................................................... 211

Contents

XI

3 Maxmin-allocation model....................................................................................213 4 Allocation comparison.........................................................................................219 5 Allocate the advertising expenses among 17 dealers...........................................221 6 Conclusion...........................................................................................................222 References...................................................................................................................222 Authors .......................................................................................................................223 Acknowledgement ......................................................................................................223 Dominik Schultheiß: Explanatory potential of John Dunning’s eclectic paradigm for the internationalization of research and development................................................224 1 Introduction .........................................................................................................224 2 Roles of international R&D-units........................................................................225 3 Application of the eclectic theory on the internationalization of R&D ...............226 4 Conclusion...........................................................................................................239 References...................................................................................................................240 Author .........................................................................................................................243 5

Intercultural aspects of technical innovation

245

Meng Fanchen: Corporate culture as success factor to new product development ............245 Abstract.......................................................................................................................245 1 Introduction .........................................................................................................245 2 Development of new products and the success factors........................................246 3 Corporate culture and its function .......................................................................248 4 The relationship between corporate culture and new product development........250 5 Conclusion and suggestion ..................................................................................255 References...................................................................................................................257 Author .........................................................................................................................258 Reinhard Meckl: Acquiring and creating knowledge through international Mergers & Acquisitions – the case of German-Chinese M&A......................................................259 1 Relevance of international Mergers & Acquisitions (M&A) for Chinese and German companies ..............................................................................................259 2 Basics and Data of German-Chinese M&A.........................................................260 3 Research question................................................................................................265 4 Conceptual background .......................................................................................266 5 Definition of research design and hypotheses .....................................................268 References: .................................................................................................................270 Author .........................................................................................................................271 Ricarda Bouncken and Viviane Winkler: Empirical results and implications for leadership of cross cultural innovation-teams............................................................272 1 Introduction .........................................................................................................272 2 Theory .................................................................................................................273 3 Qualitative study..................................................................................................275 4 Discussion and implications ................................................................................278

XII

Contents References .................................................................................................................. 281 Authors ....................................................................................................................... 283

1

Policy and innovation

Mu Rongping: The changing strategy and policy of innovation in China 1

Introduction

China has made great achievements in system reforms and economic development since it implemented the policy of reform and opening up in 1978. During the past two decades, China has taken lots of measures such as establishing the Special Economic Zones (SEZ), the Economic Technology Development Zones, the Hi-tech Industry Development Parks, and the reforms concerning the pricing system, public finance, taxation, banking and trade, which have a great impact on reforming the system of a socialist market economy. The successful policy of reform and opening up has resulted in over 20 years of high-speed economic growth in China and made China the fourth largest economy in the world. However, the economic growth of China mainly relies on increasing investment instead of innovation although China has made great progress in building up an indigenous capacity for S&T and innovation. Lots of indicators show that there are still large gaps in many aspects between China and developed countries, such as (1) the lower effectiveness of investment in China; (2) the larger gap in industrial technology capability; (3) the lower rate of valueadding in hi-tech sectors; (4) the larger gap in capacity for S&T & innovation; (5) the lower effectiveness in coupling R&D in universities and research institutes with technology development in enterprises. Besides, overpopulation and environmental as well as resourcerestrained issues are also crucial to the sustainable development of China. Therefore, the Chinese government has changed its innovation strategy and related policies, in which the building up of innovation capacity has become one of the national strategies, enterprises becoming the center of the national innovation system. This paper consists of five parts with the following goals, namely: (1) to introduce the major challenges that China is facing during the transition from a catching-up country to an innovation-driven country. (2) to review the changing innovation capacity in China since 1995. (3) to identify the key factors for innovation capacity building in China, such as the allocation of

2

1 Policy and innovation

innovation resources, R&D and innovation investment, the innovation capacity in enterprises, the linkage among S&T, Economy & Education, the IPR strategy for development and patent as well as standardization and the flow of talents for innovation. (4) to introduce the changing innovation strategy and policy in China, especially the policy measures that support the innovation capacity building in China. (5) to provide some recommendations for setting up detailed rules to ease the implementation of supportive policies.

2

Review of the changing innovation capacity in China since 1995

China has made a great achievement in building up an indigenous innovation capacity during the past 20 years. However, the data shows that there is still a very big gap in innovation capacity between China and developed countries. 2.1 Productivity of scientific and technological papers in China The number of papers published by Chinese in domestic and foreign periodicals and taken by SCI/EI/ISTP (called SCI paper, EI paper, ISTP paper respectively) has increased rapidly since 1995. The number of SCI/EI/ISTP papers published by Chinese accounts for 6.9% of the total papers in the world, only less than the U.S, Japan, and the UK. The number of SCI papers published by Chinese accounts for 5.3% of the total papers, ranking fifth in the world.

180000 160000 140000 120000 100000 80000 60000 40000 20000 0 1995

1996 SCI

1997

1998 EI

1999

2000

2001 I STP

Figure 1: S&T Papers Taken by Major Foreign Referencing System Sources: 2006 China Statistical Yearbook on Science and Technology

2002

2003

2004

2005

Tot al Number

Mu: The changing strategy and policy of innovation in China

3

The number of SCI/EI/ISTP papers published by Chinese increased from 26395 pieces in 1995 to 153374 pieces in 2005. During 1995-2005, as show in Figure 1, the number of SCI papers published by Chinese increased from 13134 pieces to 68226 pieces, with an annual growth rate of 17.91%, the position in the world ranking list went up from the 15th to the 5th; The number of ISTP papers published by Chinese increased from 5152 pieces to 30786 pieces, with an annual growth rate of 19.57%, the position in the world ranking list went up from the 10th to the 5th; The number of EI papers published by Chinese increased from 8109 pieces to 54362, with an annual growth rate of 20.96%, the position in the world ranking list went up from the 7th to the 2nd. However, productivity of SCI papers (the number of SCI papers per thousand R&D personnel) is much lower than that in developed countries, see figure 2.

400 350 300 250 200 150 100

USA

United kingdom

Switzerland

Sweden

Singapore

R.Korea

Netherlands

Japan

Italy

Germany

France

Finland

Denmark

China

Canada

Belgium

0

Austria

50

Figure 2: The Number of SCI papers per Thousand R&D Personnel

2.2 Productivity of Invention Patents in China The number of domestic applications for an invention patent received by SIPO increased from 4065 cases in 1985 to 93485 cases in 2005, while the number of domestic invention patents granted increased from 38 cases to 20705 cases in the same period. In 2005, the number of domestic applications for an invention patent received by SIPO accounts for 24.40% of the total domestic applications for three kinds of patents in China, the domestic invention patents granted by SIPO accounts for 12.06% of the total 3 kinds of domestic patents granted, as show in Figure 3.

4

1 Policy and innovation 450000 400000 350000 300000 250000 200000 150000 100000 50000 0 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

number number number number

of of of of

t he t he t he t he

domest i domest i domest i domest i

c c c c

appl i cat i ons f or pat ent r ecei ved appl i cat i ons f or i nvent i on pat ent r ecei ved pat ent s gr ant ed i nvent i on pat ent s gr ant ed

Figure 3: Domestic Applications and Grants for Patent Sources: 2006 China statistical yearbook on science and technology

However, the productivity in terms of invention patents (the number of invention patents per thousand R&D personnel) is much lower than that in developed countries, see figure4.

700 600 500 400 300 200

USA

United kingdom

Switzerland

Sweden

Singapore

R.Korea

Netherlands

Japan

Italy

Germany

France

Finland

Denmark

China

Canada

Belgium

0

Austria

100

Figure 4: the Number of Invention Patents per Thousand R&D Personnel

The productivity in terms of PCT invention patents (the number of PCT invention patents per thousand R&D personnel) is much lower than that in developed countries, see figure5.

Mu: The changing strategy and policy of innovation in China

5

120 100 80 60 40

USA

United kingdom

Switzerland

Sweden

Singapore

R.Korea

Netherlands

Japan

Italy

Germany

France

Finland

Denmark

China

Canada

Belgium

0

Austria

20

Figure 5: the Number of PCT Invention Patents per Thousand R&D Personnel

2.3 Capability for R&D Expenditure in China The intramural expenditure for R&D increased from 34.869 billion RMB Yuan in 1995 to 244.997 billion RMB Yuan in 2005, meanwhile, the proportion of the intramural expenditure for R&D to GDP increased from 0.6% to 1.34%, as shown in Figure 6. However, the proportion is still lower than that of most developed countries, as shown in figure 6.

300 250 200 150 100 50 0 1995

1996

1997

1998

1999

2000

2001

2002

experimental development applied research basic research total intratramural expenditure for R&D

Figure 6: The Intramural R&D Expenditure in China (billion RMB Yuan)

2003

2004

2005

6

1 Policy and innovation

The intramural R&D expenditure in China mainly focuses on experimental development instead of basic research. The expenditure for basic research accounts for 5.36% of the total intramural R&D expenditure in 2005, while the expenditure for experimental development accounts for 76.95% of the total, as shown in figure 6. The expenditure for basic research usually accounts for about 10-15% of the total intramural expenditure for R&D in many developed countries. The R&D expenditure in enterprises accounts for 68.31% of the total R&D intramural expenditure in 2005, which implies that enterprises have become major players in R&D activity, as shown in figure 7.

300 250 200 150 100 50 0 2000

2001

2002

2003

2004

2005

enterprises institutions of higher education independent research institutions total intratramural expenditure for R&D

Figure 7: The Intramural Expenditure for R&D by Performers

3

Key Factors for Building Innovation Capacity in China

The innovation capacity of China, in terms of the productivity of scientific papers and patents as well as the capacity of R&D expenditure, has been improved during the past ten years. However, there is still big gap in innovation capacity between China and developed countries. Therefore, it is necessary to analyze the key factors influencing the innovation capacity-building so as to adjust the national strategy and policy of innovation.

Mu: The changing strategy and policy of innovation in China

7

3.1 Investment in Innovation Capacity-building of Enterprises China has become one of the largest R&D countries in terms of the total R&D expenditure, but the productivity of HRST is still much lower than that of many developed countries, either in terms of patent application/grant per R&D personnel or international scientific papers per R&D personnel, even per R&D expenditure. Besides, the R&D expenditure in enterprises is still very limited. In 2005, R&D expenditure accounts for only 0.76% of the total sales in large and medium-sized industrial enterprises. The motivation for enterprises to invest in innovation is still not strong because the opportunity costs of innovation are relatively high in China. In practice, there are lots of other opportunities for enterprises to make profit without innovation. Therefore, it is necessary to provide more opportunities for innovators to profit by establishing an innovation-friendly system and mechanism. 3.2

Coupling R&D in University & GRIs with Technology Development in Firms China has become one of the largest countries in terms of the number of enterprises and universities as well as public research institutes. China has also become one of the largest countries in the world in terms of the quantity of published international scientific papers such as SCI/EI/ISTP papers. However, the linkage among enterprises and universities as well as public research institutes is still relatively weak. The R&D activities in universities and public research institutes have made less contribution to the technology development activities in most Chinese enterprises because of a big gap concerning technology capability among universities/GRIs and enterprises. Therefore, it is necessary to establish an effective mechanism to promote the collaboration of enterprises and universities as well as public research institutes, and to restrict universities and public research institutes to commercialize their research results and the running of business solely. 3.3 Mobility of Human Resource for Science, Technology and Innovation China has become one of the largest countries in terms of the reserves of HRST, but the intensity of HRST in China (the number of HRST per capita) is still much lower than that of most other developed countries. The structure of HRST in China is compatible neither with the global science and technology development trend nor with China’s strategy that calls for an all-round, balanced and sustainable development. There are still some barriers for mobility of HRST either from universities and public research institutes to enterprises, or from developed eastern China to western China. For example, the R&D expenditure in western China is much lower than that in eastern China which means that there is less opportunity for HRST in western China than in eastern China. 3.4 IPR Strategy and Innovation Management in Enterprises China has become one of the largest countries in terms of the manufacture scale, but a relatively weak country in terms of the intensity of IPRs (the ratio of the number of patents to GDP).

8

1 Policy and innovation

On the one hand, there is no clear national IPR strategy in general and industry IPR strategy in particular. Usually, there are no considerations on the acquisition of specific IPRs from R&D in universities and government research institutes (GRIs). Scientists usually pay more attention to scientific and technical indicators instead of the possibility of acquiring IPRs in R&D. Some universities and GRIs have set up specific departments for IPR management, few of which play an important role in making an R&D plan in consideration of IPRs issues. On the other hand, there is no clear IPR strategy in most enterprises. Usually, the purpose of technology development in enterprises is to meet the present market demand instead of developing a future market. Therefore, most Chinese enterprises try to develop or integrate technologies with respect to the development of products that are saleable in the present market, and pay most of their attention to what kinds of products are popular in the market and how to imitate or modify these products. 3.5 NIS Development and Innovation Capacity of Enterprises China has become the largest country in terms of manufacturing scale in the world, and one of the largest countries in terms of the quantity of published international scientific papers. However, the national innovation system in general is still less effective than that of developed countries because most Chinese enterprises are still not able to get enough technological supply and benefit from the R&D in universities and government research institutes. The number of independent organizations for technology development has decreased gradually since 1999 although innovation has received increasing attention from the government, industry and academia in recent years. So far, only one fourth of large and medium-sized industrial enterprises have maintained independent organizations for technology development. Most enterprises have never applied for invention patents. The innovation capacity of enterprises has become the key to the national innovation system in China. Besides, there is a big gap in terms of the orientation of universities and institutes as well as enterprises. Universities and government research institutes usually pay more attention to international scientific papers instead of market potentials, while enterprises emphasize the profits of their activity no matter if it is coming from technology innovation or not. Therefore, the linkage and the interaction among enterprises, universities and government research institutes becomes the key to fill the gap so as to improve the effectiveness of NIS and the innovation capacity in enterprises. 3.6 Educational Structure and Innovation Capacity of Enterprises China has become one of the largest countries in terms of the quantity of entrants/enrollments of undergraduates and postgraduates. However, the educational structure is still unreasonable if the gap between talents supply and demand is being taken into consideration. On the one hand, the number of entrants and the enrollments of undergraduates and postgraduates has increased rapidly since 1998. However, there are lots of graduates who are not able to find the right jobs because of relatively old subjects that are not suitable for market demands.

Mu: The changing strategy and policy of innovation in China

9

On the other hand, the capacity-building for vacation education and professional training has to some extent been slowed down since many professional schools and colleges have been transformed into normal universities which has resulted in a shortage of high-skilled technicians and technical workers.

4

Changing Innovation Strategy and Policy in China

The Chinese government initiated a large scale of strategic studies for making the medium and long term plan for the national science and technology development beginning in August 2003, and issued “the outline of medium and long term plan for national science and technology development (2006-2020)” in January 20061, with the very ambitious goal to become an innovation-driven country by 2020. Meanwhile, building up innovation capacity has become one of the national strategies2, and results in an important change in the national innovation strategy and related policies. The Chinese government promulgated “some supportive policies for implementing ‘the outline of medium and long term plan for national science and technology development 20062020’” in February 20063, and is planning to issue 99 detailed rules for implementing the above supportive policies. The major goals of the supportive policies are to establish an enterprise-centered national innovation system, especially to strengthen the innovation capacity-building of enterprises. The supportive policies mainly emphasize 6 aspects that reflect the changing innovation strategy and policy in China. 4.1 Increasing the Investment in Science, Technology and Innovation In order to increase the investment in science, technology and innovation steadily, the Chinese government has taken many measures concerning the government’s financial policy. Firstly, the supportive polices require to establish a diversified and multiple system for science and technological investment and to make governmental investment in science and technology maintain a growth rate faster than that of governmental regular revenues so as to secure the implementation of national significant special projects, national science and technology programs. Secondly, it is required to optimize the structure of government expenditure in sciences and technology, namely focusing on basic research, social Commonweal research and frontier technology research. Meanwhile, it is required to emphasize the construction of capacity for scientific research so as to solve significant scientific and technology issues in national, industrial and regional developments. It is required to encourage and support the indigenous 1

The State Council: “the outline of medium and long term plan for national science and technology development (2006-2020) ”, Guofa (the State Council issue) [2005] No. 44

2

CCCPC, the State Council: “the decision on implementing the outline of science and technology plan with a view to strengthening the indigenous innovation capacity”, 26 January 2007

3

The State Council: the message about distributing “some supportive policies for implementing ‘the outline of medium and long term plan for national science and technology development 2006-2020’”, Guofa (the State Council issue) [2006] No.6

10

1 Policy and innovation

innovation and innovation on the basis of the assimilation of imported technology in enterprises by providing necessary preferential policies such as a government fund for R&D and engineering research in enterprises. Thirdly, it is required to establish new mechanisms for managing governmental expenditure in science and technology so as to improve its effectiveness and efficacy, especially the expenditure in scientific research bases, talents, and national programs of science and technology. Besides, it is also required to establish an evaluation system of performance of governmental expenditure in science and technology. 4.2 Tax Incentives for Investment in Science, Technology and Innovation In order to encourage enterprises to increase their investment in technology development and innovation capacity-building, the Chinese government has taken many measures of tax incentives in the supportive policies. Firstly, the policy promises to provide more preferential policies for technology development in enterprises, namely: the government shares about 50% of the enterprises’ expenditure for technology development by means of tax deduction. Secondly, the policy allows enterprises to speed up the depreciation of imported facilities and instruments for scientific research and development so as to upgrade enterprises’ experimental facilities & instruments. Thirdly, the policy provides tax incentives for TDCE (technology development center of enterprise authorized by the national development and reform commission) and NERC (national engineering research center)/NETRC (national engineering technology research center), for national special projects for science and technology, national programs for science and technology as well as national research and development projects for significant technology equipments so as to promote the capacity-building for innovation of enterprises. The imported goods for scientific research and technology development and the national programs are allowed to waive tariff and related value-added tax. Fourthly, the policy promises to waive the income tax of transformed research institutes so as to strengthen the capacity-building for innovation. It is required to provide tax incentives for venture capitals and service organizations for science and technology such as university science parks and incubators of science-based enterprises so as to promote the development of SMEs. Besides, the policy also encourages social donation for innovation by means of tax deduction policies. 4.3 Government Purchase Policy for Promoting Enterprise Innovation In order to encourage enterprises to increase their investment in technology development and capacity-building for innovation, the Chinese government has taken many measures of government purchases in the supportive policies. Firstly, it is required to establish a system of purchasing indigenous innovation products with governmental funds, including to set up a system for authorizing the indigenous innovation products and to issue the list of indigenous innovation products. It is required to give high priority to the indigenous innovation products regarding the purchasing activity with governmental funds. It is also required to give high

Mu: The changing strategy and policy of innovation in China

11

priority to the indigenous innovation products in national significant construction projects and other projects that use governmental fund to purchase important equipments and products. Secondly, it is required to establish a system for the government to purchase indigenous innovation products first and to subscribe to indigenous innovation products. According to the policy, the Chinese government is ready to take the measure of first-purchase of the products that have relatively high market potentials, in accordance with the requirements of the national economic development and the trend of advanced technology development, which are made by domestic enterprises or research organizations. The Chinese government is ready to subscribe to significant innovation products and technologies to be developed and to diffuse them by establishing a system of inviting public bidding and a mechanism to diffuse the subscribed innovation products and technologies in industries. Thirdly, it is required to establish a system for authorizing domestic goods and an auditing system for purchasing foreign products. It is required to give high priority to the foreign products that are ready to transfer technology to China. Besides, defense purchases are also required to give high priority to domestic innovation products and technologies that are able to meet the demands of the national defense and national security. 4.4 Re-innovation Based on the Assimilation of Foreign Advanced Technology In order to strengthen the innovation capacity of enterprises by means of import and assimilation of foreign advanced technologies, the Chinese government has taken many measures in the supportive policies. Firstly, it is required to strengthen the management of technology import and assimilation. National important projects are required to make a plan for the assimilation of foreign advanced technology and re-innovation, with a clearly defined plan, goals, and schedule. Besides, the indigenous innovation capacity is an important indicator for assessing the import projects. Secondly, it is required to adjust the list of technologies that are encouraged to import. Encouragement is given to import advanced design and manufacture technology as foreign equipment and facilities are imported. Meanwhile, it is also required to adjust the list of technologies that are restricted to be imported. Thirdly, it is required to support re-innovation based on an assimilation of imported advanced technologies by giving high priority to national key projects that subscribe to a domestically made first set of significant equipment. Besides, the policy promises to implement a national equipment technology policy to promote the indigenous capacity for equipment manufacture and innovation. Fourthly, the policy promises to support the cooperation among industries, universities and research institutes in assimilating imported advanced technologies and re-innovation activities with a view to strengthen the capacity for re-innovation. The government is ready to give high priority to technology platforms jointly established by industries, universities and research institutes in order to undertake important projects of technology assimilation and reinnovation in the national program for science and technology infrastructures.

12

1 Policy and innovation

4.5 Building up the Capacity for the Generation and Protection of IPRs In order to support indigenous innovation, the Chinese government has taken many measures to build up capacity for the generation and protection of IPRs. Firstly, it is required to set up a list of key technologies and important products that China should hold, of indigenous IPRs concerned, in order to support the development of the listed technologies in national science and technology programs and capital investments, and to support related enterprises in the application of patents which will set up standards by participating in international trade and cooperation. Meanwhile, it is required to establish the platform of an IPR information service with a view to providing IPRs information services for generating indigenous IPR and their market exploration. Secondly, it is required to be an active take part in manipulating international standards, and to promote the development of indigenous technology standards. National science and technology programs support formulating important technology standards, encourage enterprises to develop technology standards together with universities and research institutes, and promote the standards to be integrated with scientific research and development, design, and manufacture. Meanwhile, the government is ready to establish a standard service platform to support enterprises in the participation in making indigenous standards. Thirdly, it is required to establish and improve the IPR protection system, to strengthen the effectiveness of the executing law, and build the friendly environment for protecting IPRs. The Chinese government promises to support the acquisition of indigenous IPRs abroad through national science and technology programs and various innovation foundations, and to reward IPR holders of and major contributors to IPRs. Besides, it is required to establish a special system for examining IPRs in significant economic activities, mainly focusing on mergers and acquisitions of enterprises with important indigenous IPRs and technology exports so as to avoid any loss of indigenous IPRs and misuse of IPRs. Fourthly, it is required to accelerate the examination speed of invention patents, and to establish the system for technical trade measures that accord with international regulations, the informing and coordinating mechanism and the fast response mechanism as well as the review mechanism for technical trade measures, so as to provide related precaution about possible emerging technical trade measures. 4.6

Building up the National Infrastructure and Platform for Science and Technology In order to strengthen the national innovation capacity, the Chinese government has taken many measures for building up national infrastructures and platforms of science and technology. Firstly, it is required to strengthen the construction of an experimental base, infrastructure and platforms, focusing on a batch of platforms consisting of the shared platform of the scientific basic facilities and large equipments, the shared platform of natural scientific resources, the shared platform of science data, the shared platform of science and technology documents, the platform of public service for the commercialization of research results, the platform of networks for the scientific and technological environment, which are established in the national labs, national labs for key research fields, national engineering labs, national

Mu: The changing strategy and policy of innovation in China

13

defense labs for key research fields, national engineering research centers, and enterprises’ technology development centers. Secondly, it is required to dramatically increase the support of state-owned public research institutes that have completed the classified reform since 2006. Thirdly, it is required to support enterprises, especially transformed scientific research institutes and large enterprises in establishing technology development centers and a batch of national engineering labs (NEL) in key fields by cooperating with universities and research institutes. The NELs mainly focus on developing pre-competitive technology, frontier technology, and civil and military dual-use technology. Besides, it is also required to establish an indicator system for evaluating the performance of transformed scientific research institutes. Fourthly, it is required to promote the openness and sharing of the platform and the base for science, technology and innovation by establishing an open and sharing institution and mechanism, and to evaluate the performance and effectiveness of these platforms and bases according to their openness and degree of sharing. 4.7 Cultivating and Using Talents for Science, Technology and Innovation In order to cultivate and make full use of talents for science, technology and innovation, the Chinese government has taken many measures. Firstly, it is required to bring up the world’s leading scientists and experts with an international view and systematic training in strategic research fields by implementing national projects for cultivating high level innovation talents. Secondly, it is required to train innovative talents in key projects according to the national development goals, namely to bring up strategic scientists and leading scientists as well as innovation teams in key basic research projects, and to train innovation talents and advanced engineering technologists in projects on important applied research, industrialization and engineering. Meanwhile, it is required to train talents on the national platform for science and technology and in national scientific research bases such as national labs and the national engineering research center. Thirdly, it is required to support enterprises in bringing up and attracting innovation talents by reforming income distribution and incentive mechanisms and providing some preferential policies in order to support the cultivation of practical talents for science and technology. Fourthly, it is required to set up and implement the plan for attracting excellent oversea students and talents abroad back to and in order to serve for China by providing some preferential policies such as the government’s special subsidy, taxation deduction, IPRs incentives, the priority given to the employment of their mate and the enrollment of their children, special procedures concerning visa application, resident permission, medical treatment, and social insurance. Fifthly, it is required to establish a rational system for assessing and rewarding innovation talents to establish a system for science and technology credit so as to establish an innovation-friendly culture. Meanwhile, it is required to reform the personnel system of scientific

14

1 Policy and innovation

research institutes funded by the government so as to make scientists be more active and innovative. 4.8 Supporting Indigenous Innovation with Financial Measures In order to support indigenous innovation, the Chinese government has taken many financial measures. Firstly, policy-oriented financing organizations are required to give high priority to financing national special important projects for science and technology, national projects for industrialization of science and technology, and the projects concerning the industrialization of high technology, the assimilation of imported technology, and the export of high-tech products. Meanwhile, it is required to encourage commercial banks to support indigenous innovation. Secondly, it is required to improve the financial service for SMEs’ innovation and to establish an entrepreneurial credit system so as to ease the financial services for SMEs, and to support the development of credit assurance organizations for SMEs. Thirdly, it is required to accelerate the development of venture capital, to improve the legal system of venture investment for startups so as to introduce social capital flowing to the enterprises for venture investment, and to introduce an investment in startups and the enterprises at an earlier stage. Fourthly, it is required to establish multiple capital markets to support indigenous innovation, including the stock market for SMEs, the stock market for science-based SMEs, the stock transaction of high-tech enterprises, and the regional transaction market for property rights. Fifthly, it is required to improve the policy for managing the foreign exchange of high-tech enterprises so as to support enterprises in establishing oversea organizations for R&D and design, and to support their acquisition and merger of oversea R&D organizations or hightech enterprises.

5

Conclusions

To become an innovation-driven country is a very ambitious goal for China. There are lots barriers that still restrain the development of the national innovation system in China, especially the development of innovation capacity of enterprises. The Chinese government has issued the supportive policies for implementing “the outline of medium and long term plan for national science and technology development (2006-2020)”, which consists of many measures to stimulate enterprises to increase their investment in innovation, to attract oversea talents back to and in order to serve for China. However, there are still many factors that, to a great extent, determine the efficacy and effectiveness of the supportive policies. Therefore, the Chinese government began to set up 99 detailed rules for implementing the supportive policies. The supportive policies consist of many measures which cover almost everything concerning science, technology and innovation, but which to some extent are too general to be implemented. Therefore, it is necessary to fully understand the basic guiding principles for science

Mu: The changing strategy and policy of innovation in China

15

and technology, namely: “indigenous innovation, leapfrogging in key areas, supporting development, and shaping future”, which reminds policymakers of paying more attention to the following five factors in making detailed rules. Firstly, it is necessary to pay more attention to the innovation motivation of enterprises. The innovation motivation of enterprises is very different, and determined by many factors. Therefore, in order to make enterprises invest in innovation, not only incentive policies such as governmental investment and taxation deduction should be provided; possible profitable opportunities without innovation efforts should also be restrained. Therefore, it is also very important to make detailed rules with respect to eliminating possible opportunities of making profit without innovation, or decreasing the profit rate of opportunities without innovative aspects. Secondly, it is necessary to pay more attention to setting up innovation capacity in enterprises. The nature of enterprises is to profit from any activity instead of building up their innovation capacity. The capacity-building for innovation is only a tool for enterprises to make profits. However, it also is a national interest. Therefore, it is very important to make detailed rules with respect to sharing the risks of entrepreneurial innovation instead of taking all risks. Thirdly, it is necessary to pay more attention to a systematic breakthrough instead of an arbitrary breakthrough of some technologies. The systematic breakthrough may result in a leapfrogging development of leading industries or the development of an emerging industry. Therefore, it is very important to set up detailed rules that focus on specific technological areas and specific industries and support systematic innovations and the exploration of emerging technologies. Fourthly, it is necessary to pay more attention to innovation talents and to shaping the future instead of meeting present market demands. Innovation talents are the key to an innovationdriven country. Shaping future is the process of selecting a favorable future. Therefore, it is very important to integrate the process of technological foresight with the process of making detailed rules, and to give high priority to the cultivation and mobility of innovation talents. Fifthly, it is necessary to pay more attention to the coordination among various governmental departments. Innovation resources are divided into different governmental departments in China. However, the effectiveness and efficacy of innovation are to a great extent determined by the allocation of innovation resources. The effectiveness of the coordination of governmental departments determines the effectiveness of innovation. Therefore, it is very important to coordinate all related governmental departments when setting up the detailed rules.

References 1.

National Bureau of Statistics, Ministry of Science and Technology: China Statistical Yearbook on Science and Technology 2006

2.

The State Council: “the outline of medium and long term plan for national science and technology development (2006-2020)”, Guofa (the State Council issue) [2005] No. 44

16

1 Policy and innovation

3.

CCCPC, the State Council: “the decision on implementing the outline of science and technology plan with a view to strengthening the indigenous innovation capacity”, 26 January 2007.

4.

The State Council: the message about distributing “some supportive policies for implementing ‘the outline of medium and long term plan for national science and technology development 2006-2020’”, Guofa (the State Council issue) [2006] No.6

Author Mu Rongping received his B.Sc. and M. Sc. degree from the University of Science and Technology of China, and his Ph.D. degree from Technische Universität Berlin, Germany. He had been working as teacher in Hefei Poly-technical University for four years, and has been working at the Institute of Policy and Management (IPM), Chinese Academy of Sciences (CAS) since 1990. Dr. Mu is now the director-general and professor at IPM, the director of CAS Center for Evaluation Research, Editor-in-Chief of the Journal of Science Research Management (an academic bimonthly). Besides, he is also Vice President of the Chinese Association for Science of Science and S&T Policy Research. Dr. Mu has published more than 30 papers in peer-reviewed journals and international conferences. He has led more than 20 research projects entrusted or financed by the National Development and Reform Commission, the Ministry of Science and Technology, National Natural Science Foundation of China and CAS. His current research interest is in S&T strategy and policy, S&T management and evaluation, technology foresight, and evaluation on international competitiveness of high-tech industry. Contact: [email protected], Phone: +86 (10) 62542618, Fax : +86 (10) 62542619 Institute of Policy & Management Chinese Academy of Sciences (CAS) Beijing Zhongguancun donglu No. 55 P.O. Box 8712, Beijing 100080, P.R. China

Liu Xielin and Nannan Lundin: Toward a market-based open system of innovation in China 1

Introduction

The national innovation system has become a useful tool to analyze a country’s innovative performance and to provide an input to policy development, although there are different variations of the concept and its implications (Freeman, 1987; Nelson et al. 1993; Lundvall, 1992). For a long time, the innovation system in China was a more plan-based system, in which, governmental research institutes played a dominant role, the government acted as a core coordinator from the idea creation to the final consumption of the new product by issuing a national science and technology plan annually and every five years. Companies here played a trivial role (Liu and White, 2001). Though the system was efficient for some targeted programs, such as the development of the nuclear bomb and missile system, it was an inefficient system in terms of innovation. Since the 1980s, China began its reform of the market and open economy. Companies gradually became the main players of innovation, lots of state-owned companies were transformed into an equity share or private companies, foreign affiliated companies can be seen everywhere. The market replaced the governmental plan as the core coordinator of the innovation system. So, in a sense, the system has been heading for a market-based open system of innovation. In this paper, we try to present the evolutionary process of the innovation system development in China. Some important items should be highlighted in this context: Firstly, country specific issues still matter in a world of globalization, as “most public policies influencing the innovation process or the economy as a whole are still designed and implemented at the national level” (Edquist, 2006). As China’s political and economic systems are unique, their innovation policy making needs to be grounded in this uniqueness. Secondly, the notion of linkages between academy and industry is the core of the NIS, irrespective of the country in question (Mowery and Sampat, 2006). Most policy tools derived from the NIS aime at promoting the linkages among the industry and academic institutions. Thirdly, enterprises are the most important actors in the system. Companies with different governance structures will have different capabilities to learn and to diffuse new knowledge (Lazonik, 2006). Finally, the system works in a changing environment, including political, international and regional factors.

18

1 Policy and innovation

This paper will be structured as follows. Section one is an introduction. In section two, there will be a short review of how the Chinese innovation system has evolved over the last twenty years. In section three, key linkages in the system will be discussed. In section four, the company system will be presented in more detail. In section five, we will discuss the key factors of the innovation system of China, such as the role of the government, foreign direct investment and regional diversity.

2

From the dominance of government research institutes to an enterprise centered system

2.1 Government research institutes and their role in Chinese innovation In the socialist planned regime, the innovation system in China was very unique in the world. It was dominated by a linear model of innovation with a clear-cut of division of labor. The government functioned as the key coordinator in the system. From the 1950s to 1980s, China had established different layers of governmental research institutions (GRIs) with various goals. The most important of them were at the national level, such as the Chinese Academy of Science, focusing mainly on basic research. There were hundreds of industrial research institutes under a wide range of industrial ministries, focusing on applied and developmental tasks. Regional GRIs would do research and development work defined as relevant for the regional level. At that time, enterprises were limited in terms of function to mainly being manufacturing plants. Most of them did not do any R&D, while only some large state owned enterprises had their own R&D labs and their work focused mainly on experimental issues. Hence, the innovation system of that era was rather driven according to a linear and hierarchical model. Most of the scientific research was done by the Chinese Academy of Sciences and research universities. The outcome were central and regional applied institutions doing research between the basic research and experimental work in the companies. So, during this period, the GRIs were by far the most important actors in the system (Table 1). Even in 1987, GRIs did more than half of the total R&D in China. The university system was a complimentary to GRIs. Most of the universities at that time were not involved in research, except for some research universities such as Tsinghua University and Peking University. Many specialized universities focused on industry specific technology and education. For example, there were universities specialized in light industry, metallurgy, printing, etc.. Table 1: Enterprises replaced the role of research institute as major R&D actors in China 1987

1990

1996

1997

1999

2000

2001

2002

2003

2004

Research institute

54.4

50.1

41.1

42.9

38.5

28.8

27.7

27.3

27

23

University

15.9

12.1

13.0

12.1

9.3

8.6

9.8

10.1

11

10

Enterprises

29.7

27.4

36.8

42.9

49.6

60.3

60.4

61.2

62

67

2.1

2.6

2.3

2.1

1.4

Others

Source: China Science and Technology Statistical Yearbook, 2004. Beijing: Chinese Press of Statistics

Liu and Lundin: Toward a market-based open system of innovation in China

19

Following this very division of labor in knowledge production, a key issue is how to introduce the new technologies and products into the market. This was the task of the government. The main economic tool for the government to do that is a five-year and annual economic and S&T plan. The main institutions were governmental research institutions. Even at the government level, there is an elaborated division of labor. For example, the State Planning Committee (now State Development and Reform Commission) was central in allocating production targets for the enterprises and also had the power and obligation to introduce new technologies to the economic system. The Ministry of Science and Technology would make five year and annual plans in the area of science and technology. For a long time, S&T was seen strategically as the very mechanism to cover shortages of products and strengthen China’s military position. Priority was given to targeting a few large, national projects. This reinforced the impression that great success could be achieved, albeit with huge costs. The success of the nuclear bomb, artificial insulin and other major discoveries were the result of this planning regime. These projects would involve thousands of scientists and engineers in different research institutions, universities, factories and hospitals across the country within a well planned division of labor. But overall, the whole system was less than efficient. The enterprises were output-based, with few if any incentives for efficiency and profit, and paid no attention to IPR. The research institutions and universities were funded by the government and typically produced project reports with limited industrial use. Hence the performance of innovation at that time was poor, although reverse engineering had a great impact in some sectors. Many new industries started around the same year as Korea initiated its new growth path, such as the automobile industry, ICT industry, and steel industry, but lagged behind Korea decades later. “Import, lag behind, import again, lag behind once more” was the rule of that period. So, for a quite long time, from the 1950s through the 1960s to the 1970s, technology imports laid the basis for economic development. China had imported technologies on a grand scale from the former Soviet Union, Germany, Japan and other countries. Those technologies laid down the foundation for the Chinese chemical, automobile, steel, textile and other industries. For many industrial GRIs, from 1949 to the 1980s, their main tasks were the assimilation of imported technology. In order to replace the imported technology and to save foreign currency, incremental innovations based on imported technology were implemented according to the principles of a planned economy. 2.2 The transition to an enterprise centered system In China, there is little space for curiosity driven research. S&T was generally viewed as a practical economic activity. Even now, the proportion for basic research has been kept to a level of 5-6% of the total R&D expenditure. Since the market reform had been initiated after 1978, the S&T system was soon to be exposed to market based competition. The objective of the reform was twofold: to introduce market based competition for the funding system and to establish a new governance system of S&T institutions in order to commercialize inventions as soon as possible.

20

1 Policy and innovation

One key initial change was to reform the appropriation system for funding and make the governance of the R&D institutes more flexible. It meant that government would reduce the funding for the GRIs, and that funding of the GRIs would increasingly come from other governmental or private sources. This increased the pressure on the scientists and led to more short term research projects with more immediate economic value. In order to speed up the process from research to commercial products, government firstly encouraged GRIs and universities to set up their own spin-offs and scientists to leave their research position and engage in commercial activities. Secondly, a new institution called the technical market was introduced. This new specialized market gave suppliers and users of technology a new opportunity to engage in technology transfer transactions. Lastly, special economic zones were established across China to support the development of high tech companies. In May 1988, the State Council approved and established the Beijing Municipal New Technology Industry Development Experimental Zone, granting it with 18 favored policies. In August the same year, it implemented the Torch program. Up to the end of 1992, there had been 52 National High-New Technology Industry Development Zones established all over the country. In 1993, there were 9687 high-tech enterprises registered in the zones. In the 1990s, after more than ten years of reform, the government still acknowledged that there was a great gap between the research activities of the GRIs and the needs of the industry. In the same time period, the government system underwent a significant change as most of the industry specific ministries were abolished. The new structural challenge was how to deal with the industrial GRIs affiliated to those ministries? Towards the end of 1998, the State Council decided to transform the 242 GRIs on the national level into technology based enterprises or technology service agencies, or more generally to become part of the enterprise sector. This meant that the domination of GRIs in the Chinese system of innovation was changed and the enterprises became the core part of the innovation system. Since 2000, the enterprises have performed more than 60% of total R&D activity in China. But even now, GRIs and universities are still the key players in frontier science and technological research. They can attract more talented scientists than enterprises do.

3

Industry-science linkages

The intensity and efficiency of industry-science linkages is a good indicator for the innovation capability of a system level in a country. As there was a functional division of labor in knowledge creation and diffusion for a long time in China, a strong barrier existed between knowledge creation in GRIs, universities and the use of knowledge in the enterprises. But since the introduction of market reforms in China, with stronger pressures for competition, the industry-academy linkages have been improved greatly during the past 20 years. Firstly, universities and GRIs were allowed to set up their own spin-offs so that they could commercialize their technology directly. That way, universities and GRIs would be more integrated into the economic growth process of China. Secondly, spin-off companies also served as a way for GRIs and universities to compensate for budget cuts from the government.

Liu and Lundin: Toward a market-based open system of innovation in China

21

Though the size of the spin-off industry in China has been small compared to that of the Chinese industry (Table 2), it was valuable for the high-tech industry in China. Spin-off companies gave many scientists in the universities or GRIs good opportunities to access the market knowledge. But from the end of the last century on, as the government has been continuously accelerating its support of research and education, the universities stopped to think of developing spin-off companies as their primary function. The same applies to GRIs. Table 2: University’s spin-off

1999 2000 2001 2002 2003 2004

Number of spin-off 2137 2097 1993 2216 2447 2355

Revenue (billion RMB) 26.7 36.8 44.8 53.9 66.8 80.7

Profit (billion RMB) 2.2 3.5 3.1 2.5 2.8 4.1

Sources: Statistics of University’s industry in 2004 in China, Center for S&T for Development, Ministry of Education, 2005

The result of a policy to encourage spin-offs is the birth of domestic, highly capable high tech companies, such as Lenovo, from the Chinese Academy of Sciences, and Beida Founder from Peking University. Lenovo now one of the leading companies in the IT industry in China. Most of the biotech companies are the result of spin-offs. For example, Shenyang Sunshine Pharmaceutical Co. Ltd., Beijing Shuanglu Pharmaceutical Co. Ltd., and Anhui Anke Biotechnology Co. Ltd. were all founded by former researchers from research institutes (Liu and Lundin, 2006). In general, universities and GRIs are important key factors underpinning the domestic high-tech industry of China. At the same time, universities and GRIs began to do more contract research for the industry. This activity not only is beneficial for the industry, as most of the companies, especially SMEs, have limited R&D capabilities, so, outsourcing of research to universities is their strategic ingredient in their development. For example, from 2000 to 2004, the share of universities’ industrial budget had been increasing, it was about 38% of their total research funds in 2004 (Table 3). In 2004, about 26% of the industry’s total R&D expenditure went to the universities (Table 4). Table 3: Universities’ share of research grant from industry

Share(%)

1999 10.3 5.4 52.2

2000 16.7 5.5 33.3

2001 20.0 7.2 36.2

2002 24.8 9.0 36.2

2003 30.8 11.3 36.7

2004 39.2 14.9 38.0

Share(%)

4.9 47.8

9.7 58.4

11.0 54.9

13.7 55.4

16.5 53.6

21.1 53.8

Total S&T funds (in billion) From firms From government

Source: Statistics of Science and Technology in High Education, 2000-2005, Department of S&T, Ministry of Education

22

1 Policy and innovation

Table 4: R&D outsourcing to universities and R&D institutes from large and medium-sized industrial enterprises 2000

2001

2002

2003

2004

Total R&D expenditure

35.4

44.2

56.0

72.1

95.4

Funds for university

5.5

7.2

9.0

11.2

24.9

Share of total business’ R&D (%)

15.5

16.2

16.1

15.5

26.1

Funds for R&D institutes

3.8

2.5

3.6

4.7

5.0

Share of total business’ R&D

10.7

5.6

6.4

6.5

5.2

Total outsourcing for domestic univ.and R&D inst.(%)

26.2

21.8

22.5

22.0

31.3

Source: China Science and Technology Statistical Yearbook, 2005. Beijing: Chinese Press of Statistics

Joint publishing of scientific papers between universities and industry is an alternative indicator of industry-academic linkage. For IPR and other reasons, the industry typically is reluctant to publishing papers. But from Table 5, it is clear that universities increasingly have industry engineers as their co-authors for joint publishing. Table 5: 2000－2003 co-authored papers between industry and universities First -second author

2000

2001

2002

2003

Paper

Share

Paper

Share

Paper

Share

Paper

Share

Total

51079

100

53246

100

87688

100

100310

100

Enterprises-university

4499

8.81

1123

2.11

1381

1.57

1567

1.56

University-industry

867

1.7

5301

9.96

6448

7.35

7421

7.39

Source: China Science Paper and Citation Analysis, Chinese Institute of Information, 2005

4

Company system

For a long time, enterprises typically operated as factories with few if any R&D activities or formal R&D centers. Imported technology was the main mechanisms to maintain and upgrade production capability. Table 6 shows that before 1998, enterprises spent more money on technology import than on their own R&D. Since the 1980s, SOE have been given more autonomy to invest and innovate based on their own strategic decisions. Also, millions of SME were allowed to set up by entrepreneurs. This wave of privatization and competition provided enterprises with the motivation to invest in product development and innovation on top of exploiting cost advantages or diversification. Tables 6 and 7 show that large and medium-sized companies quickly increased their R&D inputs, though comparing to that of developed countries, it is still quite low.

Liu and Lundin: Toward a market-based open system of innovation in China

23

Table 6: Expenditure of in-house R&D and technology importation and assimilation, unit: 100 million RMB

1991

58.6

90.2

Expenditure on technology assimilation 4.1

1993

95.2

159.2

6.2

1:1.67:0.06

1995

141.7

360.9

13.1

1:2.55:0.09

1998

197.1

214.8

14.6

1:1.09:0.07

1999

249.9

207.5

18.1

1:0.83:0.07

2000

353.6

245.4

18.2

1:0.69:0.05

2001

442.3

285.9

19.6

1:0.65:0.04

2002

560.2

372.5

25.7

1:0.66:0.04

Expenditure on R&D

Expenditure on technology import

Ratio 1:1.54:0.06

Source: China Science and Technology Statistics Yearbook, 1991-2003, Beijing Table 7: Ratio of R&D/sales in large and medium sized companies Year

1991

1995

1996

1997

1998

1999

2000

2001

2002

R&D/sales

0.49

0.46

0.48

0.52

0.53

0.60

0.71

0.76

0.83

Source: China Science and Technology Statistics Yearbook, 1991-2003, Beijing Table 8: The granted three patents in China domestic vs. foreign owners 1991

1995

1998

1999

2000

2001

2002

Sum of three patents granted in China

24616

45064

67889

100156

105345

114251

132399

Invention patent

4122

3393

4733

7637

12683

16296

21473

From Domestic owners

1311

1530

1655

3097

6177

5395

5868

Share of domestic owners

31.8

45.1

35.0

40.6

48.7

33.1

27.3

From foreign owners

2811

1863

3078

4540

6506

10901

15605

68.2

54.9

65.0

59.4

51.3

66.9

72.7

Utility model patent

17327

30471

33902

56368

54743

54359

57484

From domestic owners

17200

30195

33717

56094

54407

54018

57092

Share of domestic owners

99.3

99.1

99.5

99.5

99.4

99.4

99.3

From foreign owner

127

276

185

274

336

341

392

Share of foreign owners

0.7

0.9

0.5

0.5

0.6

0.6

0.7

External design patent

3167

11200

29254

36151

37919

43596

53442

From domestic owners

2667

9523

26006

32910

34652

39865

49143

Share of domestic owners

84.2

85.0

88.9

91.0

91.4

91.4

92.0

From foreign owners

500

1677

3248

3241

3267

3731

4299

Share of foreign owners

15.8

15.0

11.1

9.0

8.6

8.6

8.0

Share of foreign owners

Source: China Science and Technology Statistics Yearbook, 2003, Beijing

24

1 Policy and innovation

In terms of output, the innovation capability of Chinese companies is poor. Their capability most often is centered on incremental innovation with little radical innovation. This is why Chinese companies have a relatively high patenting activity in utility and design, but a low one in invention patents (Table 8). Comparing international patents, such as US granted invention patents, the wide gap between Chinese companies and Korean and Japanese ones becomes evident. In 2004, Korea’s patents in the USA are about 11 times that of China (Table 9). Table 9: Chinese via Korea patent registration in USA

China

number Rank

Korea

Number

2000

2001

2002

2003

2004

119

195

289

297

404

26

24

21

22

20

3331

3546

3755

4198

4590

8

8

7

5

4

Rank

The growth of SMEs is a relatively new phenomenon. The market was opened for private SMEs to enter only after the 1980s. As most of them started their business by taking a market opportunity, their innovation capability overall is quite low. Their ratio of R&D to sales is lower than that of large companies. In 2003, their R&D intensity was about 0.49, in 2004, it increased to about 0.56 (Lundin et al, 2006a) (Table 10). Table 10: A simplified comparison b/w small and large S&T-based firms, % , (2004) Small S&T based enterprises

SOE JV-HTM JV-Foreign Foreign Private

R&D / Sales

Export of new products/sales

Tech import / sales

Patent / 100 persons

1.19 0.97 1.64 1.44 1.55

0.29 4.22 4.22 6.61 3.21

0.19 0.21 0.64 0.22 0.13

0.51 0.37 0.42 0.79 0.66

Large S&T based enterprise

R&D / Sales

Export of new products/sales

Tech import / sales

Patent / 100 persons

0.91 1.01 1.30 0.99 0.74

1.55 23.01 6.44 24.37 5.90

0.32 0.40 1.18 0.15 0.05

0.06 0.41 0.74 0.25 0.90

Source: Lundin et .al, 2006a

5

Key determining factors for the Chinese innovation system

5.1 Role of government The government has been a key factor in the Chinese innovation system, even though the market force is very powerful in the real economic world. For example, the governments at different levels still control the land, large investment projects, infrastructure construction,

Liu and Lundin: Toward a market-based open system of innovation in China

25

market access to automobile and other industries, even financial markets. In terms of innovation, the government can implement its impact on science, technology and innovation by using the following tools: national R&D, policy tools and a fifth plan or long range plan. 5.1.1 National R&D program China has developed a system of programs to support innovation activities. Table 11 gives a brief overview of the main programs controlled by the Ministry of Science and Technology (MOST). Table 11: Current national R&D Programs (in 0.1 billion RMB) 1996

1997

973 Basic Research 863National High Tech R&D program (from 1986)

1998

1999

2000

2001

2002

2003

2004

3

4

5

6

7

8

9

25

35

45

55 16.144

4.5

5.054

5.2

5.372

10.37

11.66

10.32

10.63

10.634

12.464

0.51

0.51

0.51

0.51

0.5

0.5

0.5

0．5

Spark Program (1988 for rural SME)

0.39

0.39

0.39

0.4

0.4

1

1

1

Key S&T Diffusion program

0.19

0.19

0.2

0.2

0.2

0.2

0.2

0.2

Key Technologies R&D program (from 1983) Torch Program (1988, for high technology)

Source: MOST, China Science and Technology Development Report, 2006.China S&T Literature Press

Besides that, there is the National Innovation Fund for S&T based SME (about 0.5 billion RMB a year), and the National Science of Foundation for mainly curiosity driven basic research. The national program is very important not just for reasons of funding. In China, universities and GRIs all put governmental projects as their top priority. Most of the talented scientists in China are the main researchers for the program. Besides that, other regional and industrial funds quite often follow those national projects. 5.1.2 Policy After promoting the innovation became the agenda for the government, the Chinese government has also adapted various important policy instruments. The aims of the policy are to encourage R&D activities and to promote the transfer and commercialization of R&D results. One of the important policies is to establish high-tech zones and incubators to promote high-tech industry and innovation in China. This policy started at the end of the 1980s, learning from the U.S. Silicon Valley model. Now there are 53 high-tech development zones at the national level. The first high-tech zone, Zhongguancun high-tech zone, was established in Beijing 1988. The core of the policies covers the following aspects: • There is a well-functioning infrastructure so that the high-tech zones serve as a platform for innovation activities and interactions. • High-tech firms can enjoy preferential treatment in terms of a broad range of tax incentives.

26

1 Policy and innovation

• A new governance model, which is characterized by “small government, but big service”, has been implemented in these zones to reduce transaction costs. • To form a “cluster” structure to take advantage of closer co-operation and integration of the firms. In the past two decades, these high-tech zones have expanded rapidly in terms of their size and scope of activities and therefore played an important role in promoting the development of the high-tech industry in China. Up to now, more than 90% of high-tech firms and incubators are located in these high-tech zones and most of them are spin-offs from universities and GRIs, new private firms and FDI firms. In 2004, the total value added of the whole high-tech zone is 550 billion RMB, about 8.8%of GDP, its export is about 82.4 billion dollar, about 12% of the national value (MOST, 2006). The first business incubator in China was established in 1987 in Wuhan. By 2005, more than 490 incubators had been created across the country with most of them located in Beijing, Shanghai (30) and Shenzhen(32). The information technology and biomedical industries are the two most favored areas, but the number of incubators specialized in the IT industry is much larger than that of the biomedical industry. Regarding the ownership of IPR, the following important steps have been taken to facilitate the commercialization of R&D results: • Inspired by the Bay-Dole model of the U.S., the first step taken by the Chinese government is to allow IPR resulting from government-funded R&D projects to be commercialized. • Secondly, the ownership of IPR resulting from government-funded R&D projects could be transferred to the university or GRI that conducted the projects, instead of being government-owned intangible assets. • Thirdly, since 1998 individual inventors involved in government-funded R&D projects are allowed to obtain a royalty of at most 35% of the license fee when the research results are transferred. 5.1.3 Long range plan: indigenous innovation In the recent past and foreseeable future, planning was and will be an ingredient in the Chinese economic world even if the market reform has been introduced on a large scale. The ’National Plan 2006-2020 for the Development of Science and Technology in the medium and long term’ is currently the long term policy framework. The possibly most interesting result of the new plan is the declared intention to strengthen ‘independent’ or ‘indigenous’ innovation. Why indigenous? There are three different factors behind this concept. Firstly, China has been strongly dependent on foreign technology for its economic growth. In 2003, foreigninvested enterprises accounted for 85.4% of all high-tech exports (High Tech Yearbook 2006). In recent years, there has been an increasing frustration with the realization that “market for technology” policy has not resulted in the immediate and automatic knowledge and

Liu and Lundin: Toward a market-based open system of innovation in China

27

technology spillovers to Chinese enterprises that policymakers had hoped for. In the case of the automobile industry as well as other sectors, some scholars argued that joint ventures do not lead to effective transfers of knowledge. Secondly, a culture of imitation and copying is abundant not only in product development and design, but also in the field of scientific research. Hence innovations from domestic knowledge bases and intellectual property rights are badly needed in China. Thirdly, the last twenty years’ high growth rate will not be sustainable without a change in the development strategy. China needs, for example, more energy saving technology, new management skills, and new organizational practices to ensure sustainable growth for the next twenty years. There are three main policies selected to fulfill the indigenous innovation strategy. Firstly, the government plans to increase R&D by 2020 to 2.5 % of GDP (from the current level of 1.3 %). Since GDP growth is projected to increase at a similar pace, an increasing R&D expenditure as a share of GDP implies a huge increase in absolute terms. Already today, China has the third-largest expenditure on R&D in terms of purchasing power parity, trailing only the US and Japan (Serger and Breidne, 2006). Secondly, fiscal policy to activate innovation capability in the company level is assumed the most important one. The new tax policy will make R&D expenditure 150 % tax deductible, thus effectively constituting a net subsidy, as well as accelerated depreciation for R&D equipment worth up to 300 000 RMB. Thirdly, a however new policy is the public procurement of technology. This policy is the result of learning from USA and Korea’s best practices. Public procurement in China today is significant, but the policy tool itself is relatively new to China. The purpose of current public procurement practice is to cut the costs rather than promote indigenous innovation. Under the new policy, government agencies have to prioritize innovative Chinese companies by procuring their goods or services even if these are not as good or cheap as those of other companies (both Chinese and foreign). The main points of the new public procurement policy are: • Setting priority for indigenous innovative products in public procurement. • More than 30% of technology and equipments purchasing should go to domestic equipment if using public money. • Giving indigenous innovative products some price advantage in the procurement process. 5.2 Foreign direct investment After an open policy for twenty years, multinationals have been getting more and more important in the production as well as in R&D in China. During the period of 1998-2004, the number of large- and medium-sized FDI firms was steadily increasing. While the shares of value-added and exports of FDI firms in the Chinese business sector reached a relatively high level (40% and 76%, respectively in 2004), the shares of R&D expenditure and employment are still relative low (29% and 34% respectively in 2004). It implies that FDI firms’ production in the Chinese business sector has been more capital-intensive but not really R&D-intensive in manufacturing (Table 12).

28

1 Policy and innovation

Table 12 The importance of FDI firms in the manufacturing sector, 1998-2004 (Share in the manufacturing sector, %)

Year

Number of FDI firms

Share of number of LMEs

Value -added

R&D Expenditure

Tech import

Export

Employment

1998

3489

22

26

21

20

58

14

1999

3764

23

28

23

16

61

16

2000

4221

25

30

20

19

63

18

2001

4585

27

31

23

28

66

20

2002

5327

29

33

23

24

68

23

2003

6512

31

36

25

27

71

27

2004

8745

36

40

29

48

76

34

Source: Lundin et .al, 2006b

Beyond the manufacturing and with the focus on the high-tech industries, the internationalisation in the high-tech industries is of significant importance, but it also has some controversial characteristics. On the one hand, the increased trade volume shows the international competitiveness of the high-tech industries of China. On the other hand, the dominance of FDI firms and the large share of processing of imported materials as well as the reliance on foreign technology raise the questions: Are China’s high-tech industries really high-tech? And are the high-tech industries in China really Chinese? However, there are also substantial cross-industrial variations in the high-tech industries. As a well-known fact, the ICT sectors are the most internationalized high-tech industries in which value-added, technology imports and exports are dominated by FDI firms. Regarding R&D expenditure, the share of FDI firms in the computer and office equipment industry has the most remarkable increase and FDI firms in the medical equipment and instruments industry have also noticeably increased their contribution to the R&D investment at the industry level (Table 13). Table 13: The importance of FDI firms across high-tech industries, 1998 & 2004 (Share in the high-tech industries, %) Number of FDI firms

Share of number of LMEs

R&D expenditure

Tech Import

Export

Employment

Pharmaceutical products

83

16

20

4

19

11

Electronics & telecommunication

349

52

41

77

86

42

Computer & office equipment

70

59

37

94

94

51

Medical equipment & instrument

28

20

11

41

40

14

Pharmaceutical products

158

21

22

20

21

16

Electronics & telecommunication

1145

72

42

93

93

73

Computer & office equipment

336

86

82

98

98

91

Medical equipment & instrument

105

38

27

33

88

36

Source: Lundin et al., 2006b

Liu and Lundin: Toward a market-based open system of innovation in China

29

In addition to the relative importance of FDI firms at the industry level, another important, but also somehow controversial question is if FDI firms are more R&D-intensive than domestic firms in China. While the R&D intensities across different ownerships all have increased during the period 1998-2004, so far domestic firms, both state-owned and private have a higher R&D intensity than FDI firms. The implications of these observations are the following: • Domestic firms in China are strengthening their innovation capacity through increased R&D investments. This is achieved not only by the increased R&D investments in the SOEs but is also driven by an increased number of entrepreneurial and S&T-based private firms. • The lower R&D intensities in FDI firms may be explained by two types of FDI activities in China. For the first, some of FDI firms’ activities are still capital- or labour intensive manufacturing in the high-tech industries. For the second, while some foreign firms are increasing their R&D effort in China, the R&D activities are still home-based in the OECD countries. • Even though the R&D intensity in the high-tech industries has increased over time, it still is at a much lower level compared to the high-tech industries in the OECD countries. From a long-term perspective, the R&D intensities need to, and will be further boosted, driven by continued indigenous R&D efforts and intensified competition between domestic and FDI firms when the technology gaps between them are being narrowed. Furthermore, the narrowed technology gap can also facilitate strategic alliances among firms with various ownerships and thereby boost R&D investments in both domestic and FDI firms. The contribution of FDI firms to the innovation system of China can be seen from their patent outputs in China. Patents in China are classified into three categories: design, utility model and invention, where the last category is presumably most R&D-intensive. One of the biggest differences between domestic and foreign applications is the structure of the application. For domestic firms, the majority of their patent applications belong to the first two categories, although the number of invention applications has been increasing as well. For foreign applications, the invention application is the main category. The number of invention applications by domestic firms for the first time exceeded their foreign counterparts in 2003. However, the foreign firms still outperformed their Chinese counterparts significantly in terms of the numbers of granted invention patents in the past years. While the number of patent applications in China by both domestic and foreign firms has been increasing, the number of applications by Chinese firms in international patent offices is still very limited. The number of Chinese patents in the triadic patent families was only 177 in 2003, compared with 19 222 for the U.S., 15 990 for the EU and 13 564 for Japan (OECD, 2005).

30

1 Policy and innovation Domestic

Foreign

70000 60000 50000 40000 30000 20000 10000 0 1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2002

2003

2004

Figure 1: Domestic and foreign applications for invention patents Source: China Statistical Yearbook on Science and Technology, Table 6-3, 2005

Domestic

Foreign

35000 30000 25000 20000 15000 10000 5000 0 1995

1996

1997

1998

1999

2000

2001

Figure 2: Domestic and foreign invention patents granted Source: China Statistical Yearbook on Science and Technology, Table 6-4, 2005

Among foreign patent applicants, the MNCs from Japan and the U.S. are the most active applicants, while German, Korean and French companies are also applying for an increasing number of patents in China (Table 14). The distribution by field of technology reflects to a large extent the competitive strengths of these MNCs in the Chinese market.

Liu and Lundin: Toward a market-based open system of innovation in China

31

Table 14 Top ten foreign enterprises with the highest number of applications for invention patens (2003) Ranking

Country

Enterprise

Number of applications

1

Japan

Matsushita Electric Industrial Co., Ltd.

1817

2

South Korea

Samsung Electronics Co., Ltd.

1560

3

Japan

Canon Co., Ltd.

820

4

Japan

Seiko Epson Corp.

781

5

South Korea

LG Electronics Corp.

624

6

Japan

Toshiba, Inc.

583

7

United States

IBM Corporation

581

8

Japan

Sony Corp.

560

9

Japan

Mitsubishi Electric Co., Ltd.

556

10

Japan

Sanyo Electrical Motors Co., Ltd.

541

Source: China Science and Technology Indicators 2004

When regarding the spillover of the FDI, there is a controversial debate in China. For example, in the automobile industry the first mover was Volkswagen in the 1980s. Later on, Citroen, General Motor, Mazda, Nissan Motor, Honda, Ford, Hyundai Motor, Toyota and Suzuki all became key players in the Chinese market. Most of them engaged in joint ventures. In 2004, in the passenger car industry, about 2 million passenger cars were produced in China, only one tenth with local brands. They are quasi private companies: Chery and Jili. The rest are joint ventures with large TNCs in the world. The passenger car industry is dominated by TNCs. Some observers have argued that there is little spillover from them to domestic firms. In contrast, domestic companies have been losing their innovation capability following having a joint venture with foreign companies (Lu and Feng, 2005). Firstly, as a foreign partner in the joint venture cannot have an equity share of more than 50% to control the new company, they will normally not open up much to the domestic partner in terms of technology transfer. Secondly, TNCs will care more about the contract rather than the market. Thirdly, they see importing core parts from their parent structure as more cost-effective than the Chinese production. In ICT industry, the story is different. In terms of investment, FDI is more important than domestic investment. From 1990-2002, the domestic investment for the ICT industry is 22.5 billion US dollars, foreign investment is about 70.billion dollars. Some students find a positive spillover of FDI here. Thanks to this spillover, local telecommunication companies catch up very quickly (Mu and Lee, 2006). Huawei is one of them. 5.3 Regional disparity China, as a big country, has different regions with different cultures, geography and resources. This kind of diversity is also very important for innovation. Historically, the north-

32

1 Policy and innovation

eastern part used to be the core industrial base of China with heavy industries and important technologies imported from the Soviet Union. In the western region, there were some isolated industrial bases following the 1960s and 1970s’ three frontier constructions, that is, the mass transfer of the defense industry from the coater region to the west region. This made Xian, Guichou and other western cities centers of heavy industry. Currently, Chongqing, Xian and Chendu are the three examples of innovative cities in the western region. The coastal area used to be the most recently developed region. For example, the government in the planned economy era invested very little in the regions of Fujian, Zhejiang and Guandong provinces, so there are few big SOEs in those regions. But those regions used to be the commercial centers since the Ming dynasty. Once opened to the world, they adapted to the market economy much faster than other regions. Private SMEs with great entrepreneurship flourished and Guangdong, Zhejiang, Jiansu and Fujian became the main regions of the new Chinese economy. But the diversity of innovation across China also has owned to the decentralization of decision-making over both resource allocation and operational decisions from the 1980s. One result but also a serious challenge in the R&D activities of China after the decentralization is the large regional S&T disparity. The gap among east, middle and western regions in terms of R&D can be seen in Table 15. Encouraged by an unbalanced growth approach and uneven FDI inflows, the Eastern region, with Beijing, Shanghai, Guangdong and Jiangsu became the hubs for R&D activity in China. They have a much higher level of R&D expenditure and R&D intensity than other regions. Table 15 Regional gap of R&D inputs

Share of national GDP(2003) Share of national R&D

East

Middle

West

58.9%

24.6%

16.5%

71%

17%

13%

Source: China Science and Technology Statistical Yearbook, 2005. Beijing: Chinese Press of Statistics.

Being aware of the divergence among regions and the risk that the gap will increase further, the central government launched the “go west” strategy in 2000, aiming at energizing the less developed regions through a combination of fiscal, regional, FDI- and S&T policies. But even with that, the R&D expenditures and R&D intensities in the mid and the western regions have still been increasing. In addition to R&D expenditure, there are large regional disparities in a broad range of aspects, e.g. human resource, high-tech industries and openness of the regional economies in general (Liu & Lundin, 2006) (Figure 3).

Liu and Lundin: Toward a market-based open system of innovation in China

R&D/GDP(%）

R&D Expenditure (100 million yuan) 250

200

256,3

179,8

150

100

50

0

150,5 128,9 103,8 83,0 79,4 75,2 68,0 54,8 40,4 38,1 37,5 34,2 32,7 32,4 30,1 27,8 17,4 17,0 15,8 12,8 11,2 11,0 7,9 6,4 3,8 2,4 2,4 1,2 0,3

0

Beijing Guangdong Jiangsu Shanghai Shandong Liaoning Sichuan Zhejiang Shaanxi Hubei Tianjin Hebei Fujian Henan Heilongjiang Anhui Hunan Jilin Chongqi Jiangxi Shanxi Gansu Guangxi Yunnan Guizhou Inner Xinjiang Ningxia Qinghai Hainan Tibet

33

1

2

3

1,32 1,21

4

5

6

7 7,00

2,06 0,84 1,38 1,46 0,80 2,83 1,01 1,65 0,54 0,72 0,49 0,74 0,82 0,65 1,10 0,77 0,60 0,64 0,98 0,41 0,45 0,58 0,30 0,20 0,62 0,62 0,18 0,17

Figure 3: R&D expenditure and R&D/GDP by region (2003) Source: Science and Technology Indicators, 2004

The performance of regional innovation capability can be seen in a report on regional innovation capability in China (Liu, 2006). Based on indicators of knowledge creation, diffusion, company innovation, infrastructure and performance, Shanghia, Beijing, Guandong, Jiasu, Zhejinag and Shandong are the top 6 most innovative regions in China. They all come from the eastern and coastal region (Table 16). According to the report, Beijing is a science superpower in China, but the innovation capability of enterprises lags behind in relative terms. Shanghai is the number one in China on every aspect of innovation. Guangdong is similar to Shanghai. Jiangsu, Zhejiang and Shangdong score well in company based innovation and knowledge acquisition, but lag behind in knowledge creation. Liaoning scores well in other aspects, except on performance, meaning that the region has a strong base for innovation, but lacks the capacity to turn this base into a strong performance (Table 15). Liaoning is a region with significant national investment, especially in heavy industries, but is still influenced by traditions and cultures of the planned economy. The same is true for the whole north-eastern region. Chaoqing is the only western region in the top ten. Their company system performs well in innovation and knowledge acquisition, but suffers from a relatively poor infrastructure.

34

1 Policy and innovation

Table 16: Scoreboard of regional innovation capability,2005

Regions

Final score

Knowledge creation

rank

Knowledge acquisition

rank

Company innovation

rank

infrastructure performance

rank

rank

rank

Shanghai Beijing Guangdong Jiangsu

56.97 56.11 50.22 48.41

1 2 3 4

46.96 80.94 29.13 25.47

2 1 5 6

59.51 40.41 44.05 57.41

1 5 4 2

61.19 42.52 58.17 59.7

1 7 3 2

50.07 59.71 48.83 45.9

2 1 3 4

65.9 61.77 62.44 47.9

1 3 2 6

Zhejiang Shangdong Tianjing Liaoning Fujiang Chaoqing

45.29 37.96 37.43 32.05 30.74 28.63

5 6 7 8 9 10

23.21 19.92 30.56 22.5 20.19 17.66

7 13 4 8 12 16

50.5 35.98 32.19 30.68 28.35 29.45

3 6 7 8 11 9

54.36 51.64 36.19 46.02 39.09 39.7

4 5 11 6 9 8

44.21 37.09 29.17 31.97 26.79 25.11

5 6 9 8 14 18

47.95 36.99 58.37 22.89 34.94 26.8

5 7 4 17 8 12

Xiannxi

27.27

11

21.75

9

24.13

15

33.21

16

28.13

10

25.25

14

Source: LIU Xielin et al., Chinese Report of Regional Innovation Capability, Chinese Science Press, 2006

5.4 Globalization of innovation in China There is a continuous trend of globalization of innovation in China. More and more multinationals establish their R&D centres in large cities such as Beijing and Shanghai. The purpose of that is multiple: to take advantage of cheap R&D human resources in China, to locate their R&D unit near to the Chinese market. According to von Zedtwitz (2006), there were 199 foreign R&D facilities in China at the beginning of 2004. The number has increased rapidly since then and possibly has amounted to 250-300 currently.

25 20 15 10 5

19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03

0

Figure 4: Number of new establishments of foreign R&D labs in China, (1987-2003). Source: von Zedtwitz (2006)

Liu and Lundin: Toward a market-based open system of innovation in China

35

The globalization of innovation in China can also be seen from the establishing of contacts with Chinese universities and research institutes by foreign companies (Table 17). Such a cooperation is still at an initial and immature stage. It is still very difficult for foreign firms to find original ideas and sufficiently innovative projects through this kind of co-operation. At the current stage, foreign firms do not buy ready-made projects or research, they rather utilize the existing R&D research capacity and facilities (which were often purchased by the support of governmental funding and of very high-standard) to carry out the research project, which is defined by the foreign firms themselves and modified during the working process in order to adjust to the local conditions. Nevertheless, the mutual benefits generated through such co-operative efforts should not be underestimated. It will not only provide local universities and research institutes with additional funding and more advanced equipment, more importantly, it will also generate positive demonstration- and spill-over effects for the universities and allow them to get more informed about the international research frontier. Finally, it can be an efficient way for foreign firms to identify research units and personnel with high research capacity. Table 17: Selected list of research co-operations between domestic research institutes and multinationals in the biomedical industry Foreign company

Chinese partner

Details

GlaxoSmithKline

Shanghai institutes of Materia Medica (SIMM)

Chemical compound database.

Roche

Chinese National Human Genome Centre

Diabetes and schizophrenia.

Novartis

Shanghai institutes of Materia Medica (SIMM)

Herbal compounds, Chinese traditional medicine.

AstraZeneca

Shanghai JiaoTong University

Gene linked to schizophrenia.

DSM

Joined lab with Fudan university in Shanghai. JV with Chinese vitamin makers

Novo Nordisk

Collaboration with Tsinghua university in Beijing

Nutritional products activities.

Diabetes

Source: Liu & Lundin (2006a), Table 23

In recent years, a few Chinese firms, in particular in the electronics and ICT sectors, have indeed initiated their international R&D activities either by the acquisition of foreign firms/ units or through setting up R&D organisations in OECD countries. The high profile M&A deals involving Chinese firms in the high-tech sectors have caused huge attention worldwide. In these M&A deals, access to R&D centres of the western sellers is one of the key elements. For example, in the TCL- and Thomson deal, it included Thomson’s R&D centres in Germany, Singapore and the U.S. Similarly, in the Lenovo-IBM deal, Lenovo took ownership of IBM’s R&D centres in Japan and the U.S. (North Carolina)(Table 18).

36

1 Policy and innovation

Table 18: Selected M&A deals by Chinese firms (2001-2005) Chinese bidder

Target foreign firm / Unit

Industry

Holly group

Philips Semiconductors, CDM hand-set reference design (US),2001

Telecommunication

TCL International

Schneider Electronics AG (Germany), 2002

Electronics

TCL international

Thomson SA, Television manufacturing unit (France), 2003

Electronics

BOE Technology Group

Hyundai display technology,(South Korea), 2003

Electronics

Shanghai Auto Industry Corporation (SAIC)

Ssangyong Motor (South Korea), 2004

Automotive

Lenovo group

IBM ,PC Division (US), 2004

IT

Nanjing Automotive

MG Rover Group (UK), 2005

Automotive

Source: Wu (2005), The Boston Consulting Group (2005)

In addition to the acquisition of R&D centres, some Chinese firms have made greenfield investments in the form of R&D units in foreign countries. China has a total of 37 R&D operations abroad which are concentrated in the ICT sectors and 24 of them are in developed OECD countries (FIAS, 2005)(Table 19). Table 19: Selected Design, R&D labs of Chinese firms overseas Chinese firms

Location

Industry

Huawei

R&D centres in Sweden (Stockholm), U.S. (Dallas, Silicon Valley), India (Bangalore) and Russia (Moscow).

Telecom

ZTE

R&D centres in Sweden, (Stockholm), India (Bangalore).

Telecom

Glanz Group

R&D centre in U.S. (Silicon Valley)

Electronics

Konka

R&D centre in Silicon Valley

Electronics

Haier

R&D centre in Germany, US (SC) and India, design centre in Boston,

IT and electronics

Kelon

Design centre in Japan

Electronics

Foton Motor

R&D centre in Japan, Germany and Taiwan

Automotive

Source: various press reports

In a recent report of the Boston Consulting Group (BCG, 2006), among the top 100 emerging global companies from rapidly developing economies, 44 are Chinese firms, 18 of which are in the ICT sector and a few in the automobile industry (Table 20).

Liu and Lundin: Toward a market-based open system of innovation in China

37

Table 20: Selected list of Chinese firms with globalization potential Stated-owned enterprise Company Haier

Privately-owned enterprise

Industry White goods

Company Midea Group

Industry White goods

SAIC

Automobile

Huawei

Telecom equipment

BOE

Electronics

Wanxiang

Auto parts

Lenovo

Computer

SVT Group

Electronics

TCL

Electronics

CHINT Group

Electronics

ZTE

Telecom equipment

Galanz

White goods

Chery

Automobile

People Electric

Electronics

Aux Group

White goods

Lifan

Motorcycle

Geely

Automobile

Source: IBM Global Business Service, 2006. The assessment of the globalisation potential is based on multiple criteria such as key firm characteristics related to size, export and innovation capacity and industrial characteristics such R&D intensity - and competition.

Even though the number of such Chinese firms is very few and the scale of their international R&D activities is still small, a new generation of Chinese firms seem to emerge as important players in S&T-intensive (instead of labour-intensive) segments of the global market. The innovation capacities of these Chinese firms and their ability to tap into the global network have therefore generated large interest, from both research- and policymaking perspectives. In other words, are these emerging Chinese multinationals able to be global players in the near future?

6

Conclusion

The Chinese innovation system is a very dynamic one and has undergone great changes in the last twenty years. The marks of both the planned and the market economy coexist in China. The system of innovation is still in a transition from a GRIs-dominated innovation system to a more enterprise centered system. Hence GRIs and universities are still very important in R&D activity as well as in terms of attracting R&D talents. The enterprise system is still weak in innovation capacity and their innovation performance is mostly in incremental innovations. The government also shapes the system through their policy, strategy and investments. SOE has undergone reforms of governance; lots of private and limited companies have been emerging. Huawei, Lenovo and Haier are well known examples of them. SMEs have now become more important players in the economy, driven by competition and entrepreneurships. The increasingly open innovation system, spurred by FDI over many years, has created significant inducements for structural change and learning among Chinese companies. So, there are two major forces that co-shape the future of the innovation system in China. The first one is a national strategy of indigenous innovation which focuses on how to promote local innovation capability by using a more favored policy for domestic companies and research institutes. The second is a cosmopolitan innovation based on global and open linkages.

38

1 Policy and innovation

References Edquist, Charles, System of innovation: perspectives and challenge, in Jan Fagerberg, D.Mowery and R.Nelson (eds) (2006) The Oxford Handbook of Innovation. Oxford: Oxford University Press. Freeman, C.(1987), Technology policy and economic performance: lessons from Japan, London: Pinter. High tech yearbook: 2006, China High-tech Industry Statistical Yearbook, Chinese Press of Statistics. Lazonick, William, The innovative firm, in Jan Fagerberg, D.Mowery and R.Nelson (eds) (2006) The Oxford Handbook of Innovation. Oxford: Oxford University Press. Liu, Xielin, et al., Chinese Report of Regional Innovation Capability, Chinese Science Press, 2006. Liu, Xielin & Lundin, N (2006a), “Globalisation of biomedical industry and the system of innovation in China.” SNS report, (forthcoming) Stockholm. Liu, Xielin & Lundin, N. (2006b), “China’s development model: an alternative strategy for technological catch-up”, mimeo. Liu, Xielin & Steven White (2001), Comparing innovation systems: a framework and application to China’s transitional context, Research Policy 30 (2001) 1091–1114. Lu, Feng and Feng Keideng, The Policy Choice of Developing Indigenous IPR Chinese Automobile Industry, Beijing, Peking University Press, 2005. Lundin, N. Sjöholm, F., Qian, J.C. and He, P. (2006a),”The Role of Small Enterprises in China’s Technological Development”, mimeo, Stockholm School of Economics. Lundin, N. Sjöholm, F., Qian, J.C. and He, P. (2006b),” FDI, Market Structure and Innovation - A firm-level analysis of the Chinese manufacturing sector”, mimeo, Stockholm School of Economics. Lundin, Nannan and Fredrick Sjoholm, Technology development and job creation in China, paper for 10th International Convention of the East Asian Economic Association 17-18 November,2006, Beijing. Lundvall,B.A.(ed.)(1992), National Systems of innovation: towards a theory of innovation and interactive learning, London: Pinter. MOST, China Science and Technology Development Report, 2006, Chinese S&T Literature Press. Mowery,D.C. and B.N.Sampat, Universities in national innovation systems, in Jan Fagerberg, D.Mowery and R.Nelson (eds) (2006) The Oxford Handbook of Innovation. Oxford: Oxford University Press.

Liu and Lundin: Toward a market-based open system of innovation in China

39

Mu, Q and K.Lee (2005), Knowledge diffusion, market segmentation and technological catch-up: the case of the telecommunication industry in China, Research Policy, 34(2005), 759-783. Nelson,R.R.(ed.)(1993), National Systems of Innovation: A Comparative Study, Oxford: oxford University Press. OECD (2005), OECD Science, Technology and Industry Scoreboard 2005, OECD, Paris. Serger, Sylvia and Magnus Breidne, China’s 15-year plan for science and technology– a critical assessment, conference paper for New Asian Dynamics in Science, Technology and Innovation, Gilleleje, Denmark (27-29 September, 2006). von Zedtwitz, M. (2006),” Chinese multinationals: new contenders in global R&D”. Conference presentation. http://goingglobal2006.vtt.fi/programme.htm. Wu, F. (2005) “The globlalization of Corporate China”. NBR analysis. Vol. 16, No.2. Ministry of Science and Technology of the People’s Republic of China (MOST) (2005b), The Yellow Book on Science and Technology Vol.7: China Science and Technology Indicators 2004, Scientific and Technical Documents Publishing House, Beijing.

40

1 Policy and innovation

Authors Liu Xielin is Professor and director of Research Center of Management of Information and Innovation, Graduate University of Chinese Academy of Science. Born in September 25, 1957, BSc, Peking University, 1982, M.SC, Chinese Academy of Science, PHD,1994, Tsinghua University. From February of 1995 to August of 1995, he was a visiting fellow in Sloan School of Management, MIT. From December of 2004 to Feb of 2005, he was a visiting professor in Hitochibashi University of Japan. From Sept.to Dec.of 2006, he is a visiting professor in Stockholm School of Economics. From April of 1994 to Sept. of 1996, he was an assistant professor in Tsinghua University. From Oct.1996 to June of 2006, he was a professor in National Research Center for S&T for Development, Ministry of Science and Technology. Now he is a professor in Graduate University of Chinese Academy of Science. He also acts as Vice President of Chinese Association of Science of Science and S&T Policy. His research areas mainly cover innovation policy, management of technology and innovation. He has published a lot of papers in Research Policy, Technovation, Journal of Management Studies and International Journal of Technology Management. He has written about eight books in the last ten years. Contact: [email protected], Phone: +86 (10) 82680673, Fax: +86 (10) 82680804 Management School of Graduate University Chinese Academy of Science No. 80 Zhongguancun East Road, Haidian District, Beijing 100080, P.R. China www.mscas.ac.cn Nannan Lundin, Ph.D. Department of Economics, Statistics and Informatics (ESI) Örebro University 701 82 Örebro, Sweden Tel.: +46 (19) 301257 [email protected]

Fan Chunliang: The role of the national R&D program in setting up China’s national innovation system 1

Introduction

It is well-known that, to a great extent, the economic strength and competitive capability of a country is determined by the national innovation system. The international experience shows that governments play an important role in building the national innovation system, even though there still exist debates about the extent to which the governments should be involved. Through more than 20 years’ reform of S&T and the economic system, China has formed the national innovation system adapting to the market economy which has made a great contribution to China’s economic development. However, there still exist some defects in China’s national innovation system; one of them is that the enterprises are short of technological innovation capability. In recent years, the technological innovation capabilities of the enterprises have greatly increased. The proportion of R&D investment of the enterprises in the whole country’s total investment, for example, has increased from 28.48% to 56.32% during the decade from 1991 to 2001, which raised almost one time,[1] and from 2000, R&D investment of the enterprises has exceeded 50% of the total investment of the whole country. Nevertheless, it seems that the enterprises have become the principal part of innovation in the innovation system. This would be supported if a great part of the enterprises which are engaged in substantial R&D activities were transnational corporations and joint ventures in China. And the other indicators which judge the technological innovation capability of the enterprises are still very low (such as technical output). We could say that the improvement of technology innovation capability in Chinese enterprises needs to make painstaking efforts. How can the technological innovation capabilities of the enterprises be improved? The answer is to take a mutual complementary way between market and government. The main way is to create endogenetic demands of enterprises’ technological innovation through reforming the economic system and building modern enterprise institutions which lead the enterprises development on the way of depending on technological innovation. Another way is to stimulate the enterprises to invest in their own R&D and technology innovation through the guidance of the government which would be implemented by the national R&D program and relative policies. This article will discuss the role of the Chinese National High-tech R&D Program (the 863 Program) in setting up the national innovation system.

42

1 Policy and innovation

2

Theoretical framework

2.1

The national innovation system

2.1.1 The meaning of the national innovation system Since the 1980s, the increasingly international economic competitiveness has fully manifested that innovation has something to do with the characteristic of the country. On this background, the concept of the national innovation system has been brought forward. A national innovation system has been defined as follows: (1) the networks of institutions in the public and private sectors whose activities and interactions initiate, import, modify and diffuse new technology (Freeman: 1987). (2) A set of institutions whose interactions determine the innovative performance of national firms (Nelson: 1993). Other scholars also bring forwards their own different definition of the national innovation system. [2] There is no consensus definition about the national innovation system, but it has a common character: (1) There is a series of common goals that has been identified with the important development of a nation’s future; (2) it consists of some institutional actors, mainly including: First, Universities, governmental research institutions, nonprofit research institutions, which are essential factors of the R&D capability related to national innovation. Second, Industrial R&D, including formal industrial R&D laboratories and other technological capacities. Third, the educational institution that trains scientists, engineers and senior technicians, and fourth, the government which takes on the responsibilty for scientific policies making and implementing. There has existed an interaction and connection among these actors which include: the following of funds; the relationship of laws and policies, the flowing of technology, science and information, the contact and cooperation of commerce and S&T.[3] The idea of the national innovation system shows clearly that innovation is a produced process with various actors involved under a complicated regulation. Innovation not only depends on one single actor but also on the interconnection and interaction between these actors, including a series of activities in the fields of science, technology, organization, commerce, finance and law. Innovation is not decided by a function of market power but is decided by the character of the whole innovation system. 2.1.2 The system failure The innovation system has suffered failures, which is an accepted view by scholars who are studying innovation. OECD define system failure as mismatch between elements in an innovation system.[4] Scholars of an innovation study have described the system failure at different aspects. According to the views of Carlsson and Jacobsson, the system failure could be divided into network failure and institutional failure. The former means the poor contact between organizations within the innovation system, the latter means the poor performance of certain actors or laws and regulations that hindered interaction and enterprises’ spirit in the innovation system[5]. Because of the system failure, results and effects were not good concerning inno-

Fan: The role of national R&D program in setting up China’s national innovation system

43

vation. Therefore, it is necessary to overcome the failures of the national innovation system in order to strengthen the construction of the national innovation system, including the improvement of low efficiency and poor ability in the functional actors of the national innovation system, such as improving the administration efficiency of the government, strengthening enterprises’ technological capabilities, improving and strengthening the necessary interaction between these actors (such as the cooperation between industry and university), and perfecting the regulation (such as the intellectual property law). 2.1.3 The policy based on system failure For a long time, the support S&T of government has been based on the rationales of market failure which means that the market mechanism could not solve an optimized resources disposition in some S&T fields and it must be involved and supported by the government. Concerning the rationales of market failure, the S&T areas which the governments support are concentrated on the public goods areas, such as basic research, the pre-competition research on general technology and S&T infrastructure. The theory of system failure indicates that the government should consider the whole frame which improves S&T and innovation for supporting S&T, including wider factors. Because of the complicated factor of the system failure, the policy makers need to do a detailed analysis based on actual cases. The policy based system failure rationales focus on repairing the system failure and improving the structural condition in order to stimulate innovation and reap the maximum social benefits of the innovation. Because of the different innovation systems between countries and different characters of the system failure, the governmental policies based on the system failure rationales have different contents and measures, the following are common: •

Creating a common vision for the whole innovation system (through S&T planning and technological foresight).

•

Strengthening the enterprises’ technological capabilities.

•

Promoting cooperation between industry and university.

•

Improving education and training of high-level talents.

•

Perfecting old laws and regulations and making out new ones.

The policies based on system failure rationales are not to rule out the ones based on market failure rationales but rather are to replenish and perfect it. 2.2

The role of the national R&D program in the setting up of an innovation system The National R&D Program is one of the main measures with which a government directly supports R&D activities. At first, the National R&D Program was only used to support the basic research and national defense-related research. Later on, it was expanded to support the

44

1 Policy and innovation

private enterprises to strengthen their industrial competitiveness. The areas of the National R&D Program mainly include: (1) defense, public welfare (health, environment, etc.), basic research. (2) new and high technology industry. (3) The industries facing the international competitiveness. (4) energy, communication and other related areas. The problem with the first one was an external knowledge of costs, or the problems with long investment benefit which could not be solved by the market mechanism. The second and third ones had possible benefits, but there still was a great risk or great difficulties with technology which could not be solved by a single enterprise or research organization to invest in. The fourth was the one which would not have enough benefits which means that no enterprise would benefit from technological progress in the invested fields. The main purpose of setting up a National R&D Program is to solve the problems of high-level resource distribution in S&T fields where the market failed or has low efficiency. Among these, the second, third and fourth were the problems of the industrial technological innovation. Facing increasingly high international economic competition, the Naitonal R&D Program in these fields has been used as common measure by governments of many countries. The National R&D program is a resource disposition with which the government meets some national goals. Especially for the national industrial technological R&D and innovation program in which industry, universities and research institutes are involved, it is a new institutional arrangement for reducing transaction between them. It also brings out a series of common aims by which every part of the national innovation system could be cooperated and put into interaction effectively. Therefore, it implies adjustable and strengthening functions for the whole national innovation system which are applied as follows:(1) to strengthen the parts with poor efficiency and capability in an innovation system, such as to strenghen the enterprises’ technology capabilities. (2) to strengthen the interaction between all actors, such as to enhance the interrelation and interaction between industry, universities and research institutes. Facing the increasingly international competitiveness, the design of the national R&D program of many countries not only aims at developing key technologies that could play an important role in the country’s development, but also pay more attention to solving the failure of the national innovation system and strengthening the effective operation of the national innovation system.

3

The role of China’s national high-tech R&D program (863 Program) in setting up the national innovation system

China’s National High-tech R&D Program (the 863 Program) was launched in the wave of the world high-tech competitiveness during the 1980s. At first, the aim of the 863 Program was a technological purpose (to narrow the gap to the world’s advanced countries). The 863 Program has made a great contribution to the construction of the national innovation system.

Fan: The role of national R&D program in setting up China’s national innovation system 3.1

45

The general description of the program

3.1.1 background The 863 Program was launched in 1986 when America, the Soviet Union and the countries of the European Community greatly developed high technology and lead a new technological revolution and competition. Four well-known Chinese scientists – Wang Daheng, Wang Ganchang, Yang Jiachi and Chen Fangyun – keenly recognized this developmental trend and jointly proposed to the Central Committee of the Party to accelerate China’s high-tech development. With strategic vision and resolution, the Chinese leader Deng Xiaoping approved of the National High-tech R&D Program and later on named it the 863 Program. It started to be implemented in 1987. From that time, China’s high-tech research development has entered a new development period. The term of the 863 Program was originally planned to be implemented within 15 years. With the successful implementation of the “Seventh Five-Year plan”, the “Eighth Five-Year Plan” and “Ninth Five-Year Plan” within 15 years, the Chinese State Council in April 2001 approved to continue the implementation of the Program in the “Tenth Five-Year Plan”. 3.1.2 goal At the beginning, because of the weak development in most of China’s high-tech areas, the goal of the program was set up as to follow the international advanced technology development and to shorten the gap to advanced countries in a few important high-tech fields; furthermore, to strive to make a breakthrough in certain areas where China has some advantages. With the enhancement of China’s scienctific and technological level and the implementation situation of the program, the objective of the 863 Program has been readjusted. During the “Eighth-Five Year Plan”, the objective has been readjusted as follows: in the basic research fields of high-tech development, it would bring forward major high-level innovation, and obtain the complete self-intellectual property. In the transformation of highlevel results to productivity, it was reassured to concentrate on target products and to strive to transfer the major research achievements into commercial and industrial applications. During the “Fifteen Five-Year Plan” period, the goal of the program has been transformed from following tracks and copying others to self-innovation and to strengthening the hightech innovation capabilities. 3.1.3 The priority fields According to the principle of “limited targets, the key points”, the 863 Program focused on 15 topics out of 7 major fields which comprised bio-technology, space technology, information technology, laser technology, automation technology and new materials. In 1993, communication technology as a topic was listed into the 863 Program. In July 1996, the national leading group of science and technology approved marine high-technology as No.8 topic of the 863 Program. During the “Fifteen Five-Year Plan” period, 19 topics out of 6 high-tech fields and several major topics of information technology, biology and modern agricultural technology, new materials, advanced manufacture and automation technology, energy technology, resources and environment technology had been selected as key developmental research topics.

46

1 Policy and innovation

3.1.4 The evaluation of the 863 program The Ministry of Science and Technology, PRC entrusted its evaluation group to evaluate the 863 Program in 1996 and 2000. The conclusion of the first evaluation was that the goal of the Program was correct and the tasks had been basically fulfilled. In the selected high-tech fields, the gap to the advanced countries had shortened and a batch of important key technology had appeared. The contribution of the 863 Program to the industry mainly resulted in the increasing industrial technological level and strengthening industrial competitiveness, but the direct contribution to the economic benefit was not clear. [6] The conclusion of the second evaluation was that the defined strategic target and orientation conformed to national conditions and the law of S&T development by 15 years’ practice. The implementation of the Program involved the forming of an own research base over 6 important high-tech fields, played a decisive role in China’s high-tech research and development and has shortened the gap to advanced countries. [7] The second evaluation had not made any conclusion on its commercial application. 3.2

The contribution of the 863 program to the construction of the national innovation system – inducing enterprises to invest in R&D

3.2.1

From technology guidance to both technology and industrial guidance in R&D strategy In the first ten years, the contribution of the 863 Program to the industrial development was not evident. There were a few successful cases, such as a computer integrated manufacture system and an intelligentized robot. The application of the computer integrated manufacture system has increased industrial competitiveness which was successfully used to over 50 times in 1994 . In the first ten years, the 863 Program was joined mostly with research institutions and universities, only a few were enterprises. During the “Eighth Five-Year Plan” period, the percentage of enterprises joined was only 11.3% of the all projects of the 863 Program according to the investigation, [8] fewer of the enterprises hosted the project. This might have two reasons: The first was the goal of the program. The primary goal of the 863 Program was to follow the international high-tech development, but not to realize the industrialization of technology. Though from 1993 on, the aim of the 863 Program had been adjusted to industrialization, the principle mechanism and participating structure had not changed very much. The second was that, at that period, the enterprises were facing the problems of reformation and development and most of the enterprises were low at technology capabilities. In the second ten years, especially during the “Fifteen” period, the stress of the 863 Program was put on the technology breakthrough and on promoting the technological industrialization. The low participation rate of the enterprises has clearly improved. The percentage of the undertaken projects increased from 1.2% in 1994 to 30% in 2002. Some joined enterprises are well-known ones in China, such as Datang, Kelihua, Jingshan and etc..

Fan: The role of national R&D program in setting up China’s national innovation system

47

Table 1: The projects of the 863 program undertaken by enterprises Year 1994

2001

2002

2004

Project (%)

1.22

12

30

23

Fund (%)

1.03

14

31

35

personal (%)

4.97

15

37

Project by enterprises

Sources: The Promotion Center of Chinese Science and Technology; The Evaluation of China’s High Technology Development Research Plan (863),1996; The United Office of 863 Program of the Ministry of Science and Technology, PRC; The Annual Report of 863 Program (2001, 2002, 2004

3.2.2 The rules for the encouragement of enterprises joining The total proportion of the projects undertaken by enterprises has been increased by a big margin, because the 863 Program has made a series of rules to encourage enterprises to join in, which includes the classification of main projects, setting up guided projects and enhancing the industrialization projects in a special program. 3.2.3 The classification of projects In order to encourage the joining of enterprises, the program has classified its main projects into two categories: The projects of exploitation (A category) and the projects of applied research (B category). While encouraging the implementation of the forwarded projects of exploitation, the 863 Program has encouraged enterprises to join it and promote the results of transformation and industrialization. [9] The one who makes applications for applied research projects must be an enterprise or must cooperate with enterprises. The evaluation procedure of the project application must have the experts from enterprises or other users. The index of the evaluation must be beneficial to the participation of enterprises and the prospect of industrialization. The management of applied projects uses different methods in order to help enterprises to join the applied research. The enterprises undertaking the projects must invest balanced funds which will ensure to be taken into account through legal measures. 3.2.4 The guidance projects In the year of 2002, the 863 Program started guidance projects in order to guide the local government and enterprises to invest in high-tech industries that are provided with local superiority and key development and would strive to form scale, to produce benefits, and to realize the coordination between the country and locals concerning high technology and industrial development. [10] The units undertaking the projects should have a certain research and exploitation strength and industrialization capability. The responsibility for the investment of guidance projects is commonly shared between the country, locals and enterprises. While the 863 Program supports certain guidance funds, the enterprises should invest to at least 50% of the total investment in the project, the division of local science and technology

48

1 Policy and innovation

would match quota not less than national allocate funds. The 863 Program has arranged 43 guidance projects in 2002. 3.2.5 The key special projects The launching of the key special projects of the 863 Program is meant to solve the key hightech problems for meeting the needs of national important strategies. The key special projects were launched in the “Eighth Five-Year Plan” period. From 1991 to 1995, the National Scientific Committee has brought into place the implementation of rice gene, a timely transmission system of aerial remote sensing, the key technology of I-IJD-04 giant numerical program control convertible machine, superconductive technology and marine technology as special projects in line with the 863 Program. The key special projects have taken the core of the major system and engineering. During the “Fifteenth Five-Year Plan” period, the 863 Program stressed the key special projects about giant products, systems and engineering which must be mainly undertaken by enterprises to establish an integrated cooperation type of industry-university-research institute. A typical case is the project of an electric motor. The key special project of an electric motor has started by the joining of over 100 enterprises in 2001 and has drawn on 24 billion RMB mated by the country, locals and enterprises. The project has adopted the management model, one unit of which is the leader to exploit the whole motor and spares coordinated by enterprises. In order to organize an excellent group, institutions with different technology characters have been elected to have 2-3 units on making each spare and to undertake a matched project on the same key spare. Since the implementation of the project, over 200 enterprises, universities and institutions have established in China with a stable research group of 2000 technical bones. A batch of middle and young aged technical backbones has been fostered, and a group with excellent talents has been established and stabilizes the development of the electric motor industry in China. The project has promoted to make out 26 national standards, applied for 796 patents at home and abroad, among which 413 are invention patents[11]. On Jan. 16, 2006, the Ministry of Science and Technology, PRC issued “Several Ideas About the Management Reform of China’s S&T Program” pointing out that during the “Eleventh Five-ear Plan” period, the national S&T program will continue to strengthen the supporting strength for the enterprise’s technology innovation.

4

Conclusion and suggestion

“The Essentials of Middle-Lang Plan of National S&T Development”(2006-2020) (shorter form “Essentials”) point out that in the future 15 years, the whole plan of China’s S&T development will include four aspects: The first one is to set foot on the demand based on the national situation, define several key fields, break through a batch of major and key technologies, and wholly increase the supporting capability of S&T. The second is to take aim at the national target, implement several key special projects, realize the leap-over development and fill up blanks. The third one is to face the challenge, overwork forward technology and basic research, strenghen a sustainable innovation capability and lead an economic and social development. The fourth aspect is to deepen the system reform, a perfect policy and measure,

Fan: The role of national R&D program in setting up China’s national innovation system

49

to increase S&T investment, strengthen the building of talents’ groups, to promote the contribution of the national innovation system and to provide a reliable guarantee for China entering innovation countries. It can be said that at the time of promoting a key technology, the key special projects and a forward technology and applied research development, it is very important to strengthen the construction of the national innovation system, as we can even say that there is the guarantee for the fulfillment of the preceding three tasks. “Essentials” pointed out: “The aim of deepening the reformation of S&T system is to promote and perfect the construction of national innovation system…... At present period, the force of the construction of national innovation system with Chinese characters is to input patterns with industries as the principle entity, and realize the technology innovation system integrated by industry-university-research institute, and with that to be as a breakthrough to promote the construction of national innovation system….. .” We can see that the 863 Program is making a contribution on it. Since the launching of the 863 Program, industrialization has been a controvertible problem. During the “Fifth Five-Year Period”, for the implementation of the S&T Program, the nation has adopted beneficial principles for bringing enterprises into play and has encouraged the joining of enterprises in order to face the challenge of the world trade organization and promote the advance of the industrial technology. The rules and the management mechanism of the 863 Program for encouraging an active participation of enterprises has greatly promoted the self investment for R&D activities and, at the same time, has promoted the development of the joining of enterprises and the integration of industry-university-research institutes. At the time of pushing on China’s high-tech development this has made a great contribution to the construction of the national innovation system. It should be noticed that it is just the beginning for the task of the systematical promotion of the construction of the national innovation system of the 863 Program. In order to play a full role for the construction of the national innovation system, the management mechanism of the 863 Program should be further developed and perfected. One of the most important tasks is to directly strengthen the evaluation of the projects. In recent years, projects and funds of the enterprises have occupied about 1/3, which is the main body in the implementation of the 863 Program. Therefore, it is necessary to do the evaluation of the enterprises that are implementing the 863 Program, to evaluate the actual results of how the projects supported the enterprises in order to offer the scientific proof for the program’s implementation and adjustment. The content of the evaluation should include the whole distributive situation (disciplines, the type of companies and character of subjects) of the program’s support of enterprises and their results, the guidance role of the program for the enterprises (such as the investment situation of the relative resources and the development of technology capabilities of the enterprises), the research results and the promoting effect of the integration of industryuniversity-research institute due to the support of the program. It should not only take the evaluation as a measure, but also should take it as an inner regulation for the implementation of the program. According to international practices, a lot of experts who are engaged in the projects of the program should be included, and should be independent from managers and executors of the Program.

50

1 Policy and innovation

References: [1] The Analyses of R&D Project of China, the research group for the whole subjects. The Study of R&D Resource Distribution in the National Innovation System [M] Beijing: Scientific and Technology Press, 2004. 174 [2] OECD. National Innovation System[R].1997.10. [3] Niosi,J. et al. National System of Innovation : In Search a Workable Concept. Tehcnology in Society[J],1993,Vol 15.211. [4] Hauknes J. Lennart Nordren. Economic rationales of government involvement in innovation and the supply of innovation-related services(R). STEP report.1999.7. [5] Carlsson, B. and Jacobsson, S..In Search of Useful Public Policies: Key Lessons and Issues for Policy Makers” in Carlsson, B (ed.), Technological Systems and Industrial Dynamics(M), Kluwer Academic Publishers. 1997 [6] The Evaluation Group of the National S&D Program of the S&T Promotion and Development Research Centre of China, the Evaluation of China’s High-tech Research development Plan[R], 1996.8-15 [7] Chen Zhaoying The Cases of S&D Plan Evaluation – Evaluation of the 863 Program The Division of Basic Research of the Ministry of S&T, PRC. Compilation of Reference Material of the Training of the S&T Evaluation. [8] The Evaluation Group of the National S&D Program of the S&T Promotion and Development Research Centre of China, the Evaluation of China’s High-tech Research development Plan[R], 1996.8-15 [9] The Jointed Office of the 863 Program of the Ministry of Science and Technology, PRC, the Annual Report of 2002 of the National High-tech Research Developmental Plan [10] The Jointed Office of the 863 Program of the Ministry of Science and Technology, PRC, the Annual Report of 2002 of the National High-tech Research Developmental Plan [11] The Passing of Reception of the Key S&T Special Project of Electric Motor

Author Fan Chunliang is Research Fellow in the Institute of Policy and Management, Chinese Academy of Sciences. He received his Master Degree from the University of Science and Technology of China. His research on focus on Science policy, Management of R&D Project and Science and Society. He Published a book about Science and Technology Policy and more than 30 articles in Journals. Contact: [email protected], Phone: +86 (10) 62542624, Fax: +86 (10) 62542619 Institute of Policy and Management, Chinese Academy of Sciences Zhong Guan Cun Dong Lu No. 55, Beijing 100080, P.R. China

Zhao Lanxiang: The Role of ERCs: Public and Private Aspects 1

Introduction

The importance of knowledge as a competitive weapon has increased dramatically. The government-academia-industry linkages have received an increasing importance in coping with social and technological changes, which have progressed at a faster pace in the last two decades. In order to strengthen these linkages, a variety of policy measures have been developed and these measures have facilitated the development of various schemes in many countries (Ahn, 1995). In China, it has been stressed in the long and medium term plan for science and technology development that the transition from investment- to innovation-based economic development, is the strategy choice, and the policy will put more emphasis on increasing a competitive advantage through indigenous technological capability building. In line with this new emphasis on innovation intensive development, more recently, the National Development and Reform Commission announced that during the eleventh five-year economic and social development period, 100 National Engineering Laboratories(NELs) will be established in order to build the fundamental industrial innovation capability and to change the situation of the lack in support of fundamental industrial technology. The NELs program was regarded as an important part of the National technological Innovation System. Till now, three NELs have been established. It is not astonishing that policymakers need to create a more effective institutional framework to facilitate the interactions between academic and industrial research. How to make the NELs meet the needs effectively will be a challenge for the policymakers. Maybe a study on the development of ERCs programs in China, which is operating in its fifteenth year, will give us some policy implications.

2

Background for Establishing the ERCs

2.1 The Technology Contribution is Expected to Increase Effective utilization of scientific and technological resources is becoming an issue of concern all over the country. But at the beginning of the 1990s, the contribution of technology to economic growth has been much lower compared with other developed countries. How to solve this problem is the key object of the S&T system reform. The gap between academic institutions and the practical needs of society calls not only for a revision of old ideas of the

52

1 Policy and innovation

role of the universities and research institutions, but also for approaches to create more effective linkages between industry, academia and government. Bringing about an effective linkage between academic institutions and industry is of universal concern, and the government must also step in, particularly in instituting necessary policies measures. The establishment of ERCs is a governmental attempt aiming at increasing the technology contribution to practical needs. 2.2

A More Effective Approach is Needed to Bridge the Gap between Academic Institutions and the Practical Needs Although a variety of policy measures have been developed in order to strengthen these linkages, there are still some problems which led to a reconsideration of the mode and mechanism of greater coordination among industry, universities and government. In China, such linkages are often based on R&D projects sponsored by the government. The infrastructure and the capabilities for engineering research is relatively weak, which may result in the poor effectivity of the research project value to the industry. Meanwhile, academic institutions experience many problems in fulfilling their engineering research commitments as a result of inadequate facilities and funds. The recognition of this as bottleneck hindering technology transfer led to the attempts to solve the problems by enhancing the infrastructure building for engineering research and the introduction of a new mode of technology transfer. ERCs is one of the policy measures developed to eliminate such bottlenecks. 2.3 Experience from the USA The NSF-ERCs are one of the US government’s major policy innovations of the 1980s to foster an improved university–industry R&D collaboration. In 1984, the NSF published its report “New Engineering Research Centers: Purpose, Goals and Expectations.” These centers’ design features were posited as leading Firstly, to the increased industrial relevance of the academic R&D agenda, to a more rapid transfer of academic research findings to the industry, and Secondly, to the development and deployment of graduates better prepared and acculturated to work in the industry (Feller, Ailes.& Roessner, 2002).

3

Characteristic of Chinese ERC Program

3.1 Objectives of ERCs The recognition of the problems and experience from other countries led to a concern about the attempt to provide a solution through the creation of ERCs. The program involves the establishment of ERCs initiated by the National Planning Commission (NPC) in 1992, and ERCs initiated by the ministry of science and technology (MOST) in the mid-1990s.

Zhao: The role of ERCs: public and private aspects

53

The ERC program was started with the following objectives: Firstly, to promote university and industry links for the application of research results; To shorten the time of knowledge acquisition and application, Secondly, to pursue engineering research on fundamental technologies linked to industrial development. So far, 99 NDRC-ERCs and 141 MOST-ERCs have been established. Besides these 240 National ERCs, some local governments and some ministries also established ERCs at the local and ministry level respectively. 3.2 Characteristics of this New ERCs Program There are some characteristic of this new program compared with other countries which have also launched ERC programs. Firstly, nearly half of the NDRC-ERCs (49) were established with the loan financing provided by the world bank (totally $200 million). At the same time, the world bank signed a contract with China which indicated that all ERCs should establish the modern corporation system. Secondly, the ERCs program was initiated by both NDRC and MOST. Besides, there are also some ERCs sponsored by the local government. Thirdly, ERCs were established not only in major universities or government research institutes like it is the situation in the US, but also in some Large and Medium sized firms. These three key characteristics may define the current situation of Chinese ERCs and also have the impact on its development path.

4

Do the ERCs Work? A Case of CAS

After 14 years’ development of ERCs, it is time to discuss some questions about its goal and the policy measure towards this goal. Can the mode of operation help the ERCs to persue its goal? What has happened? I’d like to discuss such questions here based on Chinese Academy of Sciences (CAS) cases. 4.1 Current Status of the ERCs Program In the CAS, there are 34 ERCs at different levels, including 11 NDRC- ERCs , 14 MOSTERCs and 9 CAS- ERCs (Tab.1). There are 25 so called national ERCs, accounting for about 10% of all national ERCs.

54

1 Policy and innovation

Table 1: ERCs in Chinese Academy of Sciences. Different Sponsors NDRC- ERCs

Number of ERC 11

MOST- ERCs

14

CAS- ERCs.

9

Total

34

All these ERCs are mainly associated with 5 technological areas, including electronics /IT, new materials, advanced manufacturing, energy and resource recovery, biotechnology and medical technology (Tab. 2). Table 2: Technological fields distribution of ERCs in Chinese Academy of Sciences Technological Area Electronics / IT Biotechnology / Medical Technology New Materials Advanced Manufacturing Energy / Resource Recovery Others Total

Number of ERCs 11 6 6 5 4 2 34

4.2 Performance of ERCs From 2000 to 2004, the total sales growth rate in all these ERCs was 27% annually. During these 5 years, the portion of productive sale and technological sale in the total sale accounted for 69% and 9% respectively (Figure 1).

mil.Yuan 1600 1400

total sale

1200

technological sale productive sale

1000 800 600 400

others

200 0

2000

2001

2002

2003

2004

Figure. 1: Performance of ERCs in Chinese Academy of Sciences (2000-2004)

Zhao: The role of ERCs: public and private aspects

55

When we compare this with other ERCs at the national level, the same conclusion can be found. According to a survey among 34 NDRC-ERCs, from 2001 to 2003, the portion of technological sale in the total sale was about 20%. Another survey’s findings among 137 MOST-ERCs indicated that from 2001 to 2004, the portion was 23%.

5

Issues to be Discussed

ERC is an attempt to increase the academic institutions’ contribution to the industry by organizational forms. The issue is whether this “new” mode of R&D interaction and technology transfer is complementary for “older” forms of technology transfer. Considering the trend currently emerging in the ERCs as noted above, if ERCs still develop into the direction, the role of the ERCs will be limited. Some main problems should be considered. 5.1

An ill-defined Role as Research Institutions as well as Profit Organization Institutions The process of transforming academic research into commercializable technological innovations is complicated and needs sufficient time. The pressure of a rising impact of the Government Performance sometimes leads to ideal objectives for the R&D program. From the beginning of the ERCs program, the government has not given a well-defined role for ERCs. ERCs are expected to be a NPO so as to provide fundamental and key industrial technology (Zhao, 2006). Meanwhile, ERCs are also expected to be a firm which can survive and develop by themselves. In these 14 years, the ERCs have been wandering between research institutions and profit organization at all times. 5.2

Survival Pressure Hindering ERC’s Evolutionary Path to its Primary Goal The ending of the ERC program support curtails the ERC’s sufficient financial ability to maintain a fundamental research program, and mainly because product development is a matter of survival, the primary emphasis of ERCs has been put on “industrial product development” rather than on “technology generation and transfer”. Eventually, as one profit organization, ERCs are reluctant to transfer technology that may weaken their competitiveness. In this context, the ERCs cannot involve themselves in “technology generation and transfer” Meanwhile, the Goal of the ERCs program primarily is engaging ERCs to improve technology-generating and transferring activities. 5.3 Challenges for ERCs to Sustain Core Research Suppose the ERCs became one firm and were thus seeking to secure stable funding by providing technology in the market. What will happen? As we know, industrial firms are more concentrated on ERC activities that provide for private aspects of technological knowledge (Larson, 1998). Compared to the goals of the ERC program, it is not a satisfying solution.

56

1 Policy and innovation

Industry support of fundamental engineering research is fragile. The core funding arrangement for ERCs in effect underwrites both public and private aspects of technological knowledge. Meanwhile, most ERCs were originally established in famous universities or research institutes which provide the best research background for ERCs’ further development. As the ERCs are becoming more and more independent, in response, these universities or research institutes do not seem willing to pay much attention to the enhancement of ERCs. This will lead to the lack in fundamental and core research capability for ERCs’ further development. The loss of support both in research sources and government funding raises challenges for ERCs to sustain their core research, and their functions in the NIS.

6

Conclusions and Implications

The termination of program funding for core research activities and related infrastructure places both financial and programmatic burdens on ERCs. The ERC should be encouraged to develop even stronger ties to the industry in order to be effective in technology transfer and to be sure that research is done on real problems. But this does not implicate that the ERC itself must become a firm. The key is that the ERC program was created to develop fundamental knowledge in areas critical to competitiveness in world markets. The ability of ERCs to maintain the more strategically planned, fundamental portions of their R&D portfolios becomes problematic. ERCs are organizational innovations. Being more supported by the public sector or private firms will help to define their evolutionary path on providing the technological knowledge with public or private aspects. The distinction is best understood in terms of Nelson’s observation (Nelson, 1982) that “there is both a private and public aspect to technological knowledge and that while the lines between the two are shady, it is important to recognize both”.

References Ahn, Soon II. 1995. A new program in cooperative research between academia and industry in Korea, involving Centers of Excellence. Technovation, 15(4): 241-257 Feller, I, Ailes, C. P & Roessner, J. D. 2002. Impacts of research universities on technological innovation in industry：evidence from engineering research centers. Research Policy, 31 (3): 457–474 Larson,C..1998. R&D in iIndustry. AAAS report XXII: Research and Development FY1998. American Association for the Advancement of Science, Washington, DC: 31- 38 Marazita, C. F. 1991. Technology transfer in the united states: industrial research at engineering research centers versus the technological needs of U.S. industry. Technological Forecasting and Social Change, 39 (4) : 397- 410

Zhao: The role of ERCs: public and private aspects

57

Nelson, R., 1982. The role of knowledge in R&D efficiency. Quarterly Journal of Economics 97 (3): 453–470 Zhao, lanxiang. 2006. Contributing factors for the gap between defined ERC roles and its status quo. Science of Science and Management of S.&.T., 2006 (11): 87-92

Author Zhao Lanxiang is Professor and Deputy Director of Center for Strategic Studies, CAS. She was visiting scholar in the School of Public Policy at George Mason University, USA in 2002. Her research interests include science and technology policy, innovation policy and technology management. In recent years, she has finished some policy oriented studies in China. She has led or participated in many research projects funded by Ministry of Science & Technology of China, the State Economic and Trade Commission of China, Beijing Municipal Government, National Natural Science Foundation of China, Chinese Academy of Sciences and UNESCO. She is also Deputy Editor in Chief of Journal of Studies in Science of Science. Contact: [email protected], Phone: +86 (10) 62550680 Center for Strategic Studies Institute of Policy and Management Chinese Academy of Sciences 3D, No.33 Beisihuan Xilu Beijing 100080, P.R. China

Gao Jian and Shi Shude: The determinants of entrepreneurial activity in China – Empirical analysis by regional level data Abstract Entrepreneurship produces significant effects on the economic development. In order to boost regional entrepreneurship, two basic questions should be solved: one is how to measure entrepreneurship activity; the other is what the determinants of regional difference are. This paper finds that the China Private Entrepreneurship Activity (CPEA) Index is a suitable measurement for the regional entrepreneurial activity of China and figures out the significant difference of entrepreneurship at the regional level. Using this econometric model we find that the differences for entrepreneurial activity are explained by the regional market demand, industrial structure, unemployment rate, availability of financing, entrepreneurial culture and human capital. Technology innovation and growth rate of consumption do not have significant effects on the entrepreneurship in China. Keywords: entrepreneurship, measurement, determinants, CPEA

1

Introduction

Entrepreneurial activity provides an important impetus to regional economic development. It increases employment opportunities (Vanstel and Storey,2004; Fritsch,1997), enhances the technical innovation level (Tether,2000; Keeble,1996), and promotes economic development (Audretsch and Fritsch,1997；Fritsch and Mueller,2004). Many scholars have been carrying on researches towards the determinants of a new firm formation on the regional level. They try to explain regional differences by examining a set of regional determinants from economy, policy, culture and other social aspects. In China, however, this is a new area. Studies on regional entrepreneurial activity and new firm formation have significant meanings to recognize China’s regional entrepreneurial activity characteristics and relationship between entrepreneurial activity and regional economic development. There are two kinds of researches which can be divided up into several determinants of difference in entrepreneurial activity: one is the regional factors of entrepreneurial activity; the other is the relationship between entrepreneurial activity and special factors. Regional factors in entrepreneurial activity found to be the most important were population, income, unemployment, industry, financing and human capital. Different scholars had different analytical ways of thinking by selecting different measures to factors, and using different

Gao and Shi: The determinants of entrepreneurial activity in China

59

data sets. For example, Bartik (2001) and Fritsch and Mueller (2005) found that population density had a positive influence on entrepreneurial activity, while Armington and Acs (2002), Lee, Florida and Acs (2004), and Sutaria and Hicks (2004) found that the population growth rate had a positive effect. Similarly, Bartik (2001) and Armington and Acs (2002) found that industry density had a negative influence on entrepreneurial activity, while Sutaria and Hicks (2004) and Fritsch and Mueller (2005) found that industrial structure had a positive influence. And Fritsch and Mueller (2005) found that the unemployment rate had a strongly positive impact on entrepreneurial activity, while Sutaria and Hicks (2004) reported no such impact, but found the change of the unemployment rate had a positive impact. Armington and Acs (2002) and Lee, Florida and Acs (2004) found that income growth rate and human capital had a positive influence on the entrepreneurial activity. Fritsch and Mueller (2005) found that the regional innovation level and entrepreneurship culture had a positive influence on the entrepreneurial activity. Some pieces of literature paid attention to the effects of a special factor. Ritsilä and Tervo (2002) endeavored to simultaneously identify the separate effects of personal, regional and national unemployment on entrepreneurial activity. They found a positive and non-linear effect of personal unemployment on the likelihood of an individual to become an entrepreneur. And they also found the economic situation had an effect on entrepreneurial activity: times of low unemployment and business prosperity favored entrepreneurship. But they found no clear evidence for the regional unemployment situation affecting the likelihood of founding a business. Davidsson and Wiklund (1997) studied the impacts of cultural values and briefs on entrepreneurial activity. They surveyed large samples of 35-40 year old inhabitants in each region for cultural values and briefs data. And the results of their studies suggested that both values and briefs did have an effect on regional entrepreneurial activity. Basu and Parker (2001) thought financing was a key determinant of successful start-ups, and the largest single source of funds was self-finance, provided by the entrepreneurs’ own savings or assets. Amongst external funds, bank loans tended to be the most important source by size, followed by loans from family members. This paper analyzes the determinants of entrepreneurial activity in China at the regional level. The main aim of this paper is to identify determinants of regional entrepreneurial activity in China by conducting an empirical analysis. The paper is organized as follows: the second part establishes a measurement for the regional entrepreneurial activity level and discusses regional attributes of entrepreneurial activity in China; the third part puts forward research hypotheses of determinants of regional entrepreneurial activity; the fourth part describes the data and the empirical methodology, and examines the empirical results; the last part is the conclusion.

2

Measurements and CPEA

The entrepreneurial activity rate is an index widely used to measure the regional entrepreneurial activity level. Usually there are two approaches which can be used to calculate the entrepreneurial activity rate: the first method standardizes the number of new entrants relative to the number of establishments already in existence, which can be termed the ecological ap-

60

1 Policy and innovation

proach, because it considers the amount of start-up activity relative to the size of the existing population of business (Sutaria and Hicks, 2004); the second method, which can be characterized as the labor market approach, is to standardize the number of new firms with respect to the size of the labor force (Robson, 1994; Davidsson and Wiklund, 1997; Armington and Acs, 2002; Lee, Florida and Acs, 2004; Fritsch and Mueller, 2005). The labor market approach, however, is used more widely than the ecological approach in the existing literature. And the paper adopts the labor market approach to measure the regional entrepreneurial activity rate in China. In order to calculate the rate, we have to find two sets of data: new business number and labor force number. The GEM (Global Entrepreneurship Monitor project) has the definition that those enterprises whose established time is less than 42 months are regarded as new businesses. According to this definition, the paper views the total growth number of private enterprises within the past three years as new business number. The private enterprises number each year can be gotten from the "Chinese Statistics Yearbook". Concerning the labor force number, the paper chooses the population number of The Fifth National Population General Survey, which states a span of age from 15 to 64 years. With these two sets of data, we calculate the regional entrepreneurial activity rate in China and define this rate as CPEA index (China Private Entrepreneurship Activity). We respectively calculated CPEA data, including 31 provinces, city and autonomous regions in the mainland for eight years from 1997 to 2004 (as a result of data limiting, Taiwan, Hong Kong and Macau are not included). CPEA data presents significant differences between the areas. Taking 2004 as an example, we can divide all areas into four kinds of regions: a higher active region, of which CPEA data in 2004 is larger than or equal to 25; an active region, of which CPEA data in 2004 is larger than or equal to 12 and lower than 25; an inactive region, of which CPEA data in 2004 is larger than or equal to 7 and lower than 12; a quiet region, of which CPEA data in 2004 is lower than 7. Classified results are given as follows: Beijing and Tianjin Region, Yangtze River Delta Region (Shanghai, Jiangsu, Zhejiang), and Zhu Jiang Delta Region (Guangdong) are the most active regions. They all belong to the eastern coastal regions. Shanghai resides in the first place in all areas of China continuously from 2002 to 2004, Beijing resides in the next place, but the disparity of the two areas has been expanding. From 2002 to 2004, entrepreneurial activity rates of Shanghai, Beijing, and Tianjin keep growing faster, indicating that these three areas may have a fine future of economic development. The growth rates of entrepreneurial activity of Jiangsu, Zhejiang and Guangdong, however, have been descending in recent years. More active regions comprise eight areas: Liaoning, Fujian, Shandong, Hubei, Chongqing, Sichuan, Ningxia, and Xinjiang. The growth rate of entrepreneurial activity in these areas is comparatively stable. Of course, entrepreneurial activity rates of Liaoning, Shandong and Fujian, which are coastal areas, are larger than those of the other four inner areas. There are twelve inactive regions: Hebei, Mongolian, Shanxi, Jilin, Anhui, Jiangxi, Henan, Hunan, Hainan, Yunnan, Gansu and Qinghai. These areas have a lower growth level of entrepreneurial activity rate. Some of them even have a descending tendency, such as Hebei, Mongolian, Anhui, Hainan, Yunnan, Gansu, and Qinghai.

Gao and Shi: The determinants of entrepreneurial activity in China

61

There are five quiet regions: Heilongjiang, Guangxi, Guizhou, Tibet and Shaanxi. The entrepreneurial activity in these areas lies at the lowest level, and their entrepreneurial activity rates have very small changes. Especially in Shaanxi, its entrepreneurial activity rate has dropped continuously since 2000. In brief, there are significant differences of entrepreneurial activity in different areas of China. Obviously, the activity in eastern coastal region is higher than in the other regions, especially in Shanghai, Beijing, and Tianjin, whose entrepreneurial activity level is higher and keeps increasing steadily and quickly. In the middle regions and the northeastern regions, the entrepreneurial activity level is lower than in the eastern coastal regions. Unfortunately, in the western and north-west region, the entrepreneurial activity level is the lowest and has shown almost no changes for many years.

3

Determinants analysis

What are determinants of the regional differences? And how do these determinants influence the regional entrepreneurial activity? These are two important questions to be answered in this paper. Combining existing researches with China's actual situation, the paper splits up the determinants of regional entrepreneurial activity into seven aspects: market demand, unemployment, industrial structure, entrepreneurial culture, technology innovation, availability of financing, and human capital. Next, we analyze how these determinants affect the regional entrepreneurial activity. Market demand Market demand, which refers to the demand ability of products or services, is one of the key factors influencing regional entrepreneurial activity (Sutaria and Hicks, 2004; Armington and Acs, 2002). Generally speaking, entrepreneurs are willing to start new businesses in bigger market demand regions. We have introduced two indicators representing the local market demand level: population density and growth of inhabitant’s consumption. Fritsch and Mueller (2005), and Bartik (2001) reported that population density produced strongly positive effects on regional entrepreneurial activity. In the region with higher population density, more potential consumers can be provided for new businesses. And it is easier for entrepreneurs to see successful hopes for start-ups. On the other hand, the growth of inhabitants’ consumption also has a positive effect on the regional entrepreneurial activity (Lee, Florida and Acs, 2004; Armington and Acs, 2002). As the inhabitants’ consumption level grows, purchasing power in such regions becomes stronger and stronger. And the regions with strong purchasing power can always attract many entrepreneurs to start new businesses. China is a country whose market demand level is uneven in different regions, such as the eastern coastal regions whose level differs from that of western inner regions. Since the reform and open policy was implemented the whole market demand level in China has been increasing year by year. No doubt that the increasing market demand can produce a significant effect on the regional entrepreneurial activity. Two hypotheses are given: H1: a region’s population density is positively related to its rate of entrepreneurial activity

62

1 Policy and innovation

H2: a region’s consumption growth is positively related to its rate of entrepreneurial activity Unemployment The relationship between unemployment and regional entrepreneurial activity is not unambiguous. Existing studies prove that unemployment produces both a push and pull effect on the regional entrepreneurial activity. When a person loses his/her job and fails to find another one, he/she may well decide to create a new employment opportunity by starting a new business. Consequently, unemployment incites self-employment, and regions with high unemployment have high rates of entrepreneurial activity (Storey, 1991; Robson, 1994; Fritsch and Mueller, 2005). On the other hand, the pull effect hypothesis argues that a low local level of unemployment has a positive effect on entrepreneurial activity, conversely, high unemployment prevents starting new businesses (Ritsilä andTervo, 2002; Armington and Acs, 2002; Sutaria and Hicks, 2004). A low regional unemployment level indicates a high level of local demand as well as regional competitiveness and growth. In addition, long and frequent existence of high unemployment also tends to lead to deterioration in labor force activity, and further on, to a weakening in the quality of human capital/labor force. It can be assumed that a high unemployment region is not attractive for individuals to start new businesses. As a result, it is difficult to judge if unemployment may have the effect on regional entrepreneurial activity. When the push effect is significantly stronger than the pull effect, it expresses positive affecting results (Fritsch and Mueller, 2005; Ritsilä and Tervo, 2002); on the contrary, when the pull effect is significantly stronger than the push effect, it expresses a negative affecting result; when neither of them is not significantly stronger than the other, it doesn’t express a positive or negative affecting result (Armington and Acs,2002;Sutaria and Hicks,2004). In China, however, the pull effect of unemployment may be weakened by a quick economic development and increasing market demand. Hence, unemployment may express a positive affecting result on regional entrepreneurial activity in China. The third hypothesis is given: H3: a region’s unemployment rate is positively related to its rate of entrepreneurial activity Industrial structure With economic development and technology improvement, the industrial structure may be changed in three ways (Armington and Acs, 2002): (1) the shift from manufacturing employment to services; (2) a reduction in both firm and plant size; and (3) a shift to higher levels of technology. Accordingly, these industrial structure changes may affect entrepreneurial activity in two ways: (1) a change in market structure and the occurrence of many opportunities of starting new businesses; (2) the entrance of many entrepreneurs into the service industry to start new businesses. Dean (1993), Sutaria and Hicks (2004), Fritsch and Mueller (2005) proved that a region’s service industry employment rate is positively related to its rate of entrepreneurial activity. In the past ten years, the service industry in China has made a quick development, and more and more people have entered into it. That may bring many opportunities of starting new businesses for entrepreneurs. The fourth hypothesis is given:

Gao and Shi: The determinants of entrepreneurial activity in China

63

H4: a region’s service industry employment rate is positively related to its rate of entrepreneurial activity Entrepreneurship culture Entrepreneurship culture is defined as a social context where entrepreneurial behavior is encouraged (Johannisson, 1984). Culture accumulates over a long period and influences people’s behavior. Entrepreneurship culture, similarly, influences the behaviors of starting new businesses. In a region with lots of private enterprises and with a frequent occurrence of new businesses, people are easy to be encouraged to create a cause by starting a new business. Even some of them may view becoming an entrepreneur as their highest life pursuing. Consequently, entrepreneurship culture positively influenced regional entrepreneurial activity (Armington and Acs ,2002; Davidsson and Wiklund ,1997; and Reynolds and Bygrave, 2004). In China, entrepreneurship culture is related to regional entrepreneurial activity. In some active regions, a special entrepreneurship culture may gradually be accumulated by entrepreneurial activity. And this culture, then, promotes a regional entrepreneurial activity’s further development. For example, Zhongguancun’s technology new business cluster and Wenzhou’s private enterprises clusters are all closely related to their regional special entrepreneurship culture. In addition, the government in China invests more and more money to improve entrepreneurship education. The fifth hypothesis is given: H5: a region’s entrepreneurship culture is positively related to its rate of entrepreneurial activity Technology innovation Shumpeter (1934) advanced that a new technology may cause a market’s disequilibrium. Once a market’s disequilibrium exists, there may be opportunities of starting new businesses. Start-ups based on new technology frequently take place in regions with strong technology innovation ability. Fritsch and Mueller (2005), Shane (2001), and Dean, Meyer and DeCastro (1993) proved that technology innovation is positively related to a region’s entrepreneurial activity. The Chinese government started lots of important science and technology plans, in order to support technology innovation, such as the “Xinghuo Plan” and the “863 Plan”. These plans are very useful to produce many new technological achievements. The sixth hypothesis is given: H6: a region’s technology innovation is positively related to its rate of entrepreneurial activity Availability of financing Capital is one of the key factors of determining new business success. Paul, Sherrie and Phili (1999) found that the investment at the beginning of a new business is closely related to the enterprise's subsequent performance. However, the question is how to raise enough start-up capital? Basu and Parker (2001) thought financing was a key determinant of successful start-

64

1 Policy and innovation

ups, and the largest single source of funds was self-finance, provided by the entrepreneurs’ own savings or assets. Bates.T. (1997) thought the source of start-up capital is mainly the entrepreneur’s owned capital (savings and an entrepreneur’s other capital). The GEM China Report also found that informal investment (entrepreneur’s owned capital, loans from family members and friends, and other private capital) was the most important source of start-up capital. In brief, entrepreneurs’ self-wealth, and family and friends’ financing backing are the most important start-ups capital sources for entrepreneurs to start new businesses. However, as there is no concrete data in China to explain the proportion of private wealth in start-up capital, entrepreneurs must depend more on the entrepreneur’s owned capital or family members’ and friends’ wealth, and it is difficult for Chinese entrepreneurs to finance through external capital, because the financing channels to the small and medium-sized enterprises in China fall far behind those of developed countries. Sutaria and Hicks (2,004) found that the bank deposit level is positively related to a region’s entrepreneurial activity. Hence, the seventh hypothesis is given: H7: a region’s family saving level is positively related to its rate of entrepreneurial activity Human capital Generally, the human capital level is measured by the level of a labor force’s education. The higher the education of labor forces in a region is, the better their working quality will be. Education brings people working skills and good reasoning. Bartik (2001), Arminton and Acs (2002) proved that a region’s level of education, resulting in better skilled labor forces, on average is positively related to the rate of entrepreneurial activity. Of course, labor costs for labor forces with a better education are higher than in other regions. The increased value created by high quality employees outweighs the increased labor costs they cause. The eighth hypothesis is given: H8: a region’s rate of labor forces with higher education is positively related to the rate of entrepreneurial activity

4

Empirical results

4.1 Data The dependent variable of regression models in the paper is the entrepreneurial activity rate (CPEA). Then the explanatory (independent) variables include those given below. Population density is the regional population divided by the regional area. The data of the population and the regional areas comes from the "Chinese Statistics Yearbook" (19972004). The rate of consumption growth is the rate of an increase of the consumption level in the region (calculating the change of the consumption level within two years, and then standardizing the change with respect to the figure of the consumption level in the first year). The data of the consumption level comes from the "Chinese Statistics Yearbook" (19972004).

Gao and Shi: The determinants of entrepreneurial activity in China

65

The unemployment rate is the number of unemployed divided by the total number of labor force. The data of the unemployment rate comes directly from the "Chinese Labor Statistics Yearbook" (1997-2004). The industrial structure is the proportion of service industry employment in all industries (standardizing the figure of employment in the service industry with respect to the number of employment in all industries). The data of employment comes from the "Chinese Statistics Yearbook" (1997-2004). The availability of financing is measured by the average deposits of a family in the region (standardizing the deposits of inhabitants saved in all kinds of financing institutions with respect to the population in the region). The data of thte regional inhabitants’ deposits comes from the "Chinese Finance Yearbook" (1997-2004) and the data of regional population comes from the "Chinese Statistics Yearbook" (1997-2004). The entrepreneurship culture is a concept which is very difficult to measure. The paper uses the growth rate of the proportion of employment in private enterprises (calculating the proportion of employment of private enterprises each year, and then standardizing the changes withiin two years with respect to the proportion of the employment of private enterprises in the first year). The data of employment in private enterprises comes from the "Chinese Statistics Yearbook" (1997-2004). Technology innovation is measured by the regional R&D investment divided by the GDP in the region. The data of the R&D investment and the GDP comes from the "Chinese Statistics Yearbook" (1997-2004). 4.2 Method The paper uses the pooled cross-section data model and the panel-data model. The two models both have their advantages and disadvantages. The pooled cross-section data is pooled by cross-section data which is useful to expand data capacity when the sample data is not enough. The panel-data is combined with time series data and cross-section data. As the sample data is not complete, the paper chooses the Random effect panel-data model. For example, Beijing is not contained in the sample because its data of some explanatory variables has errors. In addition, all data of the explanatory variables lags behind the data of the dependent variable for one year. The correlation coefficients between each variable are listed in table 1. We can see from the table that CPEA is positively related to all the explanatory variables, and has big correlation coefficients with most of them (with the exception of consumption growth and entrepreneurship culture). These results are consistent with the hypotheses above. In order to detect multicollinearity, we use the Variance Inflation Factor (VIF) test. For the correlation coefficients of human capital and industry structuring, and human capital and population density, the test result shows that multicollinearity does not exist significantly. We confirm that multicollinearity is not a significant affecting factor in the regression models below. Of course, the panel-data model has the advantage of weakening the effect of multicollinearity. We can not neglect heteroscedasticity which is a common affecting factor in the crosssection data regression model. The paper first uses the Ordinary Least Squares method (OLS) to estimate the model, and then detects the regression results with the White Het-

66

1 Policy and innovation

eroskedasticity Test. The final result shows that there is a significant heteroscedasticity in the model. Therefore, the paper uses the Weighted Least Squares method (WLS) to correct heteroscedasticity. Tab 1: Correlation coefficients between variables

4.3 Results Regression results of three models are given in table 2. We can see that all predictors associated with statistically significant effects in the panel-data model are also significant in the WLS model and OLS model. While the entrepreneurship culture variable is statistically significant (0.05 level) in the results of the WLS estimation, it does not reach significance in the results of the panel-data estimation, an outcome reflecting the fundamental technical differences between the two estimation techniques. The main object in comparing the results of these alternative models is to see if the estimation size, sign and significance are comparable. Two predictors’ results in table 2 are not consistent with the hypotheses above. The coefficients for the rate of consumption growth are negative but not significant in the three models, indicating that the rate of consumption growth fails to be a significant influence. The result is surprising and not consistent with hypothesis above. As noted previously, prior empirical studies (Lee, Florida and Acs, 2004; Armington and Acs, 2002) found a positive and significant relationship, but here we did not find evidences with respect to a positive effect of the rate of consumption growth on regional entrepreneurship activity. The coefficients for technology innovation are negative in the WLS model and positive in the panel-data model, but both of them are not significant. The results are also surprising, indicating that technological innovation does not produce a significantly positive effect on the Chinese regional entrepreneurship activity. There may be two reasons: one is that entrepreneurship activity based on technology innovation is still not active in China; the other is that most of the private enterprises in China are not high-technological ones and technological innovation has not become a driving force for start-ups. No matter what is the real reason, the phenomenon that technological innovation is not the driving force of private enterprises’ start-ups is abnormal. The coefficients for population density are positive and significant in all three models. The results indicate that regions with a predominately larger population density have a higher entrepreneurial activity rate than regions with a smaller population density. This supports the

Gao and Shi: The determinants of entrepreneurial activity in China

67

hypothesis that a region’s population density is positively related to its rate of entrepreneurial activity. The result is consistent with the findings of Fritsch and Mueller (2005), and Bartik (2001). The coefficients for unemployment are positive and significant in all three models. The result is consistent with the findings of Storey (1991), Robson (1994), and Fritsch and Mueller (2005), indicating that in China the push effect of unemployment is dominantly stronger than its pull effect on regional entrepreneurial activity. And the result also indicates that regions with higher unemployment have a higher entrepreneurial activity rate, which is consistent with the hypothesis above. The coefficients for the industrial structure are positive and significant in the WLS model and the panel-data model. This is consistent with the findings of Dean (1993), Sutaria and Hicks (2004), and Fritsch and Mueller (2005), and consistent with the hypothesis above. The result indicates that as the development of the service industry advances, more and more people are employed in it, and more and more entrepreneurs come into the industry for startups. Hence, the larger the employment rate in the service industry is the higher is the entrepreneurial activity rate. Tab. 2: Regression results

68

1 Policy and innovation

The coefficients for financing availability are positive and significant in all three models, indicating high a availability of financing promotes regional entrepreneurial activity. When the regions’ financing level is higher, it is easier for the entrepreneurs to loan from their family members and friends and easier to carry out financing through private capital. The result is consistent with the hypothesis above. The coefficients for entrepreneurial culture are positive in all three models, and significant in the WLS model but not in the panel-data model. The result indicates that entrepreneurship culture has a positive effect on regional entrepreneurial activity in China. But maybe the measure of entrepreneurship culture is not appropriate which causes an insignificant result in the panel-data model. The coefficients for human capital are positive and significant in all three models, indicating that the average labors’ education attachment has positive effects on regional entrepreneurial activity. Regions with more higher education labors on average attract more entrepreneurs to start new businesses.

5

Conclusion

Regional entrepreneurial activities in China are significantly different. Eastern coastal regions, especially Beijing and Tianjin Region, Yangtze River Delta Region, Zhu Jiang Delta Region, have a higher entrepreneurial activity rate than other regions. Using the econometric model we find that the differences for entrepreneurial activity are explained by the regional market demand, industrial structure, unemployment rate, availability of financing, entrepreneurial culture and human capital. Technological innovation and the rate of consumption growth have no significant effects on the entrepreneurship in China.

References Keeble, D. 1997. Small Firms, Innovation and Regional Development in Britain in the 1990s. Regional Studies Vol. 31.3, pp. 281- 293. Ritsilä, J. & Tervo, H. 2002. Effects of Unemployment on New Firm Formation: MicroLevel Panel Data Evidence from Finland. Small Business Economics 19: 31–40. Sutaria, V. & Hicks，D.A. 2004. New firm formation: Dynamics and determinants. The Annals of Regional Science 241-262. Davidsson, P. & Wiklund, J. 1997. Values, beliefs and regional differences in new firm formation rates. Journal of Economic Psychology l8 179-199. Bartik, T.J. 2001. Small Business Start-Ups in the United States: Estimates of the Effects of Characteristics of States. Armington, C. & Acs, Z.J. 2002. The Determinants of Regional Difference in New Firm Formation. Regional Studies Vol. 36.1. pp. 33–45.

Gao and Shi: The determinants of entrepreneurial activity in China

69

Shane, S. 2001. Technology Regimes and New Firm Formation. Management Science Vol. 47, No.9 pp. 1173–1190. Dean, T.J., Meyer, G.D. & DeCastro, J. 1993. Determinants of New-Firm Formations in Manufacturing Industries: Industry Dynamics, Entry Barriers, and Organizational Inertia. Entrepreneurship Theory and Practice pp.49-60. Robson, M.T. 1994. The Determinants of New Firm Formation in UK Construction: Evidence from Quarterly VAT Registrations. Journal of the Economics of Business Vol. 1, No. 2, 1994. Audretsch, D.B. & Fritsch, M. 1999. The Industry Component of Regional New Firm Formation Processes. Review of Industrial Organization 15: 239–252. Fritsch, M. & Mueller, P. 2005. The Persistence of Regional New Business FormationActivity over Time-Assessing the Potential of Policy Promotion Programs. Working paper. Storey D. J. 1991. The Birth of New Firms- Does Unemployment Matter? A Review of Evidence. Small Business Economics 3: 167–178. Reynolds, P.D., Bygrave, W.D. & Autio, E. 2004. Global Entrepreneurship Monitor 2003 Sam Youl Lee, Richard Florida and Zoltan J. Acs, Creativity and Entrepreneurship: A Regional Analysis of New Firm Formation. Regional Studies Vol. 38.8. pp. 879–891. Bates. T. 1997. Financing small business creation: the ease of Chinese and Korean immigrant entrepreneurs. Journal of Business Venturing Vol. 12. pp. 109 24. Paul, H., Sherrie, L.M. & Philip, B. 1999. Small business finance in two Chicago minority neighborhoods Economic Perspectives Vol. 23. Jiangyanfu, Gaojian, Chenyuan & Qiuqiong. 2004. Global Entrepreneurship Monitor China & Globe 2003. Tsinghua University Publisher. Jiangyanfu, Gaojian, Chenyuan & Qiuqiong. 2003. Global Entrepreneurship Monitor China & Globe 2002 Tsinghua University Publisher.

70

1 Policy and innovation

Authors Gao Jian is currently Professor and Head of Department of Innovation and Entrepreneurship, Executive Associate Director of Research Center for Technological Innovation and Associate Director of National Entrepreneurship Research Center in Tsinghua University. He got his PhD from Tsinghua University and studied at Sloan School in MIT and GSB in Stanford University as a visiting scholar. His paper on innovation and entrepreneurship was published in Research Policy and Asia Pacific Business Review. Contact: [email protected], Phone: +86 (10) 62789932, Fax: 62789757 Shi Shude, Master student of 2004 class Department of Innovation and Entrepreneurship School of Economics and Management, Tsinghua University Beijing Beijing 100084, P.R. China www.nerc.org.cn

2

Corporate management of innovations

Hans Koller: Intercultural technology intelligence – a process and communication oriented approach 1

Introduction

The business world is characterized by an acceleration of scientific progress, a shortening of product life cycles, a diffusion of technologies between previously independent branches, and a merging of available technologies into new solutions (Carlson 2004: 51). Globalized and intensified competition forces companies in high-wage countries in particular to keep abreast of technological and competitive changes. In order to seize opportunities and counter risks in an accelerated environment such as this, management must be able to quickly recognize relevant changes in the business world and incorporate them into the process of strategic planning and decision making. Over the last decades various approaches have been presented, referred to as technology forecasting, technology foresight, and technology intelligence (Lang 1998). However, past and present studies in this field have shown that companies have difficulties mastering this challenge. One of the main problems is the amount of information on global developments that is available every day with the help of modern information and communication technology. What management needs today are less methods to collect information but practical mechanisms to make the flood of information manageable. It must be brought in a position to collect, select and process information without paralyzing the company. In this paper we present an approach which is able to handle this information overload. The concept was developed together with company partners. The approach is based on the establishment of a hierarchy of “radar groups” and the creation of a “strategy room”. The key idea

72

2 Corporate management of innovations

is to exploit existing organizational search routines for foresight to reduce the complexity of information gathering and assure broad and qualified perception and evaluation of relevant changes. The main benefit of such a process is the constant adjustement that helps the organization to face environmental complexity stepwise. The approach matches findings of the late strategy process research (Burgelman 2001) and the capability-based view of the firm (Chakravarthy/Doz 1992). Further, it also addresses intercultural aspects of business and especially technology intelligence.

2

Technology intelligence – a short overview

Neither in scholarly discourse nor in corporate practice there are standardized terms defining how to identify and how to handle information about changes affecting the competitive position of a company, especially with respect to their technology resource. At the beginning of the article I therefore differentiate between the terms in use and establish the term that is in accordance with my approach. Depending on their perspective and the depth of information collection, strategic management community speaks about environmental scanning, environmental analysis, competitive analysis, or business intelligence (Lang 1998: 20). Concentrating on technological and technologically relevant aspects, one can additionally identify a broad spectrum of terms. For a long time, the most frequently used term was "technology forecasting" (Bright 1970, Martino 1983, Millet/Honton 1991). But often "forecasting" means to identify developments in areas in which they are expected and which can therefore be monitored. Today a number of scholars reject the narrow constraints of such a pure analysis of variance and introduce distinctive approaches named "technology foresight" (Stout 1995) or "technology intelligence" (Ashton/Stacy 1995). In support of the view that the terms foresight and intelligence are nearly synonymous, I will use "technology intelligence" and define it with reference to Ashton/Klavans as "collecting, analyzing and communicating the best available information on S&T [science and technology, the authors] developments and trends occurring outside one’s own company." (Ashton/Klavans 1997: 6) One of the most important advances in technology intelligence is that it is not limited to the observation of technological changes. It also incorporates changes in society, politics and the law that in turn have an effect on technological developments (den Hond/Groenewegen 1996: 43, Koller 2002: 344, Salo/Cuhls 2003: 79). In addition, not only external changes but also manifest or latent corporate weaknesses, relevant for the way in which a company anticipates changes, are incorporated (Baum/Coenenberg/Günther 1999: 313, Ansoff 1980: 133). In the past various methodological approaches for technology intelligence have been suggested, often combining old and new models of forecasting and foresight. Literature on this subject frequently divides these into the following three levels. For the reason that this categorization has been discussed in depth elsewhere (Ansoff 1980, Baum et al. 1999, Reger 2001, Koller 2002, Lichtenthaler 2003) it will only be outlined in figure 1.

Koller: Intercultural technology intelligence

73

level

focus

examples

Corporate accounting

comparison between planned and anticipated development of accounting variables

- sales percentage - proportion of contribution margin of new products

Monitoring and Forecasting

identification and monitoring of specific early-warning indicators

-

FAQ / complaint management R&D-expenditure R&D-personal Bibliometry patent analysis reports of maintenance& service

Foresight and Intelligence

scanning all relevant areas to find weak signals indicating potential discontinuities

-

Integration of „Lead Supplier“ Integration of „ Lead Customer“ Integration of external experts Szenario analysis etc.

Figure 1: Level of information gathering within technology intelligence Source: Baum et al. 1999, Reger 2001, Koller 2002, Lichtenthaler 2003.

Today an adequate technology intelligence should see these three levels as consecutive. Whereas models in level one are based on ratio systems, like planning extrapolations and analyses of variance, and therefore primarily focus on the past, level two models make it possible to predict hidden chances and risks via monitoring early warning indicators. But even though an analysis of such indicators is useful, only those areas are monitored that have already been identified as relevant. In todays dynamic environments, where unexpected changes frequently occur in areas outside a firm’s focus, a broader foresight is needed. Hence the emphasis of models in level three is on scanning all potentially relevant environmental areas and identifying even "weak signals" (see Ansoff 1975). Tools for supporting companies in carrying out this task are for example cross-impact and vulnerability analyses (Geschka 1995). However, without a doubt about the significance of such an holistic technology intelligence, current studies are discovering that the implementation of scanning entails difficulties for companies (Reger 2001: 536, see also Martin 1995, Eerola et al. 2002). Even companies with large staff units find it hard to carry out scanning that observes all potentially relevant areas. These difficulties seem to be caused by the deficient available methodology. Attempts in the 1980s to change this situation through the use of computerized management information systems or decision-support systems usually resulted in "data cemeteries". What is needed today is an integrated approach of scanning that enables companies to identify weak signals efficiently without paralyzing the company in it's daily business. Subsequently we will introduce a concept which was designed together with corporate partners in an industry project with the purpose to develop and implement an efficient and manageable technology intelligence system.

74

3

2 Corporate management of innovations

A process and communication oriented approach of technology intelligence

The project described was situated in the field of medical engineering and therefore in a turbulent competitive environment characterized by technological change. The impetus for the project was a complaint by the company management. They felt that problems of daily business left too little time to concentrate on the strategic orientation of the company. The initial situation we found seems to be typical for the situation in many companies, particularly in small and medium enterprises. Diverse monitoring efforts have been implemented for different aspects of the environment, but the information too often remained in the divisions that collected the data or was made available in various electronic "archives" which indeed have been used only by a small group of experienced staff members. Aside there were no systematic scanning activities. Fortunately the companies personnel policy was very sustainable and some important positions were continuously occupied by the same staff over long periods of time. For this reason it was possible to identify knowledgeable and informed staff members in all divisions of the company and in the company's environment. However, the lack of foresight activities was making it difficult for the managers to respond quickly to changes and new requirements. They needed a system that offered an overview of the companies competitive position and hence enables managers to recognize long-term developments and to evaluate responding strategies. The model introduced next does not claim to be universally valid. It is, however, a proposal that has been brought into the discussion and, on account of the fact that it has been evaluated in practice, should offer valuable suggestions for further analysis of technology intelligence. 3.1 Model The proposed solution is essentially based on the knowledge that is available within the individual units of the organization. It recommends that so called "radar groups" should be formed, as consciously compositions of primarly internal experts which are formally assigned to do technology intelligence in their area of expertise. They should meet at regular intervals in order to collect information about essential developments in a defined area of the environmental conditions, to discuss, to evaluate, and to transmit their results to the radar group at the higher level in the hierarchy. This way it will be possible to take the knowledge and the contacts of the identified experts into consideration and, due to the large number of radar groups, to determine, or "scan", a broad segment of the environmental conditions. In the present case, this included, for example, regular market research reports, the analysis of patent databases, competition analyses, structured reports about meetings of suppliers, structured reports about important visits to exhibitions and conferences or reports from other company units and country divisions. The discussion of these impressions leads to a qualified evaluation and to a filtering out so that only a selected segment of the changes under discussion will be passed on to the next higher level. On this account there will be a concentration on the changes that are considered

Koller: Intercultural technology intelligence

75

to be essential and thus complexity will be reduced. Such a filtering process naturally bears the risk that there will be a selection of signals to be transmitted, based upon individual concerns. However, this may be minimized as follows: Firstly, the selection is the result of discussions among several participants. Secondly, the groups are continuously subject to the pressure of having to recognize the signals early enough and to report them as important signals. Thirdly, at the next-higher level there is a comparison of the reported signals from each group. And finally, the alternative should also not be left out of sight: There can be no objective evaluation of such signals, due to the lack of information, and the evaluation made by the respective experts is at least better than an evaluation made by outsiders. Additionally, the danger of having on "blinders" can be reduced by including externals regularly and by systematically inviting external “spin-doctors” from changing parts of the environment. In particular the inclusion of externals will integrate international and intercultural aspects. For a comprehensive reflection of the disussion in the radar groups their results should be presented personally. Furthermore it is recommended that the groups are connected by linking pins, persons who are members of multiple groups/clusters in the company (see Lickert 1967, Conway 1997, Cross and Prusak 2002). The number of the involved hierarchy levels depends on the company's size and the heterogeneity of the observed areas. The members of the radar groups will mainly be the internal knowledgeable staff and individual persons who are characterized by their special contacts to parts of the environment. The latter can either be designated as "gatekeepers" (Allen 1966, Tailor 1975) or "boundary spanners" (Keller and Holland 1975, Williams 2002). Moreover, it may also be useful to include some external knowledgeable persons, i.e. experts from research institutes or "lead customers and lead suppliers" (von Hippel 1988). In the model the aggregated information is processed to the top of the company, and thus at the top of the information processing procedure in a "strategy room". At this place the essential aspects of the company's environment and the company's own positions in these environmental areas will be described and discussed – even if only in a rough and undoubtedly schematic form. This is an attempt to satisfy an essential requirement of the company's management, namely to develop a visible expression of the company's position in the competitive environment for medium and longer term. For this purpose, instruments of strategic planning will be applied, such as business sector analysis, portfolio analysis, road maps, the clustering of essential customers and suppliers, and so forth. Thus one can partially overcome the "lack of structure, guidance and 'visualization' " (Reger 2001: 537) in the area of technology intelligence that has been criticized elsewhere. Figure 2 is a schematic illustration of the organizational structure of such a strategy room as the end of the information processing procedure. It is obvious that the descriptions in this strategy room go beyond the topics of technology intelligence. Such "business intelligence" will inevitably lead to a discussion of the effects of changes in the company's environment on the company's competitive position.

process development

product line a

product line b

…

radar group level 2 linking pin

environment

technology

environment

market

ors technology etit p co m

su

pp l ie

superior rs value chain

value chain

Conclusions Issues

suppliers

radar group level 1

strategy room

collect evaluate filter & process information

2 Corporate management of innovations

visualize, combine & conclude

76

Figure 2: Hierarchy of radar groups and the strategy room Source: own illustration.

3.2 Discussion First of all, it should be conceded that a bottom-up organized information flow does not offer surprising new elements and is already recommended in many contexts, e.g. for general planning, budgeting, reorganization, etc. (e.g. Schwarzkopf et al. 1988, Smeds et al. 2003). However, in many companies – including the one under consideration – this kind of procedure is not considered to be self-evident. Instead, the strategic planning is often left to a small staff of planning experts or even to external consultants. But under the conditions of rapidly changing environments, the area to be observed is too widespread to provide an extensive scanning and moreover, planning staff lacks the specialist knowledge that is necessary to make correct assessments. For both reasons it appears to be indispensable to use the knowledgeable staff members as "radars". They should be entrusted to analyse and evaluate the received information with regard to the effects on the company. The inclusion of the staff in the technology intelligence is not merely an option but an indispensable necessity. For this task the institutionalization of information processing and information transfer is constitutional. Although discussions about important changes in general conditions took place before as well, the company management in the current case had, however, complained that it received this information either accidentally or not at all. The concept presented here found a remedy by institutionalizing regular processes of information collection, information processing in the radar groups, and information transfer to management. So management

Koller: Intercultural technology intelligence

77

regularly considers relevant changes in general conditions and competitive strategy. Company's strategy is thus the subject of repeated discussion. Besides, another important effect of such an institutionalization of information processing is the continuing learning that increases the flexibility of companies. Since members of radar groups are tasked with the systematic observation of relevant environmental areas and take part in discussions of changes considered by others to be important, individual members are made sensitive for such changes. The ability to perceive and interpret such changes and to understand how they relate to competitive strategy is greatly improved. Of particular importance is the feedback link in the information processing system. The analysis of information in the strategy room, the way this information is analyzed, and the involvement of personnel at various levels influences to a great degree the awareness of personnel and their sensitivity for weak signals. In this way we can considerably improve the ability of individual workers and of the company as a whole to respond to such changes. It is more than evident that communication and the institutionalization of communication processes plays a central role in the approach. This applies the internal assessment of impressions about changes in radars groups, as well as the forwarding of information between the radar groups of different hierarchical levels and finally the visualization of findings in the strategy room. The institutionalized communication process forces several company members, including the management itself, to look regularly into the competitive position of the company. The strategy room itself can become an important element in the communication system as it is not used solely to present information but rather is seen as a place remote from daily problems, where the long-term strategy should be discussed and creativity is rewarded. The communication orientation is also the aspect which points out the benefit of such an approach for business and technology intelligence in multinational companies. In particular small and medium enterprises which operate in international markets have difficulties in detecting and classifying weak signals from different regions. Specialized knowledge about regional and cultural distinctions is needed to recognize and evaluate these signals. And what makes it even worse: This specialized knowledge about regional and cultural distinctions is profoundly tacit. Even after reading books about a foreign culture the management in the home country is neither able to describe the potential sources of environmental changes in a foreign region nor is it able to interpret observed signals. The amounts of the tacit knowledge about the foreign culture necessary for carrying out these tasks are tremendous. Natives are needed to recognize and interpret weak signals and play a vital role for establishing contacts to personal information sources in this foreign region. They enable the organization to acquire and process the signals with cultural background. Local radar groups seem to be the only efficient solution for this culture sensitive information scanning. Findings from this local radar group should be presented personally by a member of the group to communicate them properly and to discuss possible impacts. Finally, the integration of people in this communication process not only helps to make them more sensitive for such issues but also improves intercultural understanding. This may help in overcomming intercultural barriers which many authors regard as one of the most serious challenges in multinational companies. (Broms/Gahmberg 1983, Hofstede 2001).

78

2 Corporate management of innovations

Thus foresight activities at local levels and an institutionalized information processing simplifies and assures a broad and qualified business intelligence in multinational companies which leads to an international corporate foresight culture.

4

Reflections from an intercultural perspective

There is no doubt about the challenge of technology intelligence in todays dynamic environments. This being difficult in national contexts, it is more complicated in multinational companies, mainly because of their cultural heterogeneity (Schmid 1996: 169). Not only All in all, multinational companies have a complex cultural structure with many interactions between national cultures, corporate culture (Schein 1997), other regional subcultures, industry cultures, profession subcultures or functional cultures. (see figure 3).

CC

national culture „a“

profession culture

national culture „b“

industry culture regional culture „a1“ regional culture „a2“

Figure 3: interacting subcultures and cultural parts in multinational firms Source: in accordance with Schneider/Barsoux 1997:47.

One of the most-cited authors in cultural research is Hofstede. He demonstrated that there are national and regional cultural groupings that affect the behaviour in organizations and identified five cultural dimensions through which cultures can be described (Hofstede 2001).1 Power distance, representing the acceptance of hierarchical differences (a high level of power distance reflects autocratic decision making), uncertainty avoidance (i.e. the degree to which formal and informal rules and regulation exist is a function of uncertainty avoidance), masculinity/femininity, describing the focus on financial and performance measurements or on cooperation and relationships, collectivism/individualism which is related to group behavior, and confusian dynamism which represents the degree of a longterm profitability orientation. These cultural differences, affecting verbal communication as well as behaviour- and evaluation-schemes, influence the information processing and assessment (Hanges et al 2000).

1

There are other classifications of culture, i.e. Trompenaars/Hampden-Turner’s (1997) and Schwartz’s (1994), but most of them can actually be incorporated into the dimensions explained above.

Koller: Intercultural technology intelligence

79

In relation to technology intelligence activities in multinational companies this means, that to recognize and evaluate discontinuties and weak signals on local markets requires knowledge about the national culture and cultural particularities in branches and professions. Consequently, information collection in multinational companies has to be decentralized. Yet, its results must be interpreted and evaluated on corporate level as well. But the information transmission then is vulnerable for misinterpretation because of the above discribed cultural differences. Information coming from one country may have a different meaning in the corporate manager’s culture. Consequently, two central requirements for technology intelligence in multinational companies arise: the coordination of decentralized information processing and the development of regional staff's competence to illustrate and to argue in favor of the relevance of the identified signals and the competence to evaluate it on corporate level. Both require cross-cultural communication and intercultural competence. In the literature of intercultural management there are two mechanisms for these requirements (Beneke 1995, Welge 1999, Hofstede 2006): Intercultural competence training and meta-cultural process competence. This mechanism enables employees to communicate, builds up language skills, and informs about communication rules and differences in task performance. Further, it allows them to understand behaviors and ways of thinking in other cultures and shows them how to express themselves in international contexts transparently and properly. Basics trainings should convey knowledge about the other culture and empathy to appreciate feelings and needs. Of particular importance is the awareness of the relativness of ones own culture and different interpretation opportunities in interaction. Development of communicative structures. To delegate accountability and assure information exchange a communicative structure is needed. Therefore, committees and panels could be arranged and an internal corporate network should be established. It important that the meetings take place frequently and are bordered internationally. This supports the process of intercultural learning and assures the exchange of experience. Institutionalized communication forums also initiate informal communication which functions as the basis for new contacts (Welge 1999: 14). We truly belief, that approach supports both of these aspects. As mentioned above, a hierachy of radar groups is an institutionalized communication process which is able to bring different management levels and competences together and assures communication over and across cultural boundaries. This is why it is particularly suitable for technology intelligence in multinational companies.

5

Conclusion

Both, the theoretical analysis of technology intelligence and the results of the practical project have identified some of the difficulties that are involved in handling technological and technology-related environmental changes. The process of scanning the environment is a difficult challenge for companies. The success of a technology and business intelligence does

80

2 Corporate management of innovations

not only depend on the identification of relevant information but also on the way this information is evaluated and passed on. In the project presented here, a hierarchy of radar groups and a process of collecting, evaluating and forwarding information, which leads to a strategy room, has been institutionalized to this purpose. In an additional reflection it has been shown that the proposed solution is particularly suitable for technology intelligence in multinational companies. In sum the argumentation presented here leads to the conclusion that it is necessary to include the bearers of knowledge in technology and business intelligence and in strategic planning as a whole, as this is the only effective way to cope with extrem dynamics and intercultural challenges of organizational foresight in multinational corporations. A bundling of these activities in a central planning staff seems insufficient or at least inefficient. In contrast, the procedure developed in the practical project appears to be an efficient method of structuring this process of collecting, filtering and forwarding information. But the model presented here is just the beginning of a new way of looking at organisational foresight, thereby unfolding new questions for scholarly research. The paper has pointed out that researchers and practioneers have to be aware of cultural differences and consequently further research should investigate if such an approach is useful for every culture context. Specially in cultures like the chinese one the question arises if foresight activities based on information processing radar groups are in accordance with its strong hierachical thinking. These questions should be addressed in further research on intercultural foresight activities.

References Allen, T.J. (1966): Managing the flow of scientific and technological information. Ph.D. Thesis. Cambridge, MA. Ansoff, H.I. (1975): Managing Strategic Surprise by Response to Weak Signals. California Management Review, Vol. 18 Issue 2, pp. 21-33. Ansoff, H.I. (1980): Strategic Issue Management. In: Strategic Management Journal, Vol. 1 Issue 2, pp. 131-148. Ashton, W.B./Klavans, R. (Eds.). (1997). Keeping Abreast of Science and Technology: Technical Intelligence for Business, Columbus, Ohio. Baum, H-G./Coenenberg A. G./Günther, T. (1999): Strategisches Controlling. 2nd Ed., Stuttgart. Beneke, J. (1995): Vom Import-Export-Modell zur regionale-komplementären Zusammenarbeit: Ein Paradigmenwechsel in der internationalen Unternehmenskommunikation. In: Bolten, Jürgen (Hrsg.): Cross Culture – Interkulturelles Handeln in der Wirtschaft, Berlin, 1995, pp. 59-77. Bright J.R. (1970): Evaluating signals of technological change. In: Harvard Business Review, Jan.70, pp.62-70.

Koller: Intercultural technology intelligence

81

Broms, H./Gahmberg, H. (1983): Communication to self in organizaitons and cultures. In: Administrative Science Quarterly, Vol. 28, pp. 482-495. Carlson, L.W. (2004): Using Technology Foresight to create Business Value. In: Research Technology Management, Vol. 47 Issue 5, pp. 51-60. Conway, S. (1997): Strategic Personal Links in Successful Innovation: Link-pins, Bridges, and Liaisons. In: Creativity and Innovation Management, Vol. 6, No. 4, pp. 226-233. Dahl, S (2004): Intercultural Research: The Current State of Knowledge, Middlesex University Business School Discussion Paper. Den Hond, F./Groenewegen, P. (1996): Environmental Technolgy Foresight: New Horizons for Technology Management. In: Technology Analysis & Strategic Management, Vol. 8, No. 1, pp. 33-46. Eerola, A./Jørgensen, B.H. (2002): Technology Foresight in the Nordic Countries. Risø National Laboratory, Roskilde, Denmark. Fisher, G. (1988): Mindsets: The Role of Culture and Perceptions in International Relations, Yarmouth ME: Intercultural Press. Gemünden, H.G./Walter, A. (1995): Der Beziehungspromotor – Schlüsselpersonen für internationale Innovationsprozesse. In: Zeitschrift für Betriebswirtschaft, Vol 65, pp. 971-986. Geschka, H. (1995): Methoden der Technologiefrühaufklärung und der Technologievorhersage. In: Zahn, E. (Ed.): Handbuch Technologiemanagement, Stuttgart, pp. 623-644. Hanges P./ Lord R./ Dickson M. (2000):An Information-processing Perspective on Leadership and Culture: A Case for Connectionist Architecture. IN: Applied Psychology An International Review, Volume 49, Number 1, January 2000, pp. 133-161(29). Hippel, E. v. (1988): The Source of Innovation. Oxford University Press, NY. Hofstede, G. (2001): Lokales Denken und globales Handeln – Interkulturelle Zusammenarbeit und globales Management. 2. ed., München. Hofstede, G. (2002): Culture's consequences : comparing values, behaviors, institutions and organizations across nations. 2.ed., Thousand Oaks, Calif. [u.a.] : Sage Publ. Keller, R.T./Holland W.E. (1975): Boundary-spanning roles in a research and development organization: an empirical investigation”, Academy of Management Journal, Vol. 18 Issue. 2, pp. 388-93. Koller, H. (2002): Technologiefrühaufklärung. In: Specht, Dieter/Möhrle, Martin G. (Eds.): Gabler Lexikon Technologiemanagement. Wiesbaden, pp. 343-351. Lange H.-C. (1998): Technology Intelligence: ihre Gestaltung in Abhängigkeit der Wettbewerbssituation. Zürich. Lichtenthaler, E. (2003): Third generation management of technology intelligence processes. In: R&D Management, Vol. 33 Issue 4, pp. 361-375.

82

2 Corporate management of innovations

Martin, B.R (1995): Foresight in Science and Technology. In: Technology Analysis & Strategic Management; Vol. 7 Issue 2, pp. 139-168. Martino, J.P. (1983): Technological Forecasting – an overview. In: Management Science, Vol. 26 Issue 1, pp. 28-33. Reger, G. (2001): Technology Foresight in Companies: From an Indicator to a Network and Process Perspective. In: Technology Analysis & Strategic Management, Vol. 13, No. 4, pp. 533-553. Salo, A./Cuhls, K. (2003): Technology Foresight – Past and Future. In: Journal of Forecasting Vol. 22 Issue 2/3, pp. 79-82. Schein, E. (1997): Organizational culture and leadership. 2.ed., San Francisco: Jossey-Bass. Schwartz S. (1994). Beyond individualism-collectivism: New cultural dimensions of values. In U. Kim, H. Triandis & G. Joon (Eds.) Individualism and collectivism, NewYork: Sage. Schwarzkopf, A. B./Tersine, R. J./Morris, J.S. (1988): Top-down versus bottom-up forecasting strategies. In: International Journal of Production Research, Vol. 26 Issue 11, pp. 18331843. Smeds, R./Haho, P./Alvesalo, J. (2003): Bottom-up or top-down? Evolutionary change management in NPD processes. In: International Journal of Technology Management, Vol. 26 Issue 8, pp. 887-902. Stoppford, J.M./Wells, L.T.: Developing an Organization for Multinational Business. In Stoppford, J.M., Wells, L.T. (Hrsg.): Managing the Multinational Enterprise, New York, 1972, pp. 9-29, 201-202. Stout, D. (1995): Technology Foresight – a View from the Front. In: Business Strategy Review, Vol. 6 Issue 4, pp. 1-16. Trompenaars, F. & Hampden-Turner, C. (1997) Riding the Waves of Culture: Understanding Diversity in Global Business. New York: McGraw-Hill. Williams, P. (2002): The Competent Boundary Spanner. In: Public Administration, Vol. 80 Issue 1, pp. 103-124.

Koller: Intercultural technology intelligence

Author Hans Koller, Prof. Dr. Chair for production and technology management Holstenhofweg 85 22043 Hamburg Tel.: +49 (40) 6541-2850 Fax : +49 (40) 6541-3767 [email protected] http://www.hsu-hh.de/itm

83

Dirk Holtbrügge and Jonas F. Puck: The change of foreign investors’ ownership modes in the PRC – theoretical framework and empirical results 1

Introduction

Over the last two decades, the stock of Foreign Direct Investment (FDI) in the People’s Republic of China (PRC) has increased from US$ 57 billion in 1982 to US$ 317 billion in 2005 (UNCTAD, 2001, 2006). For 2005, the FDI inflows exceed US$ 72 billion, making the PRC the largest FDI recipient globally (UNCTAD, 2006). The consistently high GDP growth rates in combination with a relatively good infrastructure, a high political stability, and a liberalization of the trade and investment regime following the WTO entry in 2001 can be seen as the main reasons for this increase in FDI. The increasing liberalization of government regulations has led to new options with regard to ownership modes being available to foreign investors in the PRC. While international joint ventures (IJVs) were the dominant type of ownership chosen by new investors until 1997, new investors after 1997 preferred to establish wholly foreign-owned enterprises (WFOEs) (Yan and Warner, 2002). The dominance of IJVs in early years of FDI in the PRC can be attributed to the existence of regulations that prevented the establishment of WFOEs in most sectors. Although about 50% of the 510,000 FDI projects existing in the PRC in 2004 were organized as IJVs and 40% as WFOEs, only 27% of new FDI projects in 2004 were IJVs while over 70% were WFOEs (MOFCOM, 2006). Given this trend away from IJVs towards WFOEs it may be asked in how far foreign firms involved in IJVs will change their IJV into a WFOE and what the factors are that increase the likelihood of such a conversion (if legally possible). Examples of firms that have changed the ownership mode of subsidiaries in the PRC include large MNCs like Mitsubishi, Siemens, or Nestlé as well as many small and medium sized firms. We suggest that given these evident changes in foreign firms’ preferences with regard to entry mode when entering the PRC, China provides a good setting for analyzing the factors that might explain the conversion of IJVs into WFOEs. While agreat amount of research has dealt with entry mode choice in general (e.g., Agarwal and Ramaswami, 1992; Anderson and Gatignon, 1986; Tihany, Griffith and Russel, 2005; Zhao, Luo and Suh, 2004) and foreign firms entering the PRC in particular (e.g. Chen and Hu, 2002; Wei, Liu and Liu, 2005; Zhao and Zhu, 1998), there has been comparatively few interest in the change of an IJV into a WFOE and in factors that lead firms to undertake such a conversion of their ownership strategy in foreign markets. One stream of research related to this issue is the work on the instability of IJVs (e.g., Gomes-Casseres, 1987; Inkpen and Beamish, 1997; Kogut, 1989; Reuer and Tong, 2005). However, instability includes a wide

Holtbrügge and Puck: The change of foreign investors’ ownership modes in the PRC

85

range of reasons, such as the conflicts between the partners or folding of the IJV and the buyout of the local partner firm(s) is only one of several options that is not analyzed in more detail. A second stream of research that is of greater relevance to the subject of this article are studies which are based on the real-options approach (e.g. Buckley, Casson and Gulamhussen, 2002; Kogut, 1991). They explicitly focus on the idea that firms may use IJVs as they provide them with the option to convert it into a WFOE at a later stage. The focus of this research on IJVs is, however, geared towards explaining why firms choose an IJV rather than explaining why and under which circumstances firms involved in an IJV exert this option and decide to convert the IJV into a WFOE. Overall, compared to investigations into firms’ initial choice of entry modes, there has thus been little conceptual work on post-entry changes of ownership forms. However, given the continuing liberalization of the investment regime in China as well as in many other emerging markets, such as India or Russia, we discern a need for conceptual as well as empirical research in this field. This chapter contributes to filling this gap by proposing a number of relevant factors and testing them empirically against a sample of foreign firms operating in the PRC.

2

Research design and methodology

In order to analyze the factors influencing the likelihood of converting an IJV into a WFOE we carried out a questionnaire survey among foreign companies in the PRC in the beginning of 2006. We decided to focus on foreign firms headquartered in the United States, Japan, and Europe while deliberately excluding investors from Hong Kong, Taiwan, Macao, Singapore, Malaysia as well as offshore financial centers like the Virgin Islands or Western Samoa. These source countries were excluded in order to eliminate round-tripping investments (Xiao, 2004) and investments from overseas Chinese companies, which due to their cultural affinity are not readily comparable with investments from other foreign investors. We contacted the chambers of foreign trade of Japan, the United States, the UK, Germany and the European Union and were able to receive a large number of contact addresses. In addition, we analyzed company homepages and articles in newspapers (e.g., China Business Review). Overall, we compiled contact details of 1,979 IJVs or wholly-owned subsidiaries of foreign companies in the PRC. The original German language questionnaire was translated into four languages (English, Japanese, French, Spanish) by three professional translators using the translate/re-translate method to ensure the equivalency of questions (Brislin, 1970). The questionnaire was distributed via email accompanied by a cover letter explaining the aim of the study. After two weeks, we sent out a reminder to those companies that did not answer by that time. Finally, 195 usable questionnaires were received (response rate of 9.9 percent). This relatively low response rate may in part be explained with the questionnaire fatigue reported by many managers of subsidiaries of foreign firms in the PRC (one respondent who declined participating in the study, explained that he would receive more than 7 questionnaires per week). Given our interest in the likelihood that a subsidiary that had been created as IJVs has subsequently been converted into a WFOE we eliminated from our sample responses from managers of subsidiaries that had been set-up as a WFOE from the start. We also eliminated re-

86

2 Corporate management of innovations

sponses from subsidiaries in industries in which ownership restrictions still prevent foreign firms to convert their IJV into WFOEs. This filtering process left us with responses from 94 companies that could be used to empirically test our hypotheses. 67 of these companies were still IJVs while 27 of them had changed their ownership mode into a WFOE. The subsidiaries in our sample were established by companies headquartered in 13 countries, with the United States (23) accounting for the most of these companies, followed by Germany (20), Japan (10), the UK (10) and Italy (10). On average, the subsidiaries employed 689 people. The subsidiaries were mainly operating in the chemicals, mechanical engineering and computer, electronic and automotive sectors.

3

Findings

3.1 Transaction cost theory and change of ownership mode Transaction cost theory argues that firms will select an ownership mode that minimizes transaction costs (Jung, 2004). The basic idea of transaction cost theory is that a firm will internalize activities that it is able to perform at lower cost and will rely on the market for activities in which other firms have a comparative advantage (Klein, Frazier and Roth 1990). The relative advantage of internalization or a market-based solution results from a comparison of the coordination costs associated with the internalization and the transaction costs arising from the search for, negotiation with and control of a market partner (Brouthers, 2002; Erramilli and Rao, 1993). In general, Williamson (1975) suggests, that the level of transaction costs depends particularly on two factors: asset specificity and uncertainty. Asset specificity depends on the extent of losses associated with using the investment necessary for carrying out transactions in any other way. The higher the losses arising from the deployment in a new project, the higher is ceteris paribus the asset specificity. With regard to the uncertainty of a transaction, one can distinguish between internal, e.g. behavior-based or transactional, and external, e.g. context-related, insecurity. While internal uncertainty reflects the potential of transaction partners acting opportunistically (Yiu and Makino, 2002), external uncertainty reflects “the volatility (unpredictability) of the firm’s environment“ (Anderson and Gatignon, 1986, 14). From the perspective of transaction cost economics, IJVs represent a hybrid form of transaction, which lies between using the market and fully internalizing transactions in the form of a WFOE (Holtbrügge, 2004). Hennart (1988) argues, that IJVs are preferred over whollyowned subsidiaries when (1) markets for the intermediate goods held by each party are failing, and (2) acquiring or replicating the assets required to produce those goods is more expensive than obtaining a right to their use through a JV agreement. Examples for such intermediate goods are, for example, industry- or country-specific knowledge, market knowledge, access to distribution channels and resources or parts and components (Hennart, 1991; Makino and Neupert, 2000). With regard to the PRC, local knowledge can be seen as one of the most important intermediate goods that foreign firms are trying to achieve when cooperating with a local partner. Existing studies show that local knowledge is one of the major reasons for the establishment of IJVs in China (Beamish, 1989; Zhao and Zhu, 1998). Local Knowl-

Holtbrügge and Puck: The change of foreign investors’ ownership modes in the PRC

87

edge refers to the combination of knowledge of the market, the regulatory framework, the general economic conditions, the political situation and the business culture (Beamish and Inkpen, 1995; Inkpen and Beamish, 1997). Anecdotal evidence as well as empirical studies show that knowledge about the Chinese culture in general and the complex mechanisms of the Chinese business life in particular is critical for the success of an investment (Beamish and Jiang, 2002). Given this importance of cultural issues, it can be regarded as a central intermediate good required by foreign firms entering the PRC. With growing experience in the PRC, however, the need to cooperate decreases. If the foreign partner is able to absorb local knowledge from the local IJV partner over time, the need to have a local partner to provide this knowledge is greatly diminished. Thus, as Killing (1982, 127) states, the “learning process naturally weakens the desire of companies to keep their joint ventures together”. Consequently, the increase of local knowledge by the foreign IJV partner enhances the likelihood of a conversion of an IJV into a WFOE. As explained above, a second central element of transaction cost theory is the relevance of asset specificity. Thus, asset specificity is a key variable in many transaction cost theorybased studies on entry-mode choice (e.g., Anderson and Gatignon 1986; Brouthers, 2002; Chen and Hu, 2002). High asset specificity exists when firms own special technological or management knowledge (Brouthers, 2002). Bringing such knowledge into an IJV may increase the tendency of the partner to behave opportunistically. Thus, empirical studies have found a highly positive influence of asset specifity on foreign firms’ preference for WFOEs (Anderson and Gatignon 1986; Chen and Hu, 2002; Erramilli and Rao, 1993; Lu, 2002). While other studies could not confirm this impact (Brouthers and Brouthers, 2003; Hennart, 1991; Hennart and Larimo, 1998), we suggest that asset specificity may be particularly salient in the PRC given the lack of sufficient legislation to protect intellectual property rights and their enforcement (Chen and Hu 2002). Even though investors would have preferred to enter a foreign market via a WFOE, they might have not been able to do so because of legal restrictions. Thus, in line with the role of asset specificity when entering a foreign market, we thought that with the absence of these ownership restrictions firms that have invested in highly specific assets and operate IJVs are expected to be more likely to convert their IJV into a WFOE than firms with low levels of asset specificity. However, our data does not support this suggestion. Contrary to our expectations, many IJVs with a high level of asset specificity have not been converted to a WFOE. This finding is similar to a number of studies that did not find empirical support for the influence of asset specificity on entry mode decisions (Brouthers and Brouthers, 2003; Cleeve, 1997; Hennart, 1991). According to transaction cost theory, a third factor which is argued to influence changes of ownership mode is external uncertainty. External uncertainty refers to risks that cannot be influenced by firms and involves, for example, political, legal, economic, and social risks (Deng, 2001). There are two main arguments for the assumption that the degree of external uncertainty influences foreign investors’ preference for IJVs. First, the local partner possesses country-specific knowledge which may help to reduce these risks. Secondly, the market commitment of IJVs is lower than that of WFOEs since resources and risks are shared between the partners (Brouthers and Brouthers, 2003; Pan, 1996). Although the findings with

88

2 Corporate management of innovations

regard to the effects of external risks on entry mode choice are inconclusive (Davis, Desai and Fracis 2000), the existence of a positive link has received empirical support in a number of studies in the PRC (Brouthers, 2002; Kim and Hwang, 1992). For the case of post-entry ownership strategies we found a similar relationship. Firms that continue to perceive a high degree of external uncertainty will continue to cooperate with a local partner in an IJV, while firms that perceive a low degree of external uncertainty are more likely to change their IJV into a WFOE. Finally, the cultural distance between the foreign partner firms’ home country and the host country of the subsidiary is argued to influence entry mode decisions within the framework of transaction cost theory. With reference to Hennart and Larimo (1998) and Chen and Hu (2002) we suggest that the cultural distance between the transaction partners increases the uncertainty and the costs of a transaction. According to Chen and Hu (2002, 196), cultural distance can be defined as “[…] the difference in […] values and beliefs shared between home and host countries”. A large cultural distance will increase the uncertainty and thus the transaction costs. According to Erramili and Rao (1993), IJVs can reduce costs of communication and control resulting from high cultural distance. In addition, Jung (2004, 39) states that in the case of high cultural distance “firms are more likely to have difficulties in managing foreign operations alone” (for similar arguments see Hennart, 1991) While there are some studies that have found empirical support for a positive influence of cultural distance on firms’ preferences for ownership modes that allow high levels of control, i.e. WFOEs (Anand and Delios 1997), a majority of studies have found a negative relation between cultural distance and the choice of WFOE as mode of entering a foreign market (e.g., Erramilli, 1991; Hennart and Larimo, 1998; Jung, 2004; Yiu and Makino 2002). The same can be assumed when considering the likelihood of converting an existing IJV into a WFOE. Contrary to this assumption, our findings show that IJVs with foreign partners from culturally distant backgrounds are not more likely to be converted into WFOE than IJVs with foreign partners from countries that are culturally closer to China. In fact, the results reveal that firms from dissimilar cultural backgrounds are less likely to convert their IJV into a WFOE than foreign firms from culturally similar countries. This finding may thus highlight the continued importance of local IJV-partners for foreign firms wich invest in countries with culturally distant backgrounds. 3.2 Institutional theory and change of ownership mode While the contribution of transaction cost theory to explain entry mode choice is undisputed, many researchers see a need to supplement this approach with insights from other theories in order to explain entry mode decisions more comprehensively (e.g., Hennart, 1991; White and Lui, 2005). In particular, transaction cost theory has been criticized for failing to take into account non-rational entry mode decisions. In order to address this particular deficiency and to improve the analysis of entry mode choices a number of researchers have suggested to combine transaction cost economics with institutional theory. While transaction cost theory focuses explicitly on purposive and rational decisions, institutional theory does also take into account the social construction of organizational behavior and recognizes the limits imposed by social constraints on a purely economic rationale. Thus, an integration of these two ap-

Holtbrügge and Puck: The change of foreign investors’ ownership modes in the PRC

89

proaches enhances the “explanatory strengths of both theories while simultaneously accounting for their weaknesses“ (Martinez and Dacin 1999, 77). In general, institutional theory highlights the role of the institutional context for organizational decision-making. This context is usually analyzed from a technical, cognitive and sociological perspective (Lu, 2002). Organizations are assumed to face a pressure of conformity arising from this institutional context and not based on efficiency criteria as implied by transaction cost theory. From an institutional theory perspective, organizational behavior and decision-making are thus influenced by institutional pressures. With regard to the choice of organizational forms, Martinez and Dacin (1999, 78) suggest that “societal expectations of appropriate organizational form and behavior come to take on a rule-like status in social thought and action”. This reaction to regulative, normative or cognitive societal expectations is called isomorphism (Yiu and Makino, 2002). DiMaggio and Powell (1983, 149) define isomorphism as a “constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions”. Analyzing entry-mode decisions from an institutional theory perspective leads to the question, which institutional pressures are most important. Rosenzweig and Singh (1991) differentiate between internal and external isomorphic constraints. Following Davis, Desai and Francis (2000), both types of constraints can influence the decision between IJV and WFOE. The relevance of internal constraints depends, among others, on the complexity of the relation to the parent company, while external constraints originate from outside the organization, e.g., from the host country government. While institutional theory has been applied to explain the choice between IJV and WFOE (e.g. Yiu and Makino 2002), we suggest that it can a contribute to a more comprehensive explanation of IJV-to-WFOE-conversions. As explained above, internal constraints, i.e. internal isomorphic pressures, result from parent firms’ influences over the subsidiary. One reason for homogeneous parent-subsidiary behavior lies in the history and inertness of organizations (Yiu and Makino, 2002). Once made, decisions in an organization are frequently referred to and lead to the development of organizational procedures that are adopted by the subsidiaries (Lu, 2002). When a subsidiary applies the same methods to analyze situations and make decision as the parent firm, similar structures are evolving, too. This would indicate a tendency towards WFOEs as local partners may deter IJVs from following this isomorphic pressures from parent companies. The stronger this isomorphic pressures from the parent company, the less attractive IJVs become and the higher the likelihood of converting an IJV into a WFOE. While this internal isomorphic pressure may have a number of distinct facets, we suggest that the most important one is the level of control exerted by the parent company over the subsidiary (Beamish, 1989; Geringer, 1988). Our data confirms this: the intention to achieve high levels of control increases the likelihood to convert IJVs into WFOEs. This is in line with the study of Davis, Dean and Francis (2000) that found the same results analyzing entry mode decisions. With regard to external isomorphic pressures, institutional theory particularly highlights the relevance of government regulations (Gomes-Casseres, 1990). While ownership restrictions can be considered as the main explanation for the high proportion of IJVs among FDI pro-

90

2 Corporate management of innovations

jects in China until 2001 (Beamish, 1985; Mohr and Puck, 2005), we suggest that after these restrictions do not exists any longer, government regulations may still influence the likelihood of an IJV to be converted into a WFOE. As stated by Yiu and Makino (2002, 670), “the foremost concern of a multinational enterprise when entering a foreign market is to gain market legitimacy: to establish the right to do business in the new market“. This requires a foreign firm to know and to abide all regulations existing in the host country. The more complex these regulations are perceived to be, the higher the attractiveness of having a local partner to achieve legitimacy. Several studies have provided empirical support for foreign firms’ preference for IJVs in the cases of a strong regulatory framework in the host country (e.g., Brouthers, 2002; Yiu and Makino, 2002). Applied to post-entry ownership-modes of foreign firms our data reveals that the existence of complex government regulation prevents foreign firms from changing their IJV into a WFOE since the local partner is of comparatively greater use than in cases where the regulatory framework is simple and easy to manage. This result implies that firms involved in IJVs in the PRC which continue to perceive high levels of governmental regulations affecting their business activities prefer to continue their IJV, rather than converting it into a WFOE. This result is in line with arguments and results of several existing studies on entry mode decisions (e.g. Anderson and Gatignon, 1986; Brouthers, 2002; Gomes-Casseres, 1990; Yiu and Makino, 2002).

4

Conclusion and implications

This chapter discussed a number of factors that were expected to affect the likelihood of foreign firms in the PRC to convert an IJV into a WFOE. Based on an empirical survey among 94 former or actual IJVs in the PRC we come to several important conclusions which are relevant for both theory and practice. While extant conceptual and empirical research on entry mode choice abounds, our study helps to better understand post-entry changes of foreign firms’ ownership modes in the PRC. While there have been some contributions in this area, most notably the works of Kogut (1991) or Reuer and Tong (2005) and their suggestion to view IJVs as a real-option, there remains a lack of theoretical and empirical research on this subject. Given the importance of transaction cost theory for explaining entry mode choice in general (e.g., Brouthers, 2002; Lu, 2002; Yiu and Makino, 2002) and when entering the PRC in particular (e.g. Chen and Hu 2002; Zhao and Zhu 1998), we showed that this approach is also useful for analyzing post-entry changes of ownership modes. Since an increasing number of authors suggests supplementing transaction cost theory with insights from institutional theory when analyzing entry mode decisions, we also included arguments of this approach, in particular the notion of isomorphic pressures, in our analysis of IJV-to-WFOE-conversions. We found this combination fruitful and argue that a further cross-pollination between transaction cost theory and institutional theory is warranted to provide better explanations for (pre- and post-entry) choices of ownership modes. The findings of our study also contribute to practice. We have revealed several factors which enhance the likelihood of IJV-to-WFOE conversions that can be used by foreign firms involved in IJVs when they consider converting their IJV. This may prevent firms from rush-

Holtbrügge and Puck: The change of foreign investors’ ownership modes in the PRC

91

ing towards a conversion merely because other foreign firms have done the same once ownership restrictions in their industry were abolished. The results are also of interest to policy makers as they point to the potential consequences of the reduction or abolishment of ownership restrictions for foreign firms. IJVs have played and continue to play an important role in many developing countries’ strategies to acquire foreign technology and management knowhow. At the same time, ownership restrictions often act as a deterrent for foreign firms when considering FDI in a specific country. The results of our study may indicate ways for host country governments to reduce ownership caps while at the same time reaping the benefits of foreign participation in the local economy.

References Agarwal, S. and Ramaswami, S.N. (1992) ‘Choice of Foreign Market Entry Mode: Impact of Ownership, Location and Internalization Factors’, Journal of International Business Studies, 23(1): 1-28. Anand, J. and Delios, A. (1997) ‘Location Specificity and the Transferability of Downstream Assets to Foreign Subsidiaries’, Journal of International Business Studies, 28(3): 579-603. Anderson, E. and Gatignon, H. (1986) ‘Modes of Foreign Entry: A Transaction Cost Analysis and Propositions’, Journal of International Business Studies, 17(3): 1-26. Beamish, P.W. (1985) ‘The Characteristics of Joint Ventures in Developed and Developing Countries’, Columbia Journal of World Business, 20(3): 13-19. Beamish, P.W. (1989) Multinational Joint Ventures in Developing Countries, Routledge: London/New York. Beamish, P.W. and Inkpen, A.C. (1995) ‘Keeping International Joint Ventures Stable and Profitable’, Long Range Planning, 28(3): 26-36. Beamish, P.W. and Jiang, R (2002) ‘Investing Profitably in China: Is it Getting Harder?’, Long Range Planning, 35(2): 135-151. Brislin, R.W. (1970) ‘Back-Translation for Cross-Cultural Research’, Journal of CrossCultural Psychology, 1(3): 185-216. Brouthers, K.D. (2002) ‘Institutional, Cultural and Transaction Costs Influences on Entry Mode Choice and Performance’, Journal of International Business Studies, 33(2): 203-221. Brouthers, K.D. and Brouthers, L.E. (2003) ‘Why Service and Manufacturing Entry Mode Choices Differ: The Influences of Transaction Cost Factors, Risk and Trust’, Journal of Management Studies, 40(5): 1179-1204. Buckley, P.J., Casson, M.C. and Gulamhussen, M.A. (2002) ‘Internationalisation – Real Options, Knowledge Management and the Uppsala Approach’, in V. Havila, M. Forsgren and H. Hakansson (eds.) Critical Perspectives on Internationalization, Elsevier Science, Oxford: 229-262.

92

2 Corporate management of innovations

Chen, H. and Hu, M.Y. (2002) ‘An Analysis of Determinants of Entry Mode and its Impact on Performance’, International Business Review, 11(2): 193-210. Cleeve, E. (1997) ‘The Motives for Joint Ventures: A Transaction Costs Analysis of Japanese MNEs in the UK’, Scottish Journal of Political Economy, 44(1): 31-43. Davis, P.S., Desai, A.B. and Francis, J.D. (2000) ‘Mode of International Entry: An Isomorphism Perspective’, Journal of International Business Studies, 31(2): 239-258. Deng, P. (2001) ‘WFOEs: The Most Popular Entry Mode into China’, Business Horizons, 44(4): 63-72. DiMaggio, P. and Powell, W. (1983) ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields’, American Sociological Review, 48(2): 147-160. Erramilli, M.K. (1991) ‘The Experience Factor in Foreign Market Entry Behavior of Service Firms’, Journal of International Business Studies, 22(3): 479-501. Erramilli, M.K. and Rao, C.P. (1993) ‘Service Firms’ International Entry-Mode Choice: A Modified Transaction-Cost Analysis Approach’, Journal of Marketing, 57(3): 19-38. Geringer, J.M. (1988) Joint Venture Partner Selection: Strategies for Developed Countries, New York: Quorum. Gomes-Casseres, B. (1987) ‘Joint Venture Instability: Is it a problem?’, Columbia Journal of World Business, 92(2): 97-107. Gomes-Casseres, B. (1990) ‘Firm Ownership Preferences and Host Government Restrictions: An Integrated Approach’, Journal of International Business Studies, 21(1): 1-21. Hennart, J.-F. (1988) ‘A Transaction Costs Theory of Equity Joint Ventures’, Strategic Management Journal, 9(4): 361-374. Hennart, J.-F. (1991) ‘The Transaction Costs Theory of Joint Ventures: An Empirical Study of Japanese Subsidiaries in the United States’, Management Science, 37(4), 483-497. Hennart, J.F. and Larimo, J. (1998) ‘The Impact of Culture on the Strategy of Multinational Enterprises: Does National Origin Affect Ownership Decisions’, Journal of International Business Studies, 29(3): 515-538. Holtbrügge, D. (2004) ‘Management of International Strategic Cooperation: Situational Conditions, Performance Criteria and Success Factors’, Thunderbird International Business Review, 46(3): 255-274. Inkpen, A.C. and Beamish, P.W. (1997) ‘Knowledge, Bargaining Power and the Instability of Joint Ventures’, Academy of Management Review, 22(1): 177-202. Jung, J. (2004) ‘Acquisitions or Joint Ventures: Foreign Market Entry Strategy of U.S. Advertising Agencies’, The Journal of Media Economics, 17(1): 35-50.

Holtbrügge and Puck: The change of foreign investors’ ownership modes in the PRC

93

Killing, J.P. (1982) ‘How to Make a Global Joint Venture Work’, Harvard Business Review, 60(1): 120-127. Kim, W.C and Hwang, P. (1992) ‘Global Strategy and Multinationals’ Entry Mode Choice’, Journal of International Business Studies, 23(1): 29-54. Klein, S., Frazier, G.L. and Roth, V.J. (1990) ‘A Transaction Cost Analysis Model of Channel Integration in International Markets’, Journal of Marketing Research, 27(2): 196-208. Kogut, B. (1989) ‘The Stability of Joint Ventures: Reciprocity and Competitive Rivalry’, Journal of Industrial Economics, 38(2): 183-198. Kogut, B. (1991) ‘Joint Ventures and the Option to Expand and Acquire’, Management Science, 37(1): 19-34. Lu, J.W. (2002) ‘Intra- and Inter-organizational Imitative Behavior: Institutional Influences on Japanese Firms’ Entry Mode Choice’, Journal of International Business Studies, 33(1): 19-37. Makino, S. and Neupert, K.E. (2000) ‘National Culture, Transaction Costs, and the Choice Between Joint Venture and Wholly Owned Subsidiaries’, Journal of International Business Studies, 31(4): 705-713. Martinez, R.J. and Dacin, M.T. (1999) ‘Efficiency Motives and Normative Forces: Combining Transactions Costs and Institutional Logic’, Journal of Management, 25(1): 75-96. MOFCOM (2006): ‘Foreign Investment Department of the Chinese Ministry of Commerce: Statistics about Utilization of Foreign Investment in 2005 (1-12) of China’, [www document], http://www.fdi.gov.cn/common/info.jsp?id=ABC00000000000028204 (accessed 13 February 2006). Mohr, A.T. and Puck, J.F. (2005) ‘Managing Functional Diversity to Improve the Performance of International Joint Ventures’ Long Range Planning, 38(2): 163-182. Pan, Y. (1996) ‘Influences on Foreign Equity Ownership Level in Joint Ventures in China’, Journal of International Business Studies, 27(1): 1-26. Reuer, J.J. and Tong, T.W. (2005) ‘Real Options in International Joint Ventures’, Journal of Management, 31(3): 403-423. Rosenzweig, P.M. and Singh, J.V. (1991) ‘Organizational Environments and the Multinational Enterprise’, Academy of Management Review, 16(2): 340-361. Tihany, L., Griffith, D.A. and Russel, C.J. (2005) ‘The Effect of Cultural Distance on Entry Mode Choice, International Diversification, and MNE Performance: A Meta-Analysis’, Journal of International Business Studies, 36(3): 270-283. UNCTAD (2001) ‘World Investment Report 2001. Promoting Linkages’, United Nations Press: New York/Geneva. UNCTAD (2006) ‘World Investment Report 2006. FDI from Developing and Transistion Economies: Implications for Development’, United Nations Press: New York/Geneva.

94

2 Corporate management of innovations

Wei, Y., Liu, B. and Liu, X. (2005): ‘Entry Modes of Foreign Direct Investment in China: A Multinomial Logit Approach’, Journal of Business Research, 58(11): 1495-1505. White, S. and Lui, S.S.-Y. (2005) ‘Distinguishing Costs of Cooperation and Control in Alliances’, Strategic Management Journal, 26(10): 913-932. Williamson, O.E. (1975) Markets and Hierarchies. Analysis and Antitrust Implications, Free Press: New York/London. Yan, D. and Warner, M. (2002) ‘Foreign Investors’ Choices in China: Going it Alone or in Partnership’, Human Systems Management, 21(2): 137-150. Yiu, D. and Makino, S. (2002) ‘The Choice Between Joint Venture and Wholly Owned Subsidiary: An Institutional Perspective’, Organization Science, 13(6): 667-683. Zhao, H. and Zhu, G. (1998) ‘Determinants of Ownership Preference of International Joint Ventures: New Evidence from Chinese Manufacturing Industries’, International Business Review, 7(6): 569-589. Zhao, H., Luo, Y. and Suh, T. (2004) ‘Transaction Cost Determinants and Ownership-Based Entry Mode Choice: A Meta-Analytical Review’, Journal of International Business Studies, 35(6): 524-544.

Holtbrügge and Puck: The change of foreign investors’ ownership modes in the PRC

95

Authors Dirk Holtbrügge is Professor and Head of the Department of International Management at the University of Erlangen-Nuremberg, Germany. He received his Ph.D. and his postdoctoral habilitation from the University of Dortmund, Germany. His main research interests are in the areas of interna-tional management, HRM and management in Eastern Europe and Asia. He has published eight books, four edited volumes and more than 100 articles in journals such as Asian Business and Management, European Management Journal, International Business Review, Journal of International Management, Management International Review and Thunderbird International Business Review. He is a member of the editorial boards of Management International Review and Journal for East European Management Studies. Contact: [email protected], Phone: +49 (911) 5302-452 Jonas F. Puck is Lecturer in International Management at the University of ErlangenNuremberg, Germany, Department of International Management. He received his Ph.D. from the University of Erlangen-Nuremberg. His research focuses on International Human Resource Management, Management in Emerging Markets and Cross-Cultural Management. He serves on the edi-torial board of the Journal of Cross Cultural Competence and Management. Recent Publications include a book on management in China and articles in journals such as the Journal of International Management, Long Range Planning, International Business Review or European Management Journal. Contact: [email protected], Phone: +49 (911) 5302-468 Chair of International Management University of Erlangen-Nuremberg Lange Gasse 20 90403 Nürnberg Tel.: +49 (911) 5302-452 Fax : +49 (911) 5302-470 www.im.wiso.uni-erlangen.de

Chen Jin and Wang Fangrui: A research on relationships in the knowledge management system of biotechnology firms in China Abstract Chinese biotechnology firms are stuck in troubles with the deficiency of knowledge management. We built a new concept model of knowledge management systems mainly with bold suppositions. Then we gave more attention to the investigation of Chinese biotechnology firms through interviews and questionnaires. This paper presents our findings. In brief, our findings centered around two categories of systematic relationships among the organizational background, knowledge processing, and organization performances. To explore systematic knowledge management performance, Chinese firms should align knowledge management activity with organization design or transformation under the direction of clear performance objectives. Key Words: knowledge management system; biotechnology firms; Chinese firms

1

Introduction

When the industrialized era evolved to the knowledge era, knowledge comes to be the main logic of organizing rules for most firms (Miles, Snow, Mathews, Grant Miles and Coleman, 1997). Chinese firms are on the way to reformation in order to keep in line with the international management level. Challenged with the importance of knowledge management, Chinese managers seem not to have found the proper tool to cope with it and thus to be stuck in deficiency mud. What’s the problem? According to our series of interviews with managers from some biotechnology firms in China, we find that the true essence of knowledge management is a system, but Chinese managers failed in managing knowledge in a systematic way. Fortunately, they have realized this problem, and some measures have been taken by them, for example a systematic managing process of knowledge and reorganization centered on knowledge which manifest a promising future. No matter to which extent the systematization has processed, we believe some exploration and then generalization based on the Chinese practice of knowledge management will contribute much to field of knowledge management, and at least largely promote systematic knowledge managing levels. So far, a growing number of studies have examined knowledge management from various points. These include studies of knowledge managing procedures (Coombs and Hull, 1998; Leonard, 1995; Teece and Amidon, 1998; Spender, 1996), the organizational background for knowledge management (Miles et al., 1997; Gilbert and Cordey-Hayes, 1996), knowledge

Chen and Wang: A research on relationships in knowledge management system of…

97

creation based on knowledge categorization (Nonaka, 1994), and knowledge management based on the organizational capabilities perspective (Gold, Malhotra and Segars, 2001). All these studies lay the foundation for a new breakthrough on knowledge management but several issues remain. Firstly, what is the systematic relationship between knowledge managing process and organizational background? A lot of works conduct a detailed analysis of knowledge managing procedures and presented various categorizations. And another work analyzes the organizational background for knowledge management. But little has been published on the systematic relationships between them. Secondly, almost all works assume the active relationship between knowledge management and firm performance but do not give an empirical study on that. Lastly, some study is done based on the data from developing countries. Established firms in developed countries have sophisticated knowledge management rules both explicitly and tacitly. And some rules related with organizational factors such as culture play an important but embedded role in knowledge management. When starting knowledge management step by step as a project, established firms assume their culture and embedded rules and make it easy to do knowledge management. In such an environment, researches could ignore that thing and come to a partial conclusion, but maybe this ignorance is the key factor for success. So the most urgent thing is to initiate some researches in the context of developing countries such as China in the field of knowledge management and find a proper concept model based on empirical studies. In this paper, we endeavored to build a systematic knowledge management framework by exploring the systematic relationship between the knowledge managing process and the organizational background based on data collected from biotechnology firms in China. As a developing country, China lags behind developed countries both in practice and research on knowledge management. Our work will help to promote effectiveness and efficiency of knowledge management in Chinese firms and other developing counties, and we believe the results also have a profound meaning for firms in developed countries. After all, the best way is to think about a firm as a start-up, when it needs a reformation to initiate knowledge management. We organize this paper as follows. Firstly, we go over a literature review to find out elements that should be included in a knowledge management system model. Secondly, we build the concept model for the knowledge management system and give our propositions based on analyses of the relationship within the model. Thirdly, we test our hypotheses and explore two categories of relationships among the knowledge managing process, the organizational background and the organizational performance. Lastly, we give our findings and conclude with managerial implications for Chinese managers.

2

Definition of the key elements for the knowledge management system

In this part, we will define the key elements of the knowledge management system by a study of literature. Firstly, we will go through some researches on knowledge managing procedures to find out the comprehensive knowledge managing process. We believe the proper design of knowledge managing procedures is the first step for successful knowledge

98

2 Corporate management of innovations

management. Knowledge management is a kind of management process, and it depends on organizing elements such as strategy and culture. So, our next task is to find the exact organizational background for knowledge management. 2.1 Knowledge managing procedures Almost all studies on knowledge management assume the importance of knowledge and emphasize it as the source of a competitional advantage. But there is not a uniform concept for the knowledge managing process. Most works define knowledge management and the process according to their research focuses. Facing those works on the knowledge managing process, we do our literature review mainly on two levels, i.e. the firm level and the personal level. We think the two levels are the two facets of a coin. We can only grasp the true essence of the knowledge managing process when we know exactly how it goes on both levels. Coombs and Hull (1998) proposed a six-step model of knowledge management, i.e. knowledge identification, knowledge capture, altering the format, knowledge validation, contextualizing and achieving closure. However, Coombs and Hull give their focus to the relationship between knowledge management and path-dependency in innovation, and they fix the sixstep model on the micro-level. The model is more suitable for directing personal practices than designing an organizational process. Another contributor is Dorothy Leonard-Barton (1995) who bases her discussion more firmly on the core competence strategy literature and has a focus on what she calls the whole system of knowledge management, and she proposed four procedures of the knowledge managing process such as shared problem solving, implementing and integrating new technical processes and tools, experimenting and prototyping, and importing and absorbing technological and market knowledge. As we can see her discussions of knowledge managing process are concentrated in key knowledge building, i.e. innovating competence. Davenport et al. (Davenport, Jarvenpaa and Beers, 1996) proposed that knowledge management can be organized from five facets based on a field study of thirty firms. According to them, there are three ways to approve the repertory of knowledge. One is knowledge acquisition by understanding what knowledge needs and transferring knowledge investigated through multiple channels. The other two are knowledge creation by creative processes and knowledge assembling by the conjunction of various knowledge processes. And to gain profit from knowledge, there are two ways, i.e. knowledge utilizing and reutilizing. Their work is also constrained on the micro-level and focuses on personal knowledge processing. Based on the other works on knowledge managing process by literature study, Shin etc. (Shin, Holden and Schmidt, 2001) analyze elements of knowledge management, procedures and relationships between the elements and procedures. Then they propose four steps or categories of knowledge managing process, i.e. creation, storage, distribution and application. They go one step further to say that the knowledge managing process is a process from start to end and constitutes a value chain, so we need to control the whole process in order to attain value-creation. It is a good start of their work to introduce a system view into knowledge processing. But their work does not give enough attention to knowledge protection, and we think it is a bad mistake in the knowledge age faced with severe knowledge property problems.

Chen and Wang: A research on relationships in knowledge management system of…

99

As we have stated, the knowledge managing process can be studied both on the firm level and the personal level. But the two levels are not substitutional. Meyers (Meyers, 1990) figures that employee training is only one means of supporting firm knowledge management and it cannot substitute firm knowledge management. We can use the relationship between architecture and element (Henderson and Clark, 1990) of a system to define the relationship between the two levels. The firm level can act as architecture of the whole knowledge processing system and the personal level as element. The architecture of the firm level defines the boundary of knowledge processing of the whole system and the linkage between activities on the personal level. In this research, we will focus on the firm level of the knowledge managing process, but in order to facilitate activities on the personal level, we also take attributes of personal activities of the knowledge managing process into consideration. According to that purpose, we define the knowledge managing process including four procedures, i.e. knowledge acquiring, knowledge utilizing, knowledge transferring and knowledge protecting. Knowledge acquisition is the process by which firms create or outsource various aspects of knowledge needed for business operations from all related channels and by all related means. There are four facets for successful knowledge acquisition, i.e. breadth of knowledge acquisition (Davenport, Delong and Beers, 1998), quality such as speed efficiency and flexibility (Grant, 1996), capabilities of acquisition (Leonard, 1995), and management competence for knowledge acquisition activities on the person level (Quinn, Anderson and Finkelstein, 1996). Knowledge utilization is the core procedure for the whole business operation ranging from decision making to implementation. Knowledge utilization is ignored in most works on knowledge management that assume the best utilization after knowledge acquisition and transformation (Gold, 2001). Knowledge utilization can be organized into two categories for the purpose of value creation. One is knowledge utilization in daily business activities. That is the basic use of knowledge. And only when knowledge filters into every details of the whole business system, knowledge will be a kind of organization culture and knowledge utilization will be the rule for the firm. The other is knowledge utilization in innovation. It is easy to imagine where innovation will go without knowledge utilization. Here we mean innovation as broadly defined, including technological innovation and non-technological innovation. Knowledge can only create more value as a flow. Knowledge transfer is the main procedure of knowledge flows. Only when knowledge transfers can penetrate through the boundaries among all subsystems and boundaries between the whole business system and the environment, knowledge management will have the potential to be long-lasting. There is a lot of literature on knowledge flows and transfers, and almost all works emphasize the efficiency of flows in an internal system (Leonard, 1995; Spender, 1996; Appleyard, 1996; Nonaka and Konno, 1998; Szulanski, 1996). Knowledge protection embodies the knowledge managing procedures that are designed to protect the knowledge within an organization from illegal or inappropriate use or theft (Gold, Malhotra and Segars, 2001). Only when an organization establishes sound knowledge protection procedures, knowledge management can be installed into a closed loop to gain persisting competitive advantages. But not all knowledge can be defined according to property laws

100

2 Corporate management of innovations

and property rights (Porter-Liebskind, 1996). So knowledge protection needs indispensable steps and activities, and more importantly an organization needs the cultivation of necessary behaviors by incentive alignments, employee conduct rules, and even job designs. 2.2 Organizational background for knowledge management Knowledge management cannot occur without a proper organizational background. An organizational background includes all organization factors such as structure, motivation arrangements, culture and strategy etc. All organization factors must be well organized in order to find the proper background, and the comprehensive manifestation of an organizational background is an organization form. Broadly defined, an organization form does not mean structure only, it is the ways of an organization to develop and arrange all kinds of marketvalued resources such as knowledge in order to coevolve with the market (Miles, Snow, Mathews, Grant Miles and Coleman, 1997). In fact, an organization needs proper behaviors to facilitate knowledge management. An organizational background is the only and most efficient way, and an organization has to design its background for knowledge processing, including design and redesign of strategy, culture, structure, and compensation and reward programs etc. (Gilbert & Cordey-Hayes, 1996). One of the most important factors of the organizational background for knowledge management is strategy. As a kind of strategic resource, knowledge is the basic source of a competitive advantage for every organization in the knowledge age. So it is necessary for an organization to reengineer its strategic arrangement of key resources in order to deploy organizational change for knowledge management. The first step to an organizational change is to establish new hypotheses, rules and beliefs of the environment to initiate the change (Sange, 1990). That means it is the strategy that leads to the change followed by knowledge management. And when an organization gets the knowledge management run well, knowledge management can influence choosing and implementing a strategy as a matter of course (Teece, Pisano and Shuen, 1997). Another important factor is the organizational structure. To promote a smooth flowing of knowledge and ideas, an organization needs a boundless structure. When boundaries in an organization and between an organization and its environment melt down, it is easy to expand sensemaking (Dougherty, 2001; Boland and Tenkasi, 1995; Iansiti, 1995). There are many ways to be boundless. For example, an organization can design a hypertext organizational structure to promote knowledge processing (Nonaka, 1994). A third important factor is culture. Some works on knowledge management propose culture as the core element for an organization (Leonard, 1995; Davenport, Delong and Beers, 1998; Delong, 1997). To have a knowledge-supporting culture, an organization should establish a new vision (O’Dell and Grayson, 1998; Davenport, Jarvenpaa and Beers, 1998) and a new value system (Leonard, 1995; Miles et al., 1997) to cultivate proper knowledge management behaviors (O’Dell and Grayson, 1998). Compensation and reward programs also are another important factor (Quinn, Anderson and Finkelstein, 1996; Davenport and Prusak, 1998). Compensation and reward programs are one of main ways to support culture cultivation and to foster behavior. An organization should

Chen and Wang: A research on relationships in knowledge management system of…

101

help employees establish the whole concept of the business and develop personal competence. Basically speaking, no business concept can succeed without linkages to compensation and reward. Lastly, we would like to focus on the technological structure of an organization. The technological structure in the context of knowledge management does not mean product or manufacturing technology but information technology. Through information technology and communication systems in an organization, previously fragmented information and knowledge can be integrated together efficiently, and knowledge sharing, knowledge storage, and even knowledge protection can be easily done. But an organization cannot succeed in knowledge management merely with information technology and a communication system. As many surveys display, the high failure rate of knowledge management attributes to the disconnection between IT technology and soft technologies such as strategy and culture. So an organization should have a technological structure for knowledge management, but it is only a bridge and cannot be seperate from other soft elements. As we have said, an organizational background should include strategy, structure, culture, compensation and reward programs, and a technological structure. Of course, there are other elements such as processes, and they are also important for an organization. But we are not talking about organizational definition, and our purpose here is to find the key elements playing an important role in knowledge management. And managing activities will inevitably involve costs, so the preferable way is to focus on essential activities and link them to the whole system.

3

The concept model and hypotheses

Based on our definition of the four knowledge managing procedures and the five elements of an organizational background, we propose a concept model of the knowledge management system (figure 1). According to figure one, the knowledge managing process is embedded into the system of the organizational background, and runs centered with organization performance. Performance, knowledge managing process, and organizational background constitute the knowledge management system. We put performance at the center of the system, because knowledge accelerates competition in the knowledge age and every activity taken by a firm should aim at performance, no matter short-run or long-run. There are two points to be made on the logic underneath the system. Firstly, the key elements of an organization are combined into the organizational background that cultivates and supports the running of the knowledge processing, and the knowledge processing runs centralized with the organization performance through a closed loop. Lastly, the focuses of knowledge processing are geared to the development of the organizational performance, then challenges for the organizational design come forth, and this initiates changes of the organizational background resulting in the reengineering of key elements. So, the knowledge management system comprising performance, knowledge processing and the organizational background is a closed loop in itself. And the control forces are changes of external envi-

102

2 Corporate management of innovations

ronments. Organization performance plays a bridging role between the knowledge management system and the environments.

Figure 1: A concept model of knowledge management system

To sum up, there are two categories of relationships in the concept model of the knowledge management system. One is about the relationships between the elements of the organizational background and the knowledge managing procedures of knowledge processing. According to our literature review on organizational elements supporting knowledge management, organizational background elements such as strategy, culture, compensation and reward programs, organizational structure, and technology structure should have an active effect on the knowledge managing process. So, we get our first hypothesis. Hypothesis 1: Sound organizational background will actively support knowledge managing process. When we interviewed some Chinese managers on how to initiate the knowledge management project, the biggest concerns of them turned out to be architecture and the behavior of knowledge management. They give more attention to the technological structure than to the organizational structure when establishing the knowledge management architecture. And when taking employees’ behaviors on the background of knowledge managing into consideration, they mentioned only culture as facilitating factors. We propose that it is not enough with only a technological structure to fulfill knowledge management, and the organizational structure is more important for integrating dispersed knowledge and facilitate the communication of knowledge, especially tacit knowledge in Chinese firms. Furthermore, culture is the core organizational designing element to cultivate employees’ behaviors, but compensation

Chen and Wang: A research on relationships in knowledge management system of…

103

and reward can play a more important role when a proper culture is absent or blurry. All in all, considering the extent of industrialization, firms in China should give more attention to the organizational structure than to the technological structure, more to compensation and rewards than to culture when they first want to start knowledge management. Accordingly, we propose two sub-hypothesis to further explain the relationships between organizational background and knowledge managing process when the extent of industrialization is considered. Our purpose is to get some advice especially for Chinese firms. Hypothesis 1-a: The organizational structure is more important than the technological structure to establish the organizational architecture of knowledge management for Chinese firms in less industrialized sectors. Hypothesis 1-b: Compensation and rewards are more important than culture to promote proper knowledge managing behaviors for Chinese firms in less industrialized sectors. The other is about the relationship between the knowledge managing process and the organizational performance. We decompose the knowledge managing process into four procedures including knowledge acquisition, knowledge utilization, knowledge transfer, and knowledge protection. And we assume these knowledge managing procedures to have a strong and active effect on the organizational performance. Furthermore, organizational performance can only be promoted with a systematic integration of the four knowledge managing procedures. So, we get our second hypothesis as follows: Hypothesis 2: The knowledge managing process can promote organizational performance. We derive the two hypotheses mainly to explain and explore the relationships among the knowledge managing process, organizational background, and organizational performance. Our intention here is different from testing the concept model. We have to admit that it is more complex and exhausting to test a concept model. So, we assume our concept model in this paper and pay our attention mainly to the exploration of relationships. We believe some further work should be initiated on model tests.

4

Methods

4.1 Data and sample With the help from the Chinese National Science and Technology Ministry, we initiated a large survey in the middle-east of China. And we mainly chose those firms with R&D, manufacturing and marketing departments as samples in order to get an exact investigation. Our main approach of data collection was questionnaires. To guarantee the efficiency of the information collected, we chose general managers, general engineers, vise technology presidents, R&D center directors or other employees with superior positions in the organization. We sent 500 questionnaires and received 317 back. We excluded those with multiple or missing responses, then got 270 useful ones.

104

2 Corporate management of innovations

Of the useful questionnaires returned, there were 56% of small firms with revenues under thirty million RMB, and 17% of medium-sized firms with revenues from thirty million RMB to two hundred million RMB. So our research is mainly based on small and medium-sized firms. That’s consistent with our purpose stated before. 4.2 Measures We agreed that capabilities are the main source of a competitive advantage, so a part of our measures is sourced to the findings of Gold et al. (Gold, Malhotra and Segars, 2001). But we add some new measures to our questionnaire based on the literature review and appendix-A gives details on that. We categorize our measures into three constructs, i.e. knowledge managing procedures, organizational background, and organizational performance. Within the construct of knowledge managing procedures, we design four categories of questions on knowledge acquisition, knowledge utilization, knowledge transfer, and knowledge protection. The eleven questions on knowledge acquisition focus on sources of knowledge acquisition, knowledge creation and its efficiency, on the evaluation system of activities aiming at knowledge acquisition, and on the acquisitions of competence development. The other eleven questions on knowledge utilization give details on employees’ behavior, various utilizations, the appraisal of utilization, and knowledge share. The nine questions on knowledge transfer are centered on approaches, forms, rules, and direction. The last six questions of knowledge protection concentrate on the employees’ consciousness and protection means. On the background of the construct of the organizational background, we designed five categories of questions on strategy, culture, compensation and rewards, technological structure, and organizational structure. The five questions of the first category on strategy deal with vision, strategic objectives and employee cogitation. The seven questions of the second category on compensation and rewards rely on the structure of compensation and rewards, the satisfaction of skilled workers, and management rules or other means. The eight questions of the third category on culture are based on the spirit of innovation, the leader’s regard, and material regulations etc. The seven questions of the forth category on technological structure are set up on rules for knowledge categorization, hardware systems and software systems. The last seven questions of the fifth category on organizational structure deal with structure, process, teamwork, and alliance etc. Within the construct of performance, we designed twelve questions on performances of the last two years with capability development. The questions mainly focus on innovation performance such as new product development, competitive advantages such as better products, inner process innovation, and the effect of decisions. We examined the unidimensionality and reliability of each construct with the Cronbach alphas test. The result shows that all alphas coefficients, except that of the cultural construct which is 0.7238, pass over 0.8. So our measures of each construct are sound in statistics on the whole.

Chen and Wang: A research on relationships in knowledge management system of…

5

105

Results

As we have stated our purpose of this paper, we aim at exploring the relationships among the knowledge managing process, the organizational background, and the organizational performance on the backround of the assumption of our concept model of knowledge management system. And with some relationships confirmed, our concept model will be partially tested. Starting from this purpose, we go through a factor analysis, bivariate correlation and partial correlation and multivariable regression analyses in the Statistics Package for Social Science (SPSS) to test our hypotheses. To get comprehensive data, we use multiple terms to measure one object. So the value for an object has to be calculated. We design two steps to get the value. Within the first step, we use the module of the factor analysis in SPSS to extract one factor by principal components. We will accept the factor value if the result is sound, or we will turn to the second step. Within the second step, we will ask professionals to give weights for each item and then get the value by average with weights, if factor analysis fails. Luckily, all results of the factor analysis were sound. We named each factor after the object when the factor analysis closed and the result could be accepted. Table one gives details about the results of the factor analysis. Based on all factor values, we give our test of hypotheses as follows. Table 1: Results of factor analysis

5.1 Tests of hypotheses Our first hypothesis is about the relationship between the organizational background and the knowledge managing process. We proposed that knowledge processing is embedded in the organizational background, and a sound design of strategy, compensation and rewards, culture, technological structure, and organizational structure will support the running of knowl-

106

2 Corporate management of innovations

edge acquisition, knowledge utilization, knowledge transfer and knowledge protection. To test this hypothesis, we do bivariate correlation analyses between all elements of the organizational background and all knowledge managing procedures of the knowledge managing process. The result shows that active relationships do exist and the first hypothesis can be accepted. Table two shows the details of the result. Table 2: Bivariate correlation result of hypothesis 1

According to table two, we can find that there are significant relationships between the elements of the organizational background and knowledge managing procedures of the knowledge managing process at the 0.01 level. And the coefficients are all positive values surpassing 0.55. The result shows that elements of the organizational background such as strategy and culture are actively related to knowledge processing. This confirms that an organization needs to build up a sound organizational background for knowledge processing. When we look at the coefficients, we find an interesting thing. Although the coefficients are close to each other in values, there does exist a maximum. We identified them in table two. According to the coefficient values, we can see that the main element for knowledge managing procedures is statistically concentrated. The technology structure is the main element for knowledge acquisition, knowledge transfer, and knowledge protection. Compensation and rewards are the main elements for knowledge utilization. That make-up of knowledge management of the firms investigated in China goes down to daily works. And strategy, culture and organizational structure are relatively weak. As we can see, the firms in China have a weak systematic relationship among the elements of the organizational background, and they must endeavor to develop knowledge management based on the total relationships between the organizational background and knowledge processing. We think that there are two rea-

Chen and Wang: A research on relationships in knowledge management system of…

107

sons for the status quo of the knowledge management of Chinese firms. One is the advancement of information technologies and communication technologies. Almost all firms in China are pushed to introduce these technologies into their management activities. The other is a fierce competition for talented employees. Chinese biotechnology firms are busy with short-run competition and give more attention to tactical activities than strategy. At a first glance, the result of bivariate correlation analyses seems denying the two subhypotheses. So, we go through the multivariable regression analysis to test hypothesis 1-a and hypothesis 1-b. As we have said, the knowledge managing process is a closed loop by itself, and every knowledge managing procedure is integrated to the systematic process. So, we go through the factor analysis among knowledge acquisition, knowledge utilization, knowledge transfers and knowledge protection to get a factor determining the knowledge managing process (“KMO Measure of Sampling Adequacy”: 0.832; “Bartlett’s Test of Sphericity”: 0.000; “% of Variance”: 79.163). Then we put the knowledge managing process as a dependent variable and five elements of the organizational background as independent variables to do a multivariable regression analysis. Table three shows the results of the analysis of variances, and table four shows the results of the coefficients. Table 3: Result of analysis of variances in regression analysis for hypothesis 1-a

108

2 Corporate management of innovations

Table 4: Results of the coefficients in the regression analysis for hypothesis 1-a

According to the results of the multivariable regression analysis, the technological structure is the first variable entering the model defined by the knowledge managing process as a dependent variable. So, hypothesis 1-a cannot be accepted. This coincides with the results of the bivariate correlation analyses. Besides the reasons we concluded before, incompetence concerning the organizational change is the biggest problem of the knowledge management of Chinese biotechnology firms. And compensation and rewards come before organizational structure, strategy, and culture. So, hypothesis 1-b can be accepted. Remembering that our samples are mainly composed of small and medium-sized firms which account for 87% of the total samples, we can conclude that small and medium-sized firms will cultivate knowledge managing behaviors mainly by compensation and rewards when a proper culture is absent. Anyway, all five elements of the organizational background enter the model. This means that all these elements do have an effect on the knowledge managing process. That confirmed the results of the bivariate correlation analyses. But we cannot draw any further conclusion from the analysis, maybe some methodologies such as the complex system theory and benchmarking could be used to stab into all systematic relationships within the knowledge management system. Of course, we would not like to let our research lead to a causal analysis of the knowledge management of Chinese firms in this paper. To sum up, there is a strong relationship between the organizational background and knowledge processing.

Chen and Wang: A research on relationships in knowledge management system of…

109

Our last hypothesis is that knowledge managing process is actively related to organization performance. We proposed that knowledge acquiring, knowledge utilizing, knowledge transferring and knowledge protecting can help developing organization performance. We also do bivariate correlation analyses to test this hypothesis. And table five gives the details. Table 5: Bivariate correlation result of hypothesis 2

According to table five, we can find that there are significant relationships between knowledge acquisition, knowledge utilization, knowledge transfer and knowledge protection and organizational performance at the 0.01 level. And the coefficients are all positive values surpassing 0.68. This result confirms hypothesis 2, and assures that every knowledge managing procedure can promote organizational performance. The coefficient values are close to each other. But the distance between knowledge utilization and performance and that of knowledge protection and performance is a little bigger. We can see from this that the firms investigated in China give their focus to knowledge utilization and knowledge protection. There are two reasons for these problems. Firstly, as a developing country, the road of technological development of Chinese firms runs from knowledge assimilation based on technology introduction to innovation (Kim, 1997). Under this trajectory, Chinese firms are not good at indigenous innovation based on knowledge creation. The focus of knowledge management inclines to the assumption that knowledge naturally exists. Secondly, legislations on knowledge property and the implementation of these laws are not good enough in China. So, Chinese firms emphasize knowledge protection all around. For a further analysis of the relationships between the knowledge managing process and organizational performance, we select performance as a dependent variable and four knowledge managing procedures as independent variables to do a multivariable regression analy-

110

2 Corporate management of innovations

sis. Table six shows the results of the analysis of variances, and table seven shows the results of the coefficients. Table 6: Resulst of the analysis of variances within the regression analysis for hypothesis 2

Table 7: Results of the coefficients in the regression analysis for hypothesis 2

The results of the multivariable analysis confirm the findings of the bivariate correlation analyses. The variables of knowledge utilization and knowledge protection are the first and the second ones to enter the model, and this shows the imminent importance of the two variables for the organizational performance. But the variable of knowledge transfer does not enter the model, and that the result shows that the effect of knowledge transferring on the

Chen and Wang: A research on relationships in knowledge management system of…

111

organizational performance is insignificant for the Chinese firms investigated in this study. We think the result coincides with hypothesis 1-a. Chinese firms did not explore a potential advantage from the transfer of knowledge because they are incompetent with organizational change and the existing organizational structure hinders knowledge transferring. Figure two shows the distribution of our samples by the type of organizational structure. According to figure two, only 4% of all samples establish a networked structure that can promote knowledge transferring fluently, and the rest of all samples with other types of organizational structure are featured by a clear boundary.

Figure 2: Distribution of samples by type of organizational structure

To sum up, we find there do exist two categories of relationships among the organizational background, the knowledge managing process and the organizational performance in the knowledge management system by data analyses based on a survey of Chinese firms, though a part of the relationships is not significant or absent. According to the relationship found, we come to a common conclusion that only with a sound design of the organizational background the knowledge managing process can run properly and reach a good performance. 5.2 Two additional explorations Based on the concept model of the knowledge management system, we analyzed the two categories of relationships in the system based on data from Chinese biotechnology firms. Our purpose is to lay the foundation for the future study of the knowledge management system by exploring the relationships in it, but not to test our concept model. So, we pay great attention to our data analysis. And at the end of our test of the hypotheses we are sure that two other explorations have to be executed.

112

2 Corporate management of innovations

The first exploration is on the effect of the knowledge managing process on the organizational performance. After the result of the bivariate correlation analysis between knowledge managing procedures and the organizational performance comes out, the following questions come to our minds: to what extent can the knowledge managing process help to attain organizational performance and do firms really need the knowledge managing process. We believe that these are very important questions under the condition of scarce resources faced with a rapid competition. To answer these questions, we compare the results of the bivariate correlation analysis and the partial correlation analysis with respect to the relationships between the elements of the organizational background and the organizational performance controlled by knowledge processing activities. Table eight shows the result in detail. Table 8: Contrast between bivariate and partial correlation result

According to table eight, all partial correlation coefficients are much smaller than Pearson correlation coefficients. We think the margin between the two categories of coefficients can to some extent account for the net-value through the knowledge managing process for the organizational performance. So, we get a glancing answer for question one. And the result confirms again that it is necessary to introduce the knowledge managing process to enhance organizational performance. It also confirms our proposal for systematic relationships among the organizational background, knowledge processing and the organization performance. The last exploration is on the question whether there are different characteristics of the knowledge managing process among differently sized firms. Our samples are composed of small firms, medium-sized firms and big firms. Figure three shows the distribution of our samples by firm size. Will differences of size of the samples influence the consistency of our

Chen and Wang: A research on relationships in knowledge management system of…

113

result or do firms with different sizes show different characteristics in knowledge acquisition, knowledge utilization, knowledge transfer, or knowledge protection? To answer this question and explore the differences in the knowledge managing process of Chinese firms, we do a one-way analysis of variances. Table nine shows the details of the result.

Figure 3: Distribution of samples by firm size

Table 9: result of ANOVA controlled by firm size

114

2 Corporate management of innovations

According to table nine, all significant coefficients are over 0.1. So, the result shows that there are no differences of the knowledge managing process among differently sized firms. That is a very interesting finding. As we know, there should be differences at every stage or size of the development for firms. Our findings deny this common sense. And we think this will happen only when things are really new to those firms and actually pose a new start for all firms. Based on this finding on Chinese firms, we think firms should deploy the program of knowledge management as start-up businesses in order to successfully introducte a new concept or management philosophy successfully. Furthermore, we think the result of the analysis of variances confirms our proposal in this paper and definitely helps to increase the convincing power of our analysis. That is to say, a firm should establish a reformation mentality to initiate knowledge management programs and treat them systematically from the very beginning as a new start-up.

6

Managerial implications and conclusions

Knowledge management is not a new subject in the business management field. A lot of researches have worked on it and some multinational corporations really do it well. But works related to China and other developing countries are very limited, and almost all Chinese firms, even some great business groups which have their positions in the global top 500, have trouble with this topic, especially newly developed high-technological firms. Though our model has not been tested completely, our conclusions will give a great help with a new development for the study and practice on knowledge management in developing countries and also developed countries. The implication of this work is that, according to a system view, an organization should set up the knowledge management system including the organizational background, the knowledge managing process and the organizational performance. Firstly, managers need to design a proper organizational background to support the knowledge managing process. It is critical for established firms to cultivate an organizational culture for the knowledge management. But for firms in developing countries, the core task is to design a knowledge creating strategy to direct the whole knowledge management. For example, firms in China need to set up an indigenous innovation strategy to supplement new knowledge shortage. They also need to build up an integrated architecture of the organizational background to support the closed loop of the knowledge managing process. Secondly, managers will maybe design different procedures of the knowledge managing process, but all the procedures need to be integrated into a perfect closed loop without any kind of knowledge managing procedures absent, especially in Chinese firms and other firms in developing countries. Thirdly, managers need to learn about the relationships between performance objectives and the knowledge managing process. They need to put performance at the core of the knowledge management system. When the environment changes they must adapt the knowledge managing process corresponding to the environmental changes, and then reengineer the elements of the organizational background. To do this, we propose several tasks or challenges for managers in developing countries. Firstly, managers must pay most of their attention to systematic relationships in the knowledge management system and endeavor to build up a holistic framework before the initiation

Chen and Wang: A research on relationships in knowledge management system of…

115

of the knowledge management program. In this process, they should pay more attention to the “soft” facet of knowledge management than to technological architecture only. It is often just due to the absence of a proper strategy, culture and organizational structure that leads to weak activities in the knowledge managing process. For example, when strategies of developing firms focus on the utilization and protection of the existing knowledge, they will not devote resources sufficiently to create and acquire new knowledge. And when the structure is not convenient for the transfer of knowledge, it is harder to gain competitive advantages through knowledge transferring. So we would like to emphasize our proposal again that firms will only outperform in the knowledge age when they build up the knowledge management system completely. The point is the system relationships between all elements, not being good at only one activity. Secondly, as a start-up or with short development time, firms should rely more on compensation and rewards than on culture to cultivate employees’ knowledge managing behaviors. In our interviews with managers we found that with more and younger employees joined, firms are more comfortable with short-term compensation than long-term compensation in the first years. We believe that behaviors brought up by compensation in form of money will grow to rules and then culture. This process seldom happens reversely. Lastly, and maybe most importantly, firms must have a reformation mentality when they want to start knowledge management. Rules and concepts in the knowledge age are definitely different from those in the industrialized age. And an organizational reformation or reengineering will make it easier for firms to introduce a new management fashion such as knowledge management. Furthermore, it is very important for firms in developing countries or start-ups to establish their own knowledge management system by reformation instead of copying isolated practices from established firms. All in all, firms in developing countries are on the way to set up a knowledge management system where the point is to find their own systematic ways according to a common framework and rules of basic relationships. Our purpose is to introduce a systematic view into knowledge management. No matter what view a research takes, a systematic view is the basic precondition. Although it seems that our framework on knowledge management system is static and closed beyond environment, we focus on the systematic relationship between the inner activities of an organization and the outer environment through orderly ways such as the designing of performance objectives. Based on systematic relationships in an organization and those between organization and environment, an organization can successfully develop knowledge management. In this research, we approach the knowledge management system mainly by statistic relationships. It is not enough to explore the dynamic relationships between the organizational background, knowledge processing, organizational performance and the environments. Besides, we assume the concept model to be developed from a literature review without further tests and exploration. This is not enough to establish a more feasible framework to direct knowledge management building. We suggest more works with dynamic systematic methods to continue this study. And we also hope for every comment and best research on that.

116

2 Corporate management of innovations

Appendix A Study measures Measures on a five-point scale: 1. “strongly disagree”; 5. “strongly agree” Measures of Knowledge Acquisition α = .8159 Your firm can quickly gather and analyze data, information and knowledge of products and services in your industry. Your firm can efficiently gather and analyze data, information and knowledge of products and services in your industry. Your firm can gather advice on products and services from your customers. Your firm can gather industry-related information from partners. Your firm can gather industry-related information over the internet and intranet. Your firm can apply existing knowledge to create new knowledge. Your firm can derive experience from former projects to use in coming projects. Your firm can acquire new knowledge by engaging new employees. Your firm designs programs to facilitate employees’ acquiring knowledge from others. Your firm designs measures to appraise the effect of knowledge acquired. Your firm designs programs to develop sustaining capabilities of knowledge acquisition. Measures of Knowledge Utilization α = .8590 Your employees have a good understanding of the effect of knowledge on performances. Your firm highlights gaining competitive advantages through knowledge. Your firm highlights utilizing proper knowledge in a new product development. Your firm can use new knowledge to solve new arising problems quickly. Your firm adopts systematic processes in market decisions based on data analysis. Your firm appraises the effects of technologies on products and manufacturing periodically. Your firm appraises the effects of technologies on costs periodically. Your firm is good at learning from failed projects. Your employees can conveniently get the knowledge they need. Your firm designs programs to facilitate knowledge sharing among employees and departments. Your firm can quickly find related knowledge in problem solving. Measures of Knowledge Transferring α = .8707 Your firm organizes cross functional teams to tackle key problems by clear routines. Your firm builds different knowledge diffusing channels for different kinds of knowledge. Your firm regards colloquy as a routine measure to solve problems. Your firm establishes perfect processes for the categorization and processing of knowledge. Your firm designs handbooks, directories, software etc. for process description. Your firm builds procedures and channels to transfer knowledge to every department and employee. Your firm can use advanced knowledge for product design or programs.

Chen and Wang: A research on relationships in knowledge management system of…

117

Your employees can communicate with each other in informal meetings or other activities. Your firm can efficiently transfer knowledge acquired externally. Measures of Knowledge Protection α = .8539 Your employees exactly know about the importance of knowledge protection. Your firm designs programs to protect your knowledge from abuse. Your firm designs programs to protect your knowledge from theft. Your firm designs programs to reward knowledge protection. Your firm designs programs to confine core knowledge within limits. Your firm designs programs to protect your knowledge property in collaborations with partners. Measures of Strategy α = .8452 Your firm does strategic planning periodically. Your firm pays a lot of attention to knowledge when planning and implementing a strategy. Your firm has a clear vision of knowledge. Your firm has clear development objectives concerning knowledge. Your strategy is well understood by your employees. Measures of Compensation and Reward α = .8720 Your compensation and rewards are based on knowledge and competence. Your employees have the right to share the benefit of your firm’s success. Personal experience is regarded and employee satisfaction is high. Your firm encourages employees to continue learning. Your firm encourages helping each other among employees. Your firm encourages multiple contacts among employees. Your firm has specific programs to reward employees with innovative ideas. Measures of Culture α = .7238 Your firm respects innovation and builds up an open and trustful environment. Your firm cultivates the values of teamwork. Your top management level highly respects organizational learning. Your firm has specific training arrangements on the job for employees. Your firm sets up a good environment for a high participation in knowledge acquisition and transferring. Your firm has high tolerance for failure and looks at failure as an opportunity to learn. The best practice can be publicized and written down.as routines. There are good relationships among employees from different departments. Measures of Technological Structure α = .8721 Your firm has specific rules for the categorization of product knowledge. Your firm has specific rules for the categorization of manufacturing knowledge. Your firm pays attention to technological architecture to support knowledge processing.

118

2 Corporate management of innovations Knowledge from different channels can be integrated efficiently through technological architecture. Your firm owns the perfect computer system to support your businesses. Employees can conveniently attain useful knowledge by the help of computer systems. Employees from different places can communicate knowledge and information by network.

Measures of Organizational Structure α = .8996 Your organizational structure facilitates knowledge transferring and sharing. Your organizational structure facilitates knowledge creation. Your firm designs specific processes to support knowledge communication among departments. There are cross functional teams in your firm. Your objectives are perfectly prioritized. Your firm builds up useful alliances. Your managers will check knowledge processing when faced with serious mistakes. Measures of organizational performance α = .9285 In last the two years, your firm has done a lot to …… Explore new products and new services. Identify new business opportunities. Forecast potential markets for new products or services, Succeed in the quick commercialization of new inventions. Place new products quicker than your competitors. Produce better products than your competitors. Strengthen capabilities to meet an emergency. Respond to markets quickly. Optimize internal processes. Support cooperation among multiple departments efficiently. Make proper decisions based on systematic and strict evaluations.

References Andrew H Gold, Arvind Malhotra and Albert H Segars, 2001. Knowledge management: An organizational capabilities perspective. Journal of Management Information Systems, 18(1): 185. Appleyard M.M., 1996. How does Knowledge flow? Interfirm patterns in the semiconductor industry. Strategic Management Journal, 17: 137-154. Boland R.J. and Tenkasi R.V., 1995. Perspective making and perspective taking in communities of knowing. Organization Science , 6 (4): 350–372. Coombs R. and Hull R., 1998. Knowledge management practices and path-dependence in innovation. Research Policy, 27: 237-253.

Chen and Wang: A research on relationships in knowledge management system of…

119

Davenport T., Delong, D. and Beers M., 1998. Successful knowledge management projects. Sloan Management Review, 39: 43-57. Davenport T., Jarvenpaa S. and Beers M., 1996. Improving knowledge work processes. Sloan Management Review, 37: 53-65. Davenport T. and Prusak L., 1998. Working Knowledge: How Organizations Manage What They Know. Boston : Harvard Business School Press. Delong D., 1997. Building the knowledge-based organization: how culture drives knowledge behaviors. Working paper. Earnest & Young’ center for business innovation, Boston. Dougherty D., 2001. Re-imagining the differentiation and integration of work for sustained product innovation. Organization Science, 12 (5): 6221–6231. Gilbert M. and Cordey-Hayes M., 1996. Understanding the process of knowledge transfer to achieve successful technological innovation. Technovation, 16(6): 301-312. Grant R., 1996. Toward a knowledge based theory of the firm. Strategic Management Journal, 17: 109-122. Iansiti M., 1995. Shooting the rapids: managing product development in turbulent environments. California Management Review, 38 : 37–58. Leonard, D. 1995. Wellsprings of Knowledge: Building and Sustaining the Source of Innovation. Boston: Harvard Business School Press. Meyers P. W., 1990. Non-linear learning in large technological firms: period four implies chaos. Research Policy, 19: 97-115. Miles R., Snow C., Matthews J., Miles G. and Coleman H. Jr. 1997. Organizing in the knowledge age: anticipating the cellular form. Academy of Management Executive, 11(4): 724. Minsoo Shin, Tony Holden and Ruth A. Schmidt., 2001. From knowledge theory to management practice: towards an integrated approach. Information Processing and Management, 37: 335-355. Nonaka I. and Konno N., 1998. The concept of “ba”: building a foundation of knowledge creation. California Management Review, 40(3): 40-54. O’Dell C. and Grayson C., 1998. If only we knew what we know: identification and transfer of internal best practices. California Management Review, 40(3): 154-174. Quinn J.B., Anderson, P. and Finkelstein, S., 1996. Managing professional intellect: making the most of the best. Harvard Business Review, March–April: 71–80. Rebecca M. Henderson and Kim B. Clark. 1990. Architectural innovation: the reconfiguration of existing product technologies and the failure of established firms. Administrative Science Quarterly, 35: 9-30.

120

2 Corporate management of innovations

Sange P., 1990. The Fifth Discipline: The Art and Practice of Learning Organizations. New York: Doubleday. Spender, J.C., 1996. Making knowledge the basis of a dynamic theory of the firm. Strategic Management Journal, 17: 45-62. Szulanski G., 1996. Exploring internal stickiness: impediments to the transfer of best practice within the firm. Strategic Management Journal, 17: 27-43. Teece D. and Amidon D., 1998. New measure of success. Journal of Business Strategy, January/February: 20-24. Teece D.J., Pisano G. and Shuen A., 1997. Dynamic capabilities and strategic management. Strategic Management Journal, 18: 509–533.

Authors Chen Jin is Professor of Management at College of Public Administration, Zhejiang University. He is also the Deputy Director of National Institute for Innovation Management, Zhejiang University. He got his Ph.D. in 1994. In 1998, he was a Visiting Scholar in Alfred Sloan School of Management at MIT. He also visited SPRU, University of Sussex, Brighton, UK, among others. Dr. Chen is a deputy Chairman, Research Association of Science & Science and Science & Technology Policy of China. His research areas include technology and innovation management, science, technology and education policy. He is a PI of a variety of research projects in China as well as abroad. The projects he is undertaking include managing complex product systems; managing radical innovation in China; internationalization of technological innovation for Chinese enterprises; managing open innovation in China, university and industry links for indigenous technological innovation. He has published several books and more than 180 papers on management of technology and innovation. Journal outlets include IEEE trans on Engineering Management, Technovation, International Journal of Innovation and Entrepreneurship. He was awarded the Excellent Young Teacher, Ministry of Education, P.R. China, Huo Yingdong Prize, Ministry of Education, P.R. China, and Excellent Youth Fund by National Science Foundation of China. Contact: College of Public Administration, Zhejiang University, Hangzhou, China Wang Fangrui Research Center of Innovation Management (RCID), School of Management, Zhejiang University, Hangzhou, China

Torsten Eymann, Christoph Niemann and Falk Zwicker: Innovating mindfully in healthcare IT using RFID technology 1

Introduction

New information technologies usually offer new, innovative opportunities for businesses. Using the example of hospitals, innovative clinical processes can improve medical quality and thus save lives. In addition to the improvement of medical processing, information technology promises to solve economic challenges like process redesign or data handling. However, introducing innovations in the field of information technology (IT) differs from other investments by a different style of approaching and handling the technology. Many projects seem to be characterized by mindless behaviour of the management instead of being carefully considered. A mindful institution moves to IT innovation considering their own specific requirements and organizational facts. Many innovation projects, however, seem to be a result of a direct copy from other businesses, a "me too" variety, carrying the hope that the benefits of the original pertain directly also to the copy. This "trial and error" attitude is particularly dangerous, if we look at supporting life-critical processes in the healthcare sector. In this article, we will first explore differences in innovative behaviour in the field of IT innovation. Therefore we contrast the specifics of mindful and mindless behaviour and analyse their causes. Then we discuss the relevance of IT innovation and discuss the phenomenon in the context of implementing radio frequency identification (RFID) technology in hospitals. Following these elaborations, we present a layer framework in order to evaluate the success of innovating with RFID. We explain our framework using the example of an RFID implementation and provide references for a successful implementation in the field of hospitals. In the course of the discussion we clarify the advantages of RFID. We conclude our paper by giving references for further research activities.

2

The IT innovation phenomenon

2.1 Mindlessness in IT innovation Although there is hardly any doubt about the generation of competitive advantages from IT investments, it is still a complex and crucial question for managers whether and when to

122

2 Corporate management of innovations

innovate with IT (Swanson and Ramiller 2004). Many organisations do not have either the competence or the motivation to make such decisions in an appropriate manner. They often couple decisions on IT investments to the behaviour of other institutions. Such behaviour can often be observed in practical IT cases, and can be characterised as "jumping on the bandwagon" phenomenon (Abrahamson 1991). By showing "bandwagon" or "me too" behaviour, institutions tend to copy innovative behaviour or success stories from others. This should be regarded as commonplace and unsurprising. For many institutions it is an insolvable challenge to determine their specific needs for IT innovations. They are not capable to fully adopt innovations to the particular circumstances of the organisation itself. Following some widely touted "best practise" they try to avoid mistakes in a field, which they largely not understand (Swanson and Ramiller 2004). 2.1.1 Characterising mindlessness Organisations are particularly mindless in innovating with IT, when their actions lack attention to organisational specifics. The following characterisation of mindlessness is essentially based on research by Swanson and Ramiller. Neither an enterprise nor a hospital represents the full bandwidth of mindless behaviour. Often they attempt to act mindfully, however achieve the opposite because of their lack of operating experience in the field of innovating with IT. Non-innovator management Mindless institutions pay insufficient attention to identify and explore new IT innovations. For them, innovating with information technology is not classified as a crucial success factor. Therefore they wait for innovations to come, generated by prior adopters. Impressed by success stories that seem to verify the innovation as beneficial, institutions categorise the assimilation of new technologies as a simple roll out (Strang and Macy 2001). Mostly, such mindless adoption leads to mindless implementation. Mindless implementations first affect end-users, which usually are left with their fate. Frustration and confusion are not understood by the management and mostly attributed to shortcomings of the users themselves. But also if frustration und resistance takes place, often the management will not abandon the once taken approach to finish the implementation of innovation in the organisation. Rather, they will arrange themselves with the pathetic result of the implementation und try to come up with a somehow functioning system by creating workarounds for obvious shortcomings (Nelson and Winter 1982). In addition to end users' efforts, mindlessness is also influenced by conditions within an organisation. Swanson and Ramiller distinguish between three types: attention deferral, contextual insensitivity, and institutional preemption. IT innovation questions, like all managerial decisions, compete for management attention. Since IT innovation is often not a pressing issue, as the returns are more longterm, the consequence can be attention deferral. Only little time is granted by the organisation's leadership to control the adoption process and to solve unexpected problems. The result is adoption

Eymann, Niemann and Zwicker: Innovating mindfully in healthcare IT using RFID…

123

suffering from fast-paced decision making, thus resulting in bandwagon behaviour (Swanson and Ramiller 2004). Innovations in IT can lead to conflicts when an organization has to adapt to new procedures. Contextual insensitivity defines a state where the management takes much of the current organizational structure and processes for granted. Running processes, having become routine, narrow attention and thus reduce willingness for changing circumstances (Nelson and Winter 1982). In addition, mindlessness might also be caused by management attempts to force reengineering business processes by implementing new IT, thus abandoning innovation-specific questions (Champy 1997). Further reasons for mindless behaviour are institutional preemptions. Organisations sometimes can be forced by external effects to adopt IT innovations. For example, organisations must act on pressure from other organisations they are connected with to adjust their IT and/or their Enterprise Resource Planning (ERP) systems, e.g. when large companies only accept suppliers with running ERP systems (Hart and Saunders 1997). To stay with the example ERP, it can be difficult for organisations to imagine alternative ways for IT innovation than the commonly accepted strategic business imperative (Keller 1999). Often there are only limited chances for organisations to resist such preemptions which are potentially resulting in mindlessness. 2.1.2 Risks of mindlessness in healthcare Mindlessness in healthcare is potentially more dangerous than in other fields, because mistakes are usually fatal. In particular, hospitals are high reliability organisations (HROs) where the scale of consequences prohibits learning by experimentation. Erroneous or inadequate IT could mean the loss of lives. Although it is clear that mistakes resulting from mindless innovation are fatal, there are further undesired effects which are outlined in this section. The primary recipient for negative effects resulting from mindless innovation is the end user. In hospitals individual healthcare services are defined under great mental stress of the employees. If they are unconfident, frustrated and confused by the new technology, it is hard to imagine that they will totally adopt it. Due to this, reasonable adoption of IT implies reasonable running IT processes. That, however, states a potential source of errors. A further important issue is data quality. A failed adoption of IT may effect data consistency. A failed implementation of a clinical information system, for example, could implicate that the integration of clinical data fails (e.g. data from medical records). The inconsistent data resulting from such a process will cause unreliable outcomes. 2.2 Mindfulness in organizational IT innovation For some time, management research has already focused on mindful innovation and its characteristics, starting with analysing individuals' innovation to organisations, and to high reliability organisations (Weick and Sutcliffe 2001). HROs, like hospitals and other healthcare organisations operate in an environment in which it is easy to make mistakes and where

124

2 Corporate management of innovations

errors normally result in fatal consequences. Learning by experimentation is untolerable and thus organisational mindfulness should be a basic requirement (Weick et al. 1999). Conceptual foundations Originally, mindfulness is a psychological notion that reflects upon the cognitive qualities of the individual. Characteristic features of mindful behaviour, which also can be classified as mindful management behaviour, involve openness to novelty, alertness to distinction, sensitivity to different contexts, implicit awareness of multiple perspectives and orientation in the present (Langer 1989; Swanson and Ramiller 2004). Generally expressed, mindful organisations practise adaptive management of expectations in the context of the unexpected (Swanson and Ramiller 2004). Vigilance to unexpected events arises naturally from management's function. In the context of IT innovation it implies close alertness and the ability to respond to sudden events. The organisation itself avoids attention deferral. Therefore, the organisation pays attention to new information technology's fit to one's specifications. The decision for and the design of IT innovation is grounded in local facts and specifics (Swanson and Ramiller 2004). In anticipation of the unexpected the mindful organisation never decreases its alertness. It is always considering the possibility of failure. Therefore, it withstands the possibility to interpret events as simple actualities and consciously watches the complexity of IT innovation. In this regard the mindful organisation attends consequently to inconspicuous events in the operational business (Weick 1995). Mindful behaviour in organisations usually comes along with a flexible organisational structure. Because it is impossible in the context of IT innovation to count on every possibility, the mindfully organisation favours improvisation over planning and adaptation over routine (Swanson and Ramiller 2004). 2.3 Innovating mindfully in healthcare using RFID As illustrated, mindfulness in HROs should be mandatory, especially when innovating with IT and expecting the unexpected. HROs must avoid jumping on the bandwagon. A current bandwagon, which travels at high speed like an express train through the media and the consulting business, is innovation using radio frequency identification technology (RFID). Not only for industrial enterprises but also for hospitals and other healthcare organisations, RFID seems to present a simple technological solution for a lot of business problems. But how can hospitals act mindfully while innovating with RFID? We suggest a model for implementing RFID in a healthcare environment as shown in figure 1. The potentials of RFID technology can be divided into four layers, which are described using the example in the next section. While the technical challenges increase with every level, also the possible impact of an application increases.

Eymann, Niemann and Zwicker: Innovating mindfully in healthcare IT using RFID…

decreasing quantity of data increasing quality of information

medical knowledge

125

decision support

Optimisation

Integration

Context Detection

Identification

Figure 1: Different layers for the evaluation of innovation (Zwicker et al. 2006)

3

Applying the framework to patient logistics in hospitals

The presented framework offers the opportunity to introduce RFID technology into the hospital domain, while examining the value and the mindfulness on every layer.In this section, we apply a more concrete view, which takes a system to improve patient logistics as a prototype for an RFID application. The system under examination, EMIKA ("Echtzeitgestützte Mobile Informationssysteme in Krankenhaus-Anwendungen", meaning real-time controlled mobile information technology in the hospital domain), uses RFID data to track and locate resources and actors in the hospital environment. The system propagates that data through the four layers in order to achieve a solution for a operation scheduling problem. It aims to allocate all resources in the hospital in a near-optimal manner, in order to reduce waiting times for patients, and idle times for medical staff and (potentially expensive) equipment such as electroencephalograms (EEG). EMIKA applies RFID technology to track, identify and locate every actor in the hospital. While RFID allows for further benefits, such as secure identification of patients to avoid application of mixed up medications, EMIKA focuses on the logistics perspective: Patients should arrive at the right treatment room on time, which should be equipped with the right resources. The schedules of medical staff should be kept up to date and changes in the schedule should be propagated in real time. The system also needs to inform physicians and nurses about changes in their respective schedules. Medical staff needs to confirm such changes, since the information system (IS) cannot actually decide on humans' behalf. The IS

126

2 Corporate management of innovations

lacks medical knowledge, which only humans can provide. The system, therefore, needs to implement a hybrid decision making model (Rammert 2003) to allow for a split decision making process, which generates schedule proposals in the IS. Medical staff approves or discards these proposals based on their current situation and their medical assessments. The following subsections describe the four layers of the framework and apply the example application. Different techniques are used to aggregate available data and gather necessary information on the different layers. Each layer passes its information to the next layer to allow further aggregation. 3.1 Identification On the lowest layer, EMIKA introduces RFID technology into the hospital domain in order to track and identify humans moving in the hospital. RFID technology consists of RFID transponders (tags) and RFID readers. While readers have a fixed location within a building, tags identify individual actors and must be with their particular actor at all times. Since tags are very small, they could be embedded in identification badges for medical staff or in wristbands for patients. The identification layer is the interface between reality and an IS: It tracks all mobile resources in the building and maintains a logical representation of the real world, creating the "logical world". IS act within the logical world and try to optimise the given situation with regard to their specific aims. Whenever the IS reaches a result during the optimisation process, it has to communicate that result to the real world. Thus, the interface between reality and logical representation is twofold: On the one hand, an IS must sense the real world through sensors, on the other hand it must affect reality through effectors. EMIKA implements theses parts using RFID technology to sense reality and personal digital assistants (PDA) as effectors, respectively. RFID readers are positioned at passages; mostly, they are installed at doors between rooms. This way, a person moving from one room to another triggers the reader at that passage. The IS uses this data to track a person throughout the building. To make the system as fault tolerant as possible, while minimizing the costs, each passage is equipped with two readers (Hohl, Nopper, Eymann and Müller 2004). Whenever a person moves through that passage, the two readers generate a read signal in succession. The system can deduct the direction of the movement by analysing the temporal order of the read events. The identification layer of EMIKA analyses read signals and associates the unique id of every tag with a person or mobile resource in the hospital. This information is passed on to the next layer within the framework, the context detection. 3.2 Context detection The context detection layer aggregates data of different sensors of the identification layer and infers the actors' contexts. Two read events of the same unique id, but at different readers indicate a movement of one actor from one room to another. The identification layer is not able to infer these movements, since it only deals with read events from one reader at a time.

Eymann, Niemann and Zwicker: Innovating mindfully in healthcare IT using RFID…

127

Context detection becomes possible because data of all sensors are put together in this layer. Thus, two read events in short temporal succession are interpreted as one movement. To reason about movements, the context detection layer must know about the position of all readers within the hospital. By combining the location information with the read events, it is possible to infer movements. However, all deductions in this layer are possible without the intervention of humans. All data is available in the IS. As the next step, the context layer transforms the movement information into location information. The context layer needs to augment the geographical location information with semantics, such as room numbers or the role of a room (cf. Pradhan 2000). This location information about all actors is passed on to the next layer. 3.3 Integration The integration layer combines the location information of the lower layers with schedules taken from the clinical IS. It becomes possible to decide whether the originally planned schedule for operations is still valid or needs to be rescheduled. Hospitals are highly dynamic systems, which need to adapt to unforeseen circumstances very quickly: Patients cannot be forced to take their scheduled medical treatments, and thus be late or may not show up at all. On the other hand, emergencies with the need of immediate treatment may come in. To provide for such immediate treatments, other patients have to bear waiting times, since the hospital cannot follow the original schedule. The integration layer is able to invalidate the original schedule if the necessary resources cannot be allocated. Whenever the schedule becomes invalid, the optimisation layer must generate proposals for new schedules, which eliminate the error in the original schedule. 3.4 Optimisation The optimisation layer eliminates errors in the schedule by generating proposals for new schedules. The decision, that a newly computed schedule becomes the agreed upon schedule, cannot be taken by an IS automatically. Medical knowledge, which is impossible to encode, is more easily applied by evaluating the proposals by humans before applying them. Therefore, the IS generates proposals, which need confirmation of all involved actors. To generate new schedules instantly, the optimisation layer maps the decision problem to an artificial market (Eymann, Sackmann and Müller 2003). On this market, software agents negotiate the allocation of resources. The paradigm of software agents in a multi agent system (MAS) is flexible enough to model highly dynamic systems, such as hospitals. Indeed, flexibility is often seen as the most important characteristic of MAS (Kirn 2006). In the logical representation of the real world, every actor is represented by a software agent, which tries to optimise the schedule of its principal. Patients' software agents have a demand for resource usage in the hospital. The software agent tries to allocate these resources on the market with its given budget. The budget encodes two distinct characteristics of the patient:

128

2 Corporate management of innovations

• Priority of treatment: If a patient needs immediate treatment, he is given a high initial budget to be able to buy every needed resource immediately. • Waiting time: If a patient cannot buy the necessary resources at once, he has to wait until they become available. During this waiting time, the budget increases. After some waiting time the budget is high enough to enable the patient to buy all necessary resources. To match supply and demand, all resources need a price, which is an indicator for their availability. The EMIKA system does not rely on the market itself to determine a price for resources but allows resources to set their fixed price. A supplier in the artificial market (physician, nurse or medical equipment) determines its price dependant on its current activity: If a resource is idle, its price decreases with time to indicate availability. If a resource is busy in a treatment, the price increases to prevent other software agents to buy the resource on the market. A detailed description of the budget/price model is available in (Niemann and Eymann 2006). If a software agent has triggered a negotiation on the market and has secured all necessary resources, the involved software agents must communicate the decision proposal to their human principals. The human actors reserve the right to confirm or decline a decision on basis of medical knowledge. If humans decline a proposal, the software agent starts a new negotiation process until it has allocated all necessary resources for the treatment of it principal, the patient. Once the agents agree on a proposal, they communicate with their human counterparts via PDA and ask for permissions for the proposal.

4

Conclusion

This paper proposed a framework for mindful innovation with IT. The healthcare domain is in particular need for mindful innovation, since mindless innovation processes can have fatal consequences, which have to be avoided. The introduced framework consists of four layers, which allow dividing the innovation process into different stages, which can be evaluated separately. The project EMIKA aims to solve the ongoing scheduling problem in hospitals, which are highly dynamic systems, by applying the framework. First evaluations of the prototype showed promising results; the system generated schedules, which solved the scheduling problems. By applying the framework and thus dividing the problem in the four different layers, each layer accomplishes its specific goals, which are prerequisites for the higher layers. The prototypical system avoids the common bandwagon effect as each hospital has to consider its specific configuration before applying the system. If a hospital introduces such a system, it is better equipped to expect the unexpected, since the systems allows for instant rescheduling processes, which cannot be done on a manual basis, if the hospital is sufficiently large. We expect the framework to be applicable in more general terms to other domains as well, since it is not bound to any specific technology, which has been used in the EMIKA project. The RFID technology can be used for different aims, which are not the focus for the EMIKA

Eymann, Niemann and Zwicker: Innovating mindfully in healthcare IT using RFID…

129

project such as secure patient identification. Such an application would not need to implement all layers of the proposed framework, but could use only the lower two layers, because there is no need for further processing of the data. The more general application of the framework needs to be evaluated on greater scale in a more universal way. Such an evaluation needs to be done empirically and could be the focus of further research. After implementing the prototype system, the framework still needs to prove its validity in more general terms.

References Abrahamson, E. "Managerial Fads and Fashions: The Diffusion and Rejection of Innovations," Academy of Management Review Vol. 16 No. 3, 1991, pp. 586-612. Champy, J. "Packaged Systems: One Way to Force Change," Computerworld, December 22, 1997, p. 61. T. Eymann, S. Sackmann, G. Müller, I. Pippow (2003): Hayek’s Catallaxy - A Forwardlooking Concept for Information Systems? Proc. of America’s Conference on Information Systems (AMCIS-2003), August 2003, Tampa, USA. Hart, P, and Saunders, C. "Power and Trust: Critical Factors in the Adoption and Use of Electronic Data Interchange," Organization Science Vol. 8 No. 1, 1997, pp. 23-42. Hohl, A.; Nopper, N.; Eymann, T.; Müller, G.: "Automatisierte und Interaktive Kontextverarbeitung zur Unterstützung der Patientenlogistik". In: Kirn, S.; Anhalt, C.; Heine, C. (Ed.): Mobiles Computing in der Medizin – Proceedings zum 4. Workshop der GMDSProjektgruppe Mobiles Computing in der Medizin, Universität Hohenheim, Stuttgart, 19.-20. April 2004. Shaker Verlag: Aachen, 2004, pp. 14-26. Kirn, S.: Flexibility of Multiagent Systems. In: Kirn, S.; Herzog, O.; Lockemann, P.; Spaniol, O. (Ed.): "Multiagent Engineering – Theory and Applications in Enterprises". Springer: Berlin 2006, pp. 53-69. Keller, E. "Lessons Learned," Manufacturing Systems, 1999, pp. 44-46. Langer, E. "Minding Matters: The consequences of Mindlessness-Mindfulness," in Advances in Experimental Social Psychology (22), Berkow, L. (Ed.), Academic Press, 1989, pp. 137173. Niemann, C.; Eymann, T.: "Softwareagenten in der Krankenhauslogistik – ein Ansatz zur effizienten Ressourcenallokation". In: HMD: Praxis der Wirtschaftsinformatik 251 (2006), pp. 77-87. Pradhan, S.: "Semantic Location". In: Personal and Ubiquitous Computing 4 (2000), pp. 213216. Rammert, W.: "Die Zukunft der künstlichen Intelligenz: verkörpert – verteilt – hybrid". Working Paper TUTS-WP-4-2003, Technische Universität Berlin, 2003.

130

2 Corporate management of innovations

Strang, D. and Macy, M. "In Search of Excellence: Fads, Success Stories, and Adaptive Emulation," American journal of Sociology Vol. 107 No. 1, 2001, pp. 147-182. Nelson, R. and Winter, S. "An Evolutionary Theory of Economic Change," Harvard University Press, Cambridge, MA, 1982. Swanson, E.; and Ramiller, N. "Innovating Mindfully with Information Technology," MIS Quarterly Vol. 28 No.4, 2004, pp. 551-583. Weick, K. "Sensemaking in Organizations," Sage Publications, Thousand Oaks, CA, 1995. Weick, K., Sutcliffe, K. and Obstfeld, D. "Organizing for High Reliability: Processes of Collective Mindfullness," Research in Organizational Behaviour (21), 1999, pp. 81-123. Weick, K. and Sutcliffe, K. "Managing the Unexpected: Assuring High Performace in an Age of Complexity, Jossey-Bass, San Francisco, 2001. Zwicker, F., Reiher, M., and Eymann, T. "RFID im Krankenhaus – neue Wege in der Logistik," Eymann, T., Hampe, F., Koop, A, and Niemann, C. (Ed.) "Mobiles Computing in der Medizin," Shaker, Aachen, 2006, pp. 68-74

Authors Torsten Eymann, Prof. Dr. Head of department [email protected] Christoph Niemann, Dipl.-Kfm. Scientific research assistant [email protected] Falk Zwicker, Dipl.-Kfm. Scientific research assistant [email protected] Chair of Information Systems University of Bayreuth Universitätsstraße 30 95440 Bayreuth, Germany Tel.: +49 (921) 55-2807 Fax : +49 (921) 55-2216 www.wi.uni-bayreuth.de

3

Network and cooperation in innovation

Hu Lijun and Shi Junwei: Market structure, social structure and technological innovation: An interpretation of firm behavior in the transitional stage of China Abstract A bone of contention of the relationship between market structure and technological innovation is that publications focus much more on the choice of market structure. This paper induces the social structure to unpuzzle the firms’ behavior of technological innovation, and argues that a firm is a special actor embedded in the corporate social network. The technological innovation behavior of firms is, hence, a kind of social action, and affected by the interpersonal relationship and the social structure. As a result, social structure impacts on the market structure and both structures together act on the process of technological innovation, in which entrepreneurship plays quite an important role. The article discusses some characteristics of technological innovation of firms in the transitional stage of China, and provides some policy suggestions. Keywords: social structure; market structure; technological innovation; firm behavior; economic transition

1

Introduction

Since innovation was considered by Schumpeter as a major driving force in economic growth more than 60 years ago (Schumpeter, 1942), the relationship between technological innovation and market structure has been under discussion in the academic circle. Yet no

132

3 Network and cooperation in innovation

consensus has been achieved. According to classical economics and neo-classical economics, perfect competition is the most favorable market structure for technological innovation. Monopoly is another word for low efficiency in resource allocation, because in a monopolistic market structure, big players can raise the price and reduce the production to make profit without technological innovation which would only lead to failures in technological advancement. But according to the mainstream economists, there is an equilibrium of innovation between competition and monopoly. Schumpter (1942) points out that the economic development based on technological innovation is a long-range and non-equilibrium process. Contrary to the traditional idea that sizable firms lead to monopoly and monopoly leads to failure in technological innovation, he argues that the bigger the firm is the more it can provide a favorable environment for technological innovation, and technological innovation under perfect competition is inefficient. Although this is an attractive new argumentation, different voices appear. Raider (1998) argues that competition is a function of market structure which is within and beyond the market. From the perspective of network theory, he considers competition as an organizational structure of an innovation network that incorporates suppliers and consumers. When it comes to Schumpeter, he points out that technological innovation is more likely to happen in small firms. He proves his claim through empirical researches that in the highly constrained industries firms are more active in research and development under competition, therefore, technological innovation is more common in small firms. After analyzing the horizontally differentiated products under the market structures of monopoly and duopoly, Wickelgren (2004) figures out that market competition can help firms with their local non-contractible investment. He also thinks that due to the holdup problem, there is not much incentive for the firms to improve their product quality, but this could be solved to some extend with competition. Eicher and Kim (1999) introduce the variables of cost and market share to the product variety model and find out that in the high-tech industries, competition in the product market promotes technological innovation. Still there are some scholars who support Schumpeter. They argue that market concentration (on monopoly) is in favor of technological innovation. Through the Cournot Model, Yi (1999) finds out that the gains from small process innovation decrease with the increase of the number of small-and-medium firms, that is to say that market concentration is in favor of technological innovation. Liu & Wan (1997) examine the industry concentration in 16 Chinese industries with CR4 and conclude that there is a positive correlation between concentration and industry profit rate, thus highly concentrated industries can invest more heavily in R&D. In other words, there is a positive correlation between investment in innovation and concentration. The structure of their empirical study illustrates that in Chinese industries, oligopoly favors technological innovation more than monopoly. According to Nielson (2001), the statistics of patents in Denmark on manufacturing between 1996 and 1998 indicate that industry concentration urges firms to apply for patents and invest in R&D. Some neutral arguments appear sometimes. For example, after studying the technology and innovation data from 74 countries’ telecommunication industry between 1991 and 1995, Madden and Savage (1999) report that market size has a positive impact on technological innovation while there is no significant positive correlation between market structure (concentration) and innovation. When the above mentioned literature is reviewed what has been found to be interesting is the following: no consensus has been achieved on the relationship between market structure and

Hu and Shi: Market structure, social structure and technological innovation

133

technological innovation. This is especially true when it comes to the questions of which market structure favors technological innovation most. The author figures out that what is behind the disagreement is that the discussion on technological innovation has for long limited been on the framework of market structure. No other inducement of technological innovation has been discussed. Social structure is introduced to the study on technological innovation in the hope that a new explanation could be found. In this paper the author argues that market structure is a subsystem embedded in social structure, and tries to explore the following issues: How does the market structure function with respect to technological innovation? Is market structure the only factor that has an impact on technological innovation in firms? How do social structure and market structure work together? And which of these two has more impact on technological innovation in firms? Is there any difference between different environments?

2

Theoretical development: The role of the social structure in the process of technological innovation

The sociological construct of technological innovation Science and technology are playing a more and more important role in economic development, and, by far, the core of technological innovation theory is technology, which includes not only knowledge but also learning. In this paper, technological innovation will be firstly examined from the sociological perspective. Sociology is a science studying human behavior; it examines human social relations, cause and effect between different individuals and groups of individuals. It also covers the social structure, social norms, the social system and the interactions between individuals and organizations. From the sociological perspective, the technological innovations in firms not only interact with the sub-economic system, but also with the whole social system which deserves more concern. In this regard, we define firms’ technological innovation not only as a kind of technological behavior, but also as a kind of social action embedded in a social structure. Compared with market structure, social structure is more of a macro operation system through which different components and factors of the social system are solidly combined with each other. Its material characteristics indicate that social structure is in fact a rather stable network that is constructed by material social factors and their relations. Such a structure includes both institutional and non-institutional structures (Li, 1995). In relegation to the term ‘network structure’, social structure is a stable social network. Different people’s position in the network is decided by their measurable reputation and economic return (wealth, status and power) (Lin, 1978). The large amount of information, opportunities, reputations, norms, trust and rules existing in the network are what has been referred to as social capital. Unlike the economic view which considers a firm as a purely economic group, in the network structure, a firm is a special social actor that has been ‘embedded’ in the network. The different explanations of a firm indicate that firms access to, utilize and allocate favor-

134

3 Network and cooperation in innovation

able resources through the social structure as well as the market structure, and these resources naturally promote their technological innovation. Under traditional economic conditions, technological innovation is realized step by step through separated work in research, development and production. But in the new economy, successful technological innovation relies on a dynamic production relation or a network which values cooperation. Moreover, the whole process of technological innovation which includes factors as such innovation motivation, operation system and policy environment interacts inevitably with people factors (such as innovation idea, innovation ability and innovative cooperation). We can say that technological innovation is full of interactions between individuals and the social structure. They have become the internal variables of technological innovation activities and thus affect their process and consequences. In short, we do not think that technological innovation activity is a simple linear model. The complexity of people and social relations leads to the uncertainty of technological innovation. That is why the social network is decisive in the mode, direction and development of technological innovation in a firm. Technological innovation is a long and complicated process, and any part of this process is likely to be affected by other social factors. The technological innovation system of firms: social structure and market structure The innovation system theory indicates a transition from individual study to systematic research. The innovation system theory (Lundvall, 1992; Metcalfe, 1995) combines the dynamic and complexity of innovation. According to this theory, technological innovation interacts with institutional innovation, and the key factors in this system include knowledge, learning and ability. The theory also tries to interpret the complicated system behavior from a “micro base”. In the knowledge system, the technological innovation system theory focuses on the function of tacit knowledge, because it is difficult to encode and transfer such knowledge. Tacit knowledge is the most valuable link during the technological innovation process in a firm and thus is taken as the base of a firm’s competitiveness. In learning activity, in addition to “learning by doing” and “learning by using”, the functions of specialized knowledge comprise workers, basic technological structure, educational structure and communication structure. Unlike the price decision and quantity decision that can be observed in a market structure, the information (especially the tacit information) is invisible and is spread more likely through a social relation network. The technological innovation system in a firm is considered as a special network structure in this paper, which consists of some certain nodes. These nodes mainly include social actors that have a close relation with a firm’s activities, such as suppliers, dealers, competitors, cooperators, potential rivals, consumers, government, communities, universities, research institutions, and so forth. What is flowing between these nodes are factors like innovation information, knowledge, learning opportunities, social norms, trust and reputation. These factors interact with the technological innovation activities in both market structure and social structure and eventually lead to the birth of technological innovation. And the relation structure of these social actors (e.g. who is dominant) directly reflects the social structure (see Figure 1).

Hu and Shi: Market structure, social structure and technological innovation

135

Figure 1: Technological innovation with the constraint of social structure and market structure

In figure 1, the market structure works on a firm’s technological behavior by means of market variables — entry/exit, price, output, R&D competition, advertisement, etc., while the social structure utilizes information, knowledge, flow of opportunities, social norms, reputation and trust to affect a firm’s technological behavior. Although the market structure is composed of some tightly related actors including suppliers, it is influenced by the social structure factors of the government, the community, and so on, from which firms acquire information, knowledge, innovational chance, reputation and other resources. That is to say, social structure both affects directly and works indirectly on the firm’s technological innovation through the market structure. As some literature suggests, technological innovation demands for a market structure as its platform. But as this is conditional, the impacts of the market structure on the technological innovation may depend on some other factors such as technological opportunities and entry timing, so that it is more likely that they are interactive. Such a relationship is complicated. In a broad sense, the market structure is also embedded in the social structure. Therefore, social structure’s (such as social relation, institutional environment, social traditions and values) functions on technological innovation in a firm are our primary concern in this study. Two market structures are examined here, one is competition in which most players are small firms; the other is monopoly in which a large firm coexists with many small players. In the first case, being limited by their size, these firms are rather weak at independent technological innovation. There are three alternatives for these small firms: they can become stronger through a merger; or they can ask the government for innovation funds and policy support; or with the help of private relations, they can look for cooperators in the industry or research

136

3 Network and cooperation in innovation

institutions. The firms’ social relations are more decisive than the market variables in the last two alternatives. In the second case, things are more complicated. A big player could become a leader of technological innovation in the industry by investing heavily in R&D, and its status as a monopoly is strengthened. Also, it can protect its monopoly profit by seeking for governmental support at the price of an inefficient technological innovation. For the small firms in this market structure, there are two choices, either to be the market followers or, through their own effort or external support, to grow by acquiring more efficient technological innovation. The final decision depends on which is richer, the firm’s social structure resources or the market structure resources. For a firm, no matter which industry it is in, a certain market structure classifies the players in the market. The elite firms are the market leaders, and some firms form the second group, while those small players could only survive in the niches. That is, in a market structure which cannot be changed in the short term but in the long run, changes are possible. Unlike the social structure, market structure actually is a kind of dynamic relationship of quantity and technology. Social structure, as stated in sociological theories, is a rigid relation of norms. Most social behaviors do not modify the social structure they incorporate, but reflect on it. Entrepreneurial and technological innovation The entrepreneur is a core factor in a firm’s technological innovation and competitiveness; it is a necessity in technological innovation. In the economic development, innovation opportunities and potential profit exist in different industries. But such profits cannot be found, or developed, or possessed by everyone. Profit is available only through a firm’s or an entrepreneur’s innovation. That is why in Schumpeter’s point of view, innovation cannot be separated from a firm and its entrepreneur. According to Peter Drucker, innovation is the nature of entrepreneurship. When innovation activity takes place in a firm, the entrepreneur is the innovator. Although Schumpeter’s innovation theory is based on economics, his interpretation, especially his emphasis on firms and entrepreneurs, is beyond traditional economics. In fact, it is Schumpeter who first realized there is a fourth fundamental science, i.e. economic sociology. In practice, not only a firm is embedded in the social structure as an organization, but also the entrepreneurs’ innovation activities are socially embedded. That is to say that the technological innovation the entrepreneurs involve is actually a social action, too. Take innovation information for example, it is the firms that look for innovational information. There are some obstacles in innovation information spread in practice, such as production costs, information perception, ways of offering, the location of sharing and regulation. They are not doing a good job if the entrepreneurs just access information required by innovation decision through their observation of the market structure and their intuition. For we know that most of the innovation information is embedded in the social structure or the social relations network, other people are more motivated than the entrepreneurs to access this information by all means, and this is also an effective way to promote the firms’ innovation efficiency. According to our empirical study on Chinese firms, the human capital of entrepreneurs has

Hu and Shi: Market structure, social structure and technological innovation

137

made no significant contribution to the firms’ performance, while the contribution from the entrepreneurs’ social capita is huge. This could be an indirect evidence to support that the entrepreneur’s social relations are very import to a firm’s innovation performance.

3

Discussion and application: The technological innovation behavior of firms in transitional China

China has been in a transition ever since its reform and opening up. If we consider the ‘863 Plan’ as the starting point of a large-scale systematic innovation behavior, the theories and practice of technological innovation have a history of 20 years in China. During this period, both market structure and social structure have been influencing the technological innovation in firms in a different way. The change of social structure and technological innovation ‘Guanxi’ is a very important concept when people are trying to understand the Chinese social structure and social psychological behavior. In its traditional society, the Chinese social structure is a society with ‘ethics’ or ‘guanxi’ as its standards and is organized in a ‘differentiated order’ (Fei, 1947). Chinese people categorize their interpersonal relations into close and distant groups on the basis of the logic of ‘family and outsiders’. China had been employing a kind of highly centralized planned economic system since 1949, which was established on the basis of the public ownership of the means of production. In such a system, almost all of the scarce resources have been controlled by the government. Under this circumstance, the institution, administration and identification took the place of ‘family standards’ and became the allocation mode in a social resources redistribution system. Among all the social subjects, the government had become the core subject in the various social relations, thus, the Chinese social structure is characterized as a top-down vertical hierarchical authority. Therefore, administration authority dominates the different social relations that are involved around it. In such a social structure, no influence from the market structure on the technological innovation could be traced down, because the administrative intervention replaced market rules. There was no real firm or market and the law of value didn’t work. Almost all the technological innovation was promoted and distributed by the government and there was no room for the civil technological innovation force to develop. One of the main goals of reform and opening up is to establish a socialist market economy with Chinese characteristics whose core task is to reestablish firms as the subjects in the market economy and real corporations which would actively work on technological innovation. Market structure is the breakthrough point in China’s economic institutional reform. The government has put great effort into the reform of different aspects of market structure, for example market barriers, price regulation, taxation reform and technological innovation. One of the achievements of the reform is that the pressure from the market has become the motivation and incentive for the firms to innovate their technology. Still, a reform in market structure alone is not sufficient for the technological innovation system in firms; the related social structure ought to be reformed too. Our empirical study on the firms’ R&D behavior reveals that even in today’s

138

3 Network and cooperation in innovation

private firms in China, their investment strategy of R&D is directed by governmental force to a large extent (Shi, 2006). Innovation investment and technological innovation: government versus firm The Chinese government announced its development goals for future investment in technology in the National Guideline on Medium-and-Long Term Program for Science and Technology Development: the investment in R&D in 2010 will account for 2% of the GDP, and 2.5% in 2020. Statistics show that investment in technology was 196.63 billion yuan (about 24.5 billion dollars) in 2004 which is only 1.23% of the GDP. This proportion is larger than that of developing countries such as India and Brazil, and China ranks 6th in the world. The world average investment proportion is 1.6% of the GDP and in developed countries this ratio raises to 2.2%. China still has a long way to go (see Figure 2).

2.5

2.2

2 1.5

1.6 1.23

1 0.5 0 China

average of globe

average of developed countries

Figure 2: R&D/GDP (2004) of different countries (%)

In China, governmental investment played a leading role in directing and allocating social technological resources before 2000. Things have changed since the economy began to benefit from the reform. In 2000, for the first time, the firms’ investment in technological innovation accounted for more than 50%, which indicated a significant change in China’s R&D investment structure, and the firms, instead of the government, became the predominate force in R&D investment. Figure 3 shows that in 2000, the firms’ investment in R&D accounted for 57.6% of the total social R&D investment, and governmental investment decreased to only 33.4%. This ratio reached 66.8% in 2004 and investment of the government decreased to as little as 32.2%. It can be implied from these figures that in China’s economic develop-

Hu and Shi: Market structure, social structure and technological innovation

139

ment, firms are playing a more and more important role in the upgrade of national technological innovation competitiveness.

70

66.8

60 57.6 50 40 30

32.2

33.4

20 2000 share of enterprise R&D

2004 share of government R&D

Figure 3: The Change of the share of R&D investment in China: 2000-2004

Most of the R&D investment is coming from firms, but it does not mean that the firms’ behavior is the core of their technological innovation in China. Investment in technology is taken as a national strategic investment in China and the sources of fund are governmental financial input, technological tax break, venture capital, bank loans, capital market financing and firms’ investment in R&D (Shen, 2006). Five of the above-mentioned six sources are controlled by the government. Besides, the government can influence thte firms’ investment as well. Therefore, despite the increasing attention given to firms’ technological innovation, in the short run, no significant change will appear in the current social structure whose main characteristics is hierarchical authority. The roles of social culture and social capital in firm technological innovation China’s transition began in 1978 and it mainly includes the transition of the economic system and the transition of the social structure (Li, 1997). The goal of the transition of the economic system is to establish a new market economy. As to the transition of the social structure, it is far more complicated since it is a comprehensive structural transition other than a transition of a sub-system. Due to limited space, what is discussed in this paper only covers the transition of social culture and social capital. Functions of social culture Social structure transition is a must for a country’s modernization, but the decisive factor in modernization is not to deny cultural tradition but to inherit and develop the social culture

140

3 Network and cooperation in innovation

system towards the process of transition. Confucianism and ‘Golden Mean’ have been considered as the classical characteristics of the Chinese social culture, their gist is balance and rule-following which constrain innovation in Chinese society. The central government, in the recent five years, advocates strategic goals like ‘Scientific Development Philosophy’ and ‘to build an innovative country’ to cultivate a kind of innovative social culture. The cultural transition of a society usually indicates the change of its social values and such a change may have an impact on the actors from different social classes, especially the corporations. An ‘innovative country’ requires every citizen’s innovation and wisdom to make the country more energetic and competitive. For entrepreneurs, innovation is their spiritual nature; a social culture that advocates innovation will for sure cultivate a lot of entrepreneurs with innovational spirit, and these entrepreneurs will then promote the whole country’s innovation efficiency. In China, real entrepreneurs mostly do not work in state-owned-enterprises (SOEs), but in private enterprises, FDI companies and corporations. Figure 4 and figure 5 demonstrate that entrepreneur groups in China grow with the development of non-SOEs (especially private enterprises), and the number of SOEs has been decreasing rapidly during the past five years, and this phenomenon is in consistence with the reform direction of the central government. We can also think that China is training its entrepreneur group, and fostering a fundamental talent base for the technological innovation strategy. 250 200

198.2

150

132.4

100

85.9

50

45.6 40.6 19.2 15.2

37.1 30.0 13.9

0

2001 private

SOE

2005 FDI

Collective

corporation

Hu and Shi: Market structure, social structure and technological innovation

141

Figure 4: Entrepreneur group in China: 2001-2005 (unit: ten thousand)

5.5

4.7

5.8 61

12.5

10.5 private

Collective

Corporation

SOE

FDI

Others

Figure 5: Entrepreneur group in China: 2005 (%)

One more thing about innovation is that it is not just the innovation of science and technology, or the technology transition in the market competition, nor is it the innovation of certain organizations and individuals; the purpose of innovation is to cultivate a social culture that promotes innovation. When this culture of innovation settles down in the social structure, the competitiveness of technological innovation of this society will be promoted. Functions of social capital According to Coleman (1988), the so-called social capital is a kind of structural resource which will bring actors benefits. Both the reform of the market structure and the transition in the social structure share the same nature of institutional transition. Such an institutional transition will lead to the formation of an institutional structure. During China’s transition, the old institutions have already been abandoned while the new institutions have not been strong yet, thus institutional holes appear. One of the consequences is the prevailing an unofficial institution built on interpersonal relations. This is the basic logic of how the social capital works (Bian & Qiu, 2000). As discussed above, due to the government’s leading role in a firm’s technological innovation, when it is not possible for a firm to achieve technological innovation through observing the variables of the market structure, it will turn to the social structure that is rich of invisible information. When more and more organizations and individuals benefit from their social relation resources, firms will have to find technological innovation information and opportunities through their private social relation network or seek for the government’s support through their social capital. Even in the prevailing virtual

142

3 Network and cooperation in innovation

operation mode (Hu, 2001), social capital becomes the most effective glue in technological innovation cooperation among firms.

4

Concluding remarks

China is speeding up its social structure transition; it is especially true that in the context of globalization and the WTO entry, such a social structure transition becomes a powerful force in promoting the reform of social development and impacting the market structure in which firms’ technological innovation behavior exists more seriously. This will present new challenges to the industrial structure adjustment. The transition of social structure may also lead to the transition of the resources allocation and ideology in the society, thus a more positive basis could be built for the economic system reform. However, this effect neither is being paid enough attention to yet, nor is it being fully discussed. In today’s China, the government is so dominant in the economic development that the firms’ behavior depends too much on the government even in a market economy. It is not surprising to encounter these problems in a transitional economy, and with time passing by, they will be solved. After all, firms, as the subject of technological innovation, have already begun to play on the stage of economic transition in China and they will present us a satisfactory performance. As far as the government is concerned, it not only needs to provide a free, perfect “market”, but also a more effectively harmonious “society” for firms in China.

References Bian Y. & Qiu H., 2000, Social capital of firms and its efficiency, Social Sciences in China, 2: 87-99. Coleman, J. , 1990, Foundations of Social Theory. Cambridge, MA: Harvard University Press. Eicher T. & Kim S., 1999, Market structure and innovation revisited: endogenous productivity, training and market shares, working paper. Hu L. J., 2001, Virtual Enterprise, Wuhan: Wuhan Press. Li P., 1994, Another invisible hand: Structural Transition in Society, Social Sciences in China, 1: 84-94. Li P., 1995, Social Structure Transition: Sociological Analysis on Economics System Reform in China, Qiqi Ha Er: Heilongjiang People’s Press. Liu G. & Wan J., Analysis on the impact of Market structure on technological innovation, Journal of Management Engineering, 6: 10-14. Lundvall B., 1992, National system of Innovation: Towards a Theory of Innovation and Interactive Learning, New York: St. martin’s Press. Madden, G. & Savage S., 1999, Telecommunications productivity, catch-up and innovation, Telecommunications Policy, 23: 65-81.

Hu and Shi: Market structure, social structure and technological innovation

143

Metcalfe S., 1995, The Economic Foundations of Technology Policy: Equilibrium and Evolutionary Perspective, in Stoneman P. (eds.), Handbook of the Economics of Innovations of Technology Change, 409-512, Oxford Press. Nielsen A. O., 2001, Patenting, R&D and market structure: manufacturing firms in Denmark, Technological Forecasting and Social Change, 66: 47-58. Raider H., 1998, Market Structure and Innovation, Social Science Research, 27:1-21 Schumpeter J., 1942, Capitalism, Socialism and Democracy, London: George Allen & Unwin Publishers. Shen W., 2006, Scientific investment: The guarantee of independent innovation strategy, China Scientific Investment, 4: 1-4. Shi J. W., 2006, Social Capital and the Choice of Firm Behavior: A Theoretical Framework and Its Empirical Analysis in Chinese Context, Ph. D. Dissertation, ZNUEL. Wickelgren A., 2004, Innovation, market structure and the holdup problem: investment incentives and coordination, International Journal of Industrial Organization, 22: 693-713. Yi S. S., Market structure and incentives to innovate: the case of Cournot oligopoly, Economics Letters, 65: 379-388

Authors Hu Lijun is a professor of economics and management at Zhongnan University of Economics and Law, graduate school of business, and he is also the dean of Center for Modern Industrial Economy (CMIE) of the university. He received his Ph.D. from Zhongnan University of Finance and Economics. His research interests mainly include virtue enterprise, corporate strategy and industrial economics. Recent publications include three books, such as “Chinese Industrial Policies in the New Century” and more than thirty top journal articles in Chinese. Contact: [email protected], Phone: +86 (27) 88384189, Fax: +86 (27) 88384443 Shi Junwei is an associate professor of economics and management at Zhongnan University of Economics and Law. And he received his Ph.D. from this university. Professor Shi is also a postdoctoral research fellow at Guanghua School of Management, Peking University. His research interests focus on the corporate social capital, the theory of the firm, technological innovation and Chinese economic transition. He recently published a book on management effectiveness, more than ten top journal articles in Chinese and English articles in Frontier of Business Research in China and other journals. His new book Social capital and the Choice of Firm behavior: A Theoretical Framework and its Empirical Analysis in Chinese Context will be published by Peking University Press soon in 2008. Contact: [email protected], Phone: +86 (10) 62757763, Fax: +86 (27) 88384443

144 Zhongnan University of Economics & Law 114 WuLuo Road WuChang, WuHan 430060, P.R. China

3 Network and cooperation in innovation

Duan Yibing and Tang Le: Key Factors in the Successful Commercialization of Public Sector Research: Implications of the Case of CAS Spin-offs Abstract Two key channels for technology commercialization are licensing and the formation of spinoff companies. This paper focuses on spin-offs in the Chinese Academy of Sciences, investigates why and how CAS creates and operates spin-offs, and explains the impact of China’s economic development environment on the CAS spin-offs strategy. Keywords: technology commercialization; spin-off; public research

1

Introduction

Policy-makers have become increasingly interested in the commercialization of publiclyfunded research. Two key channels for commercialization are technology licensing and the formation of spin-off companies[1]. The system of patenting, marketing and licensing has the advantage that academics are able to attract social resources in order to promote the commercial development of research results, and the scientists are able to pursue their research without having to use large amounts of time to deal with unfamiliar business matters. Sometimes the new technology may be not easily patented and transacted via a license agreement which shows the importance of scientists to cooperate in order to capture the new technology’s full value. Companies which already have a reasonable level of research capability and contacts with the academic community may also be able and likely to license the new technology. Both the commitment to involve resulting benefits and the scarcity of qualified licenses lead to a second option which is to commercialize technology through a spin-off company. While there have been numerous studies of university patenting, licensing, and cooperation research and development agreement, less attention has been paid to new firm creation at public research institutions or organizations that receive public funds to conduct R&D[2]. How spin-offs help technology transfer and what the long-term impact of spin-offs on public research organizations is has not been well understood yet. In this paper, we discuss a case of spin-offs in the Chinese Academy of Sciences, investigate why and how the CAS creates and operates spin-offs, and explain the impact of China’s economic development environment on the CAS spin-offs strategy.

146

2

3 Network and cooperation in innovation

Spin-offs policy and performance in the CAS

The Chinese Academy of Sciences (CAS) is a national research institution founded by China’s government. According to the 2006 CAS Annual Statistical Report, by the end of 2005, under the CAS there were 90 scientific research institutes, 10 supporting units including one university, one graduate school, one newspaper, one publishing company, and 4 documentation and information centers. They are distributed over various parts of the country. In 2005, the CAS obtained 43.1 thousand staff and 38.1 thousand graduates, and used 10.7 billion Yuan (~1.4 billion dollar). The CAS has always played a positive role when it comes to responding to social needs and economic development in China which enhanced its public image and lead to accountability for funding. In 1985, China’s national government initiated the reform of the scientific and technologic system with the goal to establish close relations between science and technology and economic development, named “Economic development must depend on science and technology; Science and technology must tend to economic development”. Although many scientific disciplines faced tremendous pressure to contribute directly to economical matters, involvements of technology commercialization began to obtain more moral recognition and more physical support. The moral recognition is evident in the successful technology transfer having become a respectful achievement in the CAS. In 1997, China’s national government enacted the CAS to launch the ambitious Knowledge Innovation Project. According to the “3 Bases” mission revised in 1999, the CAS strives to turn itself into a scientific research base at an advanced international level, a base for fostering and bringing up advanced S&T talents, and a base for promoting the development of China’s high-tech industries. Current policies to promote the development of high-tech industries in China may be combined into 6 models: [1] Providing the primary scientific ideas & talents, sample & model, patent & technology, in accordance with the national’s target and the market need; [2] Co-operating and connecting with enterprises and society through multi-channels and multi-forms, promoting technology innovation and new products development, so as to take shape of a scale industry; [3] Co-operating with the society and enterprises to establish the project centers and R&D organizations, implementing an effective and scientific management, promoting technology transference and the steady development of enterprises; [4] Adopting venture investment to foster high-tech enterprises, integrating technology with the social capital, promoting the development the high-tech enterprises; [5] Permitting the inventors to hold shares of enterprises, encouraging them to join the technology transference; [6] Taking advantage of the talents and technology, co-operating with the small & middlescale enterprises, making contributions to the regional economic development.

Duan and Tang: Key factors in the successful commercialization of public sector research

1400

832 690

1200

147

900 800

100 Million(RMB)

620

1000

472

800

438

454

394

432

700 600 394

500

600

400

400

300 200

200

100 0

0 1997

1998

1999

2000

2001

2002

2003

2004

2005

Year Re ve n u e

Ne t As s e ts

Profi t

En te rpri s e s No.

Figure 1: The Status of CAS High-tech Enterprises

Flexible spin-off policies in CAS led to the formation of numerous spin-offs and helped CAS to generate more revenues than licensing (Figure 1). 120 billions Yuan (10 billion dollars) in revenue and 4.5 billions Yuan (600 million dollars) in profit, both of CAS enterprises in 2005, showed CAS has become a public research organization as well as a corporation majoring in technology commercialization.

3

Key factors to create and operate spin-offs in the CAS

After our observation and interviews, we found 5 overriding factors that have proven to make an essential contribution to the successful commercialization of the CAS research results over the past 20 years. These success factors are listed and discussed below. (1) Strong Administrative Support The CAS, at all levels, strongly supports the technology transfer especially to spin off technology. The support is evident in speeches from top-level administrations and in the “3 Bases” mission statements of the CAS. Former President (1986-1997) Guangzhao Zhou and President Yongxiang Lu (1997~) are enthusiastic to promote spin-offs in the CAS. Several members at the highest level of the administration are actively involved in technology transfer from industry partnering to the creation of new spin-offs. At the institute level, directors determine that an invention lacks or has commercial potential and make the decision to pursue commercialization. When a spin-off company starts to operate, all departments in the institute would provide support services for the institute-owned business. Comparing to technology licensing, startups based on institute technology tend to attract longer attention and are more likely to achieve a new product status.

148

3 Network and cooperation in innovation

(2) Stock of technology, with comparative advantage There is no commercialization without innovative research. Institutes in the CAS make efforts to promote advanced research to meet national needs and economic development. The cornerstone of its success in spin-offs is that the CAS know the local market and technology potentials very well. This is their comparative advantage. For example, as a long-term export policy, the U.S. government controls the export of high performance computers to sensitive destinations, such as Russia and China, based on foreign policy and national security concerns. A high performance computer has both civilian and military applications, so the U.S. export control actually gives a huge market space for the China Shuguang Computer Corp., a spin-off from the Institute of Computer Technology. The same situation appeared in Shanghai Nisaila Sensors Co. Ltd., a spin-off from the Shanghai Institute of Technical Physics. Nisaila is mainly engaged in research and development, the production and marketing of various kinds of sensors and inactive optical communication device products. In 2005, it achieved a sales income of 500 million Yuan (68 million dollars) and a total profit of 70 million Yuan1. (3) Routines for incentives and rewards It is common knowledge that organizations require routines relating to incentives and rewards to encourage people to perform particularly productive activities. Although the primary motivation for scientists is recognition within the scientific community, for most of the CAS institutes, the commercialization of research does not require a radical change in the way they have traditionally exploited scientific discoveries. Partly because the government’s policies and procedures ensure scientists to believe that successful commercialization also enhances reputations for themselves, research groups, the institute and the CAS. The reputations may be displayed in issues such as a more personal sharing in the spin-off, more research funding for the laboratory and more budget for the Institute. All the routines create a tolerant atmosphere to encourage institutes and scientists to engage in technology transfer. (4) Business development capabilities A fourth key to operate spin-off is its business development capabilities. At an early stage, the institute always obtains over 50% sharing to operate a spin-off. Hence, it is not surprising that the institute tries to develop specific business capabilities. After 20 years of practical spin-off training, there are a lot of commercialization-minded managesr in currently almost 400 enterprises in the CAS. For example, the Institute of Automation operates 15 spin-offs and one incubator. Although the mobility of spin-offs managers is very high, the Institute of Automation has an outstanding ability to recruit new people and involve them in spin-off activities. One source of new people is the graduates of different institutes. They have experience in the scientific field, are familiar with the local community, are able to work well with the laboratory colleagues and have the aptitude to develop technological value. The CAS technology transfer administration also organizes manager training programs and workshops. CAS spin-offs seem like a family. 1

http://www.sitp.ac.cn/jws_english/about.asp

Duan and Tang: Key factors in the successful commercialization of public sector research

4

149

Conclusion

Despite the apparently increasing praise that the CAS was enhancing local technology capability by creating new technology and ensuring new products through spin-offs, it is still unclear what the long-term impact is. As external companies have no research capability to communicate with the CAS, institutes have the motivations to operate the spin-out as a business unit, make decisions on the commercial potential of an invention, and reach a successful performance. Another reason may result from the weak intellectual property protection system in China. Comparing technology licensing, the spin-off model will be easier to protect the inventors’ benefit. If both of them get improved, partnership in technology licensing will play an important role in speeding up the process of commercialization.

References [1] Kevin Hindle, John Yencken. Public research commercialization, entrepreneurship and new technology based firms: an integrated model. Technovation, 24 ( 2004) 793 – 803. [2] Gideon D Markman, et al. Innovation speed: Transferring university technology to market. Research Policy, 34 (2005): 1058-1075. 1059 [3] Andy Lockett, Mike Wright. Resources, capabilities, risk capital and the creation of university spin-out companies. Research Policy, 34 (2005), 1043-1057. 1048 [4] Chinese Academy of Sciences. 2006 CAS Annual Statistical Report. Science Press, 2006.

150

3 Network and cooperation in innovation

Authors Duan Yibing is Associated Professor of technology management and policy at Institute of Policy and Management of Chinese Academy of Sciences. His major areas of interest are commercialisation of public research, incentives for technology transfer and the impact of basic research on technological capacity. He has been working as a policy analyst at the Institute since 1997 and served as Head of Division of Policy during 2001-2006. He also worked as Senior Visiting Scholar at SOAS, University of London and Humphrey Scholar of Fulbright Foundation at Boston University, both focusing on technology transfer. He received a PhD from the Graduate University of Chinese Academy of Sciences in Beijing. Contact: [email protected], Phone: +86 (10) 62650859 Tang Le is a graduate student for Master degree at Institute of Policy and Management of Chinese Academy of Sciences. Her specialty is Management Science and Engineering. She focuses on game-theory approach to governance of public-private collaboration in innovation. She received a bachelor’s degree in Economics. Contact: [email protected], Phone: +86 (137) 17900686 Institute of Policy and Management Chinese Academy of Science 55 Zhongguancun East Avenue Haidian, Beijing 100080, P.R. China Tel.: +86 (10) 62650859 Fax : +86 (10) 62542619 www.casipm.ac.cn

Martin Kloyer and Roland Helm: Contractual design of contract R&D: state-of-the-art of empirical research 1

Research question

The technological complexity of new product development forces innovative enterprises to externalize at least parts of their research and development (R&D) by contract R &D.1 Thus, not only capacity problems 2 of R&D can be solved. Moreover, specialization makes it possible to achieve scale effects and quality gains. Furthermore, the time-to-market can be reduced if contract R&D is combined with simultaneous engineering. In order to establish contract R&D relations and to realize potential efficiency gains it is necessary to control the specific opportunism possibilities that exist between supplier (agent) and buyer (principal).3 R&D suppliers are confronted with the danger of hold-up because they have to invest specifically at the beginning of the exchange relation.4 A supplier who anticipates this hold-updanger can be motivated to own opportunistic behaviour. Therefore, the buyer has to face the danger of moral hazard. According to normative Institutional Economics, the risk of hold-up can be limited by up-front payments, the danger of supplier opportunism by performancerelated payments. Up-front payments can be realized easily. In contrast to that, performancerelated payments are problematical in different ways. First, it is difficult to anticipate sufficiently precise milestones concerning the uncertain future development of the R&D process. Second, even if milestones can be contracted the observation of the supplier behaviour remains difficult because of the information asymmetry of hidden action that is especially strong in R&D. In particular, the observation of research processes is problematical. A supplier compensation that depends on the innovation return that is realized by the sales of final products would prevent these problems. The supplier would be motivated to refrain from opportunism in order not to endanger the market success of a final product on which his remuneration depends. Thus, milestones as well as monitoring of supplier behaviour would be unnecessary. However, in contract R&D it is not possible to agree on a continuous innovation return share at the time of contracting because there is no final product and therefore no assessment basis for such a return share. Due to uncertainty regarding the technological 1

See, e. g., Arora et al. (2001), Chiesa et al. (2004).

2

The danger of loosing core competences by specialisation are not considered here.

3

Williamson (1985, p. 47) defines opportunism as „self-interest seeking with guile“.

4

Concerning the theory of New Institutional Economics, see, e. g., Milgrom/Roberts (1992), Richter/Furubotn (2003), Picot/Dietl/Franck (2005).

152

3 Network and cooperation in innovation

development of the R&D process and concerning the evolution of markets and competition, it is even uncertain whether there will be any innovation return at all. Models of normative Institutional Economics cannot provide realizable approaches for this typical contract R&D situation because their assumptions do not exactly relate to its problems. That is why it is indispensable to find out empirically which institutions exist in the reality of contract R&D and how effectively they control opportunism. These questions are subjects of positive Institutional Economics.5 Contractual institutions are especially relevant for management practice since they can be designed deliberately and without time-lag.6 Therefore the paper examines to what extent empirical contract research, which is the branch of positive Institutional Economics that deals with single contractual regulations, identifies effective contractual arrangements for contract R&D. It will be shown that such arrangements – despite specific problems of data survey and operationalization – have been discovered. Thus, the paper has two purposes. First, addressing the practice of innovation management, it will inform about the relative effectiveness of institutions in contract R&D relations. Second, in a theoretical perspective, it shows how the young discipline of empirical contract research is able to give feed-back for models of normative Institutional Economics. In the beginning, the paper describes the mutual opportunism possibilities in contract R&D (2.1.) and points out the deficits of the approaches of normative Institutional Economics (2.2.). Chapter 3 will present the state-of-the-art of empirical contract research for contract R&D, and chapter 4 will continue with theoretical and managerial implications.

2

Contract R&D

2.1 Opportunism possibilities There are several possibilities for opportunistic behaviour in contract R&D (table 1). The problem of hidden capabilities of the supplier can be solved quite effectively by the institution of credible signals like patents. In contrast to that, the hold-up danger is more problematical. It results from two situational characteristics. First, in the beginning of the exchange relation, suppliers have to carry out unilateral specific investments. Second, since the evolution of the R&D process and the development of markets and competition are uncertain, it is impossible to anticipate contractually all the adjustments of the exchange terms that will become inevitable. R&D contracts are necessarily incomplete. One-sided specific investments and contractual incompleteness lead to the risk of hold-up, which means that the R&D buyer could appropriate the quasi-rent of the supplier. A supplier who anticipates this holdup risk has reason to behave opportunistically himself which would result in a loss of efficiency. This danger of supplier opportunism (moral hazard) is of particular importance in 5

See, e. g., Jensen (1983), p. 334; Wenger/Terberger (1988), p. 506; Eisenhardt (1989), S. 59f.; Elschen (1991), p. 1006; Ebers/Gotsch (1995), p. 195; Rüdiger (2000), p. 117f.; Masten/Saussier (2002); Richter/Furubotn (2003), p. 176ff.; Garrouste/Saussier (2005), p. 179.

6

Empirical research also underlines the strategic role of contract design (see, e. g., Rüdiger (2000)).

Kloyer and Helm: Contractual design of contract R&D

153

R&D supply relations because strong informational asymmetries (hidden action and hidden information) result in severe monitoring and assessment problems. Table 1: Opportunism possibilities in contract R&D Possibility of Supplier opportunism Buyer opportunism Hidden characteristics of the supplier with the consequence of the adverse selection risk Hidden action and hidden information resulting in moral hazard Supplier-specific investments combined with hidden intention of buyers resulting in hold-up

A remuneration that would depend on the innovation return would solve the problem of supplier opportunism. However, due to the described process-internal and –external uncertainty it is impossible to anticipate a final product as basis for an innovation return share of the supplier at the moment of contracting. Even in contract development, the basis for assessment is only recognisable after a prototype and its features have been defined – that means after contracting (table 2). The absence of a basis for assessment also implies that suppliers cannot signal capability by choosing an innovation return share instead of fix payments. Table 2: Ideal evolution of contract R&D Time of interaction t0 t1 t2 t3

Agreements, situation of interaction Contract between supplier and buyer End of the process of contract research End of the process of contract development (t3 is reached before the end of the process) Basis of assessment for the distribution of innovation returns is recognizable; decision about market entry is possible

2.2 Necessity of empirical contract research Models of normative Institutional Economics, based on its assumptions, prove the effectiveness and efficiency of institutional design options. However, there is no model which adequately depicts the presented specific conditions of contract R&D. 7 In order to justify the necessity of the empirical approach we will show some appropriate examples.8 The well-known model of incomplete contracts by Grossman and Hart (1986) shows an exchange relation whose features are similar to those of contract R&D. In a supply relation 7

Concerning the general problem of simplified assumptions in normative institutional-economic models see, e. g., Bercovitz (1999), Lafontaine/Slade (2000), Masten/Saussier (2002), p. 282ff.

8

With regard to other models, see, e. g., Morasch (1995), Goel (1999), Wielenberg (1999), Schweizer (1999), Tirole (1999), Baiman/Rajan (2002).

154

3 Network and cooperation in innovation

the buyer can improve what the supplier transferred and thereby gain an additional return. According to the model, the efficient institutional solution is based on the assumption that the costs and the additional profit associated with the improvement are already known at the time of contracting. However, this is particularly improbable in case of external R&D.9 A further example is given by Rühl (2001, p. 56) who assumes that the ratio of innovation return sharing can already be stipulated at the moment of contract conclusion. Yet, at this moment there is no final product and therefore no basis for assessment. Nöldeke/Schmidt (1995, p. 163 f.) developed a first-best-model, in which the presented problem of hold-up is solved by option contracts. Nevertheless, its application to contract R&D has to fail because of its premises. The first-best-approach assumes that opportunistic supplier behaviour could be detected by litigation. However, this is unrealistic regarding the difficulty of anticipating contractually precise milestones. These examples should have given sufficient reasons for the empirical approach this paper deals with.

3

Positive Institutional Economics and contract R&D

There are two types of positive Institutional Economics. Studies of the first type – this is the majority - examine only which institutions can be found dependent on specific informational asymmetries and opportunism possibilities in a certain type of exchange relation. The second type concentrates also on the effectiveness of institutions. Therefore, the paper will present the state-of-the-art of positive Institutional Economics concerning contract R&D according to this differentiation. 3.1

Determinants of situation and institutions

3.1.1 Market and Hierarchy The large number of quantitative-empirical studies that deal with the influence of determinants of transaction costs on the choice between market and hierarchy is immense.10 In this context, there are also studies concerning R&D.11 They focus particularly on the impact of specific investments and uncertainty on the probability of vertical integration of R&D. Some studies even consider hybrid forms of organisation as a third alternative. The following stud9

Concerning the assumptions of the model, see Richter/ Furubotn (2003), p. 276.

10

Some recent studies are, for example: Dant (1996), Aulakh/Kotabe (1997), Regan (1997), Coles/Hesterly (1998), Hanna/Maltz (1998), Jensen/Rothwell (1998), Delios/Beamish (1999), Hitt (1999), Widener/Selto (1999), Antinory (2000), Houston/Johnson (2000), Mayer (2000), Vazquez-Vicente (2000), Bréchemier/Saussier (2001), Brusoni et al. (2001), Dunbar/Phillips (2001), Jacobides/Hitt (2001), Affuso (2002), Robinson et al. (2002), Schilling/Steensmaa (2002), Woodruff (2002), Dragonetti et al. (2003), Leiblein/Miller (2003), McGahan/Villalonga (2003).

11

With regard to qualitative-empirical studies, that are not the subject of this paper, see, e. g., Butler/Carney (1983), Joskow (1985), Yarbrough/Yarbrough (1987), Hennart (1988), Tapon (1989), Pittman (1991), Globerman/Vining (1996).

Kloyer and Helm: Contractual design of contract R&D

155

ies verify a positive correlation between specific investments and the probability of internalisation or vertical integration of R&D: Masten (1984) proves a positive correlation for 1.887 systems components in aircraft industry, Pisano (1990) for 92 R&D projects in biotechnology, Lieberman (1991) for 203 enterprises in chemical industries, Masten et al. (1991) for 74 components in shipbuilding, Oerlemans/Meeus (2001) for R&D in 689 industrial companies, Love/Roper (2002) for 497 industrial companies and Roisello (2003) for 134 transaction relations in biotechnology. Moreover, Masten (1984), Masten et al. (1991), Oerlemans/Meeus (2001) and Roisello (2003) identify a positive influence of uncertainty on the probability of vertical integration of R&D. All these studies confirm that in R&D uncertainty and specific investments lead to market failure. However, the majority of these studies do not examine to what extent this market failure can be sufficiently solved by a hybrid form of organisation. 3.1.2 Number of contractual regulations, contract complexity and contract duration Studies that deal with the number of contractual regulations or with contract complexity have to be placed on an aggregation level below the form of organization and above single contractual regulations and will be considered as part of empirical contract research. Saussier’s study (2000a) confirms the positive influence of investment specificity on vertical integration. It also finds out that uncertainty leads to a smaller number of contractual regulations (in 63 transport contracts). These results may seem contradictory but in fact they support the analysis of opportunism possibilities in contract R&D above. Due to specific investments and resulting dependency there is a need for a contract. However, because of uncertainty, precise contractual regulations, for example with regard to profit sharing, are hardly possible. In a multi-sector study about 91 exchange relations, Barthelemy and Quelin (2002) deal with the dependent variable of contract complexity. The positive influence of specific investments corresponds to the assumptions. In contrast to that, the positive influence of uncertainty on contract complexity is astonishing in the first moment. However, this result can be explained by the fact that contract complexity was mainly measured by stipulations that deal with the uncertainty (incentive agreements, exit conditions, control mechanisms, adjustment clauses). These stipulations do not want to anticipate future developments. They are just reactions on anticipation problems. The variable of contract duration is in close relation with the dependent variable of contract complexity. From a theoretical perspective, it is to be expected that specific investments make an extension of a cooperation more probable because they promise partners further quasi rents. This connection is also empirically proven, for example by Joskow (1987) for 277 contracts in energy management, by Allen/Lueck (1992) for 3.104 contracts in the agriculture industry as well as by Saussier (1999, 2000b) for 29 contracts in the transport industry.12

12

There are also studies of the form of the contract (oral or written) as a dependent variable. Following the logic concerning contract complexity, it has to be assumed that there is a positive correlation between investment specificity and a negative between uncertainty and the oral instead of the written form. Both assumptions are proven, e. g., by Lyons (1994) for 101 supply relations between engineering bureaus and their buyers.

156

3 Network and cooperation in innovation

3.1.3 Single contractual regulations Studies based on Institutional Economics which examine single contractual regulations are rare so far. That is why there is still considerable need of research.13 This concerns both questions of empirical contract research, the connection between determinants of transaction costs and regulations on the one hand, as well as the influence of regulations on opportunism on the other hand. Although there are some studies dealing with the first question, only few 14 of them are relevant for contract R&D. Based on a sample of 44 contracts in the aircraft industry, Crocker/Reynolds (1993) find out that payment agreements are more flexible the higher the technological uncertainty is. With regard to contract R&D, which is characterized by a considerable technological uncertainty, it can be concluded that precise contractual regulations about return sharing are only possible in case of diminished uncertainty. This is also confirmed by the results of Yanagawa/Wada (2000) concerning 407 patentlicense contracts. They show that in case of licensing patents with a high potential for subsequent innovations15 typically invariable fees are fixed. The reason is that at the time of the contract conclusion it is impossible to anticipate the whole range of final products that will be derived from the licensed technology. Often, it is even impossible to identify only one concrete final product that could be the basis for sales-related fees. Thus, the advantage of the invariable fee is that it has to be paid as long as the originally licensed technology is used. This result shows that the licensing practice avoids stipulations of return-dependent payments if there is no basis of assessment in t0. In contract R&D, this has to be expected even more because the uncertainty is higher than in licensing. Bessy et al. (2002) demonstrate in a study of 224 technology licensing relations that the probability of sales-related fees instead of invariable fees is higher if the transferred knowledge is explicable instead of tacit. This result is consistent with our view since in contract research the transfer of tacit knowledge is more important than in contract development. Based on 161 industrial companies, a study by John/Weitz (1989) shows that in case of assessment difficulties,the sales personnel receives rather a fixed than a variable payment. This supports our view that in case of hidden action and hidden information – as in contract R&D - the buyer would rather be expected to choose a fixed payment. The role of specific investments is examined in the two following studies. The results concerning 195 cooperation agreements in biotechnology (Pisano (1989)) and 974 cooperation agreements in telecommunication (Pisano et al. (1988)) show that investment specificity leads to mutual investments in shares of the partner firm. These results illustrate that investments in a merger function as credible commitments in cases of one-sided dependencies.

13

Masten/Saussier (2002), p. 231f.

14

E. g., Crocker/Lyon (1994) about most-favoured-nation-clauses.

15

Such patents are characterized by many forward citations, i. e. citations the patent receives in subsequent patenting processes.

Kloyer and Helm: Contractual design of contract R&D

157

Finally, there are some studies that examine a contractual arrangement that is of special effectiveness in contract R&D: the bilateral monopoly based on mutual hostages. In t0, the buyer promises to negotiate an additional continuous return share for the supplier in t3. This negotiation option is based on the hostage of exploitation rights for the supplier that concern the R&D results and/or further developments conducted by the buyer after the transfer of the results. The effect of the exploitation rights is that the supplier could blockade an opportunistic buyer who would refuse the promised additional return share in t3. This credible option on additional remuneration motivates the supplier to refrain from opportunism because by opportunism he would undermine the basis for his return-dependent compensation. In order to prevent a one-sided possibility of blockade the buyer typically has an exclusive buying right. Thus, both parties could blockade reciprocally. However, they are not interested in that since the supplier would prevent his additional return share and the buyer wants to amortize those sums he payed already to the supplier. The first study that detected this institutional mechanism was that of Bessy/Brousseau (1998) who analysed technology licensing contracts in leading French companies. In 65,2% of the cases they found grant-back agreements in favour of the licensor (supplier) and in 71,7% an exclusive buying right for the licensee (buyer). Kloyer (2004, 2005a, 2005b) detected the described contract design in a sample of 76 contract R&D relations. In 80,0 % of the cases the supplier receives sufficient shares in the exploitation rights. In 88,9% of these cases the buyer gets an exclusive buying right, thus establishing a bilateral monopoly. The study gives also additional information about the payments. The part of the remuneration that is contracted in t0 consists typically of up-front payments and milestone-dependent payments, which are stipulated despite the described monitoring problems. 3.2 Institutions and the effectiveness of opportunism control This chapter deals with the empirical results concerning the effectiveness of R&D contract regulations. So far, we have some studies of institutional effectiveness on the level of the organizational form.16 Two studies focus single institutions, however not single contract stipulations. Brown et al. (2000) examine the effectiveness of vertical integration, specific investments, and “relational exchange” (sample: 395 hotel managers). Their terminology is a little bit confusing. Thus, “relational exchange” does not mean relational contracting, where long-term specific investments are decisive, but organization by values and norms. To describe relational contracting they use the term “specific investments”. The findings show that values and norms work independently and also combined with specific investments. Further research could test these effects in contract R&D. Gierl (2000) examines the influence of the contractual complexity on the buyer opportunism in a multi-sector sample of 232 component suppliers. The results show that complexity is only effective if it is combined with monitoring possibilities.

16

See, e. g., John (1984), Dwyer/Oh (1987), Anderson (1988), Provan/Skinner (1989).

158

3 Network and cooperation in innovation

So far, only the studies of Kloyer (2003, 2005a) and Helm/Kloyer (2004) examined the effectiveness of the bilateral monopoly whose importance for opportunism control in contract R&D was discovered in other studies cited above. Supplier opportunism was not measured directly17 but by looking at the perception of the cooperation risk. This approach is based on the pre-study result that a supplier who considers the profitability of his investments within the relation as lower than that of the buyer is motivated to behave opportunistically. Thus, the authors considered high levels of satisfaction with the cooperation risk control as identical to low opportunism motivation. The advantage of this kind of opportunism measurement compared to the direct one is that the construct “cooperation risk” consists of quantifiable dimensions. The results of the study show that exploitation rights for the supplier lead to higher satisfaction with the perceived control of the cooperation risk. Theoretically, a supplier could also gain bargaining power and therefore higher satisfaction with the cooperation risk control from the non-stipulation of exclusive buying rights for the buyer. However, the results do not confirm this effect. This is consistent with the explanation above where the role of the exclusive buying right as counter-weight within the bilateral was described. Furthermore, it becomes obvious that the effect of the exploitation rights is stronger in contract research than in contract development. The reason is that the only possibility for the contract research supplier to gain an additional return share is the use of his post-contractual bargaining power provided by the exploitation rights. In contrast to that, in some cases of contract development, there is an assessment basis before the completion of the contract. In these cases, the supplier gains also bargaining power from the possibility to stop the development process.

4

Conclusion

This paper demonstrates that empirical contract research is able to provide valuable feedback to theory and practice. Thus, Institutional Economists are informed about the applicability of their models. Moreover, the empirical research draws the attention on institutions that are not considered in the theoretical discussion.18 The described option on additional remuneration based on exploitation rights is just an example. The management practice can also profit from this fruitful research approach. Of course, the described institutions are wellknown in the contract R&D practice. However, management practice is not informed about the relative effectiveness of the institutions-in-use. Empirical contract research can also contribute to the discussion about the self-fulfillingprophecy-character. The main criticism is that institutions would cause the opportunism they shall control.19 The presented results concerning the institutional effectiveness in contract R&D contradict this criticism. We want to conclude by assuming that the self-fulfilling-

17

Concerning this approach, see, e. g., Anderson (1988), Dwyer/Oh (1987), John (1984), Provan/Skinner (1989).

18

See, e. g., Masten/Saussier (2002), p. 286ff.

19

See, e. g., Ghoshal/Moran (1996).

Kloyer and Helm: Contractual design of contract R&D

159

problem depends on the history of a concrete exchange relation and on personality traits of the managers who are involved in the cooperation.

References Affuso, L. (2002): An empirical study on contractual heterogeneity within the firm: The “vertical integration-franchise contracts” mix, in: Applied Economics, Vol. 34, Issue 8, pp. 931-944. Allen, D.W./Lueck, D. (1992): The “black fourty” handshake: Specific assets, reputation, and the structure of farmland contracts, in: Journal of Law, Economics & Organization, Vol. 8, Issue 2, pp. 366-376. Anderson, E. (1988): Transaction costs as determinants of opportunism in integrated and independent sales forces, in: Journal of Economic Behavior and Organization, Vol. 9, Issue 3, pp. 247-264. Antinory, C.M. (2000): Vertical integration in Mexican common property forests, Diss., University of California, Berkley. Arora, A./Fosfuri, A./Gambardella, A. (2001): Markets for technology. The economics of innovation and corporate strategy, Cambridge/Mass. Aulakh, P.S./Kotabe, M. (1997): Antecedents and performance implications of channel integration in foreign markets, in: Journal of International Business Studies, Vol. 28, Issue 1, pp. 145-175. Baiman, S./Rajan, M.V. (2002): The role of information and opportunism in the choice of buyer-supplier relationships, in: Journal of Accounting Research, Vol. 40, Issue 2, pp. 247278. Barthelemy, J./Quélin, B.V. (2002): Competence, specificity and outsourcing. Impact on the complexity of the contract, in: Paper presented at the Academy of Management Conference 2002. Bercovitz, J.E.L. (1999): An analysis of the contract provisions in business-format franchise agreements, in: Stanworth, J./Purdy, D. (Eds.): Franchise beyond the millennium: Learning lessons from the past; Proceedings of the 13th Conference of the Society of Franchising. Bessy, C./Brousseau, E. (1998): Technology licensing contracts. Features and diversity, in: International Review of Law and Economics, Vol. 18, pp. 451-489. Bessy, C./Brousseau, E./Saussier, S. (2002) : Payment schemes in technology licensing agreements : A transaction cost approach, Working paper, ATOM, Université de Paris, Panthéon-Sorbonne, Paris. Bréchemier, D./Saussier, S. (2001): What governance structure for non-contractable services? An empirical analysis, Working paper, ATOM, Université de Paris, PanthéonSorbonne, Paris.

160

3 Network and cooperation in innovation

Brown, J.R./Dev, C.S./Lee, D.-J. (2000): Managing marketing channel opportunism: the efficiacy of alternative governance mechanisms, in: Journal of Marketing, Vol. 64, pp. 5165. Brusoni, S./Prencipe, A./Pavitt, K. (2001): Knowledge specialization, organizational coupling, and the boundaries of the firm: Why do firms know more than they make?, in: Administrative Science Quarterly, Vol. 46, Issue 4, pp. 597-621. Butler, R./Carney, M.G. (1983): Managing markets: Implications for the make-buy decision, in: Journal of Management Studies, Vol. 20, Issue 2, pp. 213-233. Chiesa, V./Manzini, R./Pizzurno, E. (2004): The externalisation of R&D activities and the growing market of product development services, in: R&D Management, Vol. 34, pp. 65-75. Coles, J.W./Hesterly, W.S. (1998): The impact of firm-specific assets and the interaction of uncertainty: An examination of make-or-buy decisions in public and private hospitals, in: Journal of Economic Behavior and Organization, Vol. 36, Issue 3, pp. 383-409. Crocker, K.J./Lyon, T.P. (1994): What do "facilitating practices" facilitate? An empirical investigation of most-favored-nation clauses in natural gas contracts, in: Journal of Law and Economics, Vol. 37, pp. 297-322. Crocker, K.J./Reynolds, K.J. (1993): The efficiency of incomplete contracts: empirical analysis of air force engine procurement, in: Rand Journal of Economics, Vol. 24, pp. 126146. Dant, S.P. (1996): Ownership structure in franchising: The effect of transaction costs, production costs and strategic considerations, in: The International Review of Retail, Distribution and Consumer Research, Vol. 6, Issue 1, pp. 53-75. Delios, A./Beamish, P.W. (1999): Ownership strategy of Japanese firms: Transactional, institutional, and experience influences, in: Strategic Management Journal, Vol. 20, Issue 7, pp. 915-933. Dragonetti, N./Dalsace, F./Cool, K. (2003): A comparative test of the efficiency, focus, and learning perspectives of outsourcing, Working paper, INSEAD, Fontainebleau Cedex. Dunbar, A.E./Philips, J.D. (2001): The outsourcing of corporate tax functions activities, in: The Journal of the American Taxation Association, Vol. 23, Issue 2, pp. 35-49. Dwyer, F.R./Oh, S. (1987): Output sector munificence effects on the internal political economy of marketing channels, in: Journal of Marketing Research, Vol. 24, pp. 347-358. Ebers, M./Gotsch, W. (1995): Institutionenökonomische Theorien der Organisation, in: Kieser, A. (Ed.): Organisationstheorien, 2nd ed., Stuttgart, pp. 185-235. Eisenhardt, K.M. (1989): Agency theory: An assessment and review, in: Academy of Management Review, Vol. 14, Issue 1, pp. 57-74. Elschen, R. (1991): Gegenstand und Anwendungsmöglichkeiten der Agency-Theorie, in: Zeitschrift für betriebswirtschaftliche Forschung, Vol. 43, pp. 1002-1012.

Kloyer and Helm: Contractual design of contract R&D

161

Garrouste, P./Saussier, S. (2005): Looking for a theory of the firm: Future challenges, in: Journal of Economic Behavior & Organization, Vol. 58, pp. 178-199. Gierl, H. (2000): Opportunismus in langfristigen Geschäftsbeziehungen, in: Zeitschrift für betriebswirtschaftliche Forschung, Vol. 52, pp. 107-140. Globerman, S./Vining, A.R. (1996): A framework for evaluating the government contracting-out decision with an application to information technology, in: Public Administration Review, Vol. 56, Issue 6, pp. 577-586. Goel, R.K. (1999): On contracting for uncertain R&D, in: Managerial and Decision Economics, Vol. 20, pp. 99-106. Ghoshal, S./Moran, P. (1996): Bad for practice: A critique of the Transaction Cost Theory, in: Academy of Management Review, Vol. 21, pp. 13-47. Grossman, S.J./Hart, O. (1986): The costs and benefits of ownership. A theory of vertical and lateral integration, in: Journal of Political Economy, Vol. 94, pp. 694-719. Hanna, J./Maltz, A. (1998): LTL expansion into warehousing: A transaction cost analysis, in: Transportation Journal, Vol. 38, Issue 2, pp. 5-17. Helm, R./Kloyer, M. (2004): Controlling contractual exchange risks in r&d interfirm cooperation - An empirical study, in: Research Policy, Vol. 33, 2004, pp. 1103-1122. Hennart, J.-F. (1988): Upstream vertical integration in the aluminum and tin industries: a comparative study of the choice between market and intrafirm coordination, in: Journal of Economic Behavior and Organization, Vol. 9, Issue 3, pp. 281-299. Hitt, L.M. (1999): Information technology and firm boundaries: Evidence from panel data, in: Information Systems Research, Vol. 10, Issue 2, pp. 134-151. Houston, M./Johnson, S. (2000): Buyer-supplier contracts versus joint ventures: determinants and consequences of transaction structure, in: Journal of Marketing Research, Vol. 37, Issue 2, pp. 1-15. Jacobides, M.G./Hitt, L.M. (2001): Vertical scope revisited: Transaction costs versus capabilities & profit opportunities in mortgage banking, Working paper, Center for Financial Institutions, Wharton School, University of Pennsylvania. Jensen, M.C. (1983): Organization theory and methodology, in: The Accounting Review, Vol. 58, pp. 319-339. Jensen, M.C./Rothwell, G.S. (1998): Transaction costs, regulations, and subcontracting at nuclear power plants, in: Journal of Economic Behavior and Organization, Vol. 36, Issue 3, pp. 369-381. John, G. (1984): An empirical investigation of some antecedents of opportunism in a marketing channel, in: Journal of Marketing Research, Vol. 21, Issue 8, pp. 278-289.

162

3 Network and cooperation in innovation

John, G./Weitz, B.A. (1989): Salesforce compensation: An empirical investigation of factors related to use of salary versus incentive compensation, in: Journal of Marketing Research, Vol. 26, Issue 2, pp. 1-14. Joskow, P.L. (1985): Vertical integration and long-term contracts. The case of coal-burning electric generating plants, in: Journal of Law, Economics, and Organization, Vol. 1, pp. 3380. Joskow, P.L. (1987): Contract duration and relationship-specific investments: empirical evidence from coal markets, in: The American Economic Review, Vol. 77, Issue 1, pp. 168185. Kloyer, M. (2003): Wahrgenommene Risikokontrolle und Opportunismus in F&ELieferbeziehungen, in: Zeitschrift für Planung und Unternehmenssteuerung, Vol. 14, pp. 149-169. Kloyer, M. (2004): Opportunismus und Verhandlungsmacht in F&E-Lieferbeziehungen – Eine empirische Untersuchung, in: Zeitschrift für betriebswirtschaftliche Forschung, Vol. 56, pp. 333-364. Kloyer, M. (2005a): Informationsprobleme und Vertragsgestaltung in F&ELieferbeziehungen – Eine empirische Untersuchung, Habilitationsschrift, Friedrich-SchillerUniversität Jena. Kloyer, M. (2005b): Opportunismus und Institutionen in vertikalen Innovationskooperationen, in: Schauenberg, B./Schreyögg, G./Sydow, J. (Eds.): Managementforschung Vol. 15: Institutionenökonomik als Managementlehre?, Wiesbaden, pp. 283-326. Lafontaine, F./Slade, M.E. (2000): Incentive contracting and the franchise decision, in : Chatterjee, K./Samuelson W. (Eds.): Advances in Business Applications of game theory, Dordrecht et al., pp. 133-188. Leffler, K.B./Rucker, Randal R. (1991): Transaction costs and the efficient organization of production. A study of timber-harvesting contracts, in: Journal of Political Economy, Vol. 99, pp. 1060-1087. Leiblein, M.J./Miller, D.J. (2003): An empirical examination of transaction- and firm-level influences on the vertical boundaries of the firm, in: Strategic Management Journal, Vol. 24, Issue 9, pp. 839-859. Lieberman, M.B. (1991): Determinants of vertical integration: an empirical test, in: The Journal of Industrial Economics, Vol. 39, Issue 5, pp. 451-446. Love, J.H./Roper, S. (2002): Internal versus external r&d: A study of r&d choice with sample selection, in: International Journal of the Economics of Business, Vol. 9, Issue 2, pp. 239255. Lyons, B.R. (1994): Contracts and specific investment: An empirical test of transaction cost theory, in: Journal of Economics and Management Strategy, Vol. 3, Issue 2, pp. 257-278.

Kloyer and Helm: Contractual design of contract R&D

163

Masten, S.E. (1984): The organization of production: Evidence from the aerospace industry, in: Journal of Law and Economics, Vol. 27, pp. 403-417. Masten, S.R./Meehan, J.W./Snyder, E.A. (1991): The costs of organization, in: Journal of Law, Economics & Organization, Vol. 7, Issue 1, pp. 1-26. Masten, S.E./Saussier, S. (2002): Economics of contracts: An assessment of developments in the empirical literature of contracting, in: Brousseau, E./Glachant, J-M (Eds.): Economics of contracts: Theories and applications, pp. 273-293. Mayer, K. (2000): Transactional alignment and project performance: Evidence from information technology, Working paper, Marshall School of Business, University of Southern California, Los Angeles. McGahan, A./Villalonga, B. (2003): The choice between acquisitions and divestitures, Working paper, Boston University, Boston, Harvard Business School, Cambridge. Milgrom, P./Roberts, J. (1992): Economics, organization and management, Englewood Cliffs, New Jersey. Morasch, K. (1995): Moral hazard and optimal contract form for r&d cooperation, in: Journal of Economic Behavior & Organization, Vol. 28, pp. 63-78. Nöldeke, G./Schmidt, K. M. (1995): Option contracts and renegotiation: A solution to the hold-up problem, in: RAND Journal of Economics, Vol. 26, pp. 163-179. Oerlemans, L.A.G./Meeus, M.T.H. (2001): R&D cooperation in a transaction cost perspective, in : Review of Industrial Organization, Vol. 18, pp. 77-90. Picot, A./Dietl, H./Franck, E. (2005): Organisation. Eine ökonomische Perspektive, 4th ed., Stuttgart. Pisano, G.P. (1989): Using equity participation to support exchange: evidence from the biotechnology industry, in: Journal of Law, Economics & Organization, Vol. 5, Issue 1, pp. 109127. Pisano, G.P. (1990): The r&d boundaries of the firm: an empirical analysis, in: Administrative Science Quarterly, Vol. 35, Issue 1, pp. 153-176. Pisano, G.P./Russo, M.V./Teece, D.J. (1988): Joint ventures and collaborative alliances in the telecommunications equipment industry, Cambridge. Pittman, R. (1991): Specific investments, contracts, and opportunism: the evolution of railroad sidetrack agreements, in: Journal of Law and Economics, Vol. 29, Issue 3, pp. 565-589. Provan, K.G./Skinner, S.J. (1989): Interorganizational dependence and control as predictors of opportunism in dealer-supplier relations, in: Academy of Management Journal, Vol. 32, Issue 1, pp. 202-212. Regan, L. (1997): Vertical integration in the property-liability insurance industry: A transaction cost approach, in: The Journal of Risk and Insurance, Vol. 64, Issue 1, pp. 41-62.

164

3 Network and cooperation in innovation

Richter, R./Furubotn, E.G. (2003): Neue Institutionenökonomik. Eine Einführung und kritische Würdigung, 4th ed., Tübingen. Robinson, A./Wilson, N./Zhang, H. (2002): Employee ownership and firm-specific human capital, Paper presented at the 11th conference of the International Association for the Economics of Participation, Brüssel. Roisello, A. (2003): At the interface of governance and competence perspectives: Discussing organizational choices in the biotechnology industry, Working paper, Department of Economics, University of Strathclyde, Strathclyde. Rüdiger, M. (2000): Forschung und Entwicklung als Dienstleistung. Grundlagen und Erfolgsbedingungen der Vertragsforschung, Wiesbaden. Rühl, J. (2001): Vertragliche Gestaltung von Innovationskooperationen. Optimierung bei Informationsasymmetrie, Wiesbaden. Saussier, S. (1999): Transaction cost economics and contract duration: an empirical analysis of EDF coal contracts, in: Recherches Economiques de Louvain, Vol. 65, Issue 1, pp. 3-21. Saussier, S. (2000a): La durée des contracts interentreprises, Working Paper. ATOM Université de Paris, Phantéon-Sorbonne, Paris. Saussier, S. (2000b): Contractual completeness and transaction costs, in: Journal of Economic Behavior and Organization, Vol. 42, pp. 189-206. Schilling, M.A./Steensma, K.H. (2002): Disentangling the theories of firm boundaries: a path model and empirical test, in: Organization Science, Vol. 13, Issue 4, pp. 387-401. Schweizer, U. (1999): Vertragstheorie, Tübingen. Tapon, F. (1989): A transaction costs analysis of innovations in the organization of pharmaceutical r&d, in: Journal of Economic Behavior and Organization, Vol. 12, Issue 2, pp. 197213. Tirole, J. (1999): Incomplete contracts: Where do we stand?, in: Econometrica, Vol. 67, pp. 741-782. Vázquez-Vicente, J.H. (2000): Allocation of decision rights on the shop floor: A transaction cost approach to the Spanish food and electronics industry, Working paper, University of Vigo, Galiza. Wenger, E./Terberger, E. (1988): Die Beziehung zwischen Agent und Prinzipal als Baustein einer ökonomischen Theorie der Organisation, in: Wirtschaftswissenschaftliches Studium, Vol. 17, Issue 10, pp. 506-514. Widener, S./Selto, F. (1999): Management control systems and the boundaries of the firm: Why do firms outsource internal auditing activities, in: Journal of Management Accounting, Vol. 11, pp. 45-73.

Kloyer and Helm: Contractual design of contract R&D

165

Wielenberg, S. (1999): Die Absicherung von Kooperationen in Zulieferbeziehungen, in: Engelhard, J./Sinz, E.J. (Eds.): Kooperation im Wettbewerb, Wiesbaden, pp. 301-324. Williamson, O.E. (1985): The economic institutions of capitalism, New York. Woodruff, C.M. (2002): Non-contractible investments and vertical integration in the Mexican footwear industry, Graduate School of International Relations and Pacific Studies. Yanagawa, N./Wada, T. (2000): Post-contracting innovations and contingent payment scheme in patent licensing contracts, working paper, University of Tokyo/Gakushuin. Yarborough B.V./Yarborough, R.M. (1987): Institutions for the governance of opportunism in international trade, in: Journal of Law, Economics & Organization, Vol. 3, Issue 1, pp. 129-140.

Authors Martin Kloyer, PD Dr. rer. pol. Scientific research assistant Chair of Industrial Management Friedrich-Schiller-University Jena Carl-Zeiß-Str. 3 07743 Jena, Germany Tel.: +49 (3641) 9-43105 Fax : +49 (3641) 9-43102 Email: [email protected] Roland Helm, Prof. Dr. Head of department Unilever-Chair of Marketing and Business Administration Friedrich-Schiller-University Jena Carl-Zeiß-Str. 3 07743 Jena, Germany Tel.: +49 (3641) 9-43110 Fax : +49 (3641) 9-43112 Email: [email protected]

Liu Yun, Cheng Guo-ping and Yang Yu: Research on Influencing Factors of the Performance of Enterprises’ R&D Cooperation Abstract By analyzing the micro-mechanism of the R&D cooperation of R&D alliance enterprises, this article explores the influencing factors for the performance of R&D alliance enterprises’ R&D cooperation from four aspects: technical resource intensity of partners, organization ability of enterprises, bulwark properties of technical knowledge, and harmonization level between partners. By introducing the structural equation model (SEM) to 125 enterprises and analyzing the survey data, it demonstrates the relationship between four factors mentioned before and finally gives several suggestions to improve the performance of R&D cooperation between enterprises. Keywords: R&D alliance enterprises, R&D cooperation, Performance, Structural equation model (SEM)

1

Introduction

As the market competition becomes more and more severe, enterprises cannot meet the fast changing requirements of the customers if they only rely on their own R&D power. Correspondingly, cooperative R&D alliance development is broadly adopted due to its visible advantages: sharing the resources, sharing the research costs and evading the risks. For examples, IBM, Toshiba and Siemens co-invested 1 billion US dollars to research the new generation 256M chips. Samsung and IBM co-research the manufacture process of 45nm chips. Even in China, such examples are also frequent: Chunlan and Mitsubishi co-research colored TV CPU, Motorola and Changhong co-research digital TV and information TV, Xiaxin and Xiahua co-research TV-DVD COMBO, and so on. Although R&D alliance has grown promptly, the practical performance is quite low. Some relevant investigation shows that the failed ratio of R&D alliance reaches to 40%~70% (Park S O, Ungson G R, 2001). As the key in R&D alliance is to create a new technology by several enterprises, the valid knowledge exchanging and information sharing are necessary. Hamel & Prahalad found that the purpose and motivation out of which enterprises constitute the union are to convey the knowledge to each other (Hamel G, Prahalad C K. 1989), and Badaracco also believes that unrevealed knowledge between enterprises cannot be achieved by market transactions but by a cooperative alliance (Badaracco J. L. 1991). Therefore, knowledge exchanging and sharing between partners are not only the purpose but also the

Liu, Cheng and Yang: Research on influencing factors of the performance of…

167

foundation of the union. The knowledge sharing mechanism can impact the cooperation efficiency, and even its success.

2

Micro-mechanism and Influencing Factors of R&D Cooperation between Enterprises

2.1 Micro-mechanism When facing complex problems, enterprises would ask for external support and then become the consumers of knowledge if they cannot resolve these problems themselves. The consumers of knowledge need external opinions, judgment and understanding of an issue. Correspondingly, the suppliers of knowledge must be individuals having knowledge of this issue. Holtshouse regards knowledge as a kind of ‘flow’, that is to say, knowledge can flow from the provider of knowledge to the receiver. As far as knowledge providers are concerned, it is a process of selective ‘pushing’. With receivers, it is a process of ‘pulling’. The combination with each other generates the optimal flow of knowledge. (Holtshouse D. 1998) Maintaining the competitive position for an enterprise owes to its internal resource advantage. This kind of resource is valuable, rare and hard to imitate, such as exclusively owned technology. When the resources are unable to meet their growing demands, enterprises will seek for a resource beyond the organization, usually choosing such modes as cooperating with other enterprises and inter-depending on each other. Chung, Singn and Lee believe that enterprises tend to improve their performance and create value (Chung S. Singh H.Lee K. 2000) through combining with other enterprise resources and constituting a union. So, the need to depend on the technology and ability of partners is one of the strengths of R&D cooperation. This power comes from the enterprises’ exterior; thereby we can take it as a pulling force of enterprise cooperation. On the other hand, cooperative R&D helps the enterprises to utilize new knowledge and improve their own learning ability through sharing the cross-organizational knowledge. Kougut points out that enterprises’ setting up a cooperative alliance means to acquire the organizational knowledge of the other side, that is a special organizational convention or skill which constitutes connotative knowledge and can be achieved only through a cooperative union. (Kogut B. 1988) R&D cooperation is a chance to learn from the external, which can gain the knowledge sources and improve the ability of organizing and creating knowledge and then explore the specified fields. (Phan P H, Peridis T. 2000) So organizational learning is the internal power of enterprises’ R&D cooperation. As this power comes from the inside, we can think of it as the pushing power. Creating new technology is the common purpose of the operation among enterprises. As far as the cooperative union is concerned, every member of the union plays the roles of both knowledge provider and receiver simultaneously. The driver for enterprises to cooperate is bidirectional which includes a process of ‘pushing’ and ‘pulling’. This process structures the microcosmic mechanism of the enterprises’ cooperation.

168

3 Network and cooperation in innovation

2.2 Influencing Factors Cooperative R&D can be regarded as a bidirectional process of cross-organizational knowledge communication. According to the theory of managing communication, the basic factors determining communication results involve information, senders, coding, channels, decoding, receivers and understanding of the decoded signals, etc. (Kroue K G, Putnam, L L. 1987) These factors can be combined into three aspects: source, target and channel of the communication. Taking the R&D cooperation as the corresponding process, these factors are the technical features of sharing knowledge, of the internal properties of enterprises, and of the compatibility of the cooperative process. Meanwhile, as the reliance on resources and organization learning constitute the drivers of R&D cooperation among enterprises, technical resources owned by the partners and the learning ability of enterprises also should be two factors which can affect the cooperative results. Summarizing the analysis above, we can conclude that the main factors affecting the performance include: the technical resource intensity owned by the partners, the organizational learning ability of enterprises, technical bulwark properties of sharing knowledge, and compatible levels among partners.

3

Hypothesis and Variable Design

3.1 Technical resource intensity owned by partners Theories about resource reliance suggest that there do exist critical rare resources needed for the enterprises’ operation in the environment. The target of an R&D cooperation union is to seek these resources externally in order to cover their own shortage. Dyer and Singh believe that if two sides are willing to cooperate, share their own technical resources and have enough investment, both of them can obtain greater competitive advantages in the market. (Dyer J H, Singh H. 1998) Then we can put up hypotheses as follows: Hypothesis 1: Technical resource intensity owned by the partners is positively related to the performance of cooperation among enterprises. Based on the analysis above, we take three metrics: mutual complement of technological resources (x1), common degrees (x2) and investment degrees (x3) to measure the intensity of the technological resource. 3.2 Organizational Learning Ability of Enterprises Organizational learning is the procedure that transforms the acquired knowledge into organizational knowledge which can reform the organizational activities and be further enlarged and stored in the organization. (Argryris C, Schon D. 1996) Although based on the individual learning, organizational learning more emphasizes the transformation, diffusion and storage of knowledge. This ability is called organizational learning ability which can be taken as the velocity of absorbing new knowledge and as the results of the conversion. Then we have the hypothesis as follows: Hypothesis 2: Organizational learning ability is positively related to the performance of R&D cooperation among enterprises.

Liu, Cheng and Yang: Research on influencing factors of the performance of…

169

We use three metrics: learning desire (x4), digesting ability (x5) and integrating ability (x6) to measure the organizational learning ability. 3.3 The bulwark property of technical knowledge As R&D alliance is a cooperation of technical levels, the knowledge needed in the process of R&D design and manufacturing is technical knowledge, the same of which shared among enterprises. Therefore, technical property of knowledge will have effects on the cooperation. These effects are called ‘technical bulwark’. Then we have a hypothesis as follows: Hypothesis 3: the bulwark property of technical knowledge is negatively related to the performance of the R&D cooperation. We use three metrics: exclusive property of technical knowledge (x7), route reliance attributes (x8), and modularization degrees (x9) to measure the bulwark property of technological knowledge. 3.4 Compatibility level among partners Hagel and Singer express the relationship between enterprises more precisely by the term interaction costs. Interaction costs not only include transaction costs, but also the cost of exchanging information between enterprises. Enterprises will reach their own goals by having minimal interaction costs. (Hagel J, Singer M. 1999) In the case of certain transaction costs, the higher the compatibility among the enterprises, the lower the cost of information exchange and interaction. On the other hand, the compatibility among the partners can guarantee a harmonious and lasting cooperation and avoid the emergence of conflicts. Then we have the hypothesis as the follows. Hypothesis 4: The compatibility level is positively related to the performance of the R&D cooperation. We use three metrics: the compatibility of enterprise cultures (x10), the openness of enterprise cultures (x11) and the trusting degrees of partners (x12) to measure the compatibility level among enterprises. 3.5

Measurement for the performance of the R&D cooperation between united enterprises In the study of management science, performance is usually a resulting variable to measure the results when influencing factors are changed. The corresponding metrics are mainly financial metrics, such as profit ratio or marketing sharing ratio. As performance is the collective outcome out of all the activities, it is hard to only measure the performance knowledge sharing in the R&D cooperation brings. Thereby it is insufficient and impractical to measure the performance of the R&D cooperation only by financial metrics. In some documents, a number of scholars use subjective metrics to measure the performance of enterprise growth, such as satisfaction ratio or the degree of reaching the goals. (Roger C, David J B. 1998) On the other hand, Geringer and Hebert discovered that objective metrics are positively related to subjective metrics to a remarkable degree and that there is a distinct,

170

3 Network and cooperation in innovation

replaceable relationship between the two kinds of metrics. (Geringer, J.Michael; Hebert, Louis. 1991)

Figure 1: Conceptual Model of the Performance of R&D cooperation

Based on the consideration above, we review the performance of the R&D cooperation from objective and subjective levels. Objective metrics include: the categories of new products (y1), an increasing range of products profits (y2), and a decreasing range of production costs (y3); subjective metrics include: the promotion of the ability of research and development (y4), the satisfaction degrees of the cooperation (y5), and the extent of reaching cooperative goals (y6). From the theoretic analysis, we set the conceptual model as shown in Figure 1. The four hypotheses above are changed into eight hypotheses corresponding to the objective and subjective metrics.

4

Case Study

4.1 Questionnaire investigation and methods selection According to the analysis above, we have assigned questionnaires to over 200 enterprises located in Tianjin, Zhejiang, Anhui, etc. and reclaimed 125 valid responses. The involved R&D cooperation styles which covered the development of new products, the improvement of test methods, examinations of materials, the establishment of technological criteria, the research of critical equipments, the alteration of outdated technologies, and the supervision of technical authorization, etc. are representative and reflective of the real situations.

Liu, Cheng and Yang: Research on influencing factors of the performance of…

171

Structural Equation Modeling (SEM) has been widely applied recently which incorporates the features of regarding and processing multiple causal variables, permitting the error of independent variables and causal variables, and the flexibility of test patterns. (Hou Jietai, Wen Zhonglin, Cheng Zijuan. 2004) The structural equation here is modeled by Lisrel 8.54 and the statistical analysis of the data is handled by SPSS 11.5. 4.2 Analysis of Reliability Degree and Validity Degree The reliability degree of the questionnaires is tested by the coefficient of Cronbach α, and the validity degree of those is tested by a factor analysis (test value of KMO is 0.785, which is suitable for factual analysis). With a value greater than 0.6 of Cronbach α and higher than 0.5 of the factor of every tested variable, the tested variables are accountable to the potential variables which indicates the good quality of the questionnaires. Table 1: Analysis of Credit Degree and Validity Degree

172

3 Network and cooperation in innovation

We use Maximum Likelihood (ML) as the computing means of the model, use metrics such as the statistical amount of card square (χ2), ratio of card square (χ2), GFI, and AGFI to evaluate the results. Taking the data of Table 2, the result of the model simulation is better. Table 2: Exponent of Simulation Metrics Referenced value Value of model simulation

Statistic Amount P>0.05 187.61

GFI >0.8 0.94

AGFI >0.8 0.93

RMSEA >0.08 0.061

CFI >0.9 0.96

NFI >0.9 0.90

Table 3: The Path Modulus of SEM and t Value Hypothesis Path Modulus t Value Conclusion

H1a 0.34 2.99

H1b 0.46 4.85

H2a 0.08 1.42

H2b 0.31 2.37

H3a -0.29 -2.26

H3b -0.37 -3.17

H4a 0.17 1.69

H4b 0.22 1.99

β21 0.53 5.36

Passed

Passed

Not Passed

Passed

Passed

Passed

Not Remarkable

Passed

－

Regarding table 3, (1) the technical resource intensity of the partners is positively related to both objective performance and subjective performance. H1a and H1b are supported. (2) It is not proven that the organizational learning ability is positively related to the objective performance. H2a is not supported. But the organizational learning ability is positively related to subjective performance and H2b is supported. The possible reason for this is that it takes a long time to evaluate the effect that organization learning ability has on the objective performance. (3) The technical bulwark property is negatively related to the objective performance and subjective performance. The hypotheses of H3a and H3b are supported. (4) The compatibility level is positively related to the objective performance, but this relationship comes into existence only in the case of p 25

Figure 1 shows the age and size distribution of the firms.

N

%

Biotechnology instruments

25

41%

Biotech service provider

14

23%

Pharmaceutical biotech firms

16

26%

6

10%

61

100%

Other biotechnology firms

10%

41% Biotechnology instruments Biotech service provider Pharmaceutical biotech firms

26%

Other biotechnology firms

23%

Figure 2: Segment distribution

Dowling and Helm: Licensing strategies in the biotech industry

203

3.3 Licensing goals and strategy About over one third (36%) of the German firms had in fact used a licensing strategy. In order to measure the importance of a licensing strategy the firms were asked what percentage of their sales was generated from such an agreement and how they rated the importance of this strategy. The results are shown in Figure 3:

Generation of Revenue (in %) Selling own products

62

Out-licensing products/know-how

39

Selling patents/know-how

28

Services

36 13

Other 0

10

20

30

40

50

60

70

(N = 55, multiple answers possible)

Importance of Different Commercialization Strategies 4.9

Selling own products Out-licensing products/know-how

(N = 33)

4.4

Selling patents/know-how

(N = 24)

3.9

(N = 16)

Services

5.5

Other

(N = 20) 6

0

1

2

3

4

5

6

(N = 5) 7

(1 = not important; 7 = very important)

Figure 3: Revenue generated and importance

The data show that after “sale of own products”, licensing was the second largest generator of sales. However, in the rating of these strategies, licensing was less important than services or sale of the firms’ own products. It may be that German biotech firms are still too young to take full advantage of a licensing strategy or that the business segments of German firms have a strong effect. The segment effect is shown clearly in Figure 4, where two thirds of the pharmaceutical firms use licensing agreements, whereas for the other firms it was one third or less. This higher usage of a licensing strategy reflects both the research-intensive and high-risk nature of this segment of the industry and the lack of complementary assets such as distribution systems for new medical products. Instrument producers or service companies have neither the same R&D costs nor long development times. They can also more easily market their products and/or services to potential customers.

204

4 Tools for R&D Management Type of Licenser (N = 61) Biopharmaceutical firms (N=16)

62.5

Biotechnological instruments (N=25)

28

Biotechnological services (N=13)

21.4

37.5

72 yes no

Other (N=6)

78.6

33.3

0%

20%

66.7

40%

60%

80%

100%

Figure 4: Use of licensing by segment

There are also interesting differences between instrument and pharmaceutical firms in terms of the goals of their licensing strategies. As shown in Figure 5, gaining access to new markets and learning is more important to companies producing biotechnological instruments.

Learning from the licensee and increasing experience

4.3

2.4

4.2 4

Reducing development risks

4.6 4.3

Profiting from the licensee's reputation Gaining access to new markets

5.2

1.9

Financing further development

5.4

6.1 6.2 6.4

Earning revenue

0

1

2

3

4

5

6

7

1 = not important; 7 = very important (N = 16) Biopharmaceutical firms

Biotechnological instruments

Figure 5: Goals by segment

Legal restrictions vary between segments, which may also have an influence on licensing strategy. As shown in Figure 6, the companies in the pharmaceutical segment viewed the legal regulations as most restrictive.

Dowling and Helm: Licensing strategies in the biotech industry

205

Legal restrictions are too restrictive: 100% 80%

47 68

60%

77

83

40% 53

20%

32

23

17

0% Biopharmaceutical firms

Biotechnological instruments

yes (N = 20)

Biotechnological services

Others

no (N = 39)

Figure 6: Importance of legal regulation

Another interesting aspect of a licensing strategy is whether firms focus on core technologies or only use licensing for less important parts of their business. In addition, some firms may only award licenses exclusively to certain partners, or offer licenses to multiple customers. Figure 7 shows the data for the firms surveyed. Clearly the German firms develop licenses from their core technologies in the great majority of cases (81%) and over half of these licenses are exclusive to one partner.

Licensed Product was Part of the Core Business:

Usage of Exclusive Licenses

100%

100%

19 80%

80%

60%

42.9

60% no (N = 39) yes (N = 20)

40%

81

40% 57.1 20%

20%

0%

0% (N = 21)

Figure 7: Core technologies and exclusivity

(N = 21)

206

4 Tools for R&D Management

The source of the technology to be licensed also affects the objects pursued, as shown in Figure 8. The main objectives when licensing core products are: earning revenue, financing further development and gaining access to new markets. In contrast, when licensing noncore products, factors such as reputation, learning, and access to other patents are more important than core products.

(N = 21) Earning revenue

Financing further development

Reducing development risks

Gaining access to new markets

Profiting from licensee's reputation Learning from licensee and increasing experience -Access to patents, licences and know how of licensee

1

2

Product is not part of the core business

3

4

5

6

7

Product is part of the core business

Figure 8: Goals for core vs. non-core technologies

3.4 Comparison of Germany and the United States The use of the survey instrument by Kollmer allows for a comparison of the licensing strategies of German and US-based biotech firms. Some interesting differences and similarities were found. First, the use of exclusive licenses is much higher in the United States (99%) than in Germany (57%). This high proportion of non-exclusive licensing projects is due to rules of government-funded programs in Germany. However, when comparing the reasons for licensing, no essential differences could be found, as shown in Figure 9. Neither were there any significant differences concerning general objectives. Only small differences in licensing projects of core vs. non-core patents could be seen. The type of payments preferred by the licenser, i.e. up-front payments and milestone payments, are more common in the USA.

Dowling and Helm: Licensing strategies in the biotech industry

207

Comparison of Germany and the USA concerning Key Objectives and Reasons for Licensing Earning revenue Financing further development Gaining access to new markets Profiting from licensee's reputation Reducing development risks Learning from licensee and increasing experience -Access to patents, licences and know how of licensee

1 USA (N = 70)

2

3

4

5

6

7

Germany (N = 20)

Figure 9: Comparison of Germany and the USA on key objectives and reasons for licensing

4

Conclusions

In summary, this replicative study of biotechnology start-up firms in Germany based on the earlier research of Kollmer and Dowling (2004) has shown interesting differences and similarities between the industries in the two countries. First of all, licensing is not a commonly used strategy in German biotech firms. However, licensing is especially important to biopharmaceutical firms in Germany, which represent a lower proportion of all firms than in the United States. In terms of exclusivity, exclusive licenses are used more often than nonexclusive license agreements. However exclusive licensing is less common in Germany than in the USA, where exclusive licenses are used most of the time. We found only slight differences between Germany and the USA concerning general objectives of licensing strategies and product specific factors. The most important difference found was that US firms use upfront and milestone payments as compensation more often than German companies. Overall, the satisfaction of German biotechnology firms with their licensing strategies is high. This research suggests that licensing may become a more important strategy as the German biotechnology industry matures and more firms target the market for pharmaceutical products. As part of a larger research project we plan to collect data on other less developed biotech industries, for example in China, to enhance further our understanding of the use of licensing as a competitive growth strategy in this crucial industry.

208

4 Tools for R&D Management

Literature Adam, Y., Ong, C. H. and Pearson, A. W., 1988, Licensing as an alternative to foreign direct investment: an empirical investigation, Journal of Product Innovation Management 5, 32-49 Autio, E., 1994, New technology based firms as agents of R&D and innovation: an empirical study, Technovation 14, 259-273. Autio, E., 1997, Atomistic and systemic approaches to research on new technology-based firms: a literature study, Small Business Economics 9, 195-209. Baier, W. and Pleschak, F. (Hrsg.), 1996, Marketing und Finanzierung junger Technologieunternehmen: den Gründungserfolg sichern (Gabler, Wiesbaden, Germany) Churchill, N. C. and Lewis, V. L., 1983, The five stages of small business growth, Harvard Business Review (May-June), 33-50. Helm, R. and Meckl, R., 2004, Strategic Strengths and Weaknesses of Small and Medium Sized High-Tech Firms –A Focus on Strategy Development. Science Research Management 25(ZK 2004), 120- 124. Helm, R. and Mark, A., 2006, Signal Effects and Producer Reputation: Implications for the Marketing of New Products of Established and New Firms. International Journal of Product Development 3, (forthcoming). Kollmer, H. and Dowling, M., Licensing as a Commercialization Strategy for New Technology-Based Firms, Research Policy, 2004, 33 (2004) 1141-1151. Lerner, J. and Merges, R. P., 1998, The control of technology alliances: an empirical analysis of the biotechnology industry, Journal of Industrial Economics 46, 125-156. Ford, D., 1985, The management and marketing of technology, in: R. Lamb and P. Shrivastava (Editors), Advances in Strategic Management 3 (London) pp. 103-134. Greiner, L. E., 1972, Evolution and revolution as organizations grow. Harvard Business Review (July-August), 37-46 Oakey, R., 1991, High technology small firms: their potential for rapid industrial growth, International Small Business Journal 9(4), 31-42. O'Farrell, P. N. and Hitchens, D. M. W. N., 1988, Alternative theories of small-firm growth: a critical review, Environment and Planning 20, 1365-1383. Pisano, G., 1991, The governance of innovation: vertical integration and collaborative arrangements in the biotechnology industry, Research Policy 20, 237-249. Stankiewicz, R., 1994, Spin-off companies from universities, Science and Public Policy 21(2), 99-107. Storey, D. J. and Tether, B. S., 1998, New technology-based firms in the European Union: an introduction, Research Policy 26, 933-946.

Dowling and Helm: Licensing strategies in the biotech industry

209

Teece, D. J., 1986, Profiting from technological innovation: implications for integration, collaboration, licensing and public policy, Research Policy 15, 285-305. Telesio, P., 1979, Technology Licensing and Multinational Enterprises (Praeger, New York)

Authors Michael Dowling, Prof. Dr. Head of department Chair of Innovation and Technology Management University of Regensburg Universitätsstraße 31 93040 Regensburg, Germany Tel.: +49 (941) 943-3226 Fax : +49 (941) 943-3230 [email protected] www.wiwi.uni-regensburg.de/dowling Roland Helm, Prof. Dr. Head of department Unilever-Chair of Marketing and Business Administration Friedrich-Schiller-University Jena Carl-Zeiß-Straße 3 07743 Jena, Germany Tel.: +49 (3641) 9-43110 Fax : +49 (3641) 9-43112 [email protected]

Liang Liang and Li Yong: A lexicographic maxmin approach for allocating the fixed cost based upon data envelopment analysis 1

Introduction

Data envelopment analysis (DEA) has been proven an effective tool for performance evaluation and benchmarking since it was first introduced by Charnes et al. (1978). Following the first DEA model ----CCR, a number of different DEA models and their applications (performance evaluation for hospitals, banks, schools etc.) have now appeared in the literature (Seiford and Thrall, 1990; Seiford, 1996; Cooper et al, 2000; Zhu, 2003). Recently, one of the most important applications of the DEA technique has become to allocate the fixed cost among peer decision making units (DMUs). The fixed cost is defined as the organization’s expenses on building the common platform for its subunits, such as the investment by a bank on building a common electric trade system for its branches, or the expenses afforded by a chain super-market on building its common logistics distribution system for its subsupermarkets, etc. When the expenditure has overflowm the total budget of the organization, it may call for its peer subunits to cover the excess. Then, a problem arises as to how this cost can be assigned in an equitable and unique way to the various subunits. Cook and Kress (1999) first make a try to solve the problem under the DEA structure. In their approach, they presume that the allocated cost can be seen as a new input measure to all DMUs. Then according to the two principles efficiency invariance and Pareto-minimality, the equitable allocation is achieved by solving several linear programming problems. However, in the general case, their approach relies on finding a single efficient DMU e.g. via a cone-ratio (Charnes et al. 1989; 1990) approach. Based upon Cook and Kress’ theoretical foundation, Cook and Zhu (2005) extend the approach to other DEA models (CCR, BCC) with input and output orientations, and give a feasible (but not optimal) cost allocation. Jahanshahloo (2004), based upon efficiency invariance, also chooses a simple allocation without computational difficulties among various feasible solutions. However, the three approaches have two common shortcomings: (1) that the allocations are entirely determinated by the input side. For example, if two DMUs have the same input measures, then their allocated costs will be the same, in spite of their different outputs; (2) that the allocations, in the general case, are not unique. With the same assumption of treating the allocated cost as a new input, Beasley (2003) provides an alternative DEA-based cost allocation approach by maximizing the average efficiency across all DMUs and adding additional constraints and models to obtain a unique cost allocation. However, two shortcomings exist. One is that the proposed model, non-linear

Liang and Li: A lexicographic maxmin approach for allocating the fixed cost…

211

programming, causes some computational difficulties for decision makers. The other is that it may be difficulties in the allocation operation, since there is a huge gap between the minimal allocated cost and the maximal one. In this paper, we try to consider the problem from another perspective. The remaining part of this paper is organized as follows. The next section develops models for characterizing and measuring the efficiency of each DMU taking into account its allocated cost. It is followed by the proof that all DMUs can be DEA efficient. Then, based upon the conclusion, a “Maxmin” model is proposed in section 3. In section 4, the unique allocation of the proposed method is compared to traditional approaches from rationality and allocation operation perspectives. It is followed by an application of allocating the advertising expenditure between a manufacturer and its dealers based upon a real data set in section 5. Finally, conclusions and directions for future research are given in the last section.

2

Efficiency evaluation

Suppose we have n independent homogeneous decision making units, where each DMU j ( j = 1, 2,… , n ) consumes m inputs xij , i = 1, 2,..., m to generate s outputs yrj , r = 1, 2,..., s . To evaluate the performance of DMU 0 , its relative efficiency can be ob-

tained by solving the following standard CCR model (Charnes, et al. 1978) s

Max

∑u y r

r =1 m

r0

∑v x

i io

i =1

s

s.t.

∑u y r =1 m

r

rj

≤ 1 , ∀j

∑ vi xij

(1)

i =1

ur , vi ≥ 0, ∀r , i

where ur , vi are unknown multipliers respectively attached to outputs and inputs. Suppose that a cost R is to be distributed among n DMUs. And the allocated cost to DMU j is denoted as R j ,

n

∑R j =1

j

= R . Taking into consideration the allocated cost here, we use the

following model to evaluate the relative efficiency of DMU 0

212

4 Tools for R&D Management s

∑u y r

r =1

Max

r0

m

∑ vi xio + R0

= E0

i =1

s

∑u y r =1

s.t.

r

m

rj

∑ vi xij + R j

≤ 1 , ∀j

(2)

i =1

n

∑R j =1

j

=R

ur , vi , Ri ≥ 0, ∀r , i, j

where E0 is denoted as the relative efficiency of DMU 0 . In this definition of relative efficiency, referring to Beasley (2003), there is an explicitly set weight factor of the allocated cost to one for the sake of simplicity. From model (2), different allocation apparently causes a different efficiency score to each DMU. So, under efficiency measurement, each DMU is eager to choose an optimal allocation among various ones to maximize its efficiency score. Then a problem arises that is how high an efficiency they can get. Theorem 1: Each DMU from model (2) can be DEA efficient. Proof: Consider equations as follows: s

∑u y r

r =1

m

∑v x i =1

i ij

n

∑R j =1

j

rj

= 1 , j = 1, 2,..., n

(3)

+ Rj

=R

ur , vi , Ri ≥ 0, ∀r , i, j

Apparently, the feasible solutions to equations (3) are not only feasible, but also optimal ones to model (2). Therefore, if equations (3) have feasible solutions, then all DMUs can be DEA efficient.

Liang and Li: A lexicographic maxmin approach for allocating the fixed cost…

213

From equations (3), it shows that there are n + 1 equations, but with n+m+s variables which hints that there are feasible solutions for n+m+s>n+1. For example, let n

n

j =1

j =1

vi = 0, ∀i ; ul = R / ∑ ylj , l ∈ {1, 2,..., s}, ur = 0, r ≠ l ; R j = R × ylj / ∑ ylj , ∀j , then n

s

∑u y

Ej =

r

r =1

rj

m

∑ vi xij + R j

ul ylj

=

Rj

=

i =1

n

∑R j =1

j

ylj × R / ∑ ylj j =1 n

R × ylj / ∑ ylj

=1

j =1

n

n

j =1

j =1

, j=1,2,…,n

= R × ∑ ylj / ∑ ylj = R

ur , vi , Ri ≥ 0, ∀r , i, j n

n

j =1

j =1

Thus, vi = 0, ∀i ; ul = R / ∑ ylj , l ∈ {1, 2,..., s}, ur = 0, r ≠ l ; R j = R × ylj / ∑ ylj , ∀j is a feasible solution to equations (3), which means all DMUs from model (2) can be DEA efficient. Q.E.D.

From the first equation of equations (3), we can get a deduction as follows: s

m

r =1

i =1

Deduction 1: The allocated cost of DMU j can also be expressed as R j = ∑ ur yrj − ∑ vi xij , j = 1, 2,..., n

(4)

3

Maxmin-allocation model

Lets consider the following model: Max min R j u ,v

1≤ j ≤ n

s

m

r =1

i =1

s.t. R j = ∑ ur yrj − ∑ vi xij , j = 1, 2,..., n n

∑R j =1

j

=R

(5)

214

4 Tools for R&D Management ur , vi , R j ≥ 0 , ∀r , i, j

Apparently, all DMU fairly use a common set of weights (Cook and Zhu, 2007) in the allocation. And model (5) has two multiple objective functions. The first one, min R j , is to 1≤ j ≤ n

search a DMU to be allocated to the minimal cost among various DMUs. In the following passage, the DMU is treated as a numeraire in the allocating process. The other one is the pessimistic decision to allocate the DMU to the minimal cost. However, if min R j = β , 1≤ j ≤ n

model (5) can be translated into the following model: max β u ,v

s

m

r =1

i =1

s.t. R j = ∑ ur yrj − ∑ vi xij , j = 1, 2,..., n n

∑R j =1

(6)

=R

j

R j ≥ β , j = 1, 2,..., n ur , vi , R j ≥ 0 , ∀r , i, j

Suppose the optimal solutions to model (6) be β1* , u1*r , v1*i , R1*j , ∀r , i, j . Therefore, all DMUs can be divided into two groups: J1 = { j R1*j = β1* , ∀j ∈ J } J 2 = { j R1*j > β1* , ∀j ∈ J } = J − J1

However, the allocated costs for DMUs from J 2 are undetermined since flexibility space exists. Treating DMUs from J1 as numeraires, a further model is proposed as follows: max β u ,v

s

m

r =1

i =1

s.t. R j = ∑ ur yrj − ∑ vi xij , j ∈ J 2 s

m

r =1

i =1

R j = ∑ ur yrj − ∑ vi xij = β1* , j ∈ J1

(7)

Liang and Li: A lexicographic maxmin approach for allocating the fixed cost…

215

R j ≥ β , j ∈ J2 n

∑R j =1

=R

j

ur , vi , R j ≥ 0 , ∀r , i, j

Suppose the optimal solutions to model (7) be β 2* , u2*r , v2*i , R2* j , ∀r , i, j . Thus, J 2 can be divided into two subsets: J 3 = { j R2* j = β 2* , ∀j ∈ J 2 } J 4 = { j R2* j > β 2* , ∀j ∈ J 2 } = J 2 − J 3

If J 4 ≠ ∅ , which means there is also flexibility space in allocation for DMUs in J 4 , then further models should be used until there is no flexibility space to any DMU. Algorithm for getting the Maxmin-allocation Step 1: Let l = 1 , solve model (6) to get the optimal solutions; go to step 2 Step 2: Set l = l + 1 , solve the following general model.

max β u ,v

s

m

r =1

i =1

s.t. R j = ∑ ur yrj − ∑ vi xij , j ∈ J 2l − 2 s

m

r =1

i =1

s

m

r =1

i =1

s

m

r =1

i =1

R j = ∑ ur yrj − ∑ vi xij = β1* , j ∈ J1 R j = ∑ ur yrj − ∑ vi xij = β 2* , j ∈ J 3

R j = ∑ ur yrj − ∑ vi xij = β 2* , j ∈ J 2l − 3

(8)

216

4 Tools for R&D Management R j ≥ β , j ∈ J 2l − 2 n

∑R j =1

j

=R

ur , vi , R j ≥ 0 , ∀r , i, j

Denote the optimal solutions to model (8) as β l* , ulr* , vli* , Rlj* , ∀r , i, j . Then, J 2l − 2 can be divided into two subsets: J 2l −1 = { j Rlj* = β l* , ∀j ∈ J 2l − 2 } J 2l = { j Rlj* > β l* , ∀j ∈ J 2l − 2 } = J 2l − 2 − J 2l −1

If J 2l ≠ ∅ , then go to step 2 again, else go to step 3. Step 3: Denote the total model evolution times k, (k ≤ n), and corresponding optimal solu* , vki* , ∀r , i . And the procedure stops. tions as β k* , uklr

Therefore, all DMU sets J = {1, 2,..., n} can be divided into k groups as k

J = ∪ J 2l −1 l =1

And the unique optimal allocation is ⎧ R*j = β1* ⎪ * * ⎪R = β2 * * * ( R1 , R2 ,..., Rn ) = ⎨ j ⎪ ⎪ R*j = β k* ⎩

j ∈ J1 j ∈ J3 j ∈ J 2 k −1

where β k* > β k*−1 > ... > β1* . Theorem 2: (i) u2*r , v2*i , R2* j , ∀r , i, j also is the optimal solution to model (6). (ii) β 2* > β1* .

Liang and Li: A lexicographic maxmin approach for allocating the fixed cost…

217

Proof: (i) Since the feasible region of model (7) is a subset of the feasible region of model (6), u2*r , v2*i , R2* j , ∀r , i, j is a set of feasible solutions to model (6), while from model (7), we also s

m

r =1

i =1

have R2* j = ∑ u2*r yrj − ∑ v2*i xij = β1* , j ∈ J1 . Therefore, (i) is true. (ii) Since the optimal solutions to model (6) are also feasible solutions to model (7), it follows that β 2* ≥ β1* . While if β 2* = β1* , then J 3 ⊂ J1 . Apparently, this is inconsistent with the fact that

J 3 ⊂ J 2 , J1 ∩ J 2 = ∅ . Therefore, β 2* > β1* . Q.E.D.

Deduction 2: β k* > β k*−1 > ... > β1* , and ukr* , vki* , Rkj* , ∀r , i, j are also optimal solutions to model

(6). From the computing procedure and Theorem 2, the conclusions can be easily made. Deduction 3: The optimal allocation is unique (as from the computing procedure, this is easy to know)

Denote the common weights to all DMUs as μr* = ukr* ,υi* = vki* , ∀r , i , then the optimal allos

m

r =1

i =1

cated cost can be expressed as R*j = ∑ μ r* yrj − ∑υi* xij , j ∈ J . The proposed approach can have the following properties:

Property 1. To any two DMUs, such as

DMU l and

DMU k l , k ∈ {1, 2,..., n} , if

xil = xik , i = 1, 2,..., m, yrl ≥ yrk , r = 1, 2,..., s , then R ≥ R . * l

* k

Proof: From the condition of xil = xik , i = 1, 2,..., m, yrl ≥ yrk , r = 1, 2,..., s , we can conclude n

∑υ x i =1

* i il

=

n

s

∑υ x , ∑ μ * i ik

i =1

s

r =1

Rl* = ∑ μr* yrl − r =1

* r

s

yrl ≥ ∑ μ r* yrk , therefore,

n

r =1

∑υ x i =1

* i il

≥

s

∑μ r =1

* r

yrk −

n

∑υ x i =1

* i ik

= Rk*

218

4 Tools for R&D Management

The property shows that, with the same input profiles, the DMU with higher output profiles will afford more.

Property 2. To any two DMUs, such as

DMU l and

DMU k l , k ∈ {1, 2,..., n} , if

xil ≥ xik , i = 1, 2,..., m, yrl = yrk , r = 1, 2,..., s , then R ≤ R . * l

* k

The proof is similar to that of Property 1. It shows that, with the same output profiles, the DMU with lower input profiles will afford more.

Property 3. To any two DMUs, such as

DMU l and

DMU k l , k ∈ {1, 2,..., n} , if

xil ≤ xik , i = 1, 2,..., m, yrl ≥ yrk , r = 1, 2,..., s , then R ≥ R . * l

* k

The proof is similar to that of Property 1. It shows that, the more gaps between the output profile and its input profile, the more fixed cost the DMU will afford.

Property 4. To any two DMUs, such as DMU l and DMU k l , k ∈ {1, 2,..., n} , if xil = α × xik , i = 1, 2,..., m , yrl = β × yrk , r = 1, 2,..., s , with β ≥ α ≥ 1 , then

Rl* ≥ Rk* . Proof: from the conditions above, we have

n

∑υ x i =1

therefore, s

Rl* = ∑ μr* yrl − r =1

n

∑υ x i =1

* i il

s

n

r =1

i =1

s

n

r =1

i =1

= β ∑ μ r* yrk − α ∑υi* xik ≥ α ∑ μ r* yrk − α ∑υi* xik

= α Rk* ≥ Rk*

* i il

n

= α ∑υi* xik , i =1

s

∑μ r =1

* r

s

yrl = β ∑ μ r* yrk , r =1

Liang and Li: A lexicographic maxmin approach for allocating the fixed cost…

219

It shows that, the larger size a DMU has, the more fixed cost the DMU will afford. Especially, if β = α , then Rl* = α Rk* .

4

Allocation comparison

In this section, we illustrate the proposed method for the cost allocation problem using the same data set in Table 1 as was given in Cook and Kress (1999). It involves 12 DMUs, 2 outputs and 3 inputs with R=100. Table 1: Data sample DMU

Input 1

Input 2

Input 3

Output 1

Output 2

1

9

39

350

67

751

2

8

26

298

73

611

3

7

31

422

75

584

4

9

16

281

70

665

5

6

16

301

75

445

6

17

29

360

83

1070

7

10

18

540

72

457

8

5

33

276

78

590

9

5

25

323

75

1047

10

6

64

444

74

1072

11

5

25

323

25

350

12

6

64

444

104

1199

Alike the computational procedure given above, we have: Step 1: Solve model (6) and get the optimal solutions β1* = 3.9588,

J1 = {11} ,

J 2 = J − {11} = {1, 2,...,10,12}

Step 2: Solve model (7) and get the optimal solutions

β 2* = 5.0334,

J 3 = {5} ,

J 4 = J 2 − J 3 = {1, 2,3, 4, 6, 7,8,9,10,12} .

Step 3: Let l=2, run model (8) and so on, … Step 12: J 2×12 −1 = {12} , J 2×12 = ∅ , then the computing procedure stops. The optimal solution is β12* = 13.562 . The allocated cost to each DMU j is doneted as R*j , as given in the second column of Table 2. In order to make a comparison to traditional approaches, for the sake of simplicity, we

220

4 Tools for R&D Management

denote the allocation results based upon Cook and Kress (1999) and Beasley (2003) as R CK j , R Bj respectively. Table 2: Allocations comparison DMU

R*j

R CK j

R Bj

1

8.4945

14.52

6.78

2

6.911

6.74

7.21

3

6.6056

9.32

6.83

4

7.5218

5.6

8.47

5

5.0334

5.79

7.08

6

12.103

8.15

10.06

7

5.1691

8.86

5.09

8

6.6735

6.26

7.74

9

11.843

7.31

15.11

10

12.125

10.08

10.08

11

3.9588

7.31

1.58

12

13.562

10.08

13.97

From the Cook and Kress (1999) model, it shows that R9CK = R11CK = 7.31 . This arises since,

from Table 1, DMU 9 and DMU11 have identical input profiles ( X 9 = X 11 ). However, these DMU pairs have different output profiles ( Y9 > Y11 ). The reason for that is that DMU 9 has utilized the common platform more than DMU11 , either in using times or frequency. Then it is reasonable to claim that DMU 9 should afford more fixed cost than DMU11 . In fact, the results based upon our approach is R9* = 11.843> R11* = 3.9588. Furthermore, another pair of DMUs, such as DMU10 and DMU12 , have a similar relationship as compared to DMU 9 and DMU11 , for X 10 = X 12 , Y10 < Y12 . As from the results of Cook and Kress (1999), R10CK = R12CK = 10.08 , while from our approach, we have R10* = 12.125> R12* = 13.562. Therefore, our approach has this property (Property 1), while that of Cook and Kress (1999) does not.

Since Beasley’s results show that R9B = 15.11 > R11B = 1.58, R10B = 10.08