the dual role of universities in industrial innovation in emerging economies:
DESCRIPTION
The Dual Role of Universities in Industrial Innovation in Emerging Economies: A Comparative Study of China and the UK Xiaolan Fu Oxford University Jizhen Li Tsinghua University. Motivation - PowerPoint PPT PresentationTRANSCRIPT
The Dual Role of Universities in Industrial Innovation in Emerging Economies:
A Comparative Study of China and the UK
Xiaolan Fu Oxford University
Jizhen LiTsinghua University
Motivation
•Universities as a major contributor to advances in basic scientific research and the creation of innovation of great novelty.•Most of the received wisdom on the role of universities is based on experiences and evidence from the developed countries.•What is the role of universities in emerging economies?•How does the university-industry relationship in emerging economies differ from those in developed countries?
Proposition
Dual roles of universities in industrial innovation in emerging economies
1. Serve the traditional role of Knowledge creation2. Decipher and adapt advanced foreign technology
which is transferred to the developing countries.3. The importance of the latter role is likely to be
greater than the former in the emerging economies.
Theoretical Framework
• The role of universities in national/regional Innovation system (eg. Nelson, 1990)•The multi-faceted role of universities: education, knowledge creation, problem-solving activities and public space provision (eg. Hughes, 2010)•Firms are able to access a diversified range of knowledge sources (Kaufmann and Todtling, 2001)
•University to add industrialisation (eg. Etzkowitz and Leydesdorff, 1997; Mowery and Sampat, 2005)•Collaboration with universities in open innovation era (Chesbrough, 2003; Kitson, et al., 2001; Mowery, et al., 1996; Powell and Grodal, 2005)
Uni. & innovation in the emerging economies•The level of development of an economy will influence the role of its universities: 1. Determines the level of its indigenous tech. & innovation
capabilities and the major source of technology upgrading: foreign tech. acquisition.
2. Due to the lack of sufficient absorptive capacity in the local industry, there is a need to tap into the expertise in the universities.
3. Universities in these countries are capable for both shallow and deep assimilation through reverse engineering, but mostly not ground breaking research .
4. Affect the type of industrial and technology policy the government will adopt: promote univ. to serve the industries by use-driven science policies.
Proposition
Dual roles of universities in industrial innovation in emerging economies
1. Serve the traditional role of Knowledge creation2. Decipher and adapt advanced foreign technology
which is transferred to the developing countries.3. The importance of the latter role is likely to be
greater than the former in the emerging economies.
Innovation in China
R&D Inputs Compared: G5 and BRICKS 2006
16.8 87.5
23.2
35.6
19.6
345.7
142.2
68.7
36.3
41.7
-2,000
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
0 0.5 1 1.5 2 2.5 3 3.5 4
Spending on R&D as % of GDP
Scie
ntis
ts &
Eng
inee
rs in
R&
D Pe
r Mill
ion
Brazil China India Korea Russia US Japan Germany UK France
Brazil
Germany
China
India
Japan
USA
Korea
RussiaUK
France
+Sheet3!
7 9 11 13 16 20 25 31 35 40 51 55 68 90 104129154197
245300
371
461.6
579.19
0
100
200
300
400
500
600
700
R&D expenditure of China, 1987-2009 (bilyuan)
Universities and innovation in China
Before 1978: central planning, soviet system, universities play an important role in NIS1985: S&T system reforms -Marketisation of the S&T system (planning--market) -Focus on serving economic development (defence –eco) -Acquisition of foreign technology 1995: Acceleration of S&T development2006: Strengthen indigenous innovation
• Universities still play important role• Advocating a use-driven science policy •Marketisation provided incentives for uni-ind collaboration• University owned enterprises
Methodology
•Two steps1. Test the impact of uni-industry collaboration on
innovation of different levels of novelty (distinguish the country origin of the universities)
2. Benchmark evidence from China with that from a classic developed economy: UK
• Measurement of innovation: % of new sales1. Novel innovation: new to the world2. Diffusionary innovation: new to country / firm & sig.
improved.
Measurement of uni-industry linkage
1. Uni-industry collaboration, with: -domestic university -uni in NIEs -US & Europe -Other countries
• Controls: -other external R&D activities -In-house R&D -Firm size, age, sector, human resources
Estimation Strategy
•Tobit model
•Generalised Tobit (Hurdle) model: correction of selection bias
•IV estimates: control for endogeneity problem between collaboration and innovation performance
Data
•Chinese CIS: 1408 manufacturing firms, 2005-07•In 42 cities (both coastal and inland), 83.6% response rate•Cleaned sample: 802 firms•highly skewed to innovative firms: 95% innovators
•Benchmarking: UK CIS: 2002-04, 4296 manuf. firms•Assume consistency in trends in a matured economy•Persistence in collaboration pattern in CIS4 and CIS5 (DTI, 2008)
•Matched sample: by size, industry and innovativeness•793 matched firms due to industry structural diff.
Results: matched sample
Results: China, Tobit model estimates
Novel innovation Diffusion innovation
1 2 3 4 5 6
VARIABLES model model model model model model
co with other org. 10.40**
9.828***
(5.111)
(3.321)
co with universities
5.344
4.276
(5.205)
(3.513)
co with domestic uni.
1.345
5.954*
(5.094)
(3.532)
co with uni in NIEs
35.67***
-5.64
(11.25)
(17.1)
co with uni in US/EU/Japan
17.29
-13.58
(10.72)
(11.04)
co with uni in other countries
35.32**
-8.254
(15.97)
(16.5)
lrdin 2.106** 1.299 1.203 2.949*** 3.007*** 2.929***
(0.865) (0.937) (0.921) (0.544) (0.589) (0.595)
lrdex
1.623** 1.690***
0.0666 0.0009
(0.649) (0.647)
(0.46) (0.461)
Robustness check: Hurdle model
Novel innovation Diffusion innovation
1 2 3 4 5 6
co with other org. 2.3
6.250**
(1.442)
(2.824)
co with universities
2.111
3.967
(1.414)
(2.786)
co with domestic uni.
1.045
5.239*
(1.402)
(2.779)
co with uni in NIEs
12.09**
-0.693
(5.914)
(11.99)
co with uni in US/EU/Japan
8.429*
-8.014
(4.696)
(9.338)
co with uni in other countries
17.08**
-5.956
(8.088)
(15.87)
lrdin 0.409* 0.357 0.267 2.308*** 2.331*** 2.329***
(0.213) (0.224) (0.227) (0.427) (0.453) (0.45)
lrdex
0.105 0.119
0.197 0.108
(0.185) (0.184)
(0.363) (0.363)
Selection bias insig. Bec 95% innovators with + new sales
Robustness check: IV estimates
Novel innovation Diffusion innovation
1 2 3 4
co with other org. 18.68 -13.28
(16.23) (12.53)
co with universities
7.868
-0.775
(11.86)
(8.565)
lrdin 0.665 0.998 3.637*** 3.172***
(1.085) (1.012) (0.744) (0.66)
lrdex 1.308* 1.523** 0.466 0.164
(0.763) (0.751) (0.545) (0.518)
age -0.24 -0.24 -0.00426 -0.0181
(0.164) (0.169) (0.0657) (0.0661)
size4 3.417 4.733 -4.457 -5.526
(5.762) (5.675) (4.129) (3.923)
lack_hc1 1.606 -2.302 -10.05* -6.408
(6.547) (5.323) (5.407) (4.136)
Industry dummies yes yes yes yes
Constant yes yes yes yes
Observations 816 816 801 801 Log Likelihood -1691 -1679 -3693 -3680 Wald test of exogeneity (p-value) 0.424 0.768 0.077 0.473
IVs: location dummy, group dummy, competition, imp. of uni. Information.
Universities and firm innovation in selected university concentrated cities in China
Novel innovation Diffusion innovation
1 2
co with domestic uni. 1.297 15.31*
(14.14) (8.044)
co with uni in NIEs 5.047 3.239
(28.14) (11.00)
co with uni in US/EU/Japan 26.98 4.418
(25.52) (8.481)
co with uni in other countries 50.65*** 30.68***
(12.52) (6.164)
lrdin 0.429 2.076
(2.307) (1.375)
lrdex 0.372 -1.184
(1.603) (1.078)
UK: Tobit model estimates
Matched sample Manufacturing full sample
Novel innovation Diffusion innovation Novel innovation Diffusion innovation
1 2 3 4 5 6 7 8 9 10 11 12
co with other org. 4.514**
1.687
12.92*** 13.37***
(1.820)
(1.831)
(1.351) (1.367)
co with universities
1.641
0.650
9.834*** 7.353***
(2.128)
(2.392)
(1.685) (1.803)
co with domestic uni.
4.754**
1.326 9.613*** 6.727***
(2.222)
(2.337) (1.733) (1.850)
co with uni in Europe
-4.584
6.345 0.993 4.638
(4.061)
(6.681) (5.001) (5.559)
co with uni in other countries
-4.08
-3.866 -6.479 -2.601
(5.586)
(8.057) (5.432) (6.643)
lrdin 1.238*** 1.060*** 1.036*** 0.138 0.116 0.13 3.130*** 3.121*** 3.152*** 2.691*** 2.814*** 2.827***
(0.275) (0.284) (0.281) (0.261) (0.275) (0.2770 (0.214) (0.222) (0.222) (0.202) (0.207) (0.207)
lrdex
0.864*** 0.873***
0.179 0.13 0.880*** 0.904*** 0.711*** 0.709***
(0.32) (0.318)
(0.298) (0.293) (0.256) (0.258) (0.268) (0.268)
Conclusions
•Chinese universities have played a significant role in the promotion of the diffusion of frontier technology and the creation of diffusionary innovation.
•The contribution of domestic universities to the creation of novel innovation is limited in China.
•International innovation collaboration with foreign universities appears to be fruitful in the creation of novel innovation in China; in particular the collaboration with universities in NIEs and other countries other than US, EU & Japan.
•Collaboration with universities in the Western industrialised economies does not appear to be as fruitful as expected.
Conclusions
•Comparing China to the UK, Chinese firms appear to have more extensive, but less effective collaboration with universities.
•In the UK, collaborative linkage with domestic university is the only form which contributes significantly to industrial innovations in British firms.
•Policy implications: domestic universities in developing countries are best positioned to help local firms in assimilation, adaptation and development of foreign tech.
•Future research should investigate in-depth, in both China and UK, the reasons why international innovation collaboration with universities in the most advanced economies functions ineffectively.
Thank you!