140 likes | 309 Views
Aims of this paper. Evaluating the
E N D
1. Subsidy and networking: The effects of direct and indirect support programs of the cluster policy Junichi NISHIMURA (Hitotsubashi U.)
Hiroyuki OKAMURO (Hitotsubashi U.) 1
2. Aims of this paper Evaluating the ‘Industrial Cluster Project’ (ICP) and bringing out the conditions necessary for the effective organization of cluster policies for improving firm performance
Two research questions:
Are cluster participants that exploit support programs more successful in alliance/network formation within the cluster than the others?
Which kind of ICP support programs contributes more to firm performance? (Differences between direct R&D support and indirect networking/coordination support) 2
3. Distinction between direct and indirect support programs Direct support
Heavy (hard) government intervention
e.g. R&D consortium, R&D subsidy, incubation service
Indirect support
Light (soft) government intervention
e.g. provision of information via websites,
organizing of meetings and events (research meetings,
business/financial matching events) ,
consultancy and advisory services (technological,
managerial, and financial consultation and advise) 3
4. Literature and motivation (1) Regional innovation system has attracted many researchers, but few empirical studies on cluster policies, especially on their effective organization. (Abramo et al., 2009; Acs et al., 2002; Aldieri and Cincera 2009; Anselin et al., 1997; Audretsch and Lehmann 2005; Jaffe et al., 1993; Porter; 1998; Rondé and Hussler 2005; Yang et al., 2008)
In contrast to the preceding projects in Japan, the ICP aims at the autonomous development of regional industries and comprises both direct R&D support and indirect networking/coordination support. (METI, 2005; 2006) 4
5. Literature and motivation (2) Direct R&D support
-- Market failure (Griliches, 1992; Spence, 1984; Teece, 1986)
-- Three mechanisms (absorptive capacity, cost sharing, pump-priming effect) (David et al., 2000) and institutional-building trust (Zucker et al., 1998)
-- Empirical studies achieve no consensus on the effectiveness of public R&D support (Branstetter and Sakakibara, 2002; Czarnitzki et al., 2007; Hujer and Radic, 2005; Hussinger, 2008; Grilli and Milano, 2009)
Indirect networking/coordination support
-- Government failure (Cowling et al., 1999; Hospers et al., 2009; Porter, 2000; Wolf, 1990) and knowledge-specific failures (Dobrinsky, 2009)
-- Localized knowledge spillovers (Fujita, 2007; Malmberg et al., 1996)
-- Falck et al. (2009) empirically evaluate the cluster policy in Germany and suggest that government can support industry prosperity by providing public infrastructure and other institutions which promote network formation.
5
6. Characteristics of the ICP Purpose: Promoting industry-university-government collaboration and innovation
Project periods: 2001-2005 (1. period), 2006-2010 (2. period), 2011-2020 (3. period)
Participants and budgets: Around 10,000 firms (2009), 110 billion yen from 2001 to 2004 (0.81 billion euro/1.2 billion US$/1.3 billion A$)
Regions: 17 regional projects (2. period), most of which cover two or more prefectures
Several distinctive characteristics
1) Its policy approach is in contrast with the former policies.
2) Its policy approach is similar to that of successful European clusters.
3) The geographical scope of each regional project is considerably wider than that of any other cluster policies. 6
7. Data Questionnaire data
List up 2,668 firms that participate in the ICP
Effective responses from 511 firms (19%)
322 cluster participants (users) utilized any support programs in the ICP between 2006 and 2008, while 189 (non-users) have never used them.
Collect data on several characteristics of firms and top managers, the year and motivation of participation in the ICP, the extent of network formation before and after participation, the utilization of support programs and the outcomes based on them
7
8. Sample characteristics (users vs. non-users) 8
9. Utilization of support programs 9
10. Outcomes 10
11. Empirical models Network formation
1) Difference-in-Differences (DID) estimation
2) Propensity score method
The effect of support programs
3) Heckman’s two-step model and negative binomial model 11
12. Estimation results (1) 12
13. Estimation results (2) 13
14. Estimation results (3) 14
15. Conclusions Cluster participants that exploit support programs expand network after participating in the ICP. In particular, indirect networking/coordination support contributes to building up new collaborative network within the clusters.
Not every support program contributes to improving firm performance. Indirect support programs have an extensive and strong impact on discrete outcomes, while direct R&D support, except for R&D subsidy, has rather a weak effect. In terms of cost performance, indirect support programs (2 billion yen) are even better than direct R&D support (55 billion yen).
Only 63% of the cluster participants utilized any support programs. Participation itself does not affect firm performance (Nishimura and Okamuro, 2009). It may be necessary to encourage the exploitation of various support programs. 15