The Cluster Initiative Greenbook

Nor is there any significant difference if they are grouped by main financing source.
Government-financed CIs do not perform significantly better or worse than those
financed primarily by industry or equally by industry and government.
The only pattern emerging from the data is that the few CIs initiated primarily by the
university sector have performed somewhat better in terms of improving ties between
industry and academia, which is not surprising.
CI has increased int'l competitiveness
CI has attracted new firms to area
+.19 +.18
CI competed to get gov't financing
Not sig.
Choice of cluster based on gov't
research
Not sig.
CI competed to get gov't financing
Choice of cluster based on gov't
research
Figure 50
Relationship between
performance and early
government intervension
Source: GCIS 2003
Early government intervention
The findings are mixed regarding government actions on an initial stage to ensure the
success of a CI (see Figure 50 above). On the one hand, those CIs that went through
a process of competing with other CIs to get government financing tend to perform
better in terms of competitiveness, but not in terms of attracting new firms. On the
other hand, if government bases its choice of which cluster to support with a CI on
research identifying “attractive” industry sectors, this is related to better performance
in attracting new firms, but not to increasing competitiveness.
There are other types of government intervention, which have no significant effect
at all. If the government decides from the start which companies to involve in the CI
– instead of leaving this choice to the industry or having the CI open to all – there is no
measurable effect (not shown in graph).
Picking the right members
Limiting the scope of the CI by aiming for a certain subgroup within the cluster as
members does not help performance. (See Figure 51 below.) CIs with most members
within one hour’s travel distance, CIs with members on a particular level in the value
chain and not their suppliers or customers, CIs avoiding having direct competitors as
members, and CIs aiming at large companies rather than small ones have not per-
CI has attracted new firms to area
Not sig. Not sig. Not sig. Not sig.
Most parties within 1 hr
travel distance
Single level in value chain
No direct competitors as
members
Large companies, not
small
-.26
Domestic companies, not
foreign owned
Figure 51
Relationship between
membership scope and
performance in attracting new
firms
Source: GCIS 2003
The Cluster Initiative Greenbook • 49