PowerPoint
Characteristics and Some Cases of Cluster Evolutionary
TrajectoriesChulwoo Lee, Professor, Dept. of Geography, Kyungpook
Natl Univ., Korea
Jihye Jeon, PhD candidate, Dept. of Geography, Kyungpoon Natl
Univ., Korea
Parallel Session 1.2: Analysis of Cluster Models and Cluster
Ecosystems
Contents2. Research Background and Purpose. Adaptive Cycle
Modeland Some Cases. Modified Cluster Adaptive Cycle Model1.
Adaptive Cycle Model2. Cluster Adaptive Cycle Model1. Concepts2.
Cluster Evolutionary Trajectories3. Significances and limits
2
3
. Research Background and PurposeRecently, the development of
ICT(Information and Communications Technology) and the high-tech
industry, and then the globalization of economy and the shift of
economic paradigm so called knowledge based economy Economic space
including industrial districts is changing
dynamically.Consequently, research on industrial districts so
far...Static studies : analysis on formation factors and existence
form(Park, 1994; Lee, 2011; Lee and Lee, 1998; 2000); development
plan(Lee, 2004; Park, 2003); policy evaluation(Nam, 2004; Lee,
2005; Lee and Lee, 2007)Dynamic studies : discussion on evolution
of industrial districts (interests in a series of historical
evolution process such as emergence, growth, maturity and renewal
of industrial districts especially in the field of institutional
economic geography)
3
4An appropriate perspective is to consider cluster as complex
system, especially complex adaptive system(Martin and Sunley,
2011).Complex adaptive system is a suitable view for analyzing the
evolution of cluster, so that it is characterized by the feedback
between various components from micro to macro scale and changes to
external shocks as well as self-reinforcing and self-organizing
through internal co-evolutionary mechanismAmong the various models
which can capture complex adaptive system, the most comprehensive
model in terms of identity, stability, exogenous forcing is
adaptive cycle model(Martin and Sunley, 2011).
. Research Background and Purpose
4
5In Korea, some studies investigated the life cycle phases of
industrial districts and the characteristics of each period by
applying the life cycle model(Jeon, 2010; Koo, 2012; Jung, 2013).
Under recognizing the limitations of life cycle model, some studies
applied the adaptive cycle model to investigate cluster evolution,
and suggested growth factors for future sustainable
development(Huh, 2013; Nam, 2014).
. Research Background and PurposeIn this context, this
presentation examines the various trajectories of industrial
districts, their characteristics and some cases based on the
adaptive cycle model.
5
6
. Adaptive Cycle Model1. Adaptive Cycle Model
ConceptsClarifies the evolutionary process that super-system and
sub-system affect each other, changing the structure and function
of the entire system through complex feedback processes in
ecosystems with circularity(Holling, 2001)Posits a four-phase
process(exploitation, conservation, release, reorganization) of
continual adjustment in ecological, social and environmental
systems in terms of change of accumulation, connectedness, and
resilienceAccumulation: the potential of accumulated resources
available to the systemConnectedness: the internal connectedness of
system componentsResilience: a measure of system vulnerability to
and recovery from shocks, disturbances and stresses
6
7
. Adaptive Cycle Model
Source: Holling, 2001; Martin and Sunley, 2011Period of
experimentation and restructuring
Accumulation-low and
variedConnectedness-lowResilience-increasesPeriod of stasis and
increasing rigidity
Accumulation-slows and
stabilizesConnectedness-lowResilience-increasesPeriod of growth and
seizing of opportunities
Accumulation-rapid and
focusedConnectedness-increasingResilience-highPeriod of contraction
and decline
Accumulation-disinvestment and
destructionConnectedness-lowResilience-increases(Re)emergence and
growthStabilization, stagnation and decline< Adaptive cycle
model of the evolution of a complex system >
7
8
Martin and Sunley(2011) applied adaptive cycle model to
clusters
. Adaptive Cycle ModelThroughout each step, capital
accumulation, connectedness, resilience indicate the cycles.Three
scenarios following the Release and Decline phase are shown: A,
cluster disappears; B, the cluster undergoes a phase of renewal;
and C, a new cluster emerges and replaces the old oneCapital
accumulation: the accumulation of productive, knowledge and
institutional capital; Connectedness: the extent of traded and
untraded interdependencies among cluster firms; Resilience: the
capacity of firms to respond flexibly to shocks internal or
external to a cluster
< Stylized evolution of a cluster over an adaptive cycle
>Source: Martin and Sunley, 20112. Cluster Adaptive Cycle
Model
9
. Adaptive Cycle Model3. Significance and limits
SignificanceThe model relies on an ecosystem analogy, and also
allows for the possibility of system (cluster) renewal (recovery)
as well as replacement, or maladaptive collapse at the same time.
This seems to be valuable idea to explore cluster evolution.
The assumption that the evolution of a complex system always
occurs through a four-phase sequence is restrictive Might be open
to similar criticisms with the life cycle model
LimitsAn emphasis is not balanced between endogenous and
exogenous forcing mechanisms to move a system though these four
phase emphasis on endogenous mechanismsA rather restrictive
allowance for the two-way nature of the interaction between a
cluster and its external environment
9
10
. Modified Cluster Adaptive Cycle Model
Martin and Sunley(2011) recognized numerous development
trajectories according to complex interactions between clusters and
their environment, and contingent and strategic decision-making by
cluster-based firms.They modified and expanded the adaptive cycle
model, and then suggested six different possible sequential
trajectories.Cluster full adaptive cycle, Constant cluster
mutation, Cluster stabilization, Cluster reorientation, Cluster
failure, Cluster disappearance
< Modified cluster adaptive cycle model >Source: Martin
and Sunley, 20111. Concepts
10
112. Cluster Evolutionary Trajectories
. Modified Cluster Adaptive Cycle Model
1) Cluster full adaptive cycle (rK)Phases and
CharacteristicsEmergence, growth, maturation, decline and
replacement by a new cluster.The replacement cluster would draw
upon resources and capabilities inherited from the old clusterThe
cluster atrophies due to internal rigidities or exhaustion of
increasing returns effectsBut a new cluster emerges by utilizing
the inherited resources and capabilities
11
12
. Modified Cluster Adaptive Cycle Model
1) Cluster full adaptive cycle (rK)Example
The growth of polymers after the decline of the tyre cluster,
Akron, Ohio (Carlsson, 2001)The growth of low-carbon technology
industries, the Ruhr, GermanyThe basis of a creative district
focused on specialist retailing after the shrinkage of the
Birmingham jewellery quarter, UK (De Propris and Lazaretti, 2008)An
outdoor equipment and clothing industry after the decline of
textile and steel industries (Parsons and Rose, 2005)The Seoul
Digital Industrial Complex, Seoul, Korea (Koo, 2012)
12
13
. Modified Cluster Adaptive Cycle Model
2) Constant cluster mutation (rrr)Phases and
CharacteristicsEmergence, and growth with constant structural and
technological changeThe cluster continually adapts and evolves, by
the successive development of new branches of related activity. The
basic technology would have a comprehensive characteristicsCluster
firms are able to innovate continuously and the cluster constantly
mutates or widens in terms of industrial specialization and
technologyThere are high rates of spin-offs and spin-outs from
local firms, research institutes, or universities
13
14
. Modified Cluster Adaptive Cycle Model
2) Constant cluster mutation (rrr)
ExampleOpen network clusters, such as Silicon Valley and Medicon
Valley (Moodysson et al., 2008)The Cambridge high-technology
cluster, UK (Garnsey and Heffernan, 2005; Stam and Garnsey,
2009)The Dague Seongseo Industrial Complex, Daegu, Korea (Lee,
2007; Lee, 2008)
14
15
. Modified Cluster Adaptive Cycle Model
3) Cluster stabilization (rKK)Phases and
CharacteristicsEmergence, growth, maturation, and stabilizationThe
cluster might remain in a much reduced and restricted form for an
extended period of timeThe remaining firms in the cluster would
survive by upgrading products and/or focusing on niche or prestige
market segmentsThough the cluster retains a modest degree of
resilience, it remains potentially vulnerable to (further)
decline
15
16
. Modified Cluster Adaptive Cycle Model
3) Cluster stabilization (rKK)
ExampleLock-manufacturing cluster, the West Midlands (Bryson et
al., 2008)Transition from the production of final goods to the
production of machinery in some Italian districts (Rabellotti et
al., 2009)Diversification into export markets in Aberdeen oil
complex (Chapman et al., 2004)Machine Industrial cluster of
Changwon, Korea (Lee, 2003)
16
17
. Modified Cluster Adaptive Cycle Model
4) Cluster reorientation (rK)Phases and
CharacteristicsEmergence, growth, onset of early cluster maturation
or decline, and reorientationFirms re-orientate their industrial
and technological specialisms upon reaching or nearing maturation
or decline phase, and new cluster emerges The cluster branches into
a new formThe more innovative leading firms may play a key role by
reacting to market saturation or a rise of major competitorsA
technological breakthrough may activate reorientation
17
18
. Modified Cluster Adaptive Cycle Model
4) Cluster reorientation (rK)
ExampleRadical product diversification in the Montebelluna
sportswear cluster (Sammarra and Belussi, 2006)The Boston
high-technology cluster (Bathelt, 2001)The financial services
cluster in the City of London (Martin and Sunley, 2011)The Gumi
National Industrial Complex (Chung, 2011)
18
19
. Modified Cluster Adaptive Cycle Model
5) Cluster failure (f)Phases and CharacteristicsEmergence and
failure to take off and growAny remaining firms dont constitute a
functioning clusterThe cluster fails to achieve sufficient critical
mass, externalities or market share, the firms would create
unstable innovationNew firm formation is low and/or the firm
failure rate is high, which deters new entrants
ExampleA digital cluster in Dublin (Bayliss, 2007)
19
20
. Modified Cluster Adaptive Cycle Model
6) Cluster disappearance (rKd)Phases and
CharacteristicsEmergence, growth, maturation, decline and
eliminationNo replacement by a new clusterThe inherited resources
and competences are not sufficient or ill-suited to form the basis
of new cluster formation
ExampleSheffield steel (Potter and Watts, 2010), Dundee Jute
(MacKay et al., 2006), Como silk (Alberti, 2006)The Staffordshire
pottery and ceramics district (Sacchetti and Tomlinson, 2009)Coal
industry in Taebaek, the abandoned coal mine areas in Munkyung,
Korea
20
21
ReferencesCarlsson B., 2001, Institutions, entrepreneurship, and
growth: biomedicine and polymers in Sweden and Ohio, Small Business
Economics 19, 105121.Chung. D.C., 2011, Evolution of industrial
cluster through overcoming the lock-in effect of branch plant
agglomeration.De Propris L. and Lazaretti L., 2008, Measuring the
decline of a Marshallian industrial district: the Birmingham
jewellery quarter, Regional Studies 43, 11351154.Holling C.S.,
2001, Understanding the complexity of economic, ecological, and
social systems, Ecosystems, 4, pp.390-405.Huh, D.S., 2013, The
evolution of the IT service industry in the U.S. national capital
region: the case of Fairfax county, Journal of the Economic
Geographical Society of Korea, 16(4), pp.567-584.Koo, Y.M., 2012,
Analysis of cluster life cycle on the dynamic evolution of the
Seoul digital industrial complex in Korea, Journal of The Korean
Association of Regional Geographers, 18(3), pp.283-297.Lee. C.W.,
2003, Exploring regional system of innovation in Changwon, , 23,
pp.327-344.Lee. K.-M., 2007, The characteristic and evaluation of
the governance of the vitalization policy of metropolitan
industrial complex: the case study of Seongseo Industrial Complex
in Daegu, MA.Martin R. and Sunley P., 2011, Conceptualizing cluster
evolution: beyond the life cycle model?, Regional Studies, 45(10),
pp.1299-1318.Moodysson J., Coenen L. and Asheim B., 2008,
Explaining spatial patterns of innovation: analytical and synthetic
modes of knowledge creation in the Medicon Valley life-science
cluster, Environment and Planning A 40, 10401056.Nam, K.B., 2014,
Path dependence and resilience process of the Seoul Digital
Industrial District, Seoul, Korea, The Geographical Journal of
Korea, 48(3), pp.375-388.
21