Transcript Main point from Lecture 1

Content of the lecture



Discussion of two theories of regional clusters, (Marshallian)
industrial districts and Porters regional clusters.
The two cluster theories are discussed in the context of innovation. A
main theme is to which extent and how the two types of clusters
stimulate the innovation capability of cluster firms.
The interest in innovation is related to some ‘stylised facts’ about
firms’ innovation processes:
- Firms rely first of all on internal competence in their innovation
activity
- However, firms also often bring in ideas, information and
knowledge from external actors during the innovation process
- The ‘quality’ of the external environment and the firms’ ability to
use external knowledge determine to a large extent the innovation
capability of an economy
- Firms often find external actors of importance for their innovation
process in their local area. These actors may constitute regional
clusters or innovation systems.
A remarkable industrial development in
the 3. Italy (NEC) from the 1960s






The 3. Italy (NorthEast and Central parts of Italy) demonstrated job
growth in manufacturing in the 1960s and 70s while other parts of Italy
stagnated or declined in number of manufacturing jobs.
The growth took place in traditional (‘old’) sectors such as textile and
clothing, footwear, furniture, ceramic goods and mechanical
engineering, regarded as generally stagnating sectors in western
countries
The growth occurred in small and medium sized enterprises (SMEs),
while the dominating growth paradigm was mass production in large
firms achieving economies of scale (Fordist industrial production)
The SMEs in the same sectors were concentrated in specific areas
(such as textile areas, shoe areas etc.), where the firms formed local
networks, also often containing machine builders.
The growth of SMEs and jobs in the 3. Italy marked a deconsentration
of manufacturing from large industrialised cities in the 1. Italy (north
west) to smaller towns in the 3. Italy (diffused industrialisation).
The 3. Italia model received great interest by researchers, policy
makers and politicians as a possibly new, promising development
model.
Why did the industrial districts grow?



The growth of the industrial districts took place due to a combination of general
and specific factors. The specific factors triggered off the growth when the
general factors became favourable for the districts
General factors
a. Increased market possibilities outside Italy due to better transportation and
EU
b. Increased demand for more niche products (quality products in small batches
where SMEs have an advantage compared to LSEs)
c. Technological innovation, i.e. computerised production equipment that makes
small batch production more efficient
Specific factors in the 3. Italy
a. Tradition and knowledge in entrepreneurship and management of small firms
coming from a specific form of farming (share cropping)
b. Tradition of family firms (which are a flexible towards market changes)
c. Tradition in craft work and the existence of technical schools
d. Knowledge in trade and export (such as the ‘impannatores’ in the textile
industry in Prato)
d. A reserve of unemployed and underemployed in small scale farming that
could work for low salaries (in the development of the industrial districts)
e. Egalitarian societies characterised of local identity and solidarity, which
stimulates local collaboration. Strengthened by the activity of socialist political
parties
What is an industrial district? Definition
of the districts


The districts have many firms and jobs in one or a few adjacent
industrial sectors in a local area. (The area is overrepresented with
jobs in a sector compared with the national average, as measured
with location quotients)
Local networks between firms creating external economies, i.e. the
achievement of effective production through extensive, external
division of labour between firms specialising within various phases
of a production chain. (What can happen to achieve more effective
production when firms share the same suppliers?)

A specific, local socio-cultural dimension denoted as Marshallian
agglomeration economies consisting of three factors:
a) mutual trust between firm leaders, workers etc. that reduces
transaction costs,
b) accumulation of skill among workers (denoted as an ‘industrial
atmosphere’)
c) stimulation of incremental innovations through ‘learning by doing’
and ‘learning by using’ by factor a and b
Strength and weakness of industrial
districts as regards innovation activity


Strength: Firms in the
district frequently
develop incremental
innovations
Why? How can the
districts’ strength in
developing innovation
be explained by the
districts’ external and
agglomeration
economies?


Weakness: The districts
are slow in developing
and making use of
radical innovations?
Why? Which
characteristics of the
districts may explain
their weakness as
regards radical
innovations?
From Mark I to Mark II industrial districts


Mark I Industrial districts: Traditional industrial districts with
considerable growth in the 1960s and 70s. The local system of firms is
innovative, first of all as regards small, stepwise changes in products
and processes (incremental innovations). The innovation activity builds
on the experience based competence of entrepreneurs and skilled
workers, and on close cooperation between persons and firms in an
atmosphere of mutual trust and understanding (i.e. agglomeration
economies)
Mark II industrial districts: In the 1980s the traditional industrial districts
had to upgrade themselves to mark II districts in order for firms to carry
out more radical innovations. Radical innovations were necessary to
meet increased competition from low cost countries and to introduce
more advanced technology. Upgrading to mark II districts take place
through the establishment of centres for real service. The centres have
specialised competence for the dominating industrial sectors in the
districts. The centres offer subsidised services to firms, such as
information about markets and technological development, help in
exporting or in introducing new production equipment. The centres
supply the system of firms with professional competence that small
firms often cannot possess themselves, but competence necessary to
carry out more radical innovations
Why the large interest in industrial
districts?


The interest in industrial districts far exceeds their empirical
significance. Districts are of some importance in countries with craft
traditions like Italy, Spain, France, Denmark (and examples are found in
Norway)
The interest of the districts is rooted in:
a) the fact that the districts may constitute a possibly new and more
human-centred industrial development. The model symbolise the
possibility of small firms and craft skills in a world of impersonal big
firms and deskilling
b) the emphasizing of social and cultural factors (as traditions and trust)
in economic life, i.e. that industrial development rely on more than
‘economic factors’,
c) what the districts can tell about the mechanisms of innovation,
learning and knowledge flow, i.e. that these build on factors in the local
environment and not just on factors internal to firms,
d) what the districts have to say about the possibility of independent
SMEs to be internationally competitive as long as the SMEs collaborate,
i.e. that local areas can develop without relying on large firms
Industrial districts as a sign of wider
industrial changes in western countries: A
Second Industrial Divide





A famous book from 1984 (Piore and Sable: The Second Industrial Divide:
Possibilities for Prosperity) put the development of industrial districts in Italy
and other places (as Baden Württemberg) into a wider context.
The authors saw the development of industrial districts as empirical proofs
of wider, epoch-making changes in the world economy, i.e. as the coming of
a new, dominating way of organising industrial production, which they called
flexible specialisation, and which could at least partly replace the then
dominating Fordist production regime (mass production of standardised
goods)
Flexible specialisation is effective in the production of quality products for
quickly shifting markets
The new flexible production method reflects changing in market condition
(more uncertainty, increased demand for tailor-made products) requiring
more flexible production methods
The flexible specialised production was seen to occur in particular in local
networks of SMEs using computerised production equipment (while we now
also see diversified production in large firms)
Critics of the industrial district /flexible
specialisation hypothesis



The hypothesis of a widespread growth of industrial districts and a new
dominating production principle has been criticised. The critics point to:
a) too broad generalisation of fairly few empirical cases. The formation
of industrial districts is partly based on some place-specific factors (i.e.
socio-cultural factors) which are difficult to transfer to or copy in other
areas. That is, the Italian type of industrial districts cannot be found in
very many places.
b) poorly defined central concepts. This lead to many types of
agglomerations (clusters etc.) being classified as industrial districts
without satisfying the strict definition of the districts. Thus, the
phenomenon of a new flexible specialised economy is overstated
However, the main diagnosis by Piore and Sable of main trends in the
economy seems right, and the 1970s marked a basic change in the
economy of western countries
Important is a) the trend towards vertical disintegration (outsourcing), b)
more long-term collaboration between clients and suppliers, and c) the
growth of different types of industrial districts and regional clusters
Challenges for industrial districts in the
more global economy


The industrial districts grow in the 1960s and 70s as local networks of
SMEs, i.e. most of the firms in the value chain were found within the same
local area
A more global economy challenge the districts in several ways:
a) firms in the districts begin to source cheap components from abroad
(from low cost countries, in particular from Eastern Europe) – destroying the
local production system,
b) knowledge-intensive firms in the districts are incorporated in international
corporations – destroying the local control over important firms in the district
b) in some areas leader firms arise (from within) that are capable of
technological and market leadership, and of managing complex
subcontracting relationships – destroying the dominance of SMEs and the
symmetric relations between firms

Why do such development trends (a-c) threaten the industrial district
idea?

One important researcher in industrial districts asks what is necessary to
qualify as an industrial district: ‘Does it simply become a centre of design,
ideas and innovation (…) or does it have to contain the entire division of
labour in an industry? (Amin 2000: 166)
The boat-building industry in the ArendalGrimstad region: The growth of an industrial
district?
The development of the industry: Six historical phases
1.
2.
3.
4.
5.
The start: Tradition and knowledge about boatbuilding from the
period of sailing vessels, when Arendal was by fare the largest
shipping town in Norway
The revolution with the building of glass fibre boats in the 1950s.
This kind of boat building started in Arendal by the linking of
external knowledge and local competence
Spreading of the knowledge in the local area in the 1950s and
60s, giving rise to many start-ups and growth
The growth of giant firm Fjord Plast, and the centre for the building
of motor boats in Europe
Restructuring in the 1970s and 80s. Start-ups, specialisation on
firm level and growth of subcontractors
The background of Porter’s cluster theory



Michael Porter departs from studies of firm strategies, particularly the
question of how individual firms can compete internationally
In the 1980s he started to study what makes a nation’s firms and
industries competitive in global markets. This research led to the theory
of clusters, first national clusters (1990), then regional clusters (1998)
The central idea behind the cluster theory is:
a) that the national (or regional) environment is of central importance in
stimulating firms’ competitiveness (‘Competitive advantage is created
and sustained through a highly localized process’),
b) that the environment influences the ability to succeed in particular
industries or industry segments (nations become the home base for
competitive firms in some particular industries, and most important is
competitiveness in sophisticated industries),
c) that the international successful firms in an industry constitute
systems or clusters of firms, and
d) that firms and industries become competitive by continually
upgrading and innovation (and by developing new and sophisticated
industries)
The sources of locational competitive
advantage in a nation (‘the diamond’)

Four broad attributes of a nation form an environment that can promote or
impede the creation of competitive advantage by firms in specific industries
1) Factor conditions. Factors of production (inputs) such as skilled labour,
natural resources, capital or infrastructure. Important factors are created,
and disadvantage of factors can contribute to sustained competitive
success (e.g. automation because of high labour costs)
2) Demand conditions. The quantity and particularly quality of home
demand for the industry’s products / services (the importance of
sophisticated and demanding customers for innovation). Important in small
countries like Norway?

3) Relating and supporting industries. Supplier and related industries that
are international competitive. Support cost-effective supplies and stimulates
innovation
4) Firm strategy, structure, and rivalry. How companies are created,
organised, and managed, and the nature of domestic rivalry. How firms are
managed support some type of industries more than others (f.ex. liberal and
coordinated market economies ‘support’ different types of industries)
Two additional variables can influence the national system in important
ways; chance and government policy
The environment (‘diamond’) and clusters





The factors (the ‘diamond’) influence and reinforce each other and
combine into a dynamic system
Domestic rivalry and geographic industry concentration have
especially great power in transforming the ‘diamond’ into a system.
Domestic rivalry promotes upgrading of the whole national
‘diamond’. How? Geographic concentration magnifies interaction
within the ‘diamond’. ‘The process of clustering works best when the
industries involved are geographically concentrated within a
industry’. (Porter in 1990 then foresaw the regional cluster from
1998)
Nations succeed in industries where the national environment
(home base) supports firms’ innovation activity and accumulation of
specialised assets and skills in unique ways (which means that the
advantages are difficult to copy by firms in other places)
Nations succeed not in isolated industries, but in clusters of
industries connected through vertical (buyer/supplier) and horisontal
(common customers, technology etc.) relationships. A nation’s
economy contains a mix of clusters. Why?
The factors in the ‘diamond’ in particular influence innovation activity
and dynamism in a cluster
Putting Porter’s theory of national clusters
into practice in Norway



Reve and Jakobsen (2001) studies six selected, important industrial sectors in
Norway:
- seafood (aquaculture and fishing, including upstream activities like fodder
production, equipment and boat building, and downstream activities like
production, marketing sale, distribution)
- energy (mainly oil and gas, including searching, refining, distribution etc.)
- maritime, including shipping / shipowners (the hub), ship building, equipment,
sea insurance, classification, ship brooking, ship finance, maritime IT etc.
- ICT, including software producers, hardware producers, distributors and
traders, and network operators
- trade
- finance and insurance
Three of the sectors are seen to constitute strong national clusters: seafood, oil
and gas and maritime. These industries have in particular strong market
relations (demanding customers and export), relation to other national actors
(seafood has the highest score), and factor conditions (oil and gas has the
highest score)
The three sectors constituting clusters perform better (with some exceptions)
than the three other sectors. The clustered sectors grow faster, have higher
productivity and higher wage level. Why? Explained by upgrading.
Porter’s definition of regional clusters





Porter (1998) defines (regional) clusters in different ways and attaches
large importance to clusters
Definition: Clusters are geographic concentrations of interconnected
companies and institutions in a particular field. (…) Many clusters
include governmental and other institutions (…) that provide specialised
training, education, information, research, and technical support (p. 78).
Similarities and differences with national clusters (the ‘diamond’)?
Relation between firms: Clusters promote both (horisontal) competition
and (vertical) cooperation (p. 79)
The role of the local environment: Clusters reveal that the immediate
business environment outside companies plays a vital role as well (p.
78)
Importance: The enduring competitive advantages in a global economy
lie increasingly in local things – knowledge, relationships, and
motivations that distant rivals cannot match (p. 78)
Importance: Today’s economic map of the world is dominated by (…)
clusters: critical masses – in one place – of unusual competitive
success in particular field (p. 78)
Clusters and upgrading mechanisms, or
why clusters are critical to competition
1.
2.
3.
Dynamic regional clusters are seen to stimulate innovation activity and
the competitiveness of cluster firms. That is, clusters have three important
upgrading mechanisms (that explains why cluster firms are competitive):
Innovation pressure, which requires a) advanced, demanding customers,
b) rich communication between customers and suppliers, and c) that
clients can choose between alternative suppliers (competition between
suppliers). The innovation pressure may be spread backward in the value
chain
Complementarity, which means that firms share common resources (such
as infrastructure, experienced labour, suppliers and services). The
resources may become better and cheaper when there are a larger
market for them (f. ex. better roads, schools, or cheaper services)
Knowledge flow, when firms cooperate (f. ex in innovation projects), when
workers change jobs, when firms use the same consultants and suppliers,
and when research are commercialised
To which extent are the upgrading mechanisms dependent on the
activity of regional clusters? Or may the mechanisms function also
when firms are not co-located?
Question

What are the main differences between
Porter’s cluster theory and the theory of
industrial districts? What are the differences
as regards the definition of clusters / districts
and the mechanisms underpinning innovation
and competitiveness of clusters / districtricts?
The life-cycles of clusters
1.
2.
3.
4.
5.
Porter (and other) propose a historical development of clusters
consisting of birth, development and possibly decline. Five main
phases are proposed:
Formation of pioneer firms often based on specific local
knowledge, and specific local demand, followed by new firm
spin-offs.
Creation of a set of specialised suppliers and service firms, and
a specialised labour market.
Formation of new organisations that serve cluster firms.
Attraction of outside firms, skilled workers, and fertile grounds
for new local companies.
A period of decline for the clusters may occur due to both
external and internal factors. External factors are particularly
technological discontinuities and market changes. Internal
factors are less local rivalry, groupthink
Critique of Porter’s cluster theory





Porter’s cluster theory has received some critique from other researchers:
Are clusters a functional phenomena (related industries and firms within a
nation) or a geographical phenomena (geographic concentration of
interconnected firms? Changing perspective by Porter from 1990 to 1998.
Focus on firms’ home base, which are geographical concentrated, while
suppliers etc. are found over a larger geographical area
Very elastic definition of the ‘cluster area’. Porter writes: ‘The geographic
scope of a cluster can range from a single city or state to a country or even
a group of neighbouring countries’. Then clusters are a very divers
phenomena, giving the rise to ‘political cluster’, i.e. a cluster is what
someone decides to call a cluster
What is the role of public actors in supporting clusters? Government policy
as a less important factor in the ‘diamond’ model, or cluster as the way to
carry out industrial policy?
Why the large importance of clusters these days? Porter dos not situate his
cluster theory in any spatial and historical context, like the industrial district
theory proposing that districts as growing owing to some specific local
factors in the third Italy in particular, and as a sign of larger changes in the
world economy (the coming of flexible specialisation as new dominating way
of production)
The development of regional clusters in
Norway.


Delimiting potential regional clusters in Norway by use of
statistics. Criteria:
1) Start with labour market areas (the regions)
2) Delimit industries and regions with a location quotient (LQ)
higher than 3,0
3) Chose regions with more than 200 employees and 10 firms in
the dominating industry (LQ > 3,0)
Job growth in the regional clusters. Two development trends:
1) Relatively high job growth (small job loss) in the regional
clusters compared with the same industries in Norway. Some
support for the cluster hypothesis?
2) The exception is first of all some clusters in the large cities
demonstrating relatively large job losses. This reflects the
national deconsentration of Norwegian manufacturing industry
Tekstil og bekledning (2)
G
j
ø
Hammerfest
Vadsø




Alta
Nord-Troms
Vesterålen
Lofoten

nessjøen
øysund

Rørvik
Finnsnes

Næringsmiddel (e.g. fisk) (4)


Namsos
Brekstad
Andselv

Frøya/
Hitra



Ørsta/
Volda
Nordfjord
Ørsta/Volda

 


Mosjøen
Florø
Kraft- og vannforsyning (2)
Surnadal
Ulsteinvik Ørsta/Vol
da
Høyanger
Sogndal/
Årdal
Transportmiddel (e.g skip) (5)
N.Gudbr.dal
vik
N. Sunnhordland
Kongsb. Drammen
Elverum
Hamar
Kongsv.
Askim/Mysen
Halden
Fr.stad / Sarpsb.
Vesttelemark
Sandefj./Larvik
Skien/Porsgr.
Jæren
Lyngdal/
Farsund
Mandal
Utvinning av råolje og naturgass (2)
Elektronisk og optisk industri (3)
Bergverksdrift (1)
Papirmasse og papir (1)
Moss
Haugesund
Stavanger/
Sandnes
Trelast og varer av tre (9)
Møbler og annen
industriproduksjon (4)
Kr.sund
Ålesund
Fiskeoppdrett (17)
Kjemikalier og kjemiske produkter
(3)
Gummi- og plastprodukter (2)
Ikke-metallholdige mineralprodukter
(2)
Arendal
Metallvarer (2)
Maskiner og utstyr (3)
Off. adm. og forsvar (1)
Lessons from the cluster theory for
designing industrial policy



A shift from comparative (better access to raw material, cheap
energy sources, natural harbours, large market) to competitive
advantages (unique, place-specific factors stimulating learning
and innovation activity). Clusters are places that in particular
stimulates the formation of competitive advantages according to
Porter
Thus, policy should promote cluster formation and upgrading.
This can be done by ensuring the supply of inputs (skilled
workers, infrastructure), stimulate rivalry (how?)
Government should support all clusters, however, governement
should reinforce and build on existing and emerging clusters
rather than attempt to create entirely new ones