Content uploaded by Philip Cooke
Author content
All content in this area was uploaded by Philip Cooke on Jan 02, 2016
Content may be subject to copyright.
The Handbook of Regional Innovation and Growth
Editor: Philip Cooke
Co-editors: Björn Asheim, Ron Boschma, Ron Martin, Dafna Schwartz and
Franz Tödtling
Publisher: Edward Elgar
Chapter 1.
Introduction to the Handbook of Regional Innovation and
Growth
Philip Cooke, Björn Asheim, Ron Boschma, Ron Martin, Dafna
Schwartz and Franz Tödtling
Introduction
This book marks the maturation of a field of study that is only some twenty years old.
Regional innovation systems is taught in university undergraduate and graduate studies
courses in numerous social sciences, notably economic geography, regional planning,
business economics, innovation systems analysis, development studies, political science,
sociology of science, science technology and society and environmental studies. Moreover, it
is increasingly widely applied in regional economic governance practice by development,
innovation and enterprise support agencies. It is related to, but has a distinctive lineage from,
teaching, research and practice in national, sectoral and technological systems of innovation
as David Wolfe makes clear in Chapter 3. Its origins lie in three related strands of research
that pre-date the regional innovation systems discourse (Cooke, 2008).
These are, first, the systems view of planning, dating from the late 1960s as a comprehensive
perspective on the integrated analysis and planning of city and regional systems. The
importance of this perspective resides in its broad application and integrative perspective,
way beyond urban and regional economic processes, in biology and engineering for example.
However, with respect to spatial studies its value lay particularly in its insistence on analysis
of, for example, land use and transportation inter-connections and innovation in these, and its
interweaving of understanding spatial economic processes with better-informed aspirations to
improve them by means of policy and management. These conjoining aspects are central to
each of the systems of innovation fields noted above. They are captured by the early terms
1
technological paradigm and technological regime in for example Dosi (1982). Here the first
refers to the predominating technological profile on the macro-scale, such as ‘the Computer
Era’ or ‘the Information Age’ which is pervasive in modern working and domestic life and
tends accordingly to influence the nature, direction and pace of change (for example, Pavitt,
1984; Carlsson & Stankiewicz, 1991). Technological regime expresses the norms,
institutions, organizations and rules that tend to sustain the dominant technological paradigm
by means of standards, institutional mindsets, dominant technological discourses, government
regulations and organizational preferences. For a swift taste of this relationship, think only of
the ways in which the fossil fuels ‘paradigm’ resists the challenging discourse of a potential
renewable fuels ‘paradigm’ by its influence on the predominating fossil fuels regime of
subsidies (and their frequent absence for the challenger paradigm), regulations (low
renewables requirements), technologies (sunk costs of past and expense of new
infrastructures) and economic discourse (e.g. only ‘scale’, hence cheapness, matter).
Second, regional innovation systems thinking was influenced by empirical research findings,
initially from Italy, that showed economies of scale inside the large or multinational
corporation were not the only pathway to economic competitiveness (Bagnasco, 1977;
Becattini, 1978; Brusco, 1982). These authors pointed, in different ways, to the success of
small and medium-sized enterprises (SME) working collaboratively and competitively with
success in both mass and specialised markets (‘flexible specialisation’ after Piore & Sabel,
1984). The former included, for example, mass-consumption food production and clothing,
the latter marketed through German and American department stores; the latter included
specialist types of engineering such as agricultural irrigation systems, food packaging
machinery, spectacle frames, hiking and skiing boots, luxury footwear, luxury fashion
clothing and super-cars (also super-bikes). This was interesting because the systems of
production – diffused, often in the countryside – as Bagnasco and Brusco observed it,
concentrated in ‘industrial districts’ as Becattini saw it, were characterised by often rapid,
collectively inspired and managed innovation in products and processes, later also services,
including organizational innovation (e.g. specialised engineering or production software,
telematics and systems). Regions were important to this phenomenon in a multitude of ways.
Industrial districts, as one variant of ‘Third Italy’ were found mainly in a macro-regional belt
across north-central Italy, from Marche to Veneto and nearby regions. Some regions like
Emilia-Romagna and Tuscany were Communist-governed or ‘red’, others, like Veneto, were
Conservative (‘white’, Christian Democrat). There were different regional nuances, even
2
between the ‘red’ modes of economic governance, decentralised in Emilia-Romagna,
centralised in Tuscany. Enterprise support services were available, either through the regions
or the regionally important business associations. In the ‘white’ regions chambers of
commerce were among the most important support organizations. Thus ‘system variety’
added interest to the networked concept of SME production. Later, such regionalised SME
networks or ‘milieux’ were discovered in Germany, Switzerland, Sweden, France, Spain and
Portugal (notably by the GREMI group; Aydalot, 1986) and then in developing countries
(Schmitz, 1995). Accordingly, an influential and related research agenda had opened up.
Finally, the emergent discourse of regional innovation systems, rooted strongly in regional
science, applied regional economics and economic geography, as shown in the apparently
seminal article (Cooke, 1992) found shared conceptual interests with the national systems of
innovation literature. This was principally through connections between the Science Policy
Research Unit at Sussex University, notably its director, Chris Freeman, and a graduate
course in Regional Economic Development at Cardiff University where, in 1988 he guest-
lectured on ‘Networks of Innovators’ subsequently published as Freeman (1991). The
innovation systems perspective, developed from Friedrich List to Joseph Schumpeter, the
emphasis on overturning a hegemonic academic and policy discourse favouring the ‘linear
model’ of innovation, and the advocacy of a collective learning approach among interacting
users and producers after Lundvall (1985) resonated positively with the lengthily gestating
regional innovation systems concept. Systems, networks and interactive learning were the
common threads linking the three complementary schools of thought that came to inform the
regional innovation systems model of regional economic evolution. Both the systems
approach and Schumpeter’s economics were evolutionary and ‘learning’, along with
‘networking’ and ‘variety’ were cornerstones of the evolutionary view.1
Key Themes in this Handbook: Innovation, Productivity & Growth
No mention has yet been made of the ‘Growth’ aspect in this Handbook’s title. In reality it
was the title proposed to the lead editor when approached by the publisher with the idea for
the project in 2008. Reflecting upon it, all editors found it valuable to keep ‘growth’ in the
1 Articulating this took longer. It occurred first in the Introduction to Braczyk et al. (1998), the original
manuscript for which was written in 1995, lost by the publisher during an ownership change, then re-
discovered eighteen months later. It could easily have come second to a rival project with which at
least two of the editors of the present volume (Asheim, co-editor; Cooke) were associated. For also in
1995 a regional innovation systems seminar was held in Oslo, chapters for a book were prepared, co-
editor Keith Smith from the STEP research group in Oslo impressed on authors an evolutionary
systems approach to regional innovation. But the project was never completed.
3
title. The main reason was that it not only adds a valid dimension to the general approach to
the chapters assembled in this collection but, importantly it gives an answer to a sometimes
asked question, which is what, exactly, is innovation for (e.g. Moulaert & Sekia, 2003)? The
purpose of innovation is growth, measured in terms of productivity, efficiency and
effectiveness. The fundamental belief in Porter (1990), for example, is that ‘co-location’ by
firms increased efficiency (e.g. reducing supply chain costs; accessing talent), start-up
activity, innovation and productivity while more generic conditions for growth, such as good
infrastructure and education, were insufficient (for empirical support, see Delgado, Porter &
Stern, 2010; Spencer et al., 2010). Growth is not only an economic but a social process. It is
itself subject to critique from activists and others who seek certain ‘limits to growth’.
Reasons for this range from the moral to the environmental and the two in combination are
not unusual. From an evolutionary perspective, growth is a successful indicator of the health
of an organism, whether biological or socio-economic. It seems that capitalism, which from a
Schumpeterian perspective is fuelled by innovation, must grow in order to survive. Growth is
implicit in the notion of markets, the inefficiencies in which stimulate innovative efforts to
profit from seeking better alignments between value and price, whether of commodities,
companies or currencies. For more citizens to have access to the quality of life of the typical
middle-class household of the advanced economies is not a morally indefensible position,
given the massive inequalities that exist even in many such countries, let alone between them
and the developing world. Such a benign outcome may not be achievable, but that does not
make it undesirable. Indeed it is what democratically elected governments continue to pursue
around the world.
Hence innovation is not simply the pursuit of the new, new thing or an abstraction to satisfy
the psychology of novelty. Improving on some technology or process is what most economic
activity has been concerned with across geography and its history. Of interest to the project of
this book is the extent innovation is conceived of in this way, implicitly or explicitly. It would
be misleading to say that every contribution addresses this inter-dependence explicitly. Most
focus upon innovation and implicitly assume more or less the above argument (for a widely-
cited economic geography of endogenous growth theory, see Martin & Sunley, 1998).
However some address the innovation-growth nexus more overtly, including through the lens
of innovation and productivity (Capello’s chapter in this volume) and for the moment, it is
useful to identify some salient points in her argument. This is important for two reasons: first,
it is seldom done in innovation studies in general, let alone regional innovation studies; and
4
second, it reminds us of the two-way implications of the relationship. Thus innovation
depends on growth because it carries certain key implications regarding investment,
capabilities and organization. Most growth is asserted to come from intra-industry trade
among advanced economies, not even from inter-industry trade between them and less
developed economies (Krugman, 1991). If true, it is because, in effect, advanced economies
can afford to trade innovations made affordable by past growth, which is replenished by
returns from present investment in future innovation.
Capello’s key points about the relationship between productivity and innovation are the
following. There are three perspectives on the relationships between regional innovation and
productivity, depending what kind of knowledge determinants prevail. Taking a Functional
approach to understanding knowledge, innovation and productivity relations, the key idea is
that science based innovation in ‘scientific regions’ short-cuts the invention to innovation
circuit. Schumpeterian radical innovation may ensue with associated profits from the
increased translational productivity from invention to innovation. Geographical proximity is
accordingly crucial to this kind of productivity gain. However, research into innovation
shows it often occurs without science and is thus not confined to ‘scientific regions’ (Asheim
& Gertler, 2005). In this case, it is necessary to take a more Structural perspective on the
innovation-productivity nexus. Accordingly the regional system and its entrepreneurship
talent, knowledge and industrial value assets replace science as the determinant of a virtuous
innovation-productivity cycle. Productivity gains are reaped from the rapid recombination of
knowledge (innovation) and its commercial exploitation (entrepreneurship) which are
regional innovation system processes. Innovation is incremental and productivity gains from
the pervasive and rapid knowledge-innovation recombinations are proximate and systemic.
But neither all innovation nor associated productivity gains are confined to or derived from
the region, as an enormous amount of research literature shows (see Cooke, 2009; Ponds et
al, (2010); Breschi chapter in this Handbook). Indeed, not only may innovation involve
relational proximity over great distance but its productivity gains involve collective learning
processes. Hypothetically, these result in global network hierarchies in which typically factor
(e.g. labour) value is extracted from the ‘learning region’ and productivity increases are
lodged at the top of the hierarchy (innovative region). This is the working out of the law of
combined and uneven development, the law of increasing returns to (knowledge) scale and
that of ‘cumulative causation’. The question remains, to what extent and at what magnitude
5
may learning gains be repatriated, as seems to happen at least to some extent among returning
Taiwanese and Indian migrants in Silicon Valley (Saxenian, 2000)?
Thus treating regional innovation as a means to the end of increasing growth by enhancing
productivity opens up an important research agenda. This has been broached somewhat by
research that compares productivity gains from innovation in situ but little yet seems to have
been done in terms of inter-regional productivity magnitudes deriving from knowledge flows
of the kind exemplified by science-intensive innovation activities such as biotechnology.
Edquist et al., (2001) conducted research into product and process innovation focusing on
their comparative contribution to productivity and employment. Broadly, they concluded
productivity increased with both types of innovation although the former contributed more to
capital than labour productivity while this was reversed for process innovation. The reason
for these results is clear: by definition a product innovation creates employment while
contributing to increased efficiency (greater quality or value at lower cost), whereas process
innovation almost always involves labour-shedding. Considered from a regional viewpoint
the region where product innovation predominates should always display high job growth
(except in recessions) and capital (technological) productivity. The process innovating region
will lose jobs but gain in labour productivity. The nature and quality of regional production
then determine regional growth. Schumpeterian regions should, in theory, constantly reap
high profits from radical innovations (see discussion below on the problem of ‘what is
radical’ in innovation studies) and the region should also reap gains from the constant
demand for high-value labour. Labour-intensive production will tend to produce the opposite
result, except for bursts of labour productivity occasioned by investments in new process
technology.
A different question is addressed in comparative productivity research conducted by Van Ark
et al., (2005). Their spatial productivity question concerned differentials in productivity and
productivity growth between the US and the EU in the 2000s. Their results were interesting,
instructive and add a dimension of understanding to the centrality of innovation to regional
increasing returns in productivity and growth. They identified three key influences on the
superior productivity profile of the US over the EU in manufacturing and services. First, most
innovative technology – especially pervasive, even radical innovations related to ICT –
originate in the US, where users have an early opportunity to know and understand their
possible contributions to business enhancement. Second, budgets for investment in improved
technology are higher in the US; hence early adopters are more numerous. This was
6
particularly so in services, where EU productivity is poor in sectors like banking, which in
some large countries remains mainly paper-based. Finally, knowledge spillovers were
identified as an asset to early adopters. Thus a new system of, say, e-commerce introduced to
a firm or organization might be disruptive but collective ‘learning-by-doing’ meant solutions
to problems swiftly circulated the office. Logically, therefore, higher growth from
productivity gains associated with innovation should accrue to regions with many innovators
and early, communicative adopters – a combination of product innovation in one part of the
regional system and process innovation amongst the user community. Stasis or comparative
productivity decline would be hypothesised for the non-innovating, non-adopting, non-
communicating region.
Felsenstein’s chapter in this Handbook addresses the issue of human capital and labour
mobility in ways directly relevant to this discussion. He, too, draws attention to three
perspectives for illuminating understanding of the relationship between innovation and labour
as a factor of production. Though the fit is not perfect there is some resonance with Capello’s
taxonomy. The first perspective is that of New Economic Geography (NEG; after Krugman,
1991) which initially accounted for regional innovation in terms of labour pooling behaviour,
namely firms and workers seek out regional market-size and pecuniary externality effects,
agglomerating where they find a region where industry has a lead over everywhere else (see
also Vatne’s chapter (4) on regional agglomeration in this Handbook). This is not hugely
dissimilar to the sector-function view on regional productivity. In modelling terms, this
approach produces naive results, which Felsenstein calls ‘catastrophic’ agglomeration (and
regional desertification) that have to be corrected technically but which still produce
misleadingly over-concentrated spatial results. An alternative that does not fall into the trap of
over-emphasising a single type of knowledge determinant of regional growth (e.g. science
region) is New Growth Theory (NGT) with which Krugman (1991) is also closely associated,
along with Romer (1990) on endogenous technological change. Here, by analysing regional
externalities interactively with human capital mobility, the approach estimates the way that
human and physical capital, labour mobility and innovation impact on regional productivity
and growth. In terms of our earlier discussion on this, theory supports the deduction that the
higher the average level of human capital, the more rapid the diffusion of knowledge,
therefore the higher the level of regional productivity (including earnings). So NGT allows
different kinds of regional knowledge and innovation into the innovation-productivity
analysis, in common with Capello’s Structural approach. However, while human and physical
7
capital are found to impact regional productivity, the model results are confounded by a
‘regional innovation’ effect. Thus a third approach receives some degree of support from this
inconsistency, namely Evolutionary Economic Geography (EEG). This, like Capello’s
Cognitive approach sees institutions, organizations and cultural practices as critical in
generating regional growth. Thus cultural and institutional proximity are as important as
spatial proximity and the region represents an active innovation agent.
Linear, Interactive and Neo-linear Innovation Models
The preceding section formed a bridge from innovation via productivity to growth as an
explanation of the reasoning behind the use of the ‘regional innovation and growth’ couplet in
the title and main thematic of this handbook. This introduction next proposes to draw up a list
of a further six key issues in innovation studies that arise from subjects tackled in the chapters
that are of wider relevance although of central importance to regional innovation systems
analysis. As in the preceding section, not all chapters will be invoked in constructing analyses
and arguments about these as yet unresolved, indeed in quite large part unraised, topics for
debate. Each chapter is introduced according to the seven section structure of the book.
Chapters are alluded to where they can help illuminate the specific issues raised, alongside
the broader published innovation studies literature. They are intended to stimulate interest,
debate and
Regional Innovation Issues Under Reflection
Beyond the Interactive Innovation Model?
Radical Innovation: How Radical is Radical?
The Roles of Innovator and Entrepreneur
Path Dependence and New Path Creation
Technological or Innovation Paradigm and Regime?
System Self-organization or System Leadership?
Fig. 1. Some Conceptual Issues in Regional Innovation Studies
8
further research. The first of these concerns the core concept in evolutionary innovation
systems theory and suggests a re-examination of the received wisdom of the interactive, user-
producer critique of the traditional linear model of innovation is now overdue (Lundvall,
1985; 1992). This is for the following reasons. First, in recent research the user-producer
interactive innovation model no longer seems to capture the variety of forms innovation
processes may take in time or space. To clarify, when it was introduced in the 1980s it was
because of a perceived necessity to move away from a hierarchical administrative model of
the mode of innovation that had appeared with the rise of the modern, especially American,
large and increasingly multinational corporation (Hymer, 1976). Conceptually, innovation
contained three distinct phases consisting of invention, innovation and diffusion. In the
context of the large corporation this frequently translated into invention being done in the
corporate R&D laboratory, innovation being done by production engineers and diffusion
being the responsibility of the sales and marketing department. Each phase was a discrete
step. Much responsibility rested on the shoulders of the production engineers, responsible for
prototyping, trialling and testing an invention the scientific origins for which they may only
have had a hazy understanding. Some inventions proved unworkable, some R&D projects
could seem endless. In this model, the corporate in-house R&D scientist and team were a
privileged and expensive elite.
Firms like General Electric, AT&T and RCA were exemplary carriers of this tradition,
evolving spatial divisions of labour that separated places where R&D was conducted from
places where the other functions occurred. Of course, the needs of everyday practice eroded
such rigid conceptual boundaries. Thus university scientists might be given sub-projects
where in-house R&D capabilities were lacking, or a scientist or doctor might bring an
invention to the corporation for analysis of its potential. Much pharmaceutical invention
followed the latter course even though scientific teams were organised like armies of
molecule-hunters focused on specific diseases. In those ‘linear’ times hospital doctors
frequently found or made time to conduct patient-based research of an informal kind in the
hospital or clinic. In Le Fanu (1999) such ‘chance discovery’ by scientists and doctors outside
pharmaceuticals corporate R&D laboratories accounted for penicillin, cortisone,
streptomycin, heart pacemakers and numerous other radically, ‘paradigm shifting’ as Le Fanu
refers to them, treatments for disease. If anything, from the 1940s to the 1990s when
productivity requirements brought an end to informal medical research, and biotechnology
further ‘scientised’ it, big pharmaceuticals corporations, despite retaining major R&D
9
laboratories, could be said to be akin to prototype engineering and marketing departments in
manufacturing firms. With the aforementioned ‘scientisation’ they are primarily marketing
and financing agents for university invention and biotechnological innovation, often by
SMEs. Hence, it is arguable that the linear model was never an accurate representation of
innovation in ‘big pharma’ and was based on more of an electronics or engineering paradigm.
In engineering, too, the impact of Japanese ‘lean production’ industrial processes effectively
ended the last vestiges of the linear model wherever it might have existed, because here was
the apotheosis not of in-house design and production but the evolution of radically
decentralised production by means of outsourcing to increasingly elaborate and globalised
supply chains. The manner in which customers and suppliers, now normally independent
corporate entities, would negotiate innovation and other product and process qualities had
now moved further away from the linear conception presumed of intra-corporate ordering.
While sub-contracting had been present from the earliest days of the emergence and evolution
of capitalism, it had often developed a ‘stress’ or ‘sweating the suppliers’ culture. This
practice could extend, in some industries and economies, even into the 2000s. Not untypically
it would involve highly capricious requirements being made by customers of suppliers, such
as ordering parts, short-term, then changing the order many times before the delivery
deadline (Jones, 1984). Accordingly, as in the case of the West Midlands (UK) automotive
industry, where it reached its apotheosis, it presaged the oblivion of once important suppliers
like Lucas Automotive, on the one hand, and Rover, on the other2. Elsewhere, notably in the
Nordic countries and Germany, with their co-ordinated rather than ‘stressing’ or ‘cut-throat’
forms of supply-chain cost minimisation this was less prevalent. Nevertheless, even in
German engineering, notably automotives, despite more benign customer-supplier traditions,
organizational innovation occasioned by the rise of Asian luxury car production placed
seemingly unsustainable new burdens on suppliers. Hitherto, these had been meticulous
executors of blueprints handed down from the customer design engineering department: now,
they were themselves required to innovate in fulfilling client requirements. The threat of
2 Interestingly, this ‘stressing’ effect was not unknown in ‘lean production’ as practised by Japanese
companies. Thus Cooke & Morgan (1998) reported evidence of, for example, Panasonic requiring
annual incremental innovation of parts from its in-house keiretsu suppliers (typically a 3% cost
reduction and increased quality) to the point where it could not be achieved. The supplier then
vacated the market, the customer turning to obliging suppliers elsewhere. Acculturation to this mode
of cost control meant suppliers constantly searching for and selecting new customers, often for largely
unfamiliar products, something which is now far more widely practised as a survival strategy in
European engineering (Knie & Hård, 2010).
10
outsourcing to innovative suppliers outside the region or transplanting supply to newly-
liberated, low-wage eastern Europe acted as an incentive.
A common way of innovating organizationally to meet Asian competition was to slightly
relax corporate hierarchy, moving towards matrix management, which allowed for greater
cross-departmental team-work, and giving innovation a project-based character.
‘Simultaneous engineering’ (also known as ‘concurrent engineering’) brought representatives
from marketing departments into project teams with engineers, designers and external
suppliers to develop innovations. It is the first and last-named element that captured the core
of user-producer interactive innovation. Thus, the modern marketing manager had become
expert in framing issues relating to user needs. Meanwhile, the supplier or producer was
increasingly external to the intermediate user corporation. From an innovation perspective,
the following question was posed: ‘How can the producer know the needs of potential users,
when markets separate users from producers?’ (Lundvall, 1992, 50). Lundvall’s answer is
somewhat Nordic – ‘The relative importance of product innovations indicates that most
markets are organized markets’ (op. cit., 51). This involves information exchange, co-
operation, hierarchy, mutual trust, with durable and selective relationships. This is an
important observation of the social embeddedness of markets, against the neoclassical view
that they are atomistic and utilitarian, which is what was being critiqued. But while the point
is made that we should expect to find user-driven, or customer-driven innovation as it is now
commonly referred to, in this, it might at times be more producer-driven (e.g. see below, also
chapter (43) on design driven innovation in this Handbook). As will by now have become
clear, the ensuing complexity was in general beginning to be tackled by ‘collaborative
manufacturing’ (Sabel et al., 1991).
Into this lacuna has stepped a discourse of ‘neo-linear’ innovation models that are in key
ways anathema to the notion of durable, loyal and selective principal-agent relations between
producers and users presented above. Times have changed and the EU requires competitive
tendering, although such rules are not so difficult for firms and organisations to innovate
around and there is much regional variation in observing the letter of their law (e.g. on food
procurement; Morgan et al., 2006). But much more far-reaching than the imposition of
supranational and national transaction rules since the 1980s has been the wholesale removal
of them, particularly during the neoliberal epoch that commenced then. In Engelen &
Faulconbridge (2009), reference is frequently made to the geography and variable quality of
financial regulation between national regulatory regimes. They note the fact that California
11
and Florida were engines of sub-prime mortgage demand, but miss the fact that it was
geographically variable deregulation that both stimulated ‘securitization’ of everything, from
fish catches to student loan pools, and that it was initially only legal in a handful of US states,
mostly in the sunbelt as such. Producer driven innovation had come to dominate financial
services.
Securitization began on Wall Street when Lewis Ranieri, a utility bond manager at Salomon
Brothers, innovated the first collateralised mortgage bond (CMB) that fathered the
collateralized debt obligations (CDO), which brought the global financial system to its knees.
Much lobbying of Washington by Ranieri and others legitimated this risky trade, which along
with the Clinton administration’s repeal of the Glass-Steagal Act incentivised normal banks to
become investment banks, gambling with their investor capital. We might refer to this as
‘supply driven’ financial innovation because the normal mortgage market was starved of
capital for new loans while deregulation enabled securitization to provide mortgage firms
with instant returns from banks buying the loans and transforming them into tradeable bonds.
The user-producer interaction in this kind of innovation gives effectively zero innovation
capability to the user. Even over time, the user does not become increasingly familiar with the
inner workings of the innovation or, indeed, the codified let alone the tacit knowledge behind
it. Ultimately, even the innovators did not understand the basic modelling flaw which was
that the data utilised to estimate CDO probabilities of defaulting were inevitably historically-
based. Therefore, anything worse than the historical trend could not be predicted or, in a
cognitive sense, understood or intuited except by a few heretical ‘outsiders’ (Patterson, 2010;
Lewis, 2010). Without labouring the point, it can be seen that other areas of deregulation such
as energy and telecommunications also gave rise to supply-driven innovation opportunities
but also casualties in earlier times.
If the user-producer innovation interaction is non-existent in financial innovation, it is only
slightly less so in respect of another ‘neo-linear’ variety known as design driven innovation
(Verganti, 2006; see also chapter (43) on design driven innovation in this Handbook).
Verganti holds that design-driven innovation is akin to ‘technology push’ innovation as
conceived by Dosi (1982) as being capable of provoking regime and/or paradigm change in
dominant technologies and innovation trajectories. Verganti’s (2006) field of interest is the
Lombardy regional innovation system, particularly its overlapping and interacting design-
intensive furniture and kitchenware clusters, where the equivalent to a technological
paradigm is a socio-cultural paradigm. Instead of technology, its discourse is meaning and by
12
changing meanings designers, like technologists, are capable of changing paradigms and
regimes through innovations that entrepreneurs commercialise. This innovator-entrepreneur
division is quite pronounced here, with ‘circles’ of external as well as internal designer-
innovators iterating rounds of ‘meaning analysis’ to set an exclusive tone for new ranges of
design-intensive products to be ‘proposed’ to consumer markets (Pisano & Verganti, 2008). In
neither supply driven nor designer driven innovation models is there significant change in the
key source of impulse, the producer, over time.
A third ‘neo-linear’ innovation model is referred to as user driven innovation. First articulated
as such by Von Hippel (1988), his research showed that often technologically radical
innovations such as the semiconductor, apparently ‘producer driven’ in its design by A.T. &
T’s Bell Laboratories (as they were in the 1960s when the innovation occurred) and their
leading micro-electronics scientist William Shockley, were largely defined in the extremely
detailed specifications of their ultimate user, the US Department of Defense. Other
innovations of the time were shown to be considerably more user-driven than believed at the
time, notably gas chromatography, thermoplastics and magnetic resonance equipment.
However, many of Von Hippel’s examples include manufacturer rather than user driven
innovation, suggesting mainly that the kinds of highly technologically advanced innovations
he is concerned with are less determined by ‘producer power’ than the two preceding models.
As Verganti (2006) sees it, user driven innovation, whose main input is market research, the
data for which are often highly technical (e.g. eye-tracking equipment) and statistically
rigorous consumption analyses, predominates nowadays in mass consumer product markets
because competition is so fierce that even marginal nuances concerning packaging,
advertisement or product placement design can be advantageous. Contrasting it with what he
conceives as ‘paradigm shifting’ design driven innovation which, after Dosi (1982) regarding
radical ‘technology push’ innovation also inclines to the radical, he sees ‘market pull’ or user
driven innovation as always incremental. This, however, betrays Verganti’s product
innovation bias. Contrariwise, if organizational innovation in service markets is considered,
the radical nature of the introduction of supermarkets, fast food outlets (including drive-thru),
budget airlines, mobile telephony and Internet finance, to name a few, would suggest both
that not all mass market innovation is user driven and that such markets can sustain radical
innovation. In Von Hippel’s (2005) more recent work on the democratization of innovation,
he suggests that a notable share of contemporary innovation has become user-driven, citing
many leisure industries in support, such as mountain bikes, hiking and mountaineering gear,
13
snowboards, sailboards, microlite aircraft and microbreweries as cases in point. To some
extent regional and organic food can be subsumed here. However, the hobbyist tone many of
these entail suggest this to be a noteworthy but limited niche of primarily incremental
innovation. Moreover, the commercialisation of such products usually reveals that
entrepreneurs remain at the heart of the imitation process, expropriating the innovators or
recombiners of knowledge, much as Schumpeter predicted.
Finally, attention must be drawn to a variant of user-driven innovation, demand driven
innovation where, increasingly, what were perceived as interchangeable terms now denote
different scales and kinds of innovation. This is now clarified to mean collectively specified
demand for major public investments that may not be forthcoming through markets due to
market failure. States, national or regional must co-ordinate demand and supply as users
defining demand to producers. Hence at times user drivers supersede producer drivers and
vice-versa because many technologies and institutions must combine systemically through
‘strategic niche management’ to achieve success. This is exemplified where regions seek
transition towards sustainability. The complexity of this tends to mean that though specific
eco-innovations are incremental, in combination their effect is ‘paradigm shifting’ under a
sustainable technology regime or even production/consumption ‘landscape’. The latter is a
long-term expectation in ‘co-evolutionary transition’ theory, capable of being glimpsed in
‘transition regions’ which, nevertheless must co-exist with the broader, global, hydrocarbon
path dependence and its prevailing regulatory regime.
It is important to understand the role of discourse articulation in stimulating change from
established development paths. Just as neoliberal discourse formed the basis for a
thoroughgoing critique of the regulated character of financial markets, so in such
sustainability hot spots as Denmark, critical discourses were articulated to undermine
prevailing norms (see also the ‘green innovation’ chapter (32) in this Handbook). The first of
these was an anti-nuclear energy discourse; the second was its obverse, a pro-renewable
energy discourse. Protest movements acting out the discourse brought reversal of government
policy, suspension of nuclear energy policy and redirection of Denmark’s nuclear research
towards renewable energy research. In such demand-driven innovation settings, subsidies are
a necessary element of ‘strategic niche management’ and can be found being made to
consumption rather than only to production. The Danish case involved what ex post is
revealed as a successful consumer subsidy policy regime that more than paid for itself in tax
returns from wind-turbine production. From the early 1970s, government subsidies were
14
made available to users of first generation wind turbines. This sustained the industry, initially
based largely upon domestic demand, and enabled the north and mid-Jutland-based cluster’s
design to evolve considerably from its path dependent roots in agricultural and marine
engineering where the plough and the ship’s propeller were the inspiration.
Radical Innovation: How Radical is Radical?
In neo-Schumpeterian innovation theory, innovation can be radical or incremental. Why only
these two? The answer can be found in the intellectual origins provided by Schumpeter’s
theory of business cycles (see Andersen’s chapter (2) in this Handbook). These are waveform
fluctuations that can be traced statistically through the evolution of capitalism. Also known as
‘long waves’ they have been applied in regional science to explain the rise of ‘new industrial
spaces’ such as Silicon Valley (Hall & Markusen, 1985), spatial processes underlying
globalisation (Dicken, 1986; Knox & Agnew, 1994), regional development (Marshall, 1987)
and regional
Long
Wave
Mechan-
isation
Railroad
-isation
Electri-
fication
Motoris-
ation
Informa-
tisation
Neuro-
nanobio
Time
Period
1770-
1830
1820-
1880
1870-
1920
1910-
1970
1960-
2020
2010-
2060
New
Inputs
Water
Power
Steam
Power
Electric
Power
Oil
Micro-
process
or
Biochip,
Brain
Imaging
Driving
Industry
Textiles
Rail-
ways
Electri-
city
Automo-
tive
Micro-
comput
ers
Nano-
biotechn
ology
New
Industry
Canals,
Cotton
Mills
Steam-
ship,Tel-
egraph
Motors,
Power
tools
Aero-
space
Soft-
ware,
Neuro-
therano
stics
Six Long Waves of Techno-Economic Development (1770-2060)
Fig. 2. The Waveform Evolution of Capitalism (1770-2060)
Source: Based on Lynch, Z. www.neurosociety.com
innovation (Martin & Simmie, 2008) among many others. Fig. 2 presents a stylised
representation of these waves since the onset of the Industrial Revolution and into an
uncertain future where, nevertheless, the not unreasonable emphasis on nanobiotechnology
15
and neuroscience arises from economic geographer and evolutionary biologist Zack Lynch’s
BrainWaves weblog3. The key point in respect of the neo-Schumpeterian perspective on
innovation is that long waves are the linch-pin of its scientific theory, analysis and to some
extent predictive power. Innovation is the engine of capitalism, long waves set its course for
epochal periods and creative destruction represents the punctuation of its evolution caused by
radical innovation. Accordingly, each of the six epochs in Fig. 2, with the possible exception
of the last one, represent widely accepted as key evolutionary moments of the capitalist era.
As such, they had sometimes brutal (hand-loom weavers rapidly obsolesced by factory
production), sometimes gentler, more long-lasting transitional impacts upon societies in
which they occurred (such as the evolution of motorisation). To the extent society’s norms
had become embedded in the dominant preceding technological regime it could be its
regulatory, socio-economic governance instruments that enforced constraints to hold back the
negatively perceived impacts of such radical change. In the UK, the Factory Acts moderated
the depredations wrought on factory workers, the ‘Red Flag Act’ set motorized speed limits
of 4 mph in the country and 2 mph in towns: with a man walking ahead warning locals by
flapping a red flag, while steam still has to give way to sail at sea. Hence the concept of
technological ‘regime’ became an important accompaniment to the concept of ‘technological
paradigm’ associated with the radical innovation and its pervasive influence in itself. As
noted, this regime-paradigm distinction weaves through the main elements of the regional
innovation systems framework. There, the paradigm is the firm-based ‘exploitation’ or
‘commercialisation’ sub-system, while the regime is the socio-economic governance sub-
system that sustains it.
However, as is well-known, for the neo-Schumpeterian school radical is accompanied by
incremental innovation, which occurs during the paradigmatic epochs as marginal
improvements are made under broadly technologically equilibrium conditions. This is when
path dependence resumes, albeit renewed and redirected, following the punctuated evolution
caused by creative destruction. As Martin shows in chapter 15 of this Handbook, elements of
this may set in early, such as the QWERTY keyboard still used in iPhones or late, as in the
case of increasingly efficient automotive engines in the face of ‘peak oil’, global warming
and renewable energy vehicles. Late path dependence is a frequent response to perceived
3 The term neuro-theranostics refers to treatments that are both therapies and diagnostic treatments
in neuro-medicine.
16
paradigm shift as discussed by Dosi (1982) and Freeman et al., (1982) in respect of advances
in clipper ship technology in the face of steam, or piston engine aircraft in the face of jet
engines (Geels, 2006). But, in any case, it is a key tenet of neo-Schumpeterian innovation
studies that incremental innovation is regular, common and more of a central feature of
innovation than radical innovation. In Lundvall (1992) it is clear that this is a necessary
consequence of the rise of user-producer innovation relationships. Geographical and cultural
proximity are assets in the tacit knowledge exchange this implies:
‘...distance will play an important role. Being close to advanced users will form a
comparative advantage for the producers and vice versa.......they base their
comparative advantage upon geographical and cultural proximity’ (Lundvall, 1992,
57-8).
The contrasts between this and ‘radical innovation’ are stark. Norms, codes and standards
become inadequate, producers lack communicability and users become confused.
Geographical proximity thus becomes even more important because face-to-face contact,
trust, and even friendship are the only solvents of cognitive radical innovation blockages.
Accordingly, radical innovation is disturbing and unsettling until the new paradigm has
demonstrated its value and hitherto resistant multinationals, for example, start to make
acquisitions of new paradigm start-ups.
Hence, there is long-wave inducing innovation interspersed with epochal passages of
incremental innovation or ‘normal science’, an idea taken from Thomas Kuhn’s (1962)
pioneering research on the ‘structure of scientific revolutions’. Clearly, such a restrictive
binary division between these opposites is inadequate, except as a guide to certain interesting
cultural and geographical ‘infrastructures’ of innovation as implied by the user-producer,
interactive critique of the prevailing command model of linear innovation then prevalent.
Thus not long afterwards Christensen (1997) introduced a more cost-based, middle-range
distinction between ‘disruptive’ and ‘sustaining’ innovation, the first occurring with the
‘democratisation’ of innovations, creating mass markets by cost reduction that disrupts the
market for hitherto dominant technologies. The second is the opposite, where to sustain a
presence firms innovate upwards, making more exclusive, expensive products. However,
these are fundamentally marketing rather than innovation strategies, since in both cases the
‘innovation’ to be re-packaged already exists. Nor does this possible solution pretend to have
the theoretical depth to trace the socio-technical and economic geographic implications of
creative destruction displayed by the neo-Schumpeterian approach.
17
Verganti’s work on design driven innovation discussed above also introduces a different, less
epochal or long-wave take on radical innovation, which is nevertheless neo-Schumpeterian in
its theorisation. Design is defined as dealing with the meanings people give to products and
the messages and product languages that can be devised to convey meaning – ‘de-sign’,
‘signing it’, and giving it ‘sign-ificance’: ‘design is making sense of things’. It also has
paradigmatic representation in distinct design languages. Thus innovation of meanings is
incremental if framed within existing aesthetic norms but radical when significant
reinterpretation of meanings is achieved. Radical innovations in meaning are not immediate
and imply profound changes in socio-cultural regimes. Product ranges may be transformed
from simple tools to ‘transition objects’ intended to appeal, for example, to child-like
affections dormant in adults. The innovation model of a firm such as this means it pushes
innovative design on to the market, as ‘technology push’ does, with radical effect. The
discourse is one of making proposals to the potential consumer or user of the innovation. This
is not ‘technology push’ but ‘design push’ and conceivably ‘regime push’. Design is not
inclusive but negotiated by members of knowledgeable ‘circles’, some with hierarchical
structures (Pisano & Verganti, 2008). However, this is clearly not ‘epochal’ but more
‘episodic’ radical innovation. It is economic but it is also symbolic, invoking ‘creative
destruction’ in the relatively short-term aesthetic sphere.
Accordingly, from within the same broad theoretical perspective we have, first, an ‘epochal’,
long wave technological radicalism that has pervasive and transitional effects on communities
and markets of many if not all kinds. Contrariwise, in more specialist markets we have
‘episodic’ radical innovations that are transformative of taste and affection towards objects.
These may spawn radical innovation in related fields, even contributing to a broader ‘design-
intensive’ product ethic while drawing process and content novelty from ‘epochal’ innovation
such as widespread computerisation in society. It displays interactive qualities between the
design ‘paradigm’ and the ‘regime’ of ideas, norms and standards that envelops it cognitively.
In these respects, and drawing attention to the possibility of other varieties of radicalism in
innovation, this marks an important enrichment of the neo-Schumpeterian perspective on
regional innovation.
The Roles of Innovator and Entrepreneur
This is a lesser order of issue for the perspective informing this handbook than the preceding
ones concerning relations between innovation and growth; linear, interactive and neo-linear
18
innovation models; and ‘epochal’ versus ‘episodic innovation radicalism. It has, nevertheless
long been a grey area in innovation studies where they can be treated as the same thing or
even act as substitutes. Thus Garud & Karnoe (2001) in a widely-cited review of the literature
on path dependence (see chapters (14) by Boschma & Frenken, and (15) by Martin in this
Handbook) discuss innovation only in terms of entrepreneurship not innovators. It can, of
course seem pedantic to dwell on this, but in the absence of any discussion as to why
entrepreneurship is being privileged the reader is justified in querying the usage. For a
moment’s thought reveals that it is arguable most entrepreneurs are not innovators or indeed
innovative. Those that are can often be seen employing specialist professionals to manage
innovation. This is pronounced in fashion design as we have just seen and can be observed in
the haute couture industry studied by Wenting (2009), where signature designers are hired
routinely as in-house or freelance consultants. But this separation is not confined to fashion
for nowadays innovation in, for instance, the automotive and knowledge-intensive services
industries are subject to comparable outsourcing. In automotives, this ranges from
combustion engine consultancy A.V. List in the Steiermark region of Austria, where research,
experimentation and examination of engine technology are conducted for leading global
brands, to the production and prototype design habitually done by consulting engineers in
other parts of the automotive industry (Schamp, et al., 2004; Strambach, 2008).
This is unexceptional and yet another marker of the rise of user-producer innovation across
the economy as lean production has eroded the hitherto ‘M’ form of departmentalised
corporate hierarchy. This reached its apotheosis in the era of ambitious claims about the
virtues of economic and spatial planning. In Chapter 2 by Andersen in this Handbook, he puts
neglect of innovation in the literature down to the interest of ‘post-Schumpeterian’ regional
economists in growth modelling where innovation was forgotten because its dynamics were
also less amenable to formalisation. They were more interested in static interdependence
between different parts of the industrial system and its translation into policies of regional
development planning. According to Andersen (this volume), Schumpeter was of the opinion
that an innovation could neither be implemented nor financed by its inventor. This is because
the inventor of the innovation (or innovator, a term Schumpeter is not recorded as using
much, if at all) requires the skills of the entrepreneur to implement the innovation, including
borrowing the necessary capital and establishing a new firm to invest it. The entrepreneur
may also recombine other innovations in the new firm, but the innovations are by someone
else (i.e. the Schumpeterian ‘inventor’). So, as the word implies, the entrepreneur is the active
19
intermediary (middleman/woman) among the active knowledge source (inventor/innovator),
active financial source (bank) and the active market (attracting customers).
This makes sense, even in early capitalism, because engineering skills to invent/innovate
were specialised even then, as Uglow (2003) shows for the Boulton/Watt relationship in
marketing steam engines, most of which were sold to Cornish tin-mines, largely at the behest
of entrepreneur Boulton rather than the engineer Watt. Nowadays, in innovative industry such
as digital services or genomics, venture capitalists play a role comparable to that of the
Schumpeterian entrepreneur. Typically, they provide finance, are likely to be involved in
establishment of a start-up or spinoff firm and will oblige the inventor/innovator to accept
that the firm should have both coaching and professional management, something academic
entrepreneurs sometimes refuse, thus killing the deal (Hellmann, 2000). In reality, the
innovator may be an entrepreneur but such hybrids are unquestionably a tiny minority. Most
entrepreneurship is, again in reality, moderately routine: even venture capitalists, like
consultants have their methodology worked out and it probably does not change much in its
essentials. What may distinguish them is their relatively sophisticated knowledge of specific
technologies, necessary to help determine investment risk. As today, the Schumpeterian
entrepreneur could actually also be a serial entrepreneur although this was disallowed in his
analysis for technical reasons. Finally, this analysis is based on Schumpeter I; in Schumpeter
II oligopolistic corporations were increasingly also innovators too, meaning in-house
‘entrepreneurship’ and use of retained profits, outlawed in Schumpeter I, could be used in
innovation investments.
Path Dependence and New Path Creation
This is also something of a dependent variable in relation to the bigger picture of ‘punctuated
evolution’ and ‘creative destruction’. It arises from the perspective that equilibrium prevails
in economic development but that it is epochally ‘punctuated’ by radical innovation and
relatedness of innovation associated with the ‘carrier wave’. The idea of ‘path dependence’
(David, 1985) is intended to capture this claimed characteristic of innovation, namely that its
resonances may persist right through the long intervals and possibly beyond. This is rather
akin to the echoes of the ‘Big Bang’ that signified the birth of the universe that may still be
observed amongst the universe’s background noise by astrophysicists. But does it serve the
same purpose of confirming one theory and inclining to disprove a competing theory (e.g.
‘steady state’ origin of the universe)? What is served by showing persistence in a subject-field
20
that is largely devoted to understanding how socio-economic and technical novelty and
innovation occur? In one respect, it is a little like listening for the echoes of Big Bang in
seeking retrospectively to understand why things are as they are. In David’s (1985) study, the
focus was the QWERTY keyboard, still utilised in today’s digital micro-devices because,
although better arrangements have been innovated none has been adopted, largely for
institutional, practical reasons at the behest of users. This is helpful, as it underlines the fact
that ‘institutions matter’ in technology analysis and that it is indeed a human artifact rather
than a disembodied and societally neutral device or procedure.
Much the same can be said for the subject matter of this handbook, regional innovation and
growth. The urban and regional socio-technical paradigm changes, sometimes quite
dramatically, as in recessions or lengthier periods of industrial decline. The spatial regime
that accompanies it can, according to inherited theory, only change with difficulty and slowly,
if at all, because it is institutionally path dependent. This means the education system, the
standard social norms and expectations, the skills base, even the means to be entrepreneurial
or innovative for most firms and people are path dependent and the region in question ‘locked
in’ to an obsolescent socio-technical regime. This is rather a strong explanation that can find
good empirical support in some regions, notably older heavy industry regions formed in the
early Industrial Age. Yet, as Boschma & Frenken’s chapter (14) in this Handbook shows,
much evidence is emerging that the more accomplished regional economies are those with
relatedness to established industrial structures. In other words, new path creation is possible
and actually occurs. The acid test is, perhaps, coal-mining regions. In the UK, France and, to
a lesser extent, Germany such regions seem to illustrate the path dependent, locked-in profile
quite well. However, in the Netherlands, where much of the early work on relatedness was
conducted, its former coal mining region of Limburg prospers. Why is this? One important
reason is that the former state-owned concern Dutch State Mining (DSM) was privatised and
chose to diversify, first into chemicals, more recently into ‘biologics’ or inputs into
biotechnology and pharmaceuticals (e.g. vitamins). This ‘Nokia-type’ strategy of internal
corporate transformation has also served the region, as well as DSM’s shareholders, well.
Innovative links with regional, and later global, universities have assisted these efforts. This
is an example of what Martin (chapter 15) refers to as industry-focused evolution from path
dependence but in a context of radical renewal (not necessarily radical innovation). Indeed it
is more akin to Verganti’s version of radical innovation in that there is episodic, large firm-led
technological paradigm shift, which happens to be interactive also with a relatively
21
modernising socio-cultural regime in the Limburg region. One hypothesis about the ‘dogs
that don’t bark’ is, first, weak relatedness and, second, an inappropriate socio-cultural
(including political) regime. Of the countries listed at the outset of this section, the UK is
easily the one that most evolved regional unrelatedness in ushering into old industrial regions
branch plants in many kinds of light industry as its regional policy. Germany, by contrast, has
not entirely unsuccessfully benefited from path dependent relatedness among coal,
pharmaceuticals, steel and engineering in its policy and France is somewhere in between.
Thus relatedness offers niche understanding of ways out of negative path dependence as a
policy that encourages entrepreneurship from innovation around established industry
branches. The aim is to diversify from an ‘industrial monoculture’ but not too distantly in
terms of industry profiles. This happens more as a norm of regional innovation in the Nordic
countries as three brief sketches show. First, as the chapter (32) on Green Innovation in this
Handbook shows Jutland, Denmark’s expertise in wind turbines, followed by power station
design and other eco-innovations stems from path creation from marine and agricultural
engineering. Vastragotland’s (Gothenburg region) loss of ship construction was mitigated by
the build up of automotive engineering and development of technical expertise and
innovation in specialised ship component sub-systems sold to the Asian shipbuilders who
displaced them but still need such expertise. Finland’s pulp and paper equipment industry
sought less reliance there and in the case of ValMet (Vaasa region) diversified into luxury
sports car assembly (Porsche Boxter). When the parent repatriated assembly to south-west
Germany in the 2007-9 recession, a contract was won to assemble US firm Fisker’s electric
sports car. These are sufficiently related business moves to enable rapid adjustment, notably
in relation to the engineering skills base of the industry in question. Accordingly, while
regional innovation must pay attention to analysis of, especially, negative path dependence,
its prime focus will remain better understanding regional innovation by judicious new path
creation.
Technological or Innovation Paradigm and Regime?
With its origins in economic geography and applied regional economics, the language of
regional innovation systems research is not significantly influenced by that of the national
innovation systems approach, part of the origins of which reside in analysis of science and
technology and its relationships to society (but see Asheim & Gertler, 2005; Asheim, 2007 for
a more ‘national’ view of regional innovation). Many early definitions of that field of study
22
betray those origins, giving what Lundvall (1992) refers to as the ‘narrow’ definition of
innovation:
‘...........we may make a distinction between a system of innovation in the narrow
sense and a system of innovation in the broad sense. The narrow definition would
include organisations and institutions involved in searching and exploring – such as
R&D departments, technological institutes and universities. The broad
definition.........includes all parts and aspects of the economic structure and the
institutional set-up affecting learning as well as searching and exploring.......’
(Lundvall, 1992, 12)
The narrow definition is influenced by the American perspective of the time as conveyed by
Nelson (1993) and Rosenberg (1994) in which innovation was seen as largely technological.
As we have seen, the roles of R&D departments (e.g. Bell Labs) had been instrumental in
evolving the technological lead of US firms and the economy more generally. The
phenomenon of computerisation and its later ‘democratisation’ in the form of personal
computers owed much to military and other government contracts paid to large laboratories
from which new technology firms spun out. This, captured in the rise of territorial technology
complexes like Silicon Valley needed to be understood. Economic geographers and regional
scientists were in the vanguard in this task (e.g. Hall & Markusen, 1985; Scott, 1988) but
were typically rather technology-struck and descriptive rather than analytical.
Accordingly, the technology emphasis to innovation studies in the narrow sense hung over
the field for a long time. Even when more nuanced analyses of systems of innovation
emerged in which the enveloping regime of regulations, standards and rules gave social
content to the narrow view, thus broadening it out somewhat, the discourse was of
technological paradigms and regimes, sometimes techno-economic in the latter case (Dosi,
1982; Freeman et al., 1982). Contrariwise, the broad definition of innovation that Lundvall
alludes to in the quotation above was really far too broad, although he nuanced it somewhat
by saying that at different epochs different sub-systems of the whole economic structure and
institutional set-up would be the focus for study of innovation. Thus workshop systems would
be the focus in study of the earliest epoch of the Industrial Age; the electrification epoch
would put newly emergent R&D labs under the microscope; while the Information Age
would highlight universities and academic entrepreneurship. Yet, as is evident, the
perspective on technology remains resolutely to the forefront as the broad view of innovation
is whittled down to manageability for research purposes. Even then, accounts of national
innovation systems such as those in Nelson (1993) could be fairly sprawling and with low
analytically-based comparability across national cases as a partial consequence. To try to
23
place some further kind of manageability researchers specialised more in technological or
sectoral innovation systems analysis (Carlsson & Stankiewicz, 1991; Breschi, chapter 10 in
this Handbook) but technology again dominated and the measurements deployed in sectoral
comparisons were often not measures of innovation but invention (patents; R&D
expenditures).
Importantly, therefore, regional innovation systems research involved neither only the narrow
technology focus nor the complexities of trying to research exemplar sub-systems of epochal
technology regimes in relation to leading technological paradigms. Crucially, it depended
heavily on theory, on the one hand, and tailored empirical research on firm and organisational
innovation, on the other. This is not to say the regional perspective produced superior results
or insights to the others, mainly because the focus was different, but rather to say its
methodology was more grounded and its data for testing propositions was focused entirely on
processes and institutions responsible for innovation. This is accounted for clearly in chapters
in this volume such as Tödtling and Trippl (chapter 33; see also, ibid, 2005) and Heidenreich
& Koschatzky (chapter 39). In the former, the authors emphasise the taxonomic approach to
analysis of difference, which enabled research-based testing to eventuate in typologies that
were both valuable for comparative analysis and for the design of regional innovation
policies. Furthermore, because of the regional science and regional policy origins of the
perspective, innovation systems research examined more deindustrialising and rural regional
settings than ever occurred in the technology-focused work of national, technological or
sectoral systems research. Thus an innovation focus has paid dividends from its comparative
and primary empirical emphases, as it has from its encompassing of the study of innovative
aspects of regional governance. Heidenreich & Koschatzky (this volume) note the importance
of this in highlighting variability in regional governance powers as key elements in
understanding the structure of regional economies, especially in Europe. As they show, there
can be swift recovery from economic setbacks where there are political commitments to
regional innovation and resources to implement them and the opposite where these are
absent, citing contrasts between Spanish autonomous regions and much weaker governance
set-ups in support of their argument. This is extended a fortiori to federal governance, where
despite the traumas of transition, notable successes have been registered around regional
innovation strategies in Thuringia, Saxony and Brandenburg in former east Germany,
indicating the advantages of ‘regional experimentalism’ of the kind that regional innovation
analysis is well-attuned to (Sabel, 1995).
24
It has been noted in the section of this Introduction on neo-linear innovation models, that
there is still much to be gained from adapting foundational concepts such as technological
paradigms and socio-technical regimes to less narrow and also less epochal subjects than long
waves. Verganti’s (2006) adaptation of technological to socio-cultural regime is innovative in
the manner it shows how the broader ‘regime’ concept, first introduced in international
relations studies (Ruggie, 1975) can be usefully deployed more episodically and regionally
than hitherto. Similarly ‘design paradigm’ speeds up the application of what began life, after
Kuhn (1962), as a concept denoting less than long wave longevity. Accordingly, preference
should generally be shown for utilising ‘innovation paradigm’ and ‘innovation regime’ where
it is found analytically useful. ‘Regional innovation paradigm’ is not strictly synonymous
with ‘regional exploitation sub-system’ (one of the two sub-systems in a Regional Innovation
System, as demonstrated in Tödtling and Trippl (chapter 33) and Heidenreich & Koschatzky
(chapter 39). This is because paradigm denotes ‘dominance’ (prevailing) whereas an
‘exploitation sub-system’ denotes variety (e.g. clusters, oligopolies, supply-chain elements
co-existing regionally). But as long as usage is clear, reference to, for example, Lombardy’s
‘design driven regional innovation paradigm’ to paraphrase Verganti (2006) seems
unexceptionable. Equally, ‘regional socio-cultural regime’ adds value and takes innovation
away from its hitherto prevailing technological bias. Research on precisely this phenomenon
has been conducted in the EU 6th. Framework Programme in the CURE project
(http://www.cure-project.eu/; Heidebrink & Soul, 2007). The research reveals significant
regional socio-cultural regime distinctiveness interacting in path dependent and path creating
ways with corporate innovation practices in production, organisation and marketing.
System Self-organisation or System Leadership?
Finally, we arrive at a culminating issue that is of wide-ranging importance to regional
innovation systems studies, not least because it addresses a conceptual and real issue about
systems. This concerns the extent to which ‘practice systems,’ of which regional innovation
systems are an exemplar, are intended to achieve optimal efficiency and effectiveness through
tending towards autopoesis or self-organisation, or through a hierarchical form of directed
organisation involving system leadership (Wenger, 2000). Clearly, the latter concept has
received much attention in business school literature in the shadow of airport biographies of
the likes of ‘Neutron Jack’ Welch, former CEO of General Electric who is presented as
almost single-handedly turning around the fortunes of that company through aggressive cost
accounting, including an infamous annual cull of the company's bottom-performing 10% with
25
massive workforce reduction during his period of stewardship. Such beliefs spawned a
plethora of consultancy reports and business school articles and books on ‘leadership’
drawing on heroes as varied as football coach Vince Lombardi, Sun Yat Sen and Antarctic
explorer Sir Ernest Shackleton. The business writings of Machiavelli were even pored over
for the guidance of modern managers.
In the Welch era, GE’s pollution of the Hudson River made the company a target of the US
environmental movement, but post-Welch the firm’s two key marketing campaigns,
‘Ecomagination’ and ‘Healthymagination’ repositioned GE as a champion of green
technology and healthcare initiatives. Even so, the company’s once huge financial services
division, GE Capital, a Welch-inspired innovation supplying 55% of GE revenue, reported
$500 billion in debt occasioned by the credit crunch and bad loans. Recourse was required to
the US Treasury’s Troubled Asset Relief Programme (TARP) for a $3.5 billion loan in 2008.
Separately, GE was forced to retreat from the media business, selling its controlling stake in
NBC Universal to Comcast. An accounting scandal and dividend cut dented its reputation for
financial reliability, and in Britain it was accused of medical censorship after it took a
radiologist to court for claiming there were potentially fatal side effects to one of its
healthcare products. It could be argued that this and other hard-driving corporations, notably
Royal Bank of Scotland and Lehman Bros were not especially advantaged by charismatic
leadership in difficult times.
What about softer forms of leadership in more ‘loosely coupled systems’ typically found in
regional innovation systems? Sydow et al. (2011) study this comparatively. They conceive it
as involving motivating, involving, empowering, supporting, sense-making, mobilizing,
controlling, manipulating, legitimizing and representing. They suggest it is not so different
from internal management in the large corporation, but in one respect more than the others,
regional innovation system management, possibly more than clusters, can have parallel and
rotating leadership of action lines in pursuit of strategic aims. Stakeholder governance of
regional innovation systems means they are appropriate vehicles for such focal and temporary
management of specific actions and it is a method for keeping commitment from the
relatively high-powered individuals who are likely to find themselves invited to serve on
innovation system governance networks. This is not self management; it is leadership by a
collective organisation responsible for managing, for example industry clusters and it may
pass swiftly or rotate among incumbents who are not employed and may not even be re-
imbursed expenses. Innovation system management is unlike cluster management in being
26
this step nearer to autopoesis since it is unusual to find a cluster without a cluster-
management team in the form of a network (e.g. the Cambridge Network Ltd.) or a common
services council (e.g. Massachusetts Biotechnology Council). Variable governance of multi-
client sub-systems maintains the loose coupling and related flexibility, variety and reflexivity
of regional innovation systems.
The key problem of over-personalised leadership as revealed in the Sydow et al (2011) study
is that when the charismatic leader steps aside there may not be an equivalent available to
replicate any successes of the past; worse, the system may crumble, precisely the result of
this in the Arizona optronics case that is one of Sydow et al.’s exemplars. Thus while
collective leadership might not be autopoesis of the self-organising kind that systems
thinking tends to favour, it is far less risky than the ‘cult of the personality’ that is its
diametric opposite. This is analysed in relation to Lombardy’s design driven regional
innovation paradigm by Pisano & Verganti (2008) where a hierarchical, exclusive circle of
experts is deemed the appropriate collaboration model. Given the obvious weaknesses of
personality cult ‘leadership’, especially for non-corporate, stakeholder, loosely coupled
systems, yet the uncertainties of commitment that can be imagined from a flat-hierarchy,
network managed, loosely coupled arrangement, perhaps Wallin’s (2006) model of
‘orchestration’ is an elegant compromise, embodying the notions of ‘conducting’ as a
distinctive expertise from being the highly expert leader of the orchestrated woodwind or
violin section. The role of ‘orchestration’ is developed at greater length in chapters 23 and 42.
References
Asheim, B. T. (2007) Differentiated knowledge bases and varieties of regional innovation
systems, Innovation – The European Journal of Social Science Research, 20, 223-241
Asheim, B. & Gertler, M. (2005): The Geography of Innovation: Regional Innovation
Systems, in Fagerberg, J., Mowery, D., and Nelson, R. (eds.), The Oxford Handbook of
Innovation. Oxford University Press, Oxford
Aydalot, P. (ed.) (1986) Milieux Innovateurs en Europe, Paris, GREMI
Bagnasco, A. (1977) Tre Italia (Three Italies: the Territorial Problematic of Italian
Development), Bologna, Il Mulino
27
Becattini, G. (1978) The development of light industry in Tuscany, Economic Notes, 18, 107-
123
Brusco, S. (1982) The Emilian model: productive decentralisation and social integration,
Cambridge Journal of Economics, 6, 167-184
Carlsson, B. & Stankiewicz, R. (1991) On the nature, function, and composition of
technological systems, Journal of Evolutionary Economics, 1, 93-118
Christensen, C. (1997) The Innovator’s Dilemma, Boston, Harvard Business School Books
Cooke, P. (1992) Regional innovation systems: competitive regulation in the new Europe, Geoforum,
23, 365-382
Cooke, P. (2008) Regional innovation systems: origin of the species, International Journal of
Technological Learning, Innovation & Development, 1, 393-409
Cooke, P. (2009) The economic geography of knowledge hierarchies among internationally networked
medical bioclusters: a scientometric analysis, Tijdschrift voor Economische en Sociale Geografie, 100,
332-347
Cooke, P. & Morgan, K. (eds.) (1998) The Associational Economy, Oxford, Oxford University Press
David, P. (1985) Clio and the economics of QWERTY, American Economic Review, 75, 332-337
Delgado, M, Porter, M. & Stern, S. (2010) Clusters and entrepreneurship, Journal of Economic
Geography, 10, 495-518
Dicken, P. (1986) Global Shift, London, Harper & Row
Engelen, E. & Faulconbridge, J. (2009) Introduction: financial geographies - the credit crisis as an
opportunity to catch economic geography’s next boat? Journal of Economic Geography, 9, 587-595
Edquist, C, Hommen, L. & McKelvey M. (2001) Innovation and Employment: Process versus Product
Innovation, Cheltenham, Edward Elgar
Freeman, C, Clark, J. & Soete, L. (1982) Unemployment and Technical Innovation: A Study of Long
Waves & Economic Development, London, Pinter.
Freeman, C. (1991) Networks of innovators: a synthesis of research issues, Research Policy, 20, 363-
380
Garud, R. and Karnøe, P. (2001) Path Creation as a Process of Mindful Deviation, in Garud, R. and
Karnøe, P. (eds.) Path Dependence and Creation, London: Lawrence Erlbaum
Geels, F. (2006) Co-evolutionary and multi-level dynamics in transitions: the transformation of aviation
systems and the shift from propeller to turbojet (1930-1970), Technovation, 26, 999-1016
Hall, P. & Markusen, A. (eds.) Silicon Landscapes, London, Allen & Unwin
28
Heidebrink, L. & Soul, P. (2007) Business ethics in a regional perspective, In: Zeitschrift für
Wirtschafts- und Unternehmensethik, zfwu 8/3 (2007), 353-356Journal of Economic and Business
Ethics, 8, 353-356
Hellmann, T. (2000) Venture capitalists: the coaches of Silicon Valley, in Lee, C, Miler, W,
Hancock, M. & Rowen, H. (eds.) The Silicon Valley Edge, Stanford, Stanford University Press
Hymer, S, (1976) The international operations of national firms: A study of direct foreign
investment, Cambridge, MIT Press
Jones, D. (1984) Outsourcing in the West Midlands Automotive Industry, Birmingham, West
midlands Enterprise
Knie, A. & Hård, M. (2010) Innovation, the endless frontier: historical networks of
engineering science, Technology Analysis & Strategic Management, 22, 433-445
Knox, P. & Agnew, J. (1987) Geography of the World Economy, London, Edward Arnold
Krugman, P. (1991) Geography & Trade, Cambridge, MIT Press
Kuhn, T. (1962) The Structure of Scientific Revolutions, London, Penguin
Lewis, A. (2010) The Big Short, London, Penguin
Lundvall, B. (1985) Product Innovation & User-Producer Interaction, Aalborg, Aalborg University Press
Lundvall, B. (1992) User-producer relationships, national systems of innovation and
internationalisation, in Lundvall, B. (ed.) National Systems of Innovation, London, Pinter
Marshall, M. (1987) Long Waves of Regional Development, London, Macmillan
Martin, R.L. and Sunley, P.J. (1998) Slow Convergence? The New Endogenous
Growth Theory and Regional Development, Economic Geography, 74, 201-227
Martin, R. & Simmie, J. (2008) Path dependence and local innovation systems in city-regions,
Innovation: Management & Policy Practice, 10, 183-196
Moulaert, F. & Sekia, F. (2003) Territorial innovation models: a critical survey, Regional Studies,
37, 289-302
Morgan, K, Marsden, T. & Murdoch, J. (2006) Worlds of Food, Oxford, Oxford University Press
Nelson, R. (ed.) (1993) National Innovation Systems, Oxford, Oxford University Press
Patterson, S. (2010) The Quants, New York, Crown
Pavitt, K. (1984) Sectoral patterns of technical change, Research Policy, 13, 343-373
29
Piore, M. & Sabel, C. (1984) The Second Industrial Divide, New York, basic Books
Pisano, G. & Verganti, R. (2008) Which kind of collaboration is right for you? Harvard Business
Review, 86, 78-86
Ponds, R, van Oort, F. & Frenken, K. (2009) Innovation, spillovers and university-industry
collaboration: an extended knowledge production function approach, Journal of Economic Geography,
10, 231-255
Porter, M. (1990) The Competitive Advantage of Nations, New York, The Free Press
Romer, P. (1990) Endogenous technical change, Journal of Economic Literature, 98, 338-354
Rosenberg, N. (1994) Exploring the Black Box: Technology, Economics, & History,
Cambridge, Cambridge University Press
Ruggie, J. (1975) international responses to technology: concepts and trends, International Organization, 29,
557-583
Sabel, C., Kern, H. and Herrigel, G. (1991) Collaborative manufacturing: new supplier
relations in the automobile industry and the redefinition of the industrial corporation, In H.G.
Mendius and U. Wendeling-Schröder, (eds), Zulieferer im Netz -- Zwischen Abhängigkeit und
Partnerschaft (Suppliers in Networks – Between Dependency and Partnership) Cologne: Bund Verlag
Sabel, C. (1995) "Experimental Regionalism and the Dilemmas of Regional Economic Policy in
Europe," paper presented to the International Seminar on Local Systems of Small Firms and Job
Creation, June 1-2, Paris, OECD
Saxenian, A. (2000) Networks of immigrant entrepreneurs, in Lee, C, Miler, W, Hancock, M. & Rowen,
H. (eds.) The Silicon Valley Edge, Stanford, Stanford University Press
Schamp, E, Rentmeister, B. & Lo, V. (2004) Dimensions of proximity in knowledge-based networks:
the cases of investment banking and automobile design, European Planning Studies, 12, 607-624
Schmitz, H. (1995) Small shoemakers and Fordist giants: tale of a supercluster, World Development,
23, 9-28
Scott, A. (1988) New Industrial Spaces, London, Pion
Spencer, G, Vinodrai, T, Gertler, M. & Wolfe, D. (2010) Do clusters make a difference? Defining and
assessing their economic performance, Regional Studies, 44, 697-716
Strambach, S. (2008) KIBS as drivers of multi-level knowledge dynamics, International Journal for
Services Technology and Management, 10, 151-174
Sydow, J, Lerch, F, Huxham, C. & Hibbert, P. (2011) A silent cry for leadership: organizing for leading
(in) clusters, Leadership Quarterly 22 (2011)
30
Tödtling, F. and Trippl, M. (2005) One size fits all? Towards a differentiated regional
innovation policy approach, Research Policy 34, 1203-1219
Uglow, J. (2003) The Lunar Men: the Friends Who Made the Future, London, Faber & Faber
Van Ark, B, (2005) In search of the silver bullet for productivity growth: a review article of the power of
productivity and transforming the European economy, International Productivity Monitor, Centre for
the Study of Living Standards, vol. 10, 79-86,
Verganti, R. (2006) Innovating through design, Harvard Business Review, December, Reprint
R0612G, 1-9
Von Hippel, E. (1988) The Sources of Innovation, Oxford, Oxford University Press
Von Hippel, E. (2005) The Democratization of Innovation, Cambridge, MIT Press
Wallin, J. (2006) Business Orchestration, London, Wiley
Wenger, E. (2000) Communities of practice and social learning systems, Organization, 7, 225-246
31
