TECHNOLOGICAL AND

INDUSTRIAL CHANGE
Copyright © 2017 Division of Sustainability and Industrial Dynamics, INDEK, KTH. All rights reserved.
Unless otherwise indicated, all materials on these pages are copyrighted by the Division of Sustainability
and Industrial Dynamics, INDEK, KTH. All rights reserved. No part of these pages, either text or image
may be used for any purpose other than personal use. Therefore, reproduction, modification, storage in a
retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for
reasons other than personal use, is strictly prohibited without prior written permission.
This compendium contains nine selected chapters from the book “A Dynamic Mind. Perspectives on
Industrial Dynamics in Honour of Staffan Laestadius” (2016), Editors: Pär Blomkvist & Petter Johansson.
Division of Sustainability and Industrial Dynamics, Department of Industrial Economics and Management,
KTH. ISSN 1100-7982, ISBN 978-91-7729-170-1.

The included chapters in this compendium are as follows.

Systems thinking in Industrial dynamics Pär Blomkvist & Petter Johansson

This is Industrial Dynamics Pär Blomkvist, Petter Johansson & Staffan Laestadius

Eric Dahmén and Industrial dynamics Staffan Laestadius

A Critical View on the Innovation Systems Approach Staffan Laestadius & Annika Rickne

Diffusion of Innovations Emrah Karakaya & Pranpreya Sriwannawit Lundberg

Location as a matrix of competition Cali Nuur

Dynamics in Swedish Industrial and Political History Eric Giertz

For citations, you can use the following format:

<Authors of the chapter> (2016). <Name of the chapter>. In “A Dynamic Mind. Perspectives on
Industrial Dynamics in Honour of Staffan Laestadius” (Editors: Pär Blomkvist & Petter Johansson).
Division of Sustainability and Industrial Dynamics, Department of Industrial Economics and Management,
KTH. ISSN 1100-7982, ISBN 978-91-7729-170-1.
Systems thinking in Industrial dynamics
Pär Blomkvist & Petter Johansson

Systems thinking is an integral part of the fundamentals of industrial

dynamics. In this chapter, we explore and discuss systems thinking, what
it is and how it can be used to analyse technological, industrial and
societal change. The relevance of systems thinking have increased during
the last decades due to challenges such as globalisation and threat of
climate change. These challenges demand holistic approaches that
encompass several different interconnected areas, including government
policies, technological development, user practices, etc. In this chapter, we
present such holistic systems approaches and provide concrete examples
on how to apply them.

Introduction
Systems thinking is a way to understand interconnected parts that form a
complex whole. It provides a holistic perspective – including both social and
technical parts – that is central to researchers within industrial dynamics and
many other fields.

In this chapter we delimit our focus to how systems thinking and related
theories has evolved over time and been actively implemented in social
sciences generally and industrial dynamics specifically. After that we
introduce and discuss two systems thinking related frameworks frequently
used in industrial dynamics studies: Large Technical Systems (LTS) and the
Multi-Level Perspective (MLP). We also present a framework where we show
how LTS and MLP can be fruitfully combined.

The LTS perspective is typically used to study large physically connected

infrastructures, characterized in the LTS framework as large socio-technical
systems. The MLP perspective is used to study societal and technological
transitions and are popular in studies on socio-technical change towards
sustainability. In MLP studies the systems under investigation are not
necessarily connected by means of an infrastructural grid.

Systems thinking has a long historical background, going all the way back to
the philosophers of ancient Greece. In the following we present a brief and
condensed overview of systems terminology and the historical development of

45
systems thinking, leading up to current frameworks used in industrial
dynamics research today.

Systems thinking
The English word “system” comes from the Greek word systema, meaning
“whole compounded of several parts or members” (according to Merriam-
Webster, 2015). What does a whole mean in this sentence? Aristotle famously
expressed that 'the whole is greater than the sum of its parts', meaning that a
system whole exhibit qualities that each part of the system by themselves do
not exhibit. Today we refer to this phenomenon as emergence, also popularly
expressed as “one plus one equals three”. The more complex a system is the
more difficult it is to intuitively predict emergence from the system.

To understand what a system represents it can be fruitful to start with a simple

system that is characterised by few components with few interactions. These
are systems that are easy to understand and that we can recognize from our
everyday life, e.g. our temperature regulating thermostat at home or the cruise
control in our car.

Input Output
Process/transformation

Feedback

Boundary

Figure 1: A model of a simple system

Figure 1 shows a simple self-regulating system with the transformation

process of inputs to outputs regulated by a feedback loop and with a
permeable boundary towards the environment. If Figure 1 was to describe the
cruise control of a car then the output would be the car speed, which depends
on the input in form of the intended speed set by the driver and the actual
speed that is supplied to the transformation process through the feedback
loop.

46
It is easy to see how designed self-regulating systems, such as the cruise
control, have a very practical hands-on use. The mathematician Norbert
Wiener (1948) used the term cybernetics to describe these types of self-
regulating systems characterised by a (semi-) closed feedback signalling loop,
i.e. man-made physical systems designed to fill a specific purpose. Though
simple man-made systems are quite practical, they fall short when it comes to
describing more complex systems that we encounter in our everyday life.

A complex system is characterized by a high number of components &

subsystems and high level of interactions between these parts. The simple
system in Figure 1 does not constitute a complex whole in itself, as we have
defined it, so as we broaden our scope to the car including the driver in the
traffic system the complexity increases.

A complex system has equifinality, which means the ability to achieve its goal
in different ways. Sometimes complex systems are described as being on the
edge of chaos because the interactions of simple sub-systems can result in
unpredictable and seemingly random outputs. A complex system is also
characterized by hierarchy – as seen illustrated in Figure 2 – and showing
degrees of robustness and ability to adapt to its environment.

Figure 2. Illustration of hierarchy of systems and sub-systems

Systems thinking spread in many different research areas during the first half
of the 20th century. The biologist Ludwig von Bertalanffy, together with other
researchers such as the economist Kenneth Boulding, formulated the General
Systems Theory (GST) in an attempt to synthesize the view on systems from
different research areas (Von Bertalanffy 1968).

47
 A system is characterized by the interactions of its components and
the nonlinearity of those interactions
- Ludwig von Bertalanffy

In the work on GST von Bertalanffy highlighted the interactions between

components and the nonlinearity of these interactions in a system. He also
argued that no system that is open to interactions with the surrounding
environment, i.e. an open system, should be viewed in isolation (Von
Bertalanffy 1968). GTS was partly a response aimed towards a reductionist
approach that suggested that all parts of a system could be divided into its
individual components and each be optimised separately.

Operations research

In the 1950s and 1960s systems thinking became increasingly popular. The
promises held by systems thinking as a way to optimise operations and
decision-making made a significant impact on management research.

The wave of enthusiasm for applied systems thinking came out of the success
of Operations research efforts during the Second World War. System theories
had been used to quantify, calculate and optimize war time operations such as
ship fleet logistics, air bombing, radar defence and effective weapon systems
(Hughes & Hughes 2000). Operations research, Cybernetics and adjoining
fields promised ways to actually control large and complex technological
systems – in fact the term Cybernetics is derived from the Greek word for the
steersman on a ship (kybernētēs).

After the war operations research became widely used in industrial

management. Operations researchers addressed issues such as supply-chain
management, floor planning, staff scheduling, etc.

In the end the ambitions to control social and technical systems did not deliver
as much as first promised, but Operations research still made a big impact on
management research and its effects are visible on management studies to this
day.

As an example, social systems scientists are still using a way to define system
borders that was given by the operations researcher and system analyst West
Churchman almost 50 years ago. As the borders of a system are not given by
nature, it is the task of the analyst to define the delimitations of the system
that the analyst sets out to investigate. Churchman's approach to define what
was a part of the system and what was not was to ask the following two
questions: (1) Does “it” matter concerning the systems possibility to fulfil its
goals? (2) Can the system do anything about “it”? (Churchman 1968).

48
System dynamics

Another area, with its roots in applied systems thinking made during World
War II, is System dynamics. Influenced by the advancements of Wiener and
Cybernetics-research the pioneering computer engineer Jay Forrester
developed the field System dynamics at MIT Sloan School of Management in
the 1950's. System dynamics is an approach to understand the complex and
non-linear aspects of systems through feedback loops between interacting
components, and also flows, stocks and time-delays in the system. In 1970
Forrester created a system model to simulate the world as a socio-economic
system (called World1). Club of Rome three years later published Limits to
growth (Meadows et al, 1972) based on a later, more elaborated, version on
Forrester’s work (called World3).

The increasingly popular agent-based modelling (ABM) approach represents

more recent advancements in modelling social systems. The ABM approach is
not equation based or structured as the above mentioned system dynamics
approach, but instead focuses on the behaviour of actors and their behavioural
rules. A possibility with the ABM approach is to construct a system and then
remove particular agents one by one to detect which effect they have on the
system – something that would not be possible using system dynamics. The
hopes that today is put on ABM approaches in different fields, e.g. in the
energy field as seen in Bale et al. (2015), reminds of the promises that
operations research held during the 1950s and 1960s.

Socio-technical systems

Another area where systems thinking have had a big impact is organizational
development, with important advancements made by psychologists at the
Tavistock Institute in the UK in 1950.

Researcher Eric Trist and his colleagues at the Tavistock Institute coined the
term “sociotechnical system” to describe the organization of teams in the
mining industry and the interactions between the miners and the technology
they used in their work. By showing the importance of optimising both the
social and technical aspects of work, i.e. not only focusing on how to optimise
each working operation but also including the quality of the workers' worklife,
they provided a more holistic work design alternative to the reductionist
Taylorism-approach that was customary at the time.

Note that the Tavistock researchers used the term “sociotechnical systems” in
connection to management and organizational development, which is
different from the way the term is used in the LTS- and MLP approaches,
which will be introduced below.

49
System thinking in technological system studies

In the end of the 1970s a new way of systems thinking emerged in the fields of
social science and history of technology. The earlier applied systems thinking,
with its aim of full control had lost some of its popularity.

This new interest in system thinking was not so much aimed at the practical
control of systems, but on understanding and analysing socio-technological
systems in their wider political, industrial and societal context. It was a
reaction to the continuing growth of the interconnected technical systems that
took place after the Second World War – and also a reaction to the previously
discussed ambitions to control these systems using for example operational
analysis and cybernetics. As mentioned, these ambitions to control had not
delivered as much as they promised.

The new discussion on systems, was driven by researchers such as Bruno

Latour, Michel Callon, Wiebe Bijker, Trevor Pinch, and Thomas P. Hughes
(Bijker et al. 1987) as well as evolutionary economists Richard R. Nelson and
Sidney G. Winter (Nelson & Winter 1982). The advancements made by these
scholars have been an important base for both LTS and MLP studies.

Large Technical Systems

As mentioned systems thinking has been used in many disciplines. In this part
we turn to a physical and concrete meaning of the word system and discuss
large infrastructural systems – infrasystems (Kaijser 1994) – and their
dynamics. We present the so-called Large Technical Systems (LTS)
perspective that was introduced by Thomas P. Hughes in the beginning of the
1980s.

Hughes was a mechanical engineer turned historian. His view on

technological development in relation to the growth of large infrasystems has
had a profound influence in the fields of industrial dynamics and the history
of technology.

Thomas P. Hughes and the LTS-perspective

Industrialized nations are dependent on infrasystems. We need them for food,

transport, communication, power, heating, water, waste, etc. In fact, modern
life as we know it would be impossible without them. The systems are built up
of components such as technical artefacts (cars, electricity cables and roads);
institutions (formal and informal) and actors (professional and commercial),
connected in mutual interdependence. (Hughes 1987) It is important to notice
that each system component in itself could be viewed as a sub-system in the

50
larger infrasystem. It is the task of the researcher to define the system borders
when performing the investigation.

LTS: s is often centrally planned and managed leaving users and appropriators
with little power. The systems are not easy to change and deeply embedded is
society. Negative externalities, such as pollution and global warming effects,
are difficult to handle. Some of our infrasystems are not well adapted to the
goals of sustainable development.

It is important to understand that most LTS: s did not start as centrally

planned systems. They were not large and connected at all in the beginning of
their history. The historical origins of most, if not all, large technical systems
include some form of cooperation among its users. In Sweden, for example,
the development of the electrical and the telephone systems started as
cooperatives or other forms of joint ventures of users building the networks
from the bottom up (Kaijser 1999; Nilsson 2011). Historian of technology Arne
Kaijser (2002) describes the same process in the history of the Dutch struggle
against floods. He reveals the historical and institutional heritage created by
the appropriators’ efforts to build water control systems and its effect on
Dutch political culture. But in the end, the Dutch water system, like the
Swedish telephone and electrical systems, was transformed into a centrally
administrated and controlled, tightly integrated, national infrasystem.
Management changed from bottom-up to top-down. The appropriators at the
local level lost their influence on the journey towards a large technical system.
This centralization of the systems was also due to market mechanisms such as
economies of scale and a consolidation of the whole industry around the
system, often creating a monopoly or oligopoly market.

In Thomas P. Hughes (1983) ground breaking study, Networks of Power, on

the development of the electricity system in the US and its diffusion overseas
to Great Britain (London) and Germany (Berlin), he established the so called
Large Technical Systems (LTS) perspective (Bijker et al. 1987; Hughes 1983).
Hughes stressed that a large technical system must be analysed as a seamless
web of socio-political, economic, cultural, institutional and technical
components and thus not only technical but socio-technical in its nature.

Hughes uses the concept of technological style, or rather sociotechnical style,

when discussing the differences in system development between the countries
under investigation. The basic technological components that were
transferred from the US was not changed in any radical way, but the system
design and the institutional framing (ownership, regulations, etc.) became
quite different when the Edison system was built in London and Berlin.

An often cited definition on a large socio-technical system was presented by

the sociologist Bernward Joerges: “Such a complex and heterogeneous system
of physical structures and complex machineries 1) whose material

51
components are integrated or ‘linked’ over large areas and over time ... and,
2) supports or facilitates the operation of a large number of other technical
system, the organizations thus linked together.”21

The keywords are that the system is “large”, it is “complex” in the sense that
many parts of different character, both technical and organizational, are
included and that the parts are integrated or “linked”. In addition, it is
required that the system in question is interconnected with other systems.

The basic idea of the LTS perspective is that technological development cannot
be analysed properly by only investigating the individual component or the
different sub-systems. Here the car (the artefact/sub-system) and the road
transport infrasystem is a good example. When we buy a car or sit behind the
wheel we become, as a researcher has put it, not only drivers, but we are
“buying in to a complex road-, energy supply-, spare parts-, maintenance-,
registration- and insurance-, monitoring- and legal system” (Summerton
1998, p. 22).

The evolution of a socio-technical system can be described with the classical

S-curve – see Figure 3 – and divided into three phases: build up – expansion
– maturity. At this last stage, the LTS becomes increasingly difficult to change.
In the mature system, expansion slows down when markets are near
saturation (Hughes 1983).

When a system grows and especially in the expansion phase it can give the
impression of autonomous growth and system expansion seems inevitable –
in Hughes words: the system gains momentum. From a market perspective
system growth can be understood through its positive system externalities.
An infrasystem gets increasingly valuable when more people get access to
systems services and components – it is not very useful to be the only one
having a telephone.

21 Quotetranslated by Summerton (1998, p. 22) from Bernward Joerges’ “Large

Technical Systems: Concepts and Issues” (1988).

52
Figure 3: Illustration of adoption rate over time forming an S-shaped curve

When witnessing for example the motor road systems global expansion or the
explosive growth of the internet and mobile phone systems, it is easy to believe
in technological determinism. But it would be a mistake for managers and
policy makers to yield to this impression – in the words of Tom Hughes:

“… but such systems are not autonomous. Those who seek to control and
direct them must acknowledge the fact that systems are evolving cultural
artifacts rather than isolated technologies. As cultural artifacts, they reflect
the past as well as the present. Attempting to reform technology without
systematically taking into account the shaping context and the intricacies of
internal dynamics may well be futile. If only the technical components of a
system are changed, they may snap back into their earlier shape like charged
particles in a strong magnetic field. The field also must be attended to; values
may need to be changed, institutions reformed or legislation recast.”
(Hughes 1983, p. 465)

After the growth phase, the mature system is deeply embedded in society and
not easily changed or transformed. The large mass of an LTS – the system
inertia – consist of the physical infrastructure (roads, grids, pipes, tracks,
etc.), the organizations and institutions created for support, laws and
regulations and not the least financial and industrial corporations and
individuals (engineers, scientist, politicians), with huge vested interests in the
survival and expansion of the system. A mature socio-technical system is often
very conservative, its actors – called system builders by Hughes – unwilling
to change. It is often noted that it was not the stage coach owners that invested
in the railroad and not the railroad actors that promoted the automobile. This
feature is often called systems culture: A conservative attitude towards
change, among the actors promoting, building and living of the system.22

22 A concept used by Swedish historian of technology, Professor Arne Kaijser.

53
Thus, large infrasystems develop according to a certain path or trajectory. The
inertia acquired by the growing systems creates a “path dependence”. The
earlier choices by system actors, and also “learning by doing”, influence the
possibilities given to future actors. Donald MacKenzie takes this a starting
point when he writes: “I do not deny that paths exist in the sense of
clear/sustainable patterns of technological change.” But it is not easy to see
if the technological path depends on technology itself or if it resides in the
views of the systems stakeholders: “They invest money, careers and
credibility in being a part of “progress “and helps in this way to create the
“progress” of a certain expected form. Because the predictions turn out to be
correct, they predict with greater confidence.” MacKenzie argues that the
seemingly natural character of technological development is because “… in a
certain sense, a technological development path is a self-fulfilling prophecy…
In retrospect, it looks like to the prevailing technology possesses the internal
qualities that explain its success.” (MacKenzie 1993, p. 168).

An important concept-pair in Hughes large technical systems perspective are

the notion of Tightly coupled and loosely coupled systems. In a tightly coupled
system the connections between system components or sub-systems
(technical and institutional) are very strong and especially designed for
system purposes. The railroad and the electrical system are examples of this
type. They are centrally built, planned and managed, access for other
operators is very limited, the technical standardization of components is high
and they have a top-down perspective as distribution systems of goods. The
system builder or the system manager has one overarching goal: To avoid
technical and institutional mismatch in the system – to align system
components within the system. A tightly coupled system is highly sensitive to
disturbance, if one component fails the whole system breaks down (low
redundancy). Therefore “load factor management” is important to avoid
stress during peak usage and distributing the utilization of the system to reach
economic efficiency (economic mix). Simply put, a loosely coupled system
exhibits a lower level of the characteristics mentioned, with sea transport as
one example. The road system would be somewhere in between the two
system types. Historically it has moved from a loosely coupled system towards
a more tightly coupled system, especially since the introduction of the
automobile, and even more so with the introduction of information and
communication technologies (ICT), but it is still not a tightly coupled as the
railroad or the electric system. Advancements in ICT has transformed systems
management radically. In some systems, like electricity, ICT has opened up
the tight physical coupling between components. Auto-mobility and Air Flight
has on the other hand become more tightly coupled systems with the
introduction of ICT.

54
Changing Large Socio-Technical Systems

Tom Hughes perspective tries to solve the problem of combining the apparent,
and seemingly autonomous, system growth with the knowledge that systems
evolve, mature and decline, that they are, as any man-made institution,
subjected to contingent historical context. Socio-technical systems are
changed or being replaced by other competing systems. Gas light gave way to
electric light; road networks has changed dramatically to meet the needs of
mass motorization and today we are living in time when our global telephone
system is gradually replaced by mobile solutions.

In the LTS-perspective, system change is caused by forces, such as: new

technology, new or changed market conditions, environmental concern,
policy changes and institutional innovation or lack of primary products (like
oil). Factors that Eric Dahmén would label “transformation pressure” creating
“structural tension” (see Laestadius in this volume).

Figure 4: Illustration of a reverse salient and a salient

In Hughes' (1992) terminology, the change caused by transformation

pressure, takes the form of a “salient” or a “reverse salient” (terms similar to
Dahmén’s´ structural tension). These are metaphors borrowed from the
military, denoting a bulge that is ahead or behind of a front moving forward,
see Figure 4. The concepts are used to describe a system component or a sub-
system being ahead or lagging, compared to other components and given the
desired direction of the advance. Salients and reverse salients disturb the
system and slows down growth. They give impulses to innovate, but only if
they are recognized by key actors.

The fact, that system actors must recognize a reverse salient, is crucial when
managing a system in periods of transformation pressure. But identifying a
salient or a reverse salient can be difficult for actors deeply embedded in the
system.

55
A good example of a reverse salient in the electric system, that just recently
been identified by system actors, is the meter placed in our homes. Although
it is said to be “smart” it was originally designed to only measure electric
current as a cost for the consumer. If you produced your own electricity (by
wind or sun) it was difficult to sell surplus energy to the grid. The meter would
bill your input as a cost on your own tab. Note that the meter is a reverse
salient from the point of view of actors that would like to change the electric
system to allow for both download and upload. This was not included in the
traditional business plans of the large incumbent electricity utilities. The
meter can be identified as a reverse salient if you are a wind power actor
wanting to connect to the network. The meter can also, from the point of view
of other systems actors, be identified as a salient. Due to technical
development in other areas (ICT/Internet) the meter has become a more
advanced system component and thus pulling the system towards change.
This salient feature is the possibility of the meter to provide information on
energy consumption and user habits and its ability, when connected to the
internet, to serve as the central hub for controlling energy usage – that is to
make it possible for the consumer to time energy consumption to electricity
prices and thus become a central part in the vision of the “energy smart
home”. Another example of a present-day reverse salient is the battery
capacity in electric cars, which affects driving range. The battery is a lagging
component in the (electric) automobile system. But it is lagging only if the
range of Otto-engine cars is defined as the norm. Only if you define auto
mobility by the performance of the gasoline car the battery becomes the
reverse salient. But if you take global warming into the equation, it is in fact
the Otto-engine that becomes the lagging system component and the
battery/electric motor turns into a salient.

In the following we will concentrate on discussing the lagging components,

the reverse salients. From the arguments presented above its easy to see that
a reverse salient is not a given or natural feature of a system, it is not a dormant
failing component, like a ball bearing deteriorating and suddenly breaking.
Thus, a reverse salient is not possible to detect, using only perspectives from
inside the system and in line with the prevailing system logic. A reverse salient
is, to a certain extent, a social construction as its definition depends on the
point of view of different actors – a reverse salient is a fact in the eye of the
beholder.

The system characteristics mentioned above has primarily been used in

historical studies to analyse the growth of infra-systems and the stability of
mature systems. In the following we present our own development of Hughes
LTS-perspective focusing on system actors and change management in
infrasystems.

56
In the eye of the beholder …

In the management of infrasystems it is evident that history matters. And it

matters in two distinct ways. Firstly, when establishing a new system, the old
way of delivering the service, forces the newcomer to adjust. The old system
is, in a very concrete way, already there, and its physical infrastructure as well
as its organizational design influences any attempt to create a new or
complementing system. This feature is often summarized in the words of
Edison when he tried to exchange gas light with his new electric light systems.
To optimize the design of the system and to meet customer expectations and
established behaviour – to minimize costomer switching costs – a new system
builder had to “Dress electricity in a gas costume”.

Secondly, history and heritage is also evident within the existing system and
influences the way its actors are able to meet the need for transformation.
History creates both physical path dependence and path dependant behaviour
on behalf of the actors in the old system. The physical system components are
standardized and aligned with each other and system actors are used to deal
with system operations and business models in a certain way. Thus, every
established system has its own systems logic. When change is approaching, it
is not even certain that the incumbent system actors are able to identify the
way to handle the upcoming challenges. It is hard to think outside the systems
logic with its established architecture of standardized components, stable
hierarchies and successful business models and when trapped in a
conservative “systems culture”.

The fact that “history matters” leads us to suggest that the existing heritage of
the prevailing system must be taken into account when trying to adjust it to a
new situation. This holds true whether or not the plan is to construct a novel
system (or sub-system) or if the ambition is to expand the old system along
the lines of its particular systems logic.

Building on Hughes, we would like to differentiate between two types of

reverse salients, and thus the need to manage change, that must be handled
by actors within the infrasystem. Inspired by (Christensen 1997) we call them
sustaining respectively disruptive reverse salients. In the words of James
Utterback the conservative reverse salient is competence enhancing, given the
prevailing systems logic, and the disruptive reverses salient is (possibly)
competence destroying (Utterback 1994).

Firstly, system builders/managers have to deal with the need handle tensions
occurring in the daily operations of the system. System actors most often
handle these issues by “incremental innovation” and they usually focus on
defining the problems as a “load factor issue”. We call these day to day issues
sustaining reverse salients, and they are in often identified as simple and
obvious bottle necks in the management and maintenance of the system.

57
These bottle necks are easy to identify by the system actors. They have
established methods to deal with them and solutions follows the general logic
of the system. A sustaining reverse salient does not challenge the systems
culture nor does it threaten the business model of the system.

Secondly, system builders also need to address issues emanating from more
serious structural tensions. We label these issues disruptive reverse
salients. By this term we mean larger and more diffuse issues compared to
the simpler ones mentioned above. If these disruptive reverse salients stay
unidentified and if they are not handled, they can in some cases threaten the
very existence of the infrasystem: “When a reverse salient cannot be corrected
within the context of an existing system, the problem becomes a radical one,
the solution of which may bring a new and competing system.” (Hughes 1987,
p.75).

This differentiation between simpler sustaining and more complex disruptive

reverse salients was touched upon by Hughes in a similar way – but not made
explicit. Hughes discussed two types of innovations – conservative and radical
– that could be formulated as a solution to systems components lagging in one
way or another. Conservative innovations, corresponding to our term
sustaining reverse salients, are often produced internally within managing
organizations and thus creating improvements to the existing LTS. But
Hughes also highlighted that a radical innovation, corresponding to our usage
of the term disruptive reverse salient, often is created by external actors and
that the radical innovation may produce a competing system. The history of
Large Sociotechnical Systems shows that “…independent inventor-
entrepreneurs could be shown to specialize in identifying critical problems
and related 'reverse salients' on broad technological fronts.” (Joerges in
Mayntz and Hughes, 1988, p. 13-14).

Of course, the incumbent system actors can use another strategy to deal with
a disruptive reverse salient, in our expanded definition of the term, and that
is to view it as an “externality”. That is, to draw the border of the system in
such a way that the issues related to more serious transformation pressure is
placed outside their responsibilities. But what happens when system actors
must handle issues that they earlier saw as problems outside the realm of the
system? These are the issues facing for example transport systems based on
fossil fuels. Carbon dioxide emissions has been transformed from an
externality to become a disruptive reverse salient that must be managed by
system actors.

Tom Hughes stress the importance of identifying, in our wording, disruptive

reverse salients and trying to formulate a solution within the system logic:
“Defining a reverse salient as one or more critical problems is in itself a major
step toward a solution, or invention, for it is well known that the ability to

58
define an amorphous situation as a problem is often an anticipation of a
solution” (Hughes 1992, p.100).

But who has the right to define a disruptive reverse salient and point to a
solution? In our interpretation of this process we want to point to the privilege
of problem formulation. Which actors gains the privilege to formulate the
problem and the strategy to solve it?

One recent example is the transformation of the aviation system. Low cost
airlines such as Ryan Air defined luxury, luggage, comfort, etc., as disruptive
reverse salients and removed these components from the system. At the same
time the incumbent actors viewed all of these components as necessary and
well-functioning system parts, and were not willing to change their business
model. In fact, the low cost airlines managed to capture the privilege of
problem formulation and they redefined air travel from being associated with
comfort and leisure, into a simple and utilitarian commodity, such as a bus
ride.

The Multi-Level Perspective

Over the last years the work by Hughes has received increasing attention as
the threat of global environmental change has put increased focus on research
concerning societal and technological change. Hughes formulation of a
concrete analytical tool based on systems thinking is part of the theoretical
basis for the new research area called 'sustainability transitions'.

This research explicitly aims to analyse societal and technological change

towards increased sustainability. The challenges of achieving technological
and societal transitions at regional, national and international level are of such
complex nature that they require comprehensive systems approaches. One of
the currently most prominent systems based approaches in the 'sustainability
transitions' area is the Multi-Level Perspective (MLP).

The MLP approach was developed by Geels (2002) and Rip & Kemp (1998),
and has since been further refined to constitute an applicable cross-
disciplinary governance-oriented tool for steering societal change towards
increased sustainability (Geels 2011). MLP studies divides the studied
area/sector into three system levels (niche, regime and landscape) to explain
the dynamic change processes between the different levels causing
technological and societal transitions. The approach is illustrated in Figure 5,
and is described as following by Frank Geels (2014):

“In a nutshell, the core logic is that niche-innovations build up internal

momentum (through learning processes, price/performance

59
 improvements, and support from powerful groups); changes at the
landscape level create pressures on the regime; and destabilization of the
regime creates windows of opportunity for the diffusion of niche-
innovations. The alignment of these processes enables the breakthrough
of ‘green’ innovations in mainstream markets where they struggle with
the existing regime on multiple dimensions (economic, technical,
political, cultural, infrastructural).”

Figure 5: Illustration of the multi-level perspective, adapted from Geels (2002)

As previously stated, the ideas of Hughes are very much present in the MLP
framework. But there are some important differences to highlight. Both the
regime-concept used in MLP studies and the LTS-concept represents socio-
technical systems. But where LTS represents a (relatively) tightly coupled
physically interconnected technical systems (with social components) the
regime-concept constitutes a non-physical structure that accounts for the
stability of socio-technical systems. A regime does not necessarily represent a
infra-based system (as LTS does) but is a representation of interconnected
political, economic, cultural, social, institutional and technological elements
(Geels 2004). The MLP approach also has a wider range of applicability, (cf.
Arvidsson in this volume). The socio-technical regime also accounts for a
selection-retention mechanism for different innovations (see the introductory
chapter in this book for more about variation-selection-retention
mechanisms).

60
The other difference between LTS and MLP is the explicit division of the
studied area into three system levels to analyse the dynamics. Coevolution of
actors and institutions take place at different system levels and is influenced
by variables that change over time (Lewin & Volberda 1999), and the MLP
framework provides a structured approach to deal with these multi-level co-
evolutionary processes.

In the following part we will discuss each level individually and then provide
some examples of MLP analyses to describe the interactions between the
relatively stable socio-technical regime and the “flurry of change activities” at
niche level which takes place in the regime's environment represented by the
landscape (Geels 2012).

Niches

As explained by Geels, innovations and novelties are often developed in

niches, which are protected environments where the innovation can mature
and be tested before being included in the system. The niche level is derived
from the idea that radical novelties (innovations) are developed in niches that
enable cross-disciplinary experimentation. The niches enable specified
interactions between issues and actors. Well-developed niches are suggested
to act as building blocks for change; they are central for regime shifts.

An example of a strategic niche is a military research project which is not

exposed to the selection-retention characteristics of the market. In The
Entrepreneurial state the author Mariana Mazzucato (2013) brings up a
string of examples of important technological innovations in the iPhone –
Apple’s flagship product – that have been developed in government-protected
niches. These includes innovations such as the internet, the GPS, the touch-
screen display and the voice-activated ‘Siri’. Another niche, in the form of a so
called protected market space, is exemplified in the steam ship example below.

Niches are often subsidised to support non-profitable innovations with the

expectation of future societal benefits (Schot & Geels 2008), as the examples
from Mazzucato shows. However, niche innovation may be misaligned with
existing technological infrastructure, regulations, user behaviour, etc. Factors
in the surrounding system are crucial for an uptake of innovations, and large-
scale changes (or “regime shifts”) (Schot & Geels 2008).

Socio-technical landscape

Geels (2002) argues that radical innovations developed in niches can make a
break through when the stability of socio-technical regimes are confronted
with problems and tensions and the links between the elements within the
regime are “loosened up”. This can for example happen through external

61
pressure from landscape level, i.e. the level that constitutes the environment
of the socio-technical regime.

As explained by Schot and Geels (2007) problems as well as opportunities can

emanate from factors stemming from the landscape level, which are not
controlled by actors on the local (niche) level nor at regime level. Changes in
the landscape can influence the system development in ways difficult to
discover for the actors involved in its daily management (Blomkvist & Larsson
2013).

The sociotechnical landscape is equivalent to what other system theorists

often refer to as the system’s environment. In Hughes (1987) words the system
environment is the societal context of the system. The sociotechnical
landscape is according to Schot and Geels (2007) characterised by the set of
rules that guide technical design, the rules that shape market development
and rules for regulating these markets. The socio-technical landscape includes
the institutional and market aspects required in order for lower levels to
function.

Socio-technical regime

The term “technological regime” was first introduced by Nelson & Winter
(1982) to represent a set of shared beliefs among engineers and to describe
trajectories of innovative processes in industrial sectors.

Nelson & Winter explained the concept of technological regime using the
airplane Douglas DC-3 as an example. When the DC-3 was introduced it had
new features, such as all-metal skin, a powerful engine (a DC-3 could tow up
to three transport gliders at once), a low wing and streamlined body, etc. These
features influenced the whole industry as a basis of how an airplane should be
designed. In other words, the technological regime of the DC-3 shaped the
development trajectories and delimitations of future technological change in
aeronautics (Nelson & Winter 1982).

The discussion concerning the social construction of technology (SCOT)

during the 1980s, where the work by Hughes was central helped further
develop the concept of technological regimes to include social actors such as
researchers, policy makers, users and associations as well (Bijker 1995; Geels
& Schot 2007). The term 'technological regime' was expanded and got the new
epithet 'socio-technical regime', which strengthened the perception that
regimes (technical or socio-technical) stabilized current development trends
in several ways (Geels & Schot 2007).

According to Geels (2004) a socio-technical regime consists of three

interlinked dimensions: (a) the network of actors and social groups, (b)
formal, normative and cognitive rules and (c) physical and technical elements.

62
Examples of formal rules are building regulations and laws, examples of
normative rules are norms of behaviour, and examples of cognitive rules are
guiding principles, how a problem formulation is made, belief systems and
heuristics (Verbong & Geels 2007).

A socio-technical regime affects development paths both for technicians and

companies and can be compared with the concept innovation system – see
chapter by Rickne & Laestadius in this book – which includes elements such
as actors in the value chain, networks, institutions (which here means rules,
norms, cognition) and technology (Jacobsson & Bergek 2011).

Examples of socio-technical regime reconfigurations

To illustrate the interaction between regime, niche and landscape level Frank
Geels (2002) uses the example of steamships. Steamships were a radical
innovation that made a breakthrough in a reconfiguration process of socio-
technical regimes that happened due to landscape pressures.

Steamships began to appear in the early 19th century when sail ships
dominated the overseas transport regime. Steamships had reduced cargo
space compared to sail ships and were driven on expensive coal, but had the
advantage of being able to provide line services with both a fixed departure
and arrival time, which the wind reliant sail ships could not. Steamships were
therefore mostly used for personal and mail transport in the beginning, as well
as for high-value low-volume cargo, where speed and regularity were
appreciated.

At the time the socio-technical regime for over-seas transport was dominated
by sail ships: the ports were made for sail-ships, the ship-builders only knew
how to make sail ships, there was no infrastructure for steamships to reload
with coal, the insurance companies did not want to insure the uncertain
steamships, etc. This created stability for the sail ships while hindering the
breakthrough of steamships.

Also, at the time, the performance of steamships was not that good. The initial
steamships had paddlewheels and were made of wood. They were therefore
most suitable to traffic inland waterways. At open sea, the paddlewheel did
not stay in touch with the water all the time and the heavy steam engine
machinery made the wooden hull bend. Experiments with making iron hulls
in some cases resulted in steamships turning upside down once outside of its
dock.

In 1838 however, the British government subsidised steamships to improve

the speed of communication on special routes important to the British
Empire. These subsidies created a “protective environment” for developing
oceanic steamships. The results were increased experimentations with iron

63
hulls, screw propellers to replace the paddlewheels, more effective engines,
etc. Thus, accumulating innovations within a niche.

However, these early oceanic steamships were in the beginning a complement

to sail ships. They were in fact often sail ships with a steam engine as an
“auxiliary add-on”. Early steamship advancements therefore rather
strengthened than reconfigured the sail ship technological regime, in a hybrid
state.

In the 1850s and 60s two events on landscape level gave steamships an
unforeseen advantage. At the time, most passengers preferred the quicker
steamships compared to the time-uncertain sail ships. The emigrant-wave
from Europe to America became the first major market segment dominated
by steamships. This emigrant wave was driven on by famine and political
revolution in Europe, as well as the Californian gold rush. Events that were
clearly outside of what the actors in shipping and overseas transport could
affect.

Another important landscape event was the opening of the Suez-canal in 1869.
The canal was unsuitable for sail ships and therefore gave a great advantage
to steamships on the route Europe to India and China.

By the late 19th century steamships finally overtook sail ships as the
dominating alternative for overseas transportations. Note that the shift from
sail ships to steamships took almost a full century and was a stepwise process
of reconfiguration. Partly because the sail ships, as a defence mechanism,
adopted several radical innovations such as iron hulls (to increase cargo
space), on-board machinery (to reduce need of manpower), etc., which
increased the performance capacity compared to steamships.

So, the success of steamships was not solely dependent on technological

advancements of the steamships themselves, but dependent on a set of
developments at multiple levels that would link up and allow for steamships
to break out of its subsidised niche. In Geels own words: “Breakthroughs of
innovations thus depend on processes on the level of regimes and landscapes,
i.e. they are context-dependent. It is because of this aspect that the multi-level
perspective is useful for analysing technological transitions.”

Another more modern example of how stable socio-technical regimes can

change is seen in the Swedish residential heating sector and the shift from oil
boilers to the heat pump technology.

In the early 1970’s oil boilers dominated the Swedish residential heating
regime for small houses. The oil crises in 1973 and 1979, causing a quadrupled
price on oil, provided strong incentives for the Swedish government to reduce
the dependence on imported oil. The Swedish government launched a series

64
of subsidy programs to develop heating alternatives such as solar technology
and heat pumps. The state-owned power utility Vattenfall was assigned to be
an active part of the technological development. Vattenfall quickly turned
their attention towards heat pumps instead of solar energy, as the heat pump
fit better into Vattenfall's electricity system. In a few years over a hundred
companies joined the new and seemingly lucrative business of producing,
selling and installing heat pumps to house owners. Many of these companies
received vital support from Vattenfall. In this niche the heat pump technology
was developed to match the Swedish climatological and housing conditions
(Swedish houses are typically fitted with water-based heating systems).
However, the existing heating regime was not aligned towards heat pumps.
Problems occurred of both technical and social character, e.g. problems with
house mould due to less natural ventilation without the warm oil boilers and
lack of education of installers which caused substandard installations. As
these problems were gradually mitigated and the performance of heat pumps
increased the heating regime for small houses in Sweden stepwise
reconfigured towards heat pumps instead during the 1990’s. Today there are
more heat from heat pumps per capita produced in Sweden than any other
country. (Blomkvist & Johansson, 2014)

Multi-Level System Diagnosis: Combining LTS and MLP

In this last section of our chapter we would like to introduce a framework used
when investigating and analysing change processes in infrastructural systems.
The framework focus on the system actors point of view and on finding reverse
salients. It is primarily based on Hughes LTS-perspective and, in early stages,
developed independently of the work of Frank Geels. The framework has since
then been strongly affected and inspired by Geels and MLP.23

The framework divides the LTS under investigation into three levels and the
purpose is to locate reverse salients on these different levels. These reverse
salient, we argue, causes misalignment between system components and
slows down system expansion. Following from the discussion above, we argue
that a functioning large socio-technological system needs alignment between
different components within the system and alignment with other systems
and institutions in society. The purpose of the diagnosis is thus to identify

23 Rudimentalin Blomkvist (2004; 2006) and further developed in

Blomkvist & Larsson (2013). The framework is published in Blomkvist &
Johansson (2014).

65
misalignment between systems components or sub-systems. Each level has a
couple of general research questions to guide the investigation:

• Local alignment: How do we organize and manage the LTS at the local
level? Are there reverse salients causing misalignment on the local level
that could cause the system to slow down its development? Is there
misalignment between local system components and the next level of
the LTS – the sociotechnical regime?
• Sociotechnical regime alignment: How do we organize and manage the
LTS at the sociotechnical regime level? Are there reverse salients
causing misalignment on the regime level that could cause the system
to slow down its development?
• Landscape alignment: How do we align the whole LTS (local and regime
level) with neighbouring institutions and systems in society? Are there
reverse salients causing misalignment on the landscape level that could
cause the system to slow down its development? Problems that threaten
the development of a LTS can arise not only from within the system
(local and regime level) but also externally. Conflicts can stem from
collision with other societal institutions, rules, organizations, or
systems. Hughes (1987; 1992) uses a similar distinction when
discussing the delimitations of a large technical system, which he calls
environment. Thus, our third analytical level will focus on friction
between the two first levels in the LTS and its environment. But a note
of caution is appropriate before moving on. The “landscape” of a LTS
can be many things. Grand societal changes like industrialization,
urbanization, and modernization are concepts that can be used to
contextualize changes on the landscape level. However, in our third
analytical level our purpose is not to address these types of general and
abstract concepts. We want to keep “landscape” on a fairly concrete
level. When talking about “landscape misalignment” we mean external
factors causing direct friction when interacting with the LTS under
investigation.

In the following we will give an example of how to use the diagnostic

framework on a specific case: wind power establishment in the existing
electrical system (in this example we do not differentiate between sustaining
and disruptive reverse salients). We use local, sociotechnical regime and
landscape alignment to analyse the ongoing wind power establishment in
Sweden of today (based on Blomkvist & Sandberg 2011).

66
Wind Power and Local Alignment

Actors in the wind power industry experience problems in getting acceptance

for wind power projects in the local community in rural areas. In many
instances the problems originate from lack of experience and routines and a
lack of incentives on the local level.

Firstly, one important reverse salient is that inhabitants in the rural areas has
no direct experience in working together in large and capital intensive
projects. Knowledge and economic capability is a scarce resource. There is no
established arena where discussions can take place and the local community
(i.e. the village) has no clear and accepted institutional framework to handle
these complex questions and no legitimate procedure to arrive at a decision.
In short: the village is not a well-defined actor.

Secondly, there are no clear economic incentives to support the project if you
are not the actual landowner. The windmill affects the living environment for
many, but just a few gain economically, i.e. positive and negative externalities
are distributed unevenly. On big obstacle is the “Catchment area of wind
power”. When building a wind power-plant you inevitably reduce the
possibilities to build recreational facilities, houses or other windmills in the
area. The plant creates “wind shadow” for other windmills and reduces other
landowner’s ability to exploit their land. Actors in the industry try hard to
innovate models for partnership, sharing of profits and cooperative
ownership.

One way to solve these reverse salients and improve local alignment is to
create an organization – a distinct actor – similar to the already existing road
associations managing local roads. It is probably both economically and
socially profitable to establish local institutions for cooperation in wind power
projects. Local alignment appears when all sorts of economic and social
externalities, revenues and costs, are shared in a fair way. Local alignment can
also reduce “NIMBY-sentiments” in the community (Devine-Wright 2009).

Wind Power and Socio-Technical Regime Alignment

Local misalignment is not the only problem of wind power establishment.

Wind power mills are small compared to the centrally placed distribution
facilities in the existing electrical system and they do not always fit in the
distributive network managed by the electricity utilities.

Firstly, there is number of technical reverse salients causing miss-alignment

between the old system end the newcomer. The Swedish electricity system has
been designed as a distribution system from the centre to the periphery. Wind
powered electricity is small scale with decentralized production plants. A new

67
grid design needs to be implemented to allow for input of electricity into the
grid from windmills (i.e. “smart grid”).

Secondly, there are economic reverse salients in the difficulties of billing, due
to the contingent nature of the wind. A windmill owner can sometimes deliver
electricity to the grid and needs to get paid for this contribution. At other
times, when there is no wind, the owner must be able to use electricity from
the grid and consequently pay for that service.

Thirdly, institutions, regulations, organizations and markets are designed

exclusively for a distribution system with large central production plants. It is
hard to meet rules on security and reliability for smaller producers. And
furthermore, the dominating players, de facto an oligopoly, has no interest in
letting new players enter the market.

Concerning the socio-technical reverse salients and miss-alignment our

advice, derived from historical example, would again be to learn from the road
sector. During most of its modern history, the local roads have been an
important issue for the state and for the state road administration. Special
technical personnel were given the task to monitor and advice local road
associations. The state road administration, with its primary mandate to
manage public roads, established regional offices where designated road
engineers and legal experts handled questions of socio-technical alignment.
We believe that the wind power sector would benefit from these types of
arrangements.

Wind Power and Landscape Alignment

One surprising miss-alignment appeared when the communal income tax law
from 1928 was rediscovered, and unexpectedly, became a reverse salient in
the wind power system. It was originally aimed at farmers using surplus
goods, e.g. milk, meat, and eggs, from their business to feed themselves and
their families. The tax law postulated a stereotyped tribute for every farmer
adding the value of the benefits in kind to the yearly income. By a decision by
the central tax authority in 2008 the same rules were to be imposed on
collective ownership of wind power. Members of cooperative societies owning
shares in a windmill had to pay tribute if they were able to produce their own
electricity at a lower cost than the market price at Nordpool spot market. The
direct result was a 90% decrease in joint owned windmill projects.

It is of course possible to identify reverse salient causing landscape miss-

alignment regarding laws and customs from other areas, i.e. public right of
common (allemansrätten) animal and nature protection, cultural heritage,
etc. There are also many organizations or groups of people outside the local or
sociotechnical regime level that interpret wind power as a threat to their
interest, i.e. bird and nature organizations, recreational and sport

68
organizations and groups setting out to protect the rural landscape. To avoid
landscape miss-alignment it is important to investigate and diagnose the
environment, in a wide sense, of the wind power project.

Managerial and policy implications

Systems thinking has a wide range of applicability and the importance of
systems thinking seems to increase every year. Many of the ongoing challenges
related to globalisation and climate change involves structural problems that
demand holistic approaches where the wellness of “the whole” – and not only
the individual parts – is in focus.

Systems thinking is thus important for both policy makers and managers. A
problem that relates to complex interactions of different social and/or
technical elements can never be fixed by focusing on only one of the elements:
all parts of the system must be addressed in order to find a fulfilling solution.
The usefulness of system thinking is that it is applicable to both small and
large systems – it can be equally useful to use for the project manager at the
small company as well as for the politician engaged in bilateral discussions on
a nation's energy supply.

The LTS and MLP concepts provides concrete tools for analysing changes and
transitions in large socio-technical systems. By understanding the concept of
dynamic stability of socio-technical systems the manager and politician better
understand the challenges for engaging in transition activities. It also provides
insights to the importance of multi-level engagement. It is not enough with a
new energy-efficient technology developed in a niche if the stability of the
regime blocks the technology from achieving widespread use – alignmet is
crucial.

On methodological considerations and the use of

sources in this chapter
This chapter is mainly based on a literature review. But we have also included
exemplifications from our own research.

The scientific foundation on general systems theory is presented in Anders

Karlqvist i Teknik och Samhälle. En systemanalytisk introduktion, Tema T
Rapport 4, 1983, Universitetet i Linköping. We have also used Hughes, Agatha
C. and Thomas P. Hughes (eds.) Systems, Experts and Computers. The
System Approach in Management and Engineering, World War II and After
The MIT press, 2000 and the chapter by Arne Kaijser and Joar Tiberg in this

69
book: “From operations research to future studies: the establishment,
diffusion and transformation of the system approach in Sweden, 1945-1980”.

The general review of the LTS perspective is to large extent based on Jane
Summerton's “Stora tekniska system. En introduktion till forskningsfältet” in
Den konstruerade världen. Tekniska system i historiskt perspektiv
(Blomkvist & Kaijser 1998), Jane Summerton's “Introductory Essay. The
Systems Approach to Technological Change” in Jane Summerton (1994)
Changing Large Technical Systems, Arne Kaijser, I Fädrens spår...”Den
svenska infrastrukturens historiska utveckling och framtida utmaningar
Stockholm 1994 and Pär Blomkvist (2001) Den goda vägens vänner. Väg- och
billobbyn och framväxten av det svenska bilsamhället 1914–1959, Symposion
2001 (diss.)

The MLP perspective has generated a lot of research in the last decade. We
would like to point at one of the most influential articles that can be said to
have initiated this field: Geels, F.W., 2002. “Technological transitions as
evolutionary reconfiguration processes: a multi-level perspective and a case-
study.” Research Policy.

Suggested further readings

• The Social Construction of Technological Systems: In this seminal book

a discussion, together with several important insights and ideas about
technology development, is captured. In this book the concepts of Large
technical systems (LTS), Social Construction of Technology (SCOT) and
the Actor-Network Theory (ANT) are presented (Bijker et al. 1987).
Concepts that still are of relevance for both managers and researchers
still.
• Hughes, T.P., 1992. “The dynamics of technological change: Salients,
critical problems, and industrial revolutions.” In G. Dosi, R. Gianetti, &
P. Angelo Tonineili, eds. Technology and Enterprise in a Historical
Perspectiv. Oxford: Clarendon Press.
• Hughes, T.P., 1987. “The Evolution of Large Technical Systems.” In W.
E. Bijker, T. P. Hughes, & T. Pinch, eds. The Social Construction of
Technological Systems: New Directions in the Sociology and History
of Technology. Cambridge, MA: MIT Press.
• Hughes, Agatha C. and Thomas P. Hughes (eds.) Systems, Experts and
Computers. The System Approach in Management and Engineering,
World War II and After The MIT press, 2000.

70
 • Geels, F.W., 2002. Technological transitions as evolutionary
reconfiguration processes: a multi-level perspective and a case-study.
Research Policy, 31(8-9), pp.1257–1274. Available at:
[Link]
• Geels, F.W., 2011. The multi-level perspective on sustainability
transitions: Responses to seven criticisms. Environmental Innovation
and Societal Transitions, 1(1), pp.24–40. Available at:
[Link]
[Accessed April 29, 2014].
• Geels, F.W. & Schot, J., 2007. Typology of sociotechnical transition
pathways. Research Policy, 36(3), pp.399–417. Available at:
[Link]

References
Bale, C.S.E., Varga, L. & Foxon, T.J., 2015. Energy and complexity: New ways
forward. Applied Energy, 138, pp.150–159. Available at:
[Link]

Von Bertalanffy, L., 1968. General System Theory: Foundations,

Development, Applications, New York: Geroge Braziller, Inc.

Bijker, W.E., 1995. Of Bicycles, Bakelites, and Bulbs, Cambridge, MA.

London, England: The MIT Press.

Bijker, W.E., Hughes, T.P. & Pinch, T. eds., 1987. The Social Construction of
Technological Systems – New Directions in the Sociology and History of
Technology Anniversar., Cambridge, MA: MIT Press.

Blomkvist, P., 2001. Den goda vägens vänner. Väg- och billobbyn och
framväxten av det svenska bilsamhället 1914-1959, Symposion (diss.)

Blomkvist, P., 2004. “Transfering Technology – Shaping Ideology. American

Traffic Engineering, Experts and Commercial Interests in Establishment
of a Swedish, and European, Car Society in the Post War Period.”
Comparative Technology Transfer and Society.

Blomkvist, P., 2006. “Roads for peace – roads for flow. Lobbying for a
European Highway System.” In A. Kaijser & E. van der Vleuten, eds.
Netwoking Europe. Transnational Infrastructures and the Shaping of
Europe, 1850-2000. Sagamore Beach MA, USA: Science History
Publications.

71
Blomkvist, P. & Johansson, P., 2014. Introducing a socio-technical analysis
framework; Analysing the Swedish heat pump development in three steps.
In Energy Systems Conference. London.

Blomkvist, P. & Kaijser, A., 1998. Den konstruerade världen. Tekniska

system i historiskt perspektiv, Stockholm.

Blomkvist, P. & Larsson, J., 2013. An Analytical Framework for Common-

Pool Resource–Large Technical System (CPR-LTS) Constellations.
International Journal of the Commons, 7(1), pp.113–139.

Blomkvist, P. & Sandberg, T., 2011. Vindkraft och samfällighetsförvaltning.

ASPECT.

Christensen, C., 1997. The Innovator’s Dilemma: When New Technologies

Cause Great Firms to Fail, Harvard Business School Press.

Churchman, W., 1968. The Systems Approach, New York: Dell Publishing
Co.

Devine-Wright, P., 2009. Rethinking NIMBYism: The role of place

attachment and place identity in explaining place-protective action.
Journal of Community & Applied Social Psychology, 19(6), pp.426–441.

Geels, F.W., 2012. A socio-technical analysis of low-carbon transitions:

introducing the multi-level perspective into transport studies. Journal of
Transport Geography, 24, pp.471–482. Available at:
[Link]

Geels, F.W., 2004. From sectoral systems of innovation to socio-technical

systems. Research Policy, 33(6-7), pp.897–920. Available at:
[Link]
[Accessed March 19, 2014].

Geels, F.W., 2014. Regime Resistance against Low-Carbon Transitions:

Introducing Politics and Power into the Multi-Level Perspective. Theory,
Culture & Society, 31(5), pp.21–40. Available at:
[Link]

Geels, F.W., 2002. Technological transitions as evolutionary reconfiguration

processes: a multi-level perspective and a case-study. Research Policy,
31(8-9), pp.1257–1274. Available at:
[Link]

Geels, F.W., 2011. The multi-level perspective on sustainability transitions:

Responses to seven criticisms. Environmental Innovation and Societal
Transitions, 1(1), pp.24–40.

72
Geels, F.W. & Schot, J., 2007. Typology of sociotechnical transition
pathways. Research Policy, 36(3), pp.399–417. Available at:
[Link]
[Accessed May 23, 2014].

Hughes, A.C. & Hughes, T.P. eds., 2000. Systems, Experts and Computers
The System Approach in Management and Engineering, World War II
and After, The MIT Press.

Hughes, T.P., 1983. Networks of Power: Electrification in Western Society,

1880-1930, Baltimore: Johns Hopkins University Press.

Hughes, T.P., 1992. The dynamics of technological change: Salients, critical

problems, and industrial revolutions. In G. Dosi, R. Gianetti, & P. Angelo
Tonineili, eds. Technology and Enterprise in a Historical Perspectiv.
Oxford: Clarendon Press.

Hughes, T.P., 1987. The Evolution of Large Technical Systems. In W. E.

Bijker, T. P. Hughes, & T. Pinch, eds. The Social Construction of
Technological Systems: New Directions in the Sociology and History of
Technology. Cambridge, MA: MIT Press.

Jacobsson, S. & Bergek, A., 2011. Innovation system analyses and

sustainability transitions: Contributions and suggestions for research.
Environmental Innovation and Societal Transitions, 1(1), pp.41–57.

Kaijser, A., 1999. Den hjälpsamma handen”: Om den institutionella

utformningen av svenska infrasystem. Historisk tidsskrift, 119, pp.397–
417.

Kaijser, A., 1994. I fädrens spår : den svenska infrastrukturens historiska

utveckling och framtida utmaninga, Stockholm.

Kaijser, A., 2002. Systembuilding from below: Institutional change in Dutch

water control systems. Technology and Culture, 43(3), pp.521–548.

Lewin, A. & Volberda, H., 1999. Prolegomena on for Coevolution : A

Framework for Research on Strategy and New Organizational Forms.
Organization science, 10(5), pp.519–534. Available at:
[Link]

MacKenzie, D., 1990. Inventing Accuracy: A Historical Sociology of Nuclear

Missile Guidance,

Mazzucato, M., 2013. The Entrepreneurial State: Debunking Public vs.

Private Sector Myths, Anthem Press.

73
Meadows, D., Randers, J. & Medows, D., 1972. Limits to Growth,

Nelson, R.R. & Winter, S.G., 1982. An Evolutionary Theory of Economic

Change, Cambridge, MA: The Belknap Press of Harvard University Pres.

Nilsson, D., 2011. Pipes, Progress, and Poverty: Social and Technological
Change in Urban Water Provision in Kenya and Uganda 1895–2010.
KTH Royal Institute of Technology.

Rip, A. & Kemp, R., 1998. Technological change. In Human Choice and
Climate Change – Resources and Technology. Columbus: Battelle Press,
pp. 327–400.

Schot, J. & Geels, F.W., 2007. Niches in evolutionary theories of technical

change. Journal of Evolutionary Economics, 17(5), pp.605–622.

Schot, J. & Geels, F.W., 2008. Strategic niche management and sustainable
innovation journeys: theory, findings, research agenda, and policy.
Technology Analysis & Strategic Management, 20(5), pp.537–554.

Summerton, J., 1994. Changing Large Technical Systems, Oxford: Westview

Press.

Summerton, J., 1998. Large technical systems. An introduction to the

research field. In R. Maynts & T. P. Hughes, eds. Large Technical
Systems: Concepts and Issues. Boulder, Colorado.

Utterback, J.M., 1994. Mastering the Dynamics of Innovation, Harvard

Business School Press.

Verbong, G. & Geels, F., 2007. The ongoing energy transition: Lessons from
a socio-technical, multi-level analysis of the Dutch electricity system
(1960–2004). Energy Policy, 35(2), pp.1025–1037.

Wiener, N., 1948. Cybernetics: Or Control and Communication in the

Animal and the Machine, MIT Press.

74
This is Industrial Dynamics
Pär Blomkvist, Petter Johansson & Staffan Laestadius

The chapters in this book present a set of analytical Perspectives on

Industrial Dynamics (ID). In this introductory chapter, we give a
background on how the subject area is interpreted at our department
(SID) at INDEK, KTH. We also discuss how our field has evolved by
presenting the three most influential origins of Industrial Dynamics: Neo-
classical economics, Innovation theory and Evolutionary economics.

We start by a quote from the first chapter, that may serve as a motivation for
the book project and for the usefulness of the research area of Industrial
Dynamics:

“This is a period when the world economy has to transform

fundamentally in order to rapidly reduce the impact of modern
industrial activities and life styles on the climate as well as to adapt to
the climate change on the way. To a large extent that transformation has
to take place within and between dominating firms and industries which
during two hundred years of development have become highly
dependent on fossil fuels. This transformation will – and must – impact
on technologies and industrial capabilities and it will necessitate
management activities as well as policy interventions.” (Laestadius in
this volume)

The great challenge ahead for industrial scholars is finding the tools for
analysing and managing technology and industrial transformation in climate
change. Many projects in the industrial dynamics group at INDEK are related
to climate change: The automotive industry is transforming but the electric
car is facing a “chasm”. New conditions are created for sustainable energy
business. The wind power industry is transforming: in industrial structure, in
production costs, in acceptance, in development of new forms of ownership.
The transformation of pulp & paper to bio refineries is partly driven by the
need to substitute for oil. The diffusion of photovoltaics (the most expensive
form of sustainable energy) is also – in many of its applications – the most
suitable for developing regions; how to manage that?

These are projects on the meso level – i.e. on the level of

technological/innovation systems and industrial structures. The areas under
investigation are relevant for strategic management decisions, industrial and
technology policy and for understanding transformation to sustainability and
green growth: Thus, Industrial dynamics, to a large extent, relates to strategic

13
management and policy making on a meso level. The theoretical base of our
research, as a branch of the discipline industrial economics and management,
has strong relations to a variety of other disciplines such as economics,
business administration, economic history and history of science &
technology, industrial sociology and economic geography. We use quantitative
as well as qualitative research methods.

Innovation is the creation of new combinations, and processes of problem-

solving activities constitute the building blocks of innovative creativity. Thus,
innovation is a cognitive learning process, irreversible and path dependent.
From our research perspective we emphasize that management of knowledge
production in its systemic context is a complex process that need much more
of integrated research connecting knowledge, learning and innovation. When
studying technological trajectories and industrial transformation, we bring
science and institutional change into the theory of innovation.

Industrial dynamics is thus a truly multi-disciplinary area of knowledge. At

KTH it is characterized by a strong technology focus topic: in research and
teaching of industrial phenomena, entrepreneurship and their implications on
management and policy our ambition is to integrate that with in depth
analyses on the role of technology and the mechanisms and knowledge
processes behind technological change. Below we discuss the origin and
heritage of industrial dynamics from three interdependent perspectives.

The heritage from Economics

Industrial dynamics has one of its origins in the microeconomic theory within
the tradition of neoclassical economics. Or rather in the shortcomings of
orthodox economics with its strong focus on markets, on prices and on perfect
conditions for equilibrium. What is assumed to constitute the perfectly
competitive economy – i.e. homogenous products, perfect information, many
actors, no externalities, no economies of scale, etc. – are not only rare
situations but also conditions which economic actors strive to avoid. What is
often classified as market failures in the textbooks in microeconomics is in
fact the normal landscape where industrial and technical actors have to
navigate. Economic development does not evolve from everybody producing
and selling the same thing but from the creativity taking place behind the
markets in structures, like firms and universities, which are more complex
than just transmitting market signals.

Before we return to those perspectives in the sections on Innovation theory

and Evolutionary economics below, we turn to prof. Bo Carlsson – one of the
early introducers of the Industrial Dynamics tradition – and who´s
perspective on these issues is as follows:

14
 “Industrial Dynamics (ID) is a new and rapidly growing field of
research. Its theoretical roots are similar to those of Industrial
Organization (IO), but the questions addressed are different. IO is
based on equilibrium (static or comparative static) analysis; there is no
causal analysis. In ID the emphasis is on dynamics: seeking to identify
and understand the reasons why things are as they are. ID focuses on
the causes (driving forces) of economic transformation and growth, and
on understanding the underlying processes of transformation, not just
the outcomes. The transformation is viewed in its wider historical,
institutional, technological, social, political, and geographic context.
This means that the analysis often has to transcend disciplinary
boundaries and involve multiple dimensions and levels. Economic
growth can be described at the macro level, but it can never be explained
at that level […] Economic transformation is a matter of experimental
creation of a variety of technologies that are confronted with potential
buyers in dynamic markets and hierarchies. Economic growth results
from the interaction of a variety of actors who create and use
technology, including demanding customers. The interaction takes
place in an evolving institutional setting.”1

According Bo Carlsson there are five broad themes that constitute the basic
questions in industrial dynamics:
1. The causes of industrial development and economic growth, including
the dynamics and evolution of industries and the role of
entrepreneurship
2. The nature of economic activity in the firm and the dynamics of
supply, particularly the role of knowledge.
3. How the boundaries and interdependence of firms change over time
and contribute to economic transformation.
4. Technological change and its institutional framework, especially
systems of innovation.
5. The role of public policy in facilitating adjustment of the economy to
changing circumstances at both micro and macro levels.

Research and teaching at the division of sustainability and industrial

dynamics (SID) at KTH touch upon all these five themes. Many SID
publications, however, have a strong focus on the role of knowledge and
technological activities in systems, industries and firms. To analyse these
knowledge formation processes in technology, the technological trajectories
emerging and their contribution to changing conditions for economic and

1 BoCarlsson “Industrial Dynamics: A Review of the Literature 1990-2009”, Industry and

Innovation, Vol 23, Issue 1, pp. 1-61.

15
social actors it is necessary to go beyond the understanding of prices and
markets. This is also what Industrial Dynamics is about.

The heritage from Innovation theory

Innovation theory is central to the subject area. In short, innovation theory
has its origin in the tradition emerging from the economist Joseph
Schumpeter's research around the two world wars of the 20th century. It was
Schumpeter who introduced the concepts of “innovation” and “entrepreneur”
as he was the first to point out the need for venture capital2. In the beginning
of his research the entrepreneurial problem was in focus; later he studied
whether the economy, which over time became more dominated by large
corporations could preserve its creativity and capacity for innovation3.

The modern innovation research – with roots in the 1970s and 1980s – has
largely inherited those old research areas. Not the least Christopher Freemans
The Economics of Industrial Innovation (1974)4 provided inspiration to many
innovation researchers. The management aspects of innovation – which also
are core issues in the research and teaching at SID/KTH, have their origin in
writings of Burns and Stalker (1961)5 although James Utterback´s popular
Mastering the Dynamics of Innovation (1994) 6 , synthesizing much of
innovation research from the late 20th century, may be looked upon as an
important milestone in the field.

Significant parts of the management oriented innovation research have

focused on identifying and classifying the variety of innovations as regards
their technology and function in the industrial system in order to understand
their challenges to management and policy. Among all papers published in
this area three are frequently quoted:

In The Patterns of Industrial Innovation (1978) William Abernathy & James

Utterback develop the distinction between radical and incremental
innovations as well as product and process innovations and analyse how they
change in character when industries/firms mature.7

2 Schumpeter, J., (1911/1934) The Theory of Economic Development, Cambr., Mass: Harvard Univ.
Pr.
3 Schumpeter, J., (1943/2000) Capitalism, Socialism and Democracy, London & New York:

Routledge.
4 Freeman, C., (1974) The Economics of Industrial Innovation, Harmondsworth: Penguin.
5 Burns, T. & Stalker, G.M., (1961) The Management of Innovation, London: Tavistock.
6 Utterback, J. (1994) Mastering the Dynamics of Innovation, Boston, Mass., Harvard Business

School Pr.
7 Abernathy, W & Utterback, J., (1978), ”The Patterns of Industrial Innovation”, Technology

Review, Junw/July, pp 41-47.

16
A second important contribution in this field, can be found in Henderson and
Clark´s distinction between Modular and Architectural innovation. If a
company changes the architecture of a product without changing the core
design an architectural innovation is created. This kind of innovation is often
triggered by a small modification of a component. Although the alteration of
the architecture may not seem as a large alteration it can have a massive
impact on the company and result in a leap not unlike the case of radical
innovation. One example is the ceiling fan, if a company decides to take the
next step and introduce small portable fans it would be an architectural
innovation as it involves assembling the components differently. The new
product is different from the previous and the acceptance by existing
customers might be hard to predict. This new product could be so different
from previous products that it requires different sales channels and
distribution, possibly attracting new customers. The danger lies in not
realizing that an architectural innovation is in fact not incremental.8

Clayton Christensen´s The Innovator´s Dilemma (1997) provides a third

contribution to this research area with his dichotomy between sustaining and
disruptive innovations.9 In short sustaining innovations are those that fall in
line with existing capabilities and thought styles of a firm and can, thus, even
if being radical, be handled by the existing structure of management and
engineering. Disruptive innovations, on the other hand, are those that fall
outside the fundamental skills and experiences of a firm making it more or
less impossible for the firm to adapt to the new technology.

The original Schumpeterian concept of innovation covered all creative

combinations that were introduced on the market or in the
industrial/economic system. Innovations could thus be organizational as well
as technical, related to markets as well as institutions etc. However, for a long
time, innovation research has been dominated by a more or less explicit focus
on classical mechanic industry. This is now changing.

IKEA's combination of logistics, design and production with customer

involvement is thus entirely in Schumpeter's spirit as well as its development
of new business concepts. Creativity, design and new business models in the
digital economy of today also contribute with new aspects of innovative
behaviour.

8 Henderson, R. M. & K. B. Clark, (1990) “Architectural Innovation: The Reconfiguration of Existing

Product Technologies and the Failure of Established Firms”, Administrative Science Quarterly 35:1
(1990), 9–30.
8 See e.g. the editors’ introductory chapter in Albert de la Bruhèze, Adri A. & Oldenziel, Ruth (eds.)

(2009), Manufacturing technology, manufacturing consumers: The making of Dutch consumer

society, Amsterdam: Aksant, 2009.
9 Christensen, C. (1997) The innovator´s dilemma – when new technologies cause great firms to

fail, Boston, Mass: Harvard business School Pr.

17
This “design discussion” also reveals that innovations are not identical with,
or necessarily the result of, science. And innovation theory cannot limit its
analytical scope to R&D processes in universities and industrial research labs.
The role of science in innovation processes is an open question and a research
task for the innovation analyst. Some industries are more R&D based than
others. The pulp & paper industry e.g. which has been studied frequently at
our division, is historically an industry with low R&D intensity (see Novotny
2016 – this volume). For the ICT industry, of course, the situation is different
(see Long 2016 – this volume, on ICT dynamics).

The analyses of innovation processes have a significant impact on our

understanding of firms. Firms that learn to innovate can develop capabilities
to avoid the price trap of competitive equilibrium which dominates neo-
classical analysis. The new theory of the firm, which focuses on how
companies may allocate and develop their resources and create a climate
contributing to dynamic capabilities is a key starting point in modern
enterprise-based analysis of industrial change. The companies' differences as
regards capabilities create variety when it comes to competitiveness: also
seemingly low-tech businesses can creatively build up competencies that
become competitive in global scale. This is basically what successful
entrepreneurship and management of innovation and technology is about.

This leads us to the context or embeddedness of industrial and technical

change. This embeddedness can be understood – and analysed – from a
territorial perspective. We may call this geographical or physical proximity.
But context can also be analysed from a cognitive perspective, i.e. proximity
of thought styles. This may include engineering communities and cultures
more or less related to various industries and technologies and not necessarily
localized to a specific region. Embeddedness is also a question of culture,
institutions and physical infrastructure. There are several research traditions
in this area with strong family resemblances. Some are focused on clusters,
others on innovation systems: national and regional, among others (see
Laestadius and Rickne in this volume).

At SID/KTH the focus is also on technological systems and their

transformation, a transformation where the incentives sometimes come out of
necessities, sometimes out of opportunities. Especially in these times of
globalization and the emergence of India and China as industrial giants. How
to establish and maintain innovative activities and production at the global
level? There has also been a focus on development blocks, a Schumpeterian
inspired concept developed by Erik Dahmén (See Laestadius in this volume).
Neither of these concepts are necessarily restricted to certain territories. The
embeddedness, relations and forces may well be characterized by non-
territorial proximity.

18
The heritage from Evolutionary Economics
Industrial Dynamics also has its roots in Evolutionary Economics. Although
this tradition of economic theory has its origin in the analyses of Alfred
Marshall, Thorstein Veblen and, not the least, Joseph Schumpeter the modern
approach to non-equilibrium and evolutionary processes of industrial change
came with Nelson & Winter´s An evolutionary theory of economic change
(1982).10

In short, adopting an evolutionary approach in the analysis makes it possible

to get rid of most of the conditions framing the conditions for competitive
equilibrium and to focus on paths of industrial and technical transformations.
The heritage from evolutionary theory also means that history matters. In
Industrial dynamics the time line is very important. We argue that earlier
choices by actors and historically acquired inertia (sunk costs, technical
standards, already built infrastructure, etc.) strongly influences the
possibilities for present actors to engage in transformation of technology,
technological systems and industry. This focus on historical (and institutional
heritage) is mirrored in the use of concepts like these: learning, structural
tensions, co-evolution, technological paradigms/trajectories/regimes, path
dependence, product cycles.

Fundamental concepts in this analytical approach, and imported into

economic theory by Nelson & Winter are: Variation, Selection environment,
Adaption, Retention.

“Variation is the creation of a novel technical or institutional form within

a population under investigation. Selection occurs principally through
competition among the alternative novel forms that exists, and actors in
the environment select those forms which optimize or are best suited to
the resource base of an environmental niche. Retention involves the
forces (including inertia and persistence) that perpetuate and maintain
certain technical and institutional forms that were selected in the past”11

In a case study on the dynamic behind the introduction of the moped in

Sweden, these evolutionary inspired concepts were used like this:

“In our case the niche is the space between the bicycle and the motorcycle
as defined by the government in 1952 – the perceived need for a bicycle
with a help motor. The variation is all the different moped types

10 Nelson, R. & Winter, S. (1982) An Evolutionary Theory of Economic Change, Cambr., Mass &
London: The Belknap Press of Harvard University Press.
11 Van de Ven, Andrew H. & Ragu Garud (1994), “The Coevolution of Technical and Institutional
events in the Development of an Innovation”, in Baum Joel A.C. and Jitendra V. Sing (eds.),
Evolutionary Dynamics of Organizations (Oxford: Oxford U.P. 1994)

19
 launched at the market by various producers and firms. Selection takes
place on the market performed by different user groups (consumers)
showing their preferences choosing among the marketed moped types –
thus establishing the eight stable moped families. The selection process
is also influenced by the regulations and requirements put on the moped
by the government during different phases of our history – i.e. the
moped laws of 1952 and 1961. Retention is shown on the level of
technical design in the stable moped families and on an institutional
level by the persistence of the rule from 1952 that the mopeds had to have
pedals like a bicycle – a rule that persisted a long time after the moment
when technical design had made the pedals obsolete.”12

Inspiration from evolutionary theory (and Biology) can also be applied to the
sorting out of various technical artefacts by constructing what is called a
“Phylogenetical Tree” by biologists and paleontologists. It is used when
investigating family relations between different species. This method of
sorting species or artefacts into family groups has been used by W. Bernard
Carlson when analysing Edison’s sketches on the telephone. He and his
associates defined every sketch as a “fossil”. By sorting the chronologically and
looking for family resemblance they found both mechanical similarities and
family ties on a more structural level:

“…devices that shared an underlying principle of operation, a common

mental mode or meme. As we found telephones sharing an underlying
principle of operation, we began to group them chronologically in
horizontal rows on a map, letting them constitute different lines of
[Edison’s] research”13

Today´s evolutionary approaches to industrial and technical change may be

looked upon as a third wave of evolutionary theory within social science. The
first wave of biologism, emerged in late 19th century and disappeared around
the period of WW1. That wave was by many labelled Social Darwinism
although its primus motor was the sociologically interested engineer Herbert
Spencer rather than Charles Darwin and had a strong influence on
engineering communities and theories on industrial organization. This was
also a period when evolutionary economics was founded by economists like

12 Blomkvist, P. & Emanuel, M. (2009) Från nyttofordon till frihetsmaskin. Teknisk och
institutionell samevolution kring mopeden i Sverige 1952–75 (From Utility to Freedom: The Co-
evolution of Technology and Institutions in the History of the Swedish Moped 1952–75), Division
of Industrial Dynamics, Royal Institute of Technology, Stockholm (Stockholm 2009), TRITA-IEO
2009/16
13 Carlson,W. Bernard (1988) “Invention and evolution: the case of Edison’s sketches of the
telephone”, in Ziman, John (ed.), Technological Innovation as an Evolutionary Process
(Cambridge: Cambridge U.P., 2000), 150. The most influential study on analyzing technical
artefacts as “fossils” can be found in: Basalla, G., (1988): The Evolution of Technology, Cambr.
Univ. Press

20
Alfred Marshall (industrial dynamics), Joseph Schumpeter (entrepreneurial/
innovation theory) and Thorstein Veblen (institutional theory). Of these
Marshall struggled with a strong biological influence.

The shortcomings of the early and narrow biological analogies in social science
– before the evolutionary theory had matured and settled also among
biologists – may have contributed to the difficulties for evolutionary
perspectives to challenge the dominating equilibrium paradigm in economic
theory in general but also in other social sciences. Although there are
exceptions, it may be argued that the evolutionary approaches introduced by
Schumpeter before WW1 as a foundation for innovation/entrepreneurship
theory (the second wave) did not take off in theoretical development until well
after WW2. And, as mentioned, this take off came with Nelson & Winter who
showed the usefulness of general evolutionary thinking and analogies
deliberated from 19th century biological overtones.

There was also – around WW2 – an important Swedish track of institutional

and structural analyses with a clear evolutionary perspective; Erik Dahmén,
among others, may be argued to belong to that tradition with his structural
theory of industrial and technical transformation based on the development
block concept, (Laestadius in this volume).

Today´s research on industrial and technical transformation (development,

innovation, entrepreneurship and management – also organizational theory,)
is thus based on a significant foundation of theories, partly conflicting, but to
a large extent related to an evolutionary perspective which implicitly or
explicitly challenge the equilibrium dominance in standard economics and
social theory. We argue that researchers – and policy makers – with ambition
to understand processes of industrial and technical transformation – and how
to manage them – also must develop an understanding on the foundation of
these concepts and theories.

Not the least is this important in a period like this when mankind is facing
challenges to fundamentally transform the economy away from its addiction
to carbon and towards sustainable life styles and means of production.
Reducing GHG emissions with approx. 90% of the present will be necessary
within a few decades. Old development paths have to be abandoned. New
technological trajectories must be entered upon. New innovative models for
business, for transport, for resource efficiency and for organizing the economy
and society must be developed. This is probably the most far reaching process
of industrial dynamics and transformation ever seen.

21
 Eric Dahmén and Industrial Dynamics
Staffan Laestadius

Industrial and technical change takes place in a cognitive space and a field
of forces where imbalances result in tensions that create incentives for
change. Uneven development of an industrial transformation process
may cause necessities as well as opportunities for actors in the system.
This is the foundation for the theory of development blocks (DB:s) as
developed by the Swedish economist Erik Dahmén. His analysis, inspired
by Joseph Schumpeter, may be used to analyse the transformation of an
economy – and its industrial and technical structure – on a meso level and
without any assumptions of equilibrium which are common in orthodox
economic theory.

The notion of development block catches the essential relations in an

industrial/technological system but is not necessarily related to a certain
industry (as defined in public statistics) or technology. The actors and
forces of a DB may change over time as new technologies emerge and old
ones fade away. This chapter not only introduces the DB concept and its
theoretical embeddedness in detail to make it useful as a tool for analysis.
It also illustrates how the DB concept can be used for managerial and
policy purposes in the great transformations ahead for our society.

Introduction
This is a period when the world economy has to transform fundamentally in
order to rapidly reduce the impact of modern industrial activities and life
styles on the climate as well as to adapt to the climate change on the way. To
a large extent that transformation has to take place within and between
dominating firms and industries which during two hundred years of
development have become highly dependent on fossil fuels. This
transformation will – and must – impact on technologies and industrial
capabilities and it will necessitate management activities as well as policy
interventions.

To contribute to this change process, we need a theoretical understanding on

how these systems fit together, how they develop, and on the mechanisms of
change. There is the market of course, but industrial transformation is much
more than just prices. This chapter introduces a classic Swedish approach to
the study of industrial and technical transformation – the development block

22
analyses by Erik Dahmén. The basic argument – which will be revealed in the
text that follows – is that his approach still contributes with sharp and useful
tools for policy makers as well as for industrialists and
entrepreneurs/innovators for the understanding and change of our industry.

The idea behind this paper is that a qualified analysis cannot be performed by
just grasping a set of anonymous tools from the tool box. The advanced analyst
is also aware of the potential and shortcomings of the tools used as well as why
the tools were developed. Knowing this sharpens the analysis and reduces the
risk of walking in the wrong direction.

The chapter is organised as follows. Section two contains an analysis of an in

depth reading of Dahmén´s Ph.D. dissertation from 1950. Section three
focuses on the development block concept and how the Dahménian analysis
developed after his dissertation.14 In section four I discuss Dahmén's impact
(and sometimes lack of impact) on the intensified research in industrial and
technical change which has taken place in the 1980s and the following
decades. Section five illustrates how the Dahménian concepts can be used to
analyse the industrial transformation and creation of new development blocks
that may take place during the decades ahead. This section thus illustrates the
most important managerial and policy implications of the Dahménian
approach.

Erik Dahmén’s dissertation from 1950

Erik Dahmén´s Ph.D. thesis – Svensk Industriell Företagarverksamhet
(Swedish Industrial Entrepreneurial Activity) – from 1950 is probably the first
empirical study ever based on a Schumpeterian analytical framework. It is
fully related to the mechanisms behind economic – or rather industrial and
technical – transformation. In that process Dahmén also introduces a set of
concepts which still are highly relevant for research on industrial and
technical change. These concepts may be looked upon as important parts of
the theory of industrial dynamics; in short: the analytical tools that connect
evolutionary economic theory with industrial and technical analysis.

The dissertation is written in a period when the academic discourse among

economists basically followed two tracks: on the one hand the main stream of
analyses was since long focused on the characteristics of and conditions for
equilibrium, on the other there was a strong focus on highly aggregated
models that (against the background of the crises in the 1930s) focused on
economic cycles. Not the least the second discourse (what Dahmén labelled

14 Thereare some other texts on the Dahménian approach. See eg. Dahmén (1991); Lindgren (1996);
Pålsson-Syll (1997) and Karlsson (2007). The intention here is to have a stronger focus on the
industrial-technical dimensions than is the case with those texts.

23
 konjunkturteori – cycle theory) dominated Swedish economic analysis: even
if John Maynard Keynes (Keynes, 1936/70) is the internationally most well-
known representative for this track, this line of thinking also had strong roots
among Swedish economists and within the Swedish political system (Berman,
2006).15

Dahmén positions his study in relation to business cycle theory which, he

argues, in spite of advanced econometric methods does not get close to the
core issues of economic development. He notes that new products, new
methods, new forms of organization, and new markets – i.e. those innovations
Schumpeter (1934) lists in his Theory of Economic Development – can
destroy the old ones and still not be detected in the aggregates which cycle
theorists analyse. Behind a vague up- or downturn there can be a significant
structural transformation the tendencies of which business cycle analysis,
with its focus on aggregate units like savings, consumption, wage level etc.,
run the risk of neglecting: “in such an analysis it has instead been most
convenient to handle the transformation as datum” i.e. as something given –
and not to explain (Dahmén, 1950, p. 5).

It may be argued that also John Maynard Keynes was well aware of these
limitations in his General Theory. His discussion of “effective demand” is
related to a “given situation of technique, resources and factor costs” (Keynes,
1936/1970, p. 24) and he also, in another of his works, makes a reference to
Schumpeter as regards development in the long run, i.e. when technologies
and industries are transformed or changes through innovative investments
(Keynes, 1930/1960, p. 95f).

Dahmén also positions his analysis in relation to research in economic history,

which, on the one hand has had historical transformation processes as one of
its primary research objects but on the other, from various reasons, had not
developed theoretical tools suitable for understanding the transformation
mechanisms

As a consequence, following Dahmén, a gap developed between the

theoretically advanced economists, who did not grasp or were not interested
in the mechanisms of transformation, on the one hand, and the economic
historians, on the other, who were interested in the processes of
transformation but had not developed methods to analyse the causalities in
these processes. That is the gap Dahmén intends to fill through analysing
factual historical processes based on explicit, theoretical problems, i.e. what
he labels a causal analysis. Although this analysis can be performed
independently from the study of business cycles, Dahmén argues that the

15 For the use of sources and references in this chapter, see appendix A.

24
deeper understanding of transformation processes can contribute to business
cycle analysis (Dahmén, 1950, p. 6f).

Three thought traditions or discourses on economic development may be

argued to constitute the foundation for Dahmén´s dissertation. The first has
its origin in Thorstein Veblen, whom in a critical way analyses the lack of
evolutionary perspectives in economic theory, in the essay “Why is economics
not an evolutionary science?” (1898) and in his book The Theory of Business
Enterprise (1904). While the first essay is more general in nature, Veblen is in
his book from 1904 more focused on the fact that entrepreneurial and
industrial activities in themselves are lacking in economic theory. That was in
fact what became the core issue in Schumpeter's dissertation from
(1911/1934).

The second intellectual track for the Dahménian analysis is found in the
research by his Ph.D. supervisor Johan Åkerman (also influenced by Veblen).
Dahmén refers primarily to the more superficial text Ekonomiskt
framåtskridande och ekonomiska kriser (1931) and the more theoretically
developed Ekonomisk teori 1 & 2(1939 & 1944) where Åkerman develops the
distinction between “alternative analysis” and “causal analysis” (in Åkerman,
1944). While the former relates to the selection/choice between alternatives
which economic actors are assumed to do all the time in an ahistorical context,
the second concept relates to a historical process where time is important and
the set of alternatives changes over time.

The third pillar in Dahmén´s analysis is the works by Schumpeter. In fact,

Åkerman had put Schumpeter on the reading list for Dahmén in his Ph.D.
work. If Åkerman contributes with the general background for the analysis
Dahmén admits that Schumpeter´s text from 1911 contributes with the
specific ideas. Not the least, Dahmén argues, this is the case with the “business
aspect” put forward in Schumpeter (1911) where he makes a clear distinction
between the technological dimension and the business dimension thus
opening for “industrial activity” and “new combinations of production
factors”. He also mentions Schumpeter´s skepticism towards aggregated
analyses (Dahmén, 1950, p. 8f)

Dahmén describes the Schumpeterian analysis as a three-step process: first

the comparison of the stationary and circular flow, secondly the disturbances
in the stationary flow due to the innovations introduced by entrepreneurs and
thirdly the adaption processes taking place following from the new
combinations introduced in the system and which may cause a crowding out
of old technologies and businesses.

He discusses the fact that the Schumpeterian approach can be used in the
analysis of business cycles – which is what Schumpeter does in his Business
Cycles (Schumpeter, 1939) – but that the most important implication of

25
 Schumpeter´s work is that he does not restrict himself to the aggregated
analysis of most economic theorists but to the fundamental character of the
transformation process: primarily, according to Dahmén, on the micro
character of the process, i.e. in the transformation of companies, of
technologies and industries (Dahmén, 1950, p. 10f).

Dahmén did not include Schumpeter´s Capitalism, Socialism and

Democracy (Schumpeter, 1943) in his Ph.D. thesis work. As a consequence
the Schumpeterian metamorphoses in the view on the character of the
innovation process and the dynamics of capitalism prior to WW2 (often
labelled “Mk1” and “Mk2”) is not mentioned and still less analysed by him.16

After two chapters of detailed descriptive texts of the economic development

before World War 1 and the interwar period, Dahmén, in his fourth chapter,
introduces some of his concepts and his research questions. A core issue is
that his causal analysis is related to “a non-reversible process and not to the
time-less cause-consequence relation” (Dahmén, 1950, p. 45). Basically this is
what Åkerman wrote in his Ekonomisk kausalitet (Åkerman, 1936) and
further developed in Ekonomisk teori 2 (Åkerman, 1944).

The double character of the transformation process is discussed in detail: on

the one hand the positive fact that innovations are introduced, on the other
the negative fact that the old is destroyed: technologies become obsolete and
firms are closed down. Basically this is what Schumpeter (1943, chpt VII)
labels “creative destruction” but that book is not referred to by Dahmén in his
analysis. Methodological innovations (i.e. what we nowadays normally label
process innovations) are analysed in relation to the standard concepts in
micro economic theory: the production function and the isoquant. The choice
between factor inputs following variations in factor prices but within the
framework of a known technology is interpreted as a movement along the
isoquant. Innovations, i.e. new methods, must – although it can sometimes be
difficult to make a clear distinction – following Dahmén, be looked upon as a
shift of the isoquant, i.e. a movement towards origo in the normal presentation
of the production function as illustrated in fig. 1 (Dahmén, 1950, p. 46 ff).

16 In
short: the further we moved into the 20th century it became obvious that the capitalist economy
was transforming away from an “entrepreneurial capitalism” (MK1) into what may be labelled
”corporate capitalism”; (MK2) i.e. became dominated by monopoly like structures created through
processes of ”creative destruction” combined with mergers and acquisitions. The Schumpeterian
(1943) problem thus became how these great corporations with their monopoly power and
bureacracies could uphold their entrepreneurial spirits.

26
Dahmén is clear that this discussion is far from unproblematic. The isoquant
is (in standard theory) normally assumed to be known making it possible for
firms to locate themselves along it depending on relative factor costs. In cases
when the isoquant is not known this Dahménian distinction is not applicable.
This problem – whether the isoquant is or can be known and/or can be
smoothly followed – has been observed in many critical analyses of the
production function (see eg. Rosenberg, 1976).

The analysis of the interwar period gave Dahmén a solid empirical ground for
his distinction between advancing, stagnating and disappearing industries.
For the advancing he makes a distinction between what he calls market
suction and market expansion, where the former is connected to demand
mechanisms outside the industry itself (i.e. what today often is labelled
“demand pull”) and the latter is basically driven by internal mechanisms like
methodological (process) innovations, within the industry, i.e. what we may

27
 label “technology push”. As regards stagnating and disappearing industries
Dahmén argues that what is interesting from his analytical point of view are
those industrial transformations which take place due to new processes and
products, not due to cyclical phenomena or simple “malinvestments”
(Dahmén, 1950, p. 49ff). Following the clustering in time of several advancing
industries/technologies is, following Dahmén, what drives business cycles.
This is very close to the arguments of Schumpeter (1943) and is also one of
Dahmén’s main arguments against the shortcomings of aggregate analyses in
the understanding of these cycles.

According to Dahmén, the transformation analysis is an argument also for

institutional analyses in relation to the process, i.e. studies of the economic
and political structure and the social structure from a sociological perspective
(like educational level, income and power relations etc.). It is also a motive for
analyses of the industry structure (ibid. p. 52 ff). This is in fact an argument
for what his supervisor Johan Åkerman label “structural analysis”. Here
Dahmén also formulated his main research question: to analyse the interwar
process of business formation, business development and closures in Sweden.
His ambition is to put flesh on the bones for the theoretical reflections
(assumptions) provided by Alfred Marshall and Joseph Schumpeter (ibid. p.
55f)

Although Dahmén already in chapter 4 (ibid. p. 52) identifies the tensions that
may take place due to imbalances in the innovation process that aspect is not
developed further in the theoretical chapters 1 and 4. These problems, and the
concept development block (in this text frequently labelled DB) are introduced
for the first time in chapter 5 more or less ad hoc in the discussion on the
unbalanced nature of economic development. 17 Based on the very detailed
analysis performed by Åkerman on the first half century of Asea (Åkerman,
1933) Dahmén concludes that Asea solved its balance problems in promoting
subsidiary companies to engage in electrification of Swedish industry thus
creating an industry in need of large power systems: “first through completing
a full electrical “development block” did they manage to successfully create an
electric industry” (Dahmén, 1950, p. 66ff).

This is also the chapter where the incentives for the transformation of
development blocks, what Dahmén label “structural tensions”, are analytically
developed in the dissertation. Dahmén is clear that these tensions appear –
and can be studied – on at least two analytical levels. The first level is related
to company, industry or institutional level: structural tensions can in this
perspective be analysed in terms of over production, malinvestments and
cultural and market related inertia. This may create difficulties for innovative
companies to establish themselves. This level also includes the bottle necks

17 Both varieties of spelling – ”bloc” and ”block” – are used in this chapter.

28
which may occur due to limits set by communication structure and/or limited
local markets (ibid. p. 68ff).

Secondly Dahmén relates his structural tensions to technological and

organizational development. Even if it, following Dahmén, can be difficult to
make a clear distinction between the economic/institutional level and the
technological/organizational one he is of the opinion that we – on this level –
face the challenges of completing the development block through bringing
technology and organization into fitness between each other. These later
balancing problems, which Dahmén found important, has earned remarkably
little attention within economic theory although they probably are as
important than the former as regards incentives for economic change (ibid. p.
70ff See also chpt 3 in this volume).

This later variety of problems reveal themselves industrially in production

chains which, as a consequence of technical change somewhere in the system
become unbalanced. Dahmén illustrates with steel manufacturing, a
technology which changed rapidly during the 19th century and thus created
structural tensions on many levels within Swedish industrialization: “ingot
steel processes could sometimes not easily be introduced because the blast
furnace technology was not developed in parity to produce enough of cheap
pig iron to feed the Bessemer converters or Martin-Owens” (ibid. p. 71).

In summary Dahmén is of the opinion that development blocks have a strong

cumulative importance, not the least when companies learn to think in blocks
themselves and also to finance whole blocks: this, he argues, created the
prerequisites for a cumulative expansion of industry, often with high profits.
In short, this may be interpreted as Dahmén’s more empirically based
contribution to the analysis of economic development once introduced by
Joseph Schumpeter (ibid., p. 74ff).

The development of the development block theory

The transformation analysis and structural tensions were core elements in the
thought style of Erik Dahmén. These conceptual elements were also
something he struggled with already in the late 1930s. In his licentiate
dissertation the concepts seems to have been integrated to a coherent
conceptual world which also included the concept “development block”. 18
There were also the “positive” and “negative” aspects of the development
process, i.e. the double edged phenomenon Schumpeter (1943) labelled
“creative destruction”.

18 The details of this have not been checked in Dahmén (1942), however.

29
 It is far from obvious how concepts develop in the history of ideas and when
or by whom they finally are introduced. That is the case here. It cannot be
excluded that the term “development block” has its origin in Johan Åkerman
or in discussions between the supervisor and the Ph.D. candidate (cf. Pålsson-
Syll, 1997).

The structural tensions, causing uneven development between and within

sectors (and technologies) may be identified as a corner stone in the
development block theory. In fact, this non-equilibrium process, related to the
innovation sequences, is found also in Schumpeter (1939):
“Industrial change is never harmonious advance with all elements of the system
actually moving, or tending to move, in step. At any given time, some industries move
on, other stay behind; and the discrepancies arising from this are an essential element
in the situations that develop. Progress – in the industrial as well as in any other sector
of social life – not only proceeds by jerks and rushes but also by one-sided rushes
productive of consequences other than those which would ensure in the case of
coordinated rushes…. We must recognize that evolution is lopsided, discontinuous,
disharmonious by nature – that disharmony is the very modus operandi of the factors
of progress ... Evolution is a disturbance of existing structures and more like a series
of explosions than a gentle, though incessant, transformation.” (p 101f).

Dahmén early connected his development block analysis to the ex ante – ex

post discourse which developed among Swedish economists during the 1940s.
This distinction, or dichotomy, can be used in several ways: the most common
is to – like Gunnar Myrdal – use it to describe the selection set economic
actors are facing and the expectation based decisions they make ex ante (in
advance) in the beginning of a period on the one hand and what occurs ex post
(afterwards) when the real effects of all actors´ decisions have worked trough
the economy, and eventually oscillated to a new equilibrium. Dahmén’s
approach was somewhat different, however. He connected his ex ante concept
to the industrial entrepreneurs which identified advancing industrial
development blocks and thus proactively through their investments
contributed to their growth. The ex post concept he connected to the reactive
actions based on actors reading of price signals which reflect emerging
unbalances and, which in their searching for profitable niches, contribute to
fill in empty holes (Dahmén, 1991, p. 140).

Erik Dahmén needed seven years to finish his dissertation after the licentiate.
To some extent this was a consequence of his large empirical work, partly
performed during his years at the Swedish Industrial Research Institute, IUI.
It is, however, his conceptual development rather than his empirical work that
has become used by later analysts of industrial and technical transformation.
That conceptual world did also become more developed and more strictly
formulated in the later, although short and few, papers which he published.

30
Two of these papers deserve some comments here (both available in Dahmén,
1991, p. 126-148).

The paper on “Schumpeterian Dynamics” (ibid. p. 126-135) is interesting from

two aspects: the first because Dahmén synthesizes his inspiration from
Schumpeter four decades earlier, and secondly due to its very strong focus on
transformation and “dynamics” rather than growth:

“Schumpeterian dynamics is characterized primarily by its focus on economic

transformation rather than on economic growth, defined as an increase in “national
product”, “capital stock” and other related aggregates. It contrasts not only with
Walrasian macroeconomic equilibrium theory but also with neoclassical and
postkeynesian macroeconomic growth models. Though “dynamic” according to
generally accepted terminology such models do not analyse underlying processes at
the micro level and in markets but instead relations between a number of broad
aggregates and the result of such processes …” (p. 127)

Analysing the consequences and mechanisms of Scumpeterian innovations

and their role in “creative destruction” Dahmén continues that

“… transformation thus includes both economic growth and decline but a conceptual
distinction is instrumental. This is because transformation analyses focus on causal
chains outside the scope of growth analyses, namely on disequilibria and chain effects
created inter alia by entrepreneurial activities, market processes and competition as a
dynamic force. The micro underpinnings of such analyses therefore differ from those
of growth models where the main interest is in aggregates, such as investment and
saving, productivity, income distribution … Seen through Schumpeterian glasses, the
micro units have no well-defined generalizable “propensities”, and they are not fully
informed calculators reacting in a mechanical way to prices they cannot influence.
Instead, firms continuously seek new information and often search for projects which,
if carried out, exert transformation pressure on the markets.” (p. 128)

In the paper titled “Development Blocs in Industrial Economics” (ibid. p. 136-

148) transformation appears as a historical process of sequential movements
between possibilities (or opportunities) and necessities. It is defined as
sequences of complementarities which through series of structural tensions,
i.e. situations of non-equilibrium, may result in a continuous transformation
rather than a final equilibrium. Dahmén also makes a distinction between the,
according to him, static concept competitiveness and the, dynamic, concept
transformation power. While the former relates to the conventional price
competition which prevails in equilibrium theory the later relates to the
transformation potential inherent in the development blocs. Basically these
concepts, close to those of Schumpeter (1911/34), argue that the essential
competition is not based on prices and costs but on innovation and
transformation.

32
It may be argued that Dahmén is of the opinion that the elimination of
structural tensions, i.e. the filling of the development blocks, may result in
situations of balance and equilibrium. But nothing in the rest of his texts on
this matter demands or even implies equilibrium. It is thus, in analogue with
global weather, possible to imagine a system of continuous disturbances,
which all the time tend to fill out existing disequilibria/tensions occurring in
the system but which during these processes continuously recreates new
imbalances on micro level. In fact, equilibrium theory – to which Schumpeter
devoted his first chapter in his Economic Development (1911/1934) – plays a
very limited role in Dahmén´s analyses.

In addition, it may be argued that the industrialists and entrepreneurs, which

are in focus in the Dahménian analysis, are not restricted to the restrictive
assumptions in the core theory of economics. They may make wrong
investments, they may have tunnel vision, and they may be wrong in their
market forecasts and selection of technologies. These very human and non-
perfect actions contribute to the structural tensions and create incentives for
other actors in the economy.

Already in his licentiate Dahmén criticizes Schumpeter that he – in spite of his

evolutionary approach – still is too stuck in his equilibrium thinking to draw
the full consequences of his view on the essential nature of industrial activity
in the capitalist economy. Also here, in analysing the dynamics of
industrialism, Dahmén is well in line with his supervisor (see eg. Åkerman,
1932, p. 49).

The role of development blocks in modern theory on

economic, industrial and technical change.
New ideas and concepts often develop in several varieties and in many
different contexts more or less independent from each other. That is also the
case with the development block concept which has “family resemblance” with
other analytic approaches. One is the metaphor model based on the concepts,
salients, reverse salients and critical problems, introduced by Tom Hughes
(1992; see also chpt 3 in this volume). In the Hughes model the salient –
originally a fortification term indicating a (fortified) position ahead of the
frontier – is a technological component which has developed faster than other
components in the system and thus represents an advancement along the
development frontier. The salient thus indicates a disequilibrium, a critical
problem, providing opportunities for carriers of the technology/system ahead
to exploit their position. The reverse salient in the Hughes terminology is a
system component that may have faced strong resistance and is lagging
behind the frontier. Analogously, the reverse salient creates incentives for the
carriers of the lagging system to catch up.

33
These reactions among the carriers of technology on salients and reverse
salients do not necessarily end up in equilibrium. Solving the bottle neck
created by a reverse salient may well end up in an over shoot, a salient, thus
contributing to sequential equilibrium which may follow a certain direction,
path or trajectory.

The historian of technology Tom Hughes analyses technical systems rather

than industrial ones but that distinction is not always easy to keep clear. Both
Dahmén and Hughes illustrate their technical arguments with the well-known
imbalances between the spinning and Weaving industry in the eve of the
industrial revolution. And, as mentioned, Dahmén explicitly argues that the
structural tensions on the technological- organizational level probably are the
most important and also those most neglected by economists. Here Hughes
and Dahmén are very close.

Analysing the similarities, and differences between Hughes and Dahmén takes
us to at least two problems worth more in depth discussion. The first is the
systems approach, the second relates to the actors –
entrepreneurs/innovators/industrialists – which drive the system forwards.

First, and maybe most important, is the fact that the transformation process
in both approaches is analysed – and can only be understood – at a systems
level which is less aggregated than the whole economy. Such an approach is
natural for historians of technology as Tom Hughes but, as revealed in section
two above – far from obvious for economists. We are not here talking on
“micro level” which in orthodox economic theory relates to a level when the
unit of analysis is “firm” or “household”, but a level between the standard
aggregates in economic theory.

In his analyses Hughes works with a systems concept – “technological

systems” – incorporating the relevant system components which cooperate to
the functioning of the system. In short Hughes argues that his technological
systems are socially constructed artefacts and that “inventors, industrial
scientists, engineers, financiers, and workers are components of but not
artefacts in the system” (Hughes, 1997).

In the early 1990s innovation researchers inspired by Dahmén (and

independent from, Tom Hughes) developed a similar systems approach with
elements from Dahmén’s structural analysis and with a clear technological
profile. In the paper “On the nature, function and composition of
technological systems” Bo Carlsson and Richard Stankiewicz (1991), against a
background of emerging neo Schumpeterian innovation research, make an
attempt to identify technological system as “a network of agents interacting in
a specific economic/industrial area under a particular institutional
infrastructure or set of infrastructures and involved in the generation,
diffusion, and utilization of technology. Technological systems are defined in

34
terms of knowledge/competence flows rather than flows of ordinary goods and
services…. In the presence of an entrepreneur and sufficient critical mass, such
networks may be transformed into development blocks, i.e. synergistic
clusters of firms and technologies within an industry or a group of industries.”
(See chpt 4 in this volume).

The technological systems concept was further developed almost a decade

later by Bo Carlsson (Carlsson, 2000). The conceptual development must be
understood against the conceptual congestion which has emerged within what
may be labelled the innovation discourse: national innovation systems,
regional innovation systems, sectoral innovation systems, competence bocks,
clusters, industrial district and various network approaches all claim to have
relevance for the understanding of this field of research. As mentioned, these
systems approaches are further discussed in chapter 4 of this volume. Here I
restrict myself to the general reflection that it is far from obvious that these
innovation systems approaches give a fair understanding of Dahmén. As
mentioned in section 2 above, Dahmén already in his dissertation identified
two analytical dimensions in his development block concept: one more
industrialist related and one more technological/organizational. The
technological dimension is more clear and important in Dahmén’s own texts
than is the case in much of the innovation systems approaches.

Let us so turn to the second dimension of similarities and differences between

Dahmén and Hughes: the role of the industrialist/entrepreneur/innovator.
For Hughes, system building is the important process, and the innovators play
the main roles. Although Hughes does not use the term himself, we can
conclude that Hughes´s innovator heroes are entrepreneurial, they innovate
and their systems also develop into profitable companies/industries (cf.
Hughes, 1989). Hughes´s actors are labelled and identified as system builders
(cf. chpt 3 in this volume).

Here we can identify a difference as regards the unit of analysis between

Hughes and Dahmén. Although the industrialists are present in the analysis –
not only in the title of his magnum opus – the analytical focus is on the
structure and system rather than on the role of the individual in the creation
of the system. Here, and somewhat paradoxically, it may be argued that
Hughes is more close to the young Schumpeter (1911) than what Dahmén is.
But this difference should not be overestimated: while Schumpeter´s heroes
are more close to launching the innovations on the market, Hughes´s are more
close to developing them technologically.

Industrial analyses in our time have a strong spatial or territorial focus. Not
the least is that reflected in the innovation systems traditions discussed in
chapter 4 in this volume.

35
It is maybe too easy to connect Dahmén´s works to some of these tracks. Not
the least the Scandinavian innovation system researchers have normally read
Dahmén or at least heard about his analytical approach. There are, however,
strong reasons to stay clear in the analytical comparisons. Dahmén’s
development blocks have no explicit spatial/territorial connection. The
Development block is related to structural tensions which to a large extent –
although embedded in socio-cultural mechanisms – basically are of technical
nature: a bottle neck as regards process technology may well be solved with
the help of imported artefacts or solutions from China.

This a-regional understanding of the original concept “development block”

becomes clear in chapter 16 in the dissertation where Dahmén explicitly
discusses industrial companies in relation to the small villages where they
have emerged (a very Swedish “bruks” history). In this chapter he introduces
the concept “economic-geographical development block” – a variant on the
theme but here, when explicitly pointing to the spatial dimension, he uses
another phrase (Dahmén, 1950, p.375).

This is not unimportant. Development blocks and technological systems,

whether in the form of Dahmén, Hughes or Carlsson are not primarily
geographical concepts. They are defined on a technological/cognitive space
with now necessary spatial topological anchoring. The spatial dimension
becomes an empirical question and something which – if necessary – can be
handled in the policy analysis and as a challenge for management.

Development blocks, industrial transformation and

climate change – on the managerial and policy
implications of the Dahménian approach in our time.
Although man made climate change has been analysed during more than a
century the awareness of the dramatic implications for our life styles and our
industry and technology is more recent (and still difficult for many of us to
accept) After two centuries of growing CO2 emissions in the magnitude of
3%/yr., mankind has to reduce these emissions with approx. 5-6%/yr. in the
decades to come to avoid a disastrous increase of global temperature in the
magnitude of 2°C above late 19th century level (see Laestadius, 2013 & 2015,
for further references).

When dealing – on a more concrete policy level – with this coming

transformation the Dahménian tool box is highly relevant; and not only for
understanding of the industrial/technical connections in society but also to
change and transform them. First of all: GHG emissions mainly have their
origin in a set of energy transforming development blocks which may be the

36
targets for “bottom up policies” with a strong focus on new technology and
systems solutions rather than restricting policies on the macroeconomic level
(i.e. general taxation and pricing systems).

Secondly: the DB approach facilitates the identification of core

relations/connections – economic, technological, cognitive – across the
borders of traditional sectors and industries thus revealing the important
parameters to focus on. Development block borders do not necessarily follow
the borders of industry classification: the DB for pulp and paper may include
related segments from many different industries and technologies. And, as
mentioned above, DB borders may change when old technologies are
substituted by new ones.

Thirdly: the DB approach is fundamentally related to the dynamics within the

blocks: the tensions, i.e. the necessities and the opportunities created during
the non-balanced transformation. This is probably a core issue in the years to
come: a giant transformation of industrial systems and technologies cannot
be a process in balance. Neither was this the case in our industrial history of
the past nor can this be the case in the future. Transformation policy must not
be based on a smooth and balanced change. Entrepreneurial/innovative
activities will identify existing tensions as well as create new ones. This was
Schumpeter´s idea with creative destruction and so is also the case with the
approach by Dahmén.

Fourthly: the DB approach has no necessary relation to “green growth”, a

concept which has become popular in recent years (OECD 2011 & 2015). The
transformation of blocks can take place irrespective of whether the economy
as a whole grows or not. At best “green growth” is an outcome of the sum of
block transformations. And at best there are – on aggregate economic level –
self propelling growth mechanisms of handling the necessities and exploiting
the opportunities. But there is no guarantee that the net effect of the
transformation of the development blocks can be accounted for as green
growth (Laestadius 2013 & 2015). In fact, as mentioned in section 2 above, the
empty space left due to the neglect by macro oriented economists and
economic historians, who did not deliver explanations, on the transformation
processes occurring beneath the cyclical processes was a core issue for
Dahmén in his early writings already. Also in his later writings (cf. Dahmén
1991 and section 3 above) he was clear that transformation was the core issue
– not economic growth. And this is still the case.

This argument can be illustrated from three sectors we may identify as

development blocks (or even systems in the meaning of Hughes): energy,
transport and steel.19 We may also face necessities and opportunities related

19Biomass is analysed separately in chapter 12 in this volume. The transformation of biomass may
well have connections to the transformation we discuss here.

37
to these sectors which may turn them into new development blocks. In fact, I
argue, this is not only something we may study (ex post) as academic analysts.
We may also (ex ante) act as managers or policy makers and try to create or
transform these blocks!

As regards energy the systems/bloc character is obvious not the least because
significant parts (electricity) are connected via the grid. The grid is always in
balance (frequency stability) due to the character of electricity production.
Industrial activity and household life styles are based on the continuous
availability to electricity historically provided by large scale power stations
(hydro, nuclear, fossil fuels and biomass/waste). Climate change now forces
us to close down all fossil fuels (of which Sweden already has very few) while
biomass partly will be allocated to other uses; and this parallel to a situation
when there is strong resistance to nuclear power. One solution to this problem
in addition to more energy efficiency is a large expansion of sustainable forms
of energy like wind power and photovoltaics. But the sun does not always
shine and the wind does not always blow. So there is need for more.

We are thus facing strong tensions in this development block. The

immediately available technologies of sustainable energy production do not fit
the existing structure of demand. This creates necessities as well as
opportunities for energy providers as well as energy users. We cannot in
advance know in detail how this block will transform although we can foresee
very high prices for the most extreme demand for stability among a few users,
new technical solutions for adaption to less stability (in grid frequency) among
many users to avoid price peaks and new creative solutions to transform the
base load system. New technologies for balancing and storing energy, not the
least in the combination of heat pumps, district heating, biogas and hydrogen
may develop.

The transport block, which may be defined wide or narrow, shows similar
characteristics. A widely defined block for personal transport will of course
include all subsystems related to the more “narrow” automotive block (which
in itself is far from small as 82% of all European personal transport is carried
out with cars) plus other systems related to personal transport. Climate
change mitigation necessitates that all fossil fuels in car transport are
eliminated. The only long term sustainable solution for fuelling the car is
electricity, in some cases in combination with biofuels. But that
transformation must probably be combined with a sharp reduction of car
transport, growth of other sustainable means for transport like buses and light
distribution trucks, overall reduction of personal transport, new models for
ownership and organization of transport related services etc.

Also this transformation will create tensions with necessities and

opportunities for various actors in the automotive block, which constitutes a
large part of the industrial sector in Sweden. In addition, these tensions will

38
also be present for all those households (not included in Dahmén´s DB
concept) which objectively or subjectively base their life styles on cars using
fossil fuels. And these tensions are not far in the distant: if the car fleet will be
fossil free in 2030 – a declared goal for Swedish policy and an important step
to reach CO2 targets as formulated in the COP21 agreement in Paris 2015 –
approximately 300 000 cars from the present fleet have to be scrapped yearly
and substituted by other transport solutions in combination with an
immediate stop of new registrations of fossil fuelled cars.20 Also aviation faces
a strong necessity to transform. Swedes fly more than average in Europe and
the growth of air travel is larger than European average. And aviation is a
significant emitter of CO2 which has to be eliminated.

The steel block is, since the eve of Swedish industrialization one of the
backbones of the Swedish economy. Sweden has a higher production of iron
ore and steel per capita than most other countries in the world. We also have
a larger export specialization in iron and steel than most other countries; still
more remarkable compared to other industrialized countries. Swedish iron
ore and steel is in addition of higher quality than the global average: in short
Swedish producers get more money per ton sold than most other producers.

The steel block is also one of the most GHG-emitting sources in Sweden thus
making it to one of the core problems in the Swedish transformation away
from dependence of fossil fuels. The reason for this is that carbon is used for
the chemical process in the reduction of ore based iron, not only for “energy
proper”. Scrap steel, however, can be melted and transformed with electricity
which can be based on renewable resources. In short: large scale blast-furnace
based iron production is not compatible with climate change mitigation. The
problem is that Swedish steel (and iron ore) are the probably least polluting
process in the world, per ton delivered and still more per function performed.
Closing down Swedish steel will thus, in a world that still uses steel, only
contribute to a leakage effect and other more dirty producers to invade the
open space (Kander, [Link]., 2015).

Of interest now, from a Dahménian perspective, is whether these tensions in

form of necessities and opportunities can contribute to new development
blocks. Not the least is this a challenge for industrial policy makers and
industrialists to proactively develop new paths for a post carbon Swedish
industry. Let us illustrate how this may take place in a new development block
around hydrogen.

Sweden can probably invest in renewable electricity – primarily wind power

and photovoltaics – for a production of additional 100-150 TWh/year. When
the wind is blowing and the sun is shining the excess capacity can be stored in

20 Many of today’s cars can drive on biobased substitutes for gasoline and diesel. But there will never
be biofuels enough to feed a transport system with the structure and growth of today.

39
many ways, one of which is hydrogen. This hydrogen can be used in many
ways: one is top/reserve capacity for energy shortages, another is for parts of
the transport system. A third, and potentially significant use is for the
reduction of iron ore to steel.

As regards reserve capacity in the grid hydrogen will probably never be a large
scale battery, but it may be developed for local systems (maybe
complementary to biogas turbines) like hospitals etc. This is an opportunity
which still has to be developed.

As regards transport, fuel cell technology is already a working technology for

cars although presently more expensive than battery based electricity. It is,
however, an alternative way to store energy for electric cars. Also this is an
opportunity that may develop from large scale hydrogen storing in an
unbalanced system with cheap renewable electricity. In addition, we do not
know the potential for hydrogen in shipping. What we know is that electricity
based on present battery technology is no future for ocean shipping.

So, back to the steel issue: a hydrogen based reduction of iron ore to steel is
probably the only realistic sustainable alternative to carbon. In fact, the
process will emit steam water rather than CO2. It is also innovative: in reality
it means allocating resources on a technology of tomorrow rather than
spending money on Carbon Capture and Storage (CCS) to preserve a
technology from yesterday.

What we see here is the potential creation of new interdependencies between

changing technologies and transforming industries, new tensions in the form
of necessities and opportunities. We can, ex ante, imagine that we are
contributing to the formation of new development blocks and that the old ones
may change in character. At this stage we cannot imagine the details, but as
we know that climate change mitigation and adoption is necessary (there is no
Business As Usual as climate change will transform the conditions for human
activities) this kind of creative analysis, entrepreneurship and management,
based on Schumpeterian and Dahménian concepts and theory may be a good
starting point for action: i.e. on how to destroy the old carbon based structures
and how to create new development blocks for the future.

We do not have to know the details that may emerge. As regards the hydrogen
– steel block we are just in the beginning of an unknown process. Enough is
to grasp the core imbalances that have to be addressed. The entrepreneurs,
researchers, innovators and policy makers of the future will create new –
hopefully sustainable – tensions and new business necessities and
opportunities along the way.

Neither do we know whether this giant transformation will create aggregate

economic growth. Following Erik Dahmén we focus on the transformation of

40
industry and technology which is the foundation of all economic change.
Whether this ends up in growth is another question (Laestadius, 2013 & 2015).

On the use of sources and references in this chapter

In his dissertation Dahmén makes frequent references to his intellectual
sources. These references are often mentioned in this chapter when the origin
in and inspiration for Dahmén’s conceptual development is described and
analysed. The sources/references mentioned in this text are, however,
sometimes those – younger – editions which have been used in when locating
Dahmén in his intellectual context and when checking his understanding of
his inspirateurs. The exact editions used by Dahmén is available in his
reference list. An illustration: Dahmén refers to Schumpeter (1911) which is
the original German Jena edition. In this chapter I have used the updated and
translated Harward edition from 1934 and reprinted 1962 (Schumpeter,
1934/1962).

References made by Dahmén in his text are not referred to as sources here
unless they are cross checked for the purpose of this chapter.

References
Berman, Sheri, 2006, The Primacy of Politics: social democracy and the
making of Europe´s twentieth century

Carlsson, Bo, 2000. “Teknologiska system som analytisk ram – en

introduktion”, in Carlsson et al., Innovationssystem, kluster och
kompetensblock, Stockholm: Nutek & Rådet för arbetslivsforskning.

Carlsson, Bo, & Stankiewicz, Richard, 1991, On the nature, function and
composition of technological systems, Journal of Evolutionary Economics,
Vol. 4, p. 93-118.

(Dahmén, Erik, 1942, Ekonomisk strukturanalys. Begreppet felinvestering

som kojunkturteoretiskt instrument. Lic.-avhandling, Lund: Lunds
universitet. (reprinted in condensed version in Statsvetenskaplig Tidskrift.
This reference is listed only – not used or checked).

Dahmén, Erik, 1950, Svensk industriell företagarverksamhet, Band 1 & 2,

Stockholm: IUI. (For an English translation of volume 1: see Dahmén, Erik,
1970,

41
Dahmén, Erik, 1970, Entrepreneurial Activity and the Development of
Swedish Industry 1919-1939, AEA translation series 99-1314337-3,
Hmewood, Ill: Irwin.

Dahmén, Erik, 1991, Development Blocks and Industrial Transformation (ed.

by Bo Carlsson & Rolf G.H. Henriksson), Stockholm: IUI.

Dahmén, Erik, 1994, “Towards Research on the Technology of Economic

Development, Summary Remarks 1”, in Granstrand, Ove, ed., 1994,
Economics of Technology, Amsterdam: North-Holland.

Edquist, Charles, 1997, Systems of Innovation – Technologies, Institutions

and Organizations, London: Pinter.

Eklund, Klas, 1986, “Det våras för Dahmén”, Skandinaviska Enskila Bankens
Kvartalskrift, Nr 2/86.

Eliasson, Gunnar & Åsa, 1996, “The Biotechnological Competence Bloc”,

Revue d´Economie Industrielle, Vol. 78, p. 7-26.

Fagerberg, J; Laestadius, S. & Martin, B., eds., 2015, Challenegs for Europe –
Economic Development, Climate Change and Governance, Oxford: Oxford
Univ. Pr.

Hughes, Thomas P., American Genesis – A Century of Invention and

Technological Enthusiasm, New York: Viking.

Hughes, Thomas P., 1992, “The Dynamics of Technological Change: Salients,

Critical Problems, and Industrial Revolutions”, in Dosi, Giovanni et al.,
Technology and Enterprise in a Historical Perspective, Oxford: Clarendon
Press.

Hughes, Thomas P., 1997, “The Evolution of Large Technical Systems” i

Bijker, Hughes & Pinch, eds., The Social Construction of Technological
Systems, Cambridge, Mass.: The MIT Press.

Kander, A., et al., 2015, “National greenhouse-gas accounting for effective

climate policy on international trade”, Nature Climate Change, Vol. 5, May,
pp. 431-435.

Keynes, J.M.,1930/1960, A Treatise on Money, Vol. II, London: Macmillan &

Co.

Laestadius, Staffan, 1992, Arbetsdelningens dynamik, Lund: Arkiv.

Laestadius, Staffan, 2013, Klimatet och välfärden, Umeå: Borea bokförlag.

42
 Laestadius, Staffan, 2015, “Transitions paths – assessing conditions and
alternatives” in Fagerberg, Jan; Laestadius, Staffan & Martin, Ben.,2015

Laestadius, Staffan & Rickne, Annika, 2012, “The Theoretical Foundation for
Swedish Innovation Policy”, in Rickne, A.; Laestadius, S. & Etzkowitz, H.,
eds., 2012.

Lindgren, Håkan, ed., 1996, Economic Dynamism in Honour of Erik

Dahmén, Inst.f. ekon. forskn., Stockholm: Handelshögskolan

Marshall, Alfred, 1890/1990, Principles of Economics, London: Macmillan.

OECD, 2011, Towards Green Growth, Paris: OECD.

OECD, 2015, Towards Green Growth? – Tracking Progress, Paris: OECD.

Porter, Michael, 1990, The Competitive Advantage of Nations, London &

Basingstoke: Macmillan.

Pålsson Syll, Lars, 1997, Den strukturanalytiska traditionen, Lund:

Studentlitteratur.

Rickne, A.; Laestadius, S. & Etzkowitz, H., eds., 2012, Innovation Governance
in an Open Economy- Shaping Regional Nodes in a Globalized World,
London: Routledge.

Rosenberg, Nathan, 1976, “Problems in the Economist´s Conceptualization of

Innovation” i Perspectives on Technology, Cambridge: Cambridge U.P.

Saxenian, AnnaLee, 1994, Regional Advantage, Cambridge, Mass.: Harvard

Univ. Press.

Schumpeter, Joseph, 1911/68, The Theory of Economic Development,

Cambridge, Mass.: Harvard U.P.

Schumpeter, Joseph, 1939, Business Cycles, Vol. I., New York: Mc Graw-Hill.

Schumpeter, Joseph, 1943/2000, Capitalism, Socialism and Democracy,

London & New York: Routledge.

Stankiewicz, Rikard, 2000, “The concept of ´design space´, i Ziman, John,

red., Technological Innovation as an Evolutionary Process, Cambridge:
Cambridge U.P.

Veblen. Thorstein, 1898, “Why is Economics not an Evolutionary Science?”,

The Quarterly Journal of Economics, Vol. xii, July.

43
Veblen, Thorstein, 1904/1932, The Theory of Business Enterprise, New York:
Scribner´s.

Åkerman, Johan, 1931, Ekonomiskt framåtskridande och ekonomiska kriser,

Stockholm: Kooperativa förbundets bokförlag.

Åkerman, Johan, 1933, Ett elektriskt halvsekel: översikt över ASEAs

utveckling 1883-1933, Stockholm: ASEA.

Åkerman, Johan, 1936, Ekonomisk kausalitet, Malmö: Förlagsaktiebolagets i

Malmö Boktryckeri.

Åkerman, Johan, 1939, Ekonomisk teori I, Lund: Gleerup.

Åkerman, Johan, 1944, Ekonomisk teori II, Lund: Gleerup.

44
A Critical View on the Innovation Systems
Approach
Staffan Laestadius & Annika Rickne

In light of large and difficult technological and industrial

transformations, pertaining to diverse challenges as environmental
sustainability, health for an aging population or food for all, policy,
industry and academia alike are in need of solid analytical tools. It is
a daunting task to understand and come to grips with multifaceted
and global realities that ever change.

In this chapter we address one major movement that started already

in the 1980s, and which still holds strong relevance for grasping
knowledge and innovation dynamics: innovation systems. This is a
conceptual framework for the analysis of innovation processes and
innovation policy dynamics. We scrutinize the theoretical
foundations of the approach, review the state of the art, and discuss
the benefits, shortcomings and potential uses of the analytical tool.

Introduction
The aim of this chapter is to scrutinize the theoretical foundations of
innovation systems, which is a conceptual framework for the analysis of
innovation processes and innovation policy within a certain context (defined
as a system). The chapter reviews the state of the art, i.e. our ambition is
pedagogical and synthetic rather than to be fundamentally critical or to
develop theory. In extension, the aim is to sharpen the intellectual and
analytical tools of innovation systems, to make them still better and useful for
the analyst to understand – and handle – the great industrial transformations
we have reasons to expect for decades ahead.

Let us, for the sake of convenience, already at this stage provide a broad
definition: an innovation system (IS) is, in this chapter, defined as an
interrelated structure of institutional and actor based condensations in an
economic space. These actors are engaged in generation, diffusion, and
utilization of innovation, in an area which has specific industrial and
innovative properties compared to its context with which it has exchange.
Commonly, the components of an IS are defined as all the various types of
actors within the system which are relevant for innovation processes to come

75
about, the networks and knowledge flows between them, the knowledge areas
(i.e. technologies), and artefacts involved in the specific system, and the
institutional set up guiding the behaviour of the components as well as of the
system as such. The various innovation systems approaches – be they
sectorally, technologically, nationally or regionally defined – thus focus on the
fact that emergence, specialization and competitiveness of sectors /knowledge
areas/countries/regions relate to the institutional conditions, the setup of
actors and their specific, path-dependent, knowledge base and how this is
shared.

The chapter is structured in the following way. We start with laying the
foundation on which these system concepts rests: an evolutionary view of
innovation (section 2). Indeed, we see modern innovation theory as a reaction
to the often static views of economic theory with its focus on equilibrium
rather than dynamic changes, selection mechanisms, learning processes and
path dependencies. Next, in section 3, we return to the innovation system
concept presenting its basic characteristics, paying particular attention to the
importance of institutions, the problems of openness and intentions in
systems, as well as the identification of systems borders.

Paving way for a deeper understanding of the policy uses of the IS concept, we
dig, in section 4, into the various IS related concepts. That section is primarily
of a typological character with the intention to give an overview of the
conceptual jungle of partly overlapping concepts.

Section 6 focuses on two problems for which the IS approach has been
criticized. First, one set of critique says that in spite of claiming to address
evolutionary aspects of innovation there is a risk of rather static and structure
based analyses. Several authors have addressed this critique by developing a
functional approach to the analysis. Second, critiques assert that while the
institutional set up is used to delimit the systems analysed, institutional
changes and influences, as well as governance aspects, have more seldom been
captured by IS analyses.

An evolutionary economic view of innovation

Our analysis focuses on the innovative properties of innovation systems (IS).
We can in the definition of IS above find an implicit criticism of orthodox
economic theory with its focus on equilibrium. Innovation theory, with a
cultural heritage back to the Austrian economist Joseph Schumpeter, centres
on the breaking up from equilibrium, on the creative destruction when new
products and processes substitute for old ones, and on the transformation and
growth taking place as a result (Schumpeter, 1934/62).

76
In the innovation literature, entrepreneurship and innovation are different,
and complementary, sides of the same coin. The entrepreneur is the actor who
introduces the innovations on the market; be they science based,
technological, service related, organizational or market related innovations.
In the literature, an entrepreneur without innovations is not perceived, even
though in common jargon anyone stating a new venture can be called an
entrepreneur. However, the strong interrelation between innovation and
entrepreneur is an important part of the original theory pointing towards the
action oriented element in the innovative economy, not only the exploration
aspect (focusing on inventions and R&D).

Basically, innovation theory is related to what may be labelled evolutionary

economics. The logic is simple: the essentials of the Schumpeterian message
100 years ago was that innovations continuously destroyed the equilibrium
process and forced the actors of the economy to a behaviour which ended in
industrial transformation and change. In evolutionary theory, equilibrium
thus is a special case, an attractor on systems level for sure, but never realized
as the actors in the system always strife to avoid “the circular flow of economic
life” as Schumpeter formulated it (Schumpeter, 1934/62, chpt II).
Competition in this world is not primarily a price competition between
homogenous products among independent and fully informed actors, but a
continuous struggle to use different solutions (e.g. advanced design) to create
variety that attracts certain consumers. As an outcome this creates temporary
monopolies in all areas, thus providing a short period of profitability before
the competitors catch up with new solutions; more or less influenced by, or
deviating from, the original ones (Schumpeter, 1942/2000, chpt. VII).

The attempts to combine the strengths of evolutionary theory with

equilibrium theory have a long tradition among economists and in fact
anticipate the contributions of Schumpeter. Most interesting, and still
relevant, are probably the serious attempts made by Alfred Marshall in late
19th century already (Marshall, 1890/1990; Laestadius, 1992; 1999). He did
not solve the problem of combining evolutionary theory with equilibrium;
during the process of trying he, however, introduced the concept
“externalities”, which became the cornerstone in his analysis of learning
effects, knowledge spillovers and cultural influences, transmitted outside the
market but over the borders between firms as “were they in the air”. For
Marshall this was a core concept in his analysis of the economic and industrial
dynamics in industrial districts. Although he did not use the concept
“innovation”, what he analysed was in fact the – cultural and knowledge based
– conditions for innovative processes.

That takes us back to the IS concept. Originating in the analysis of Marshall,

there is an insight that externalities is one of the core concepts for
understanding the learning and knowledge formation process characterizing
dynamic systems; be they industrial districts or innovation systems of

77
“higher” order. On all levels actors – professional craftsmen and engineers as
well as firms and R&D units – learn and develop new knowledge more or less
influenced by each other, contributing to development of the actors
themselves as well as of the whole system. This analytical approach is an
essential part of understanding the processes that occur in industrial districts
(and the innovation system) and in the economy. Learning, however, is not
the same as innovation, as we will see below.

Knowledge formation and learning are thus core processes in the

transformation of economies. Not least the American economists Nelson and
Winter (1982), who have written the probably most important book on
Evolutionary Theory and Economic Change, have a strong focus on
knowledge, skills and competences on individual and organizational level.
They were among the first scholars who focused on the role of non-codified or
tacit knowledge – a concept originally developed by Polanyi (1967) – in the
learning processes of the economy.

The evolutionary perspective of Nelson and Winter focuses on the tension, or

balance, between the creation of variety (the invention process) on the one
hand; on the selection process on the other and, on the third, on the retention
created in the system by the development of routines to keep the system (and
production) going. In particular, it is important not to ignore the selection
processes, something that is often done by innovation researchers as well as
policy units. Most innovations fail, and have to fail, before, during or after
their introduction to the market. Nelson and Winter argue that we here find
one of the classical failures among innovation researchers: the neglect of the
demand side.

Actors (networks, systems) who do not – through learning – have the

capability to adapt their routines, products and processes to compete with
innovations in their fields of activities (markets) have to close down or
restructure their activities. There is, thus, a connection between learning
systems and innovative ones (Lundvall, 1994), but they are not identical.
Although innovations, following Schumpeter, on the one hand may be
characterized as new creative combinations of knowledge which may be
recently acquired as well as known since long – and thus has to be (re)learned
– they, on the other, must be looked upon as breakaways from ingrained and
learned paths; i.e. processes of unlearning. Innovations destroy existing
structures and create something new on the ruins of the old. This destruction
may be more or less radical: nevertheless, it represents an act of creation –
and creative destruction – that is more than just adapting to old routines
(Schumpeter, 1942).

Marshall realized that intra-district (local) learning and knowledge transfer

from one generation of craftsmen to another, under certain circumstances,
could contribute to cementation of old competencies and to an incapacity to

78
innovate new products or processes. The diligent and quick to learn pupil is
far from always the most innovative, and the innovative violinist breaks at
least some of the routines learned from the master. In this context. “ learning”
should not be identified as identical with, or enough for, the variety – selection
– retention processes.

This view on innovation as a creative process combining learning and

unlearning also challenges the conventional view – still common among policy
makers and some parts of academia – on the innovation process as
fundamentally linear. Strongly simplified, the linear model of innovation
displays an understanding of innovation activities as did they start with basic
knowledge (e.g. basic science), followed by more applied knowledge (e.g.
applied science), product or service development (e.g. by engineering) and
later close-to-market-activities. As argued since long by many innovation
researchers this model – strongly connected with the US post WWII report
Science the Endless Frontier (Bush, 1945) – ignores, the fact that innovation
processes take place over and between different systems levels and domains
of knowledge. Innovation is in essence a highly interactive and iterative
process, where various knowledge components (e.g. market knowledge)
repeatedly come onto the scene, interact, change and shape new
understanding (Rickne, 2000). For example, the design elements can be, and
often are, substantial and (recent) scientific knowledge may be either present
or absent in this creative process (Utterback et al, 2006). Although there are
many academic contributions to this discourse (e.g. Faulkner, 1994) the
probably most well-known challenge of the linear model has been written by
Kline and Rosenberg (1986). Of central importance for their analysis is the
stock of knowledge to which all activities – also R&D – contributes and which
may be utilized in all phases of the innovation process which in its turn is
characterized by significant feedback loops and engineering and design
activities.

Nevertheless, is the linear model – with its strong focus on the analytically
biased basic sciences – still alive as a blueprint for significant parts of the
innovation and technology policy discourse. To some extent this is probably
because parts of the academic community has knowledge interests in it,
because the linear model is easy to communicate and finally because it is
attractive for politicians and policymakers who can argue that money to basic
science contributes to the long run development of industry. Naturally, basic
science may indeed, and often do, contribute to and is essential to innovative
change, but our point here is that there is not a linear or necessarily causal
relationship.

Some may argue that there is a part of the IS literature which more or less
uncritically adopts and/or basically ends up in the linear model although, as
argued here, this mode is just a sub set of all possible forms of innovation. It
may, however, also be argued that the IS approach may be used as a

79
framework for identifying the complexity and network character of different
innovation processes.

The innovation system concept defined

Let us return to the IS definition given in the introduction: an interrelated
structure of institutional – incl. cultural – and actor based condensations in
an economic space and which has specific industrial and innovative properties
compared to its environment with which it has exchange. The “innovation
system framework” may in fact be looked upon as a set of concepts with family
resemblance, 24 containing terms with more or less similar meaning. 25
Members of the core family are notions like “national innovation system”
(NIS), “regional innovation system” (RIS), “technological (innovation)
system” (TS),26 and “sectorial innovation system” (SIS). We may also include
notions like “development block (DB)”, “competence block (CB)”, “industrial
district (ID)” and, not least the “diamond” and “cluster” terms once
introduced by Michael Porter (1990).27 Worth mentioning – but not part of
the IS family as such – are the so called multi-level perspective (Geels, 2002;
2004), and sociotechnical system concepts such as those presented by Bijker
(1995) or Hughes (1992), even though they will not be explicitly discussed in
this paper. (see Blomkvist and Johansson in this volume)

Let us then turn to a more detailed presentation and analysis of the different
members of the IS family. National innovation system (NIS) was the original
concept introduced around 1986/87 first, probably, by the British economist
Christopher Freeman (who in fact had used the concept already in 1983, cf.
Freeman, 1987, 1983/2004) in his analyses of the Japanese economy. The NIS
concept was also – in connection to Freeman – used by the Swedish-Danish
economist Bengt Åke Lundvall in his analyses of the competitiveness of small
European economies (Freeman and Lundvall, 1988). Around 1990 many texts
were published where the NIS concept was used (Dosi et al., 1988; Lundvall,
1992; Nelson, 1993)

It is easy to recognize that the 1980s was an ideal and logical foundation for
this kind of theories. By then the post WWII catching-up processes from the

24 The IS approach may perhaps not be seen as a formal ‘theory’. Instead, as Edquist (1997) expresses,
the IS concept should primarily be understood as a “conceptual framework”.
25 This is in itself a concept originally formulated by Wittgenstein in his Philosophical Investigations

(1953/89).
26 This notion, as defined by Carlsson and Stankiewicz (1991) should not be confused with Hughes’

(1992) term ‘technological systems’. Some literature now refers to the type of innovation systems
called ‘technological systems’ as ‘technological innovation systems’, to avoid such confusion (see e.g.
Hillman et al, 2009).
27 The diamond based analysis and the cluster concept are perhaps the most well-known, being

cornerstones in Porter´s analysis of the ”competitive advantage of nations” (and regions).

80
first wave of Asian tigers was maturing: not only Japan but also South Korea,
Taiwan, Hong Kong and Singapore challenged the incumbent industrial
nations. Several of the newly industrialized countries were not neo-liberal
unfettered market economies but rather characterized by strong institutions
with high ambitions to favour industrial dynamics. The success stories from
these catching-up countries are in sharp contrast to the stagnation and lack of
policy that was revealed in the incumbent world (Freeman, 1987; Nelson,
1993; Dertouzoz et al, 1988).

There is also a connection between the NIS concept, formulated in the late
part of the 80s, and the heritage from the German economist Friedrich List.
Already in the first known text on NISs there are significant references to the
national production system concept once developed by List in his analyses of
the industrial competition between the incumbent and free trade oriented
United Kingdom – benefitting from economies of scale and being far ahead
on the learning curve – and the emerging German and US catching-up
economies (Freeman, 1983/2004).

The 1980s was, in addition, a period characterized by globalization as well as

emerging pervasive technologies, like the ICT, which in the extension were
assumed to threaten the small countries´ possibilities to stay competitive
(Laestadius, 1980; Sharp, 1983; Freeman and Lundvall, 1988). From a
northern European perspective there was also the problem of whether small
countries, like Norway, Denmark, Sweden and Finland, could maintain
competitiveness on country level within an integrated Europe. All in all, this
contributed to give the national innovation systems discourse a “small country
bias” (Freeman and Lundvall, 1988; Lundvall, 1992).

Somewhat simplified it may be argued that this also is mirrored in the two
dominating NIS texts from the early 90s: one European and one American.
The comparative approach in the American anthology (Nelson, 1993) is more
focused on the analysis of the existence of innovation systems than the
European anthology, edited in Aalborg/Denmark (Lundvall, 1992), where
most papers focus on the character of the systems. Several of the papers in
the American anthology argues for caution as regards the potential to
understand, still less to manage, innovation systems (Mowery and Rosenberg,
1993).

The national innovation system obviously fulfilled a need (for academics) in

analytically knitting evolutionary innovation theory with institutional theory
as well as to connect industrial transformation/dynamics with policy, here
illustrated with an influential text by Stan Metcalfe (1995, p. 462-63).
“A national system of innovation is that set of distinct institutions which jointly and
individually contribute to the development and diffusion of new technologies and which
provides the framework within which governments form and implement policies to influence
the innovation process. As such it is a system of interconnected institutions to create, store

81
 and transfer the knowledge, skills and artefacts which define new technologies. The element
of nationality follows not only from the domain of technology policy but from elements of
shared language and culture which bind the system together, and from the national focus of
other policies, laws and regulations which condition the innovative environment. In the
operation of national systems, governments play an important part in their support of science
generally and in their procurement of technologies to meet the needs of the executive. To
define such a system empirically one must locate the boundaries, its component institutions
and the ways in which they are linked together.”

The NIS concept developed into an organizing metaphor that successively

colonized more and more space in international policy documents (OECD,
1997; 1999). Not least in the OECD (1999) document – Managing Innovation
Systems – national innovation systems appear to be a question primarily of
governance. By then – in the late 90s – the original NIS concept had
developed into a general concept but also bifurcated into a family of varieties
(Edquist, 1997). In a world where knowledge formation/learning as well as
industrial activity takes place on a global scale it is far from obvious that the
essential and systemic condensations are national or even territorial in any
real sense. With introducing the concept sectorial innovation system, the
Italian economists Breschi and Malerba intend to focus on the industrial
rather than the territorial dimension (Breschi and Malerba, 1997; Malerba,
2004).

Analogously, although with focus on technology rather than industrial sectors

there also emerged a research tradition on technological systems, or
technological innovation systems (TIS) (Carlsson and Stankiewicz, 1991;
1995).28 The difference, in short, between these approaches can be explained
by the fact that the logic of technological change does not necessarily follow
the logic of an industry. ICT, e.g., may penetrate and influence various
industries in different ways: Some advanced process industries may, to a
certain extent be more influenced by ICT than the IT industry itself! Also
biotechnology is often realized in unexpected industries: functional food is,
for example, the result of biotechnology that has colonized a small niche
within a traditional low-tech industry. Technological cognition processes, in
addition, do not necessarily take place within “communities” or “thought
worlds” that follow industrial classification or regional or national borders.
Although craftsmanship and technological cultures (as once described by
Marshall, see below), may develop locally, fundamental technological
paradigms and regimes develop globally among researchers and engineers
and condition and influence the direction of technological change (Dosi, 1988;
Nelson & Winter, 1982; Laestadius, 2000). Today the TIS literature is closely
linked to that of socio-technical transition management and a multi-level
perspective (Hillman et al, 2011).

28 Severalanalysts within this research tradition changed the “brand name” to

technological innovation systems (TIS to make it more in line with the other IS-
approaches.

82
This distinction is far from unimportant and has policy implications: if
industries are targets for industrial policy we have reasons to expect other
results compared to if the policy focus is on technologies. Not only will there
be different receivers of the resources connected to the policy; it will also
develop needs for other policy tools and different routines.

The concept development block (DB), formulated by the Swedish economist

Erik Dahmén (1950), deserves to be mentioned in this context although his
block concept is discussed in detail by Laestadius (in this volume). The DB
concept contributed to the development of the NIS concept among the Danish
economists. Still more has it influenced the technological systems approach
(Carlsson, 2000). The development block – which theoretically may be argued
to slide between a sectorial and a technological logic – is based on the notion
of structural tensions, i.e. imbalances occurring due to the very character of
the innovation process. There are always – in every period – necessities and
problems that have to be handled, as well as opportunities that may be
utilized. These imbalances may develop on the level of technology in the form
of bottlenecks versus local breakthroughs, as analysed by Tomas Hughes with
the concepts salients, reverse salients and critical problems (Hughes, 1992).
But they may also occur on industry/sector level: the rapidly increasing
electricity output due to large scale construction of hydro power plants in
Sweden contributed to the establishment of energy consuming TMP pulp and
paper plants. The DB concept – which never explicitly was formulated in
territorial terms – may be looked upon as a means of making Schumpeterian
dynamics empirically useful and adds a structural dimension to Schumpeter´s
entrepreneurial focus (see Laestadius, in this volume). Dahmén also stated
that development blocks may be used for ex post analyses by academics as well
as intentionally ex ante by industrialists (or policy makers) to create and/or
exploit systems imbalances.

The geography of innovation

Clearly for the sectorial and technological approaches of IS, innovation

processes are not non-territorial, but exists in a geographical reality: The
system is seen as global but composed of – and often analysed in terms of –
geographically distinct nodes (e.g. a country or a region). In the regional
innovation system (RIS) approach geography is at the core of the analysis
(and knowledge areas of sectors secondary). All the IS concepts, but perhaps
best the RIS concept illustrates the difficulties to distinguish a distinct level of
territorial analysis in a globalized economy. RIS is an approach with a clear
inspiration from the NIS writings and is to large extent based on a similar
analytical approaches (Asheim and Gertler, 2005), but also bear resemblance
to Porter’s reasoning.

The regional dynamics dimension has been there all the time although
branded differently. Specifically, Alfred Marshall´s concept industrial district

83
is highly relevant in this context (Marshall 1890/1990). The Marshallian
district was e.g. an important tool for the understanding of the industrial
dynamics of northern Italy (Brusco, 1982; Amin and Thrift, 1994).

Also the rapid industrial development in Silicon Valley – which from many
aspects differed from the Italian – has contributed to the growing interest for
the local and for the territory as well as for analysing the socio-cultural
networks and processes which contribute to an innovative atmosphere – or to
the absence of such climates (Saxenian, 1994). Silicon Valley – the dynamics
of which to a large extent coincidence with, or even was the carrier of, the ICT
boom during the 1970s and 80s – has been the victim of often too far reaching
generalizations based on one extreme case: not least the interpretation of
Silicon Valley as a “science park” has contributed to exaggerated expectations
– often connected to policy failures – on how to use policy to create industrial
dynamics based on local universities.

The differences between, among other things, the low-tech industrial

creativity in some regions on the one hand and the significant clustering of
R&D labs in some localities on the other has contributed to a family of RIS
concepts. One way to capture these differences is to make a distinction
between institutional regional innovation systems (IRIS) and
entrepreneurial ones (ERIS) (Asheim and Gertler, 2005). Although this
distinction is not necessarily exhaustive – both systems may in fact have a low-
tech/high-tech dimension – it serves the purpose to reveal the fact that
government, academia, firms and other institutions play different roles in
different forms of regional innovation systems. This is, of course, highly
relevant for policy institutions: the variety of regional innovation systems may
respond differently to various types of policies.

Globalization, it may be argued, paradoxically has increased the interest for

the regional dimension in industrial dynamics. The existence of “sticky places
in slippery space”, as formulated by Markusen (1996) two decades ago, brings
up the question on the role of regional mechanisms – if any – in creating
industrial creativity in a totally open economy.

Just to assume that innovativeness grows in a region as a consequence of

territorial co-location of firms does not mirror reality in the globalization
process of our time (Cooke, 2005). Although the presence of social interaction,
trust and local institutions is essential for the development of clusters this
does not exclude that also significant non-local knowledge may be necessary
to maintain or develop competitiveness (Maskell et al., 1998; Asheim and
Herstad, 2003; Cooke et al., 2000; Isaksen, 2005). Regional clusters must
develop interfaces with other parts of the world (Rickne, 2000). Interactivity
as regards learning and knowledge formation, thus, must not be restricted to
geographic (territorial) proximity only (Coenen et al., 2004; Torre and Gilly,
2000; Boschma, 2005; Malmberg and Power, 2005; Gustavsson, 2009).

84
The complex interrelationship in the regional system is characterized by
interdependence on several levels of which the territorial level is just one of
many possible (Howells, 1999). Within the framework of these
interrelationships – which are more or less open towards the rest of the world
– certain activities tend to locate to specific regions (Asheim and Coenen,
2006; Cooke et al., 2000). A regional innovation system is thus not a national
innovation system writ small. Even if there are similarities between the NIS
and RIS level of analysis – e.g. in the institutional structure – we argue that a
significant part of the difference is in the “embeddedness” (Lundvall and
Maskell, 2000). In short: the national innovation system is primarily
institutional also from a strict point of view: politics, laws, authorities, school
system, culture politics, incentive structure in the economy, etc. The regional
system, on the other hand, is primarily social and cultural and contains co-
operation, exchange of experiences (transfer of knowledge). The regional level
– it is argued – is more based on social relations than on institutions
(Granovetter, 1985). However, this is far from always obvious: as will become
clear below the distinction between NIS and RIS become blurred when
comparing countries of very different size, like China and Sweden.

Even if the world has become “slippery” human activities have to be grounded
somewhere on the globe, and in a “globalized” world – with global access to
competence localised elsewhere – all kinds of even marginal competitive
advantages in a certain location may serve as a mechanism for competitive
advantage, and maybe initiate a path dependent process. From these
mechanisms, tensions are created between the local and the global. It is, thus
no paradox that the interest for territorial analyses increases parallel to the
globalization process.

The competitive analyses by Michael Porter also belong to this research

territory; his analyses, in fact, span between the national and the regional
(Porter, 1990). Although the concept “innovation system” is missing in his
magnum opus his analyses show family resemblance with the IS approach.
The Porter diamond may be looked upon as a pedagogically attractive
formulation of a (national) innovation system although the “government” in
the original diamond (Porter, 1990, chpt 3) is given a much more limited role
than what European analysts probably would assume. In the more empirical
based part of the book the concept cluster – which Porter may be given the
credit for having introduced into the analyses of industrial dynamics – has a
strong family resemblance with the Marshallian industrial district and the
regional innovation system. The clusters a la Porter, which are analysed
through the diamond approach, are characterized by competition and rivalry
as well as co-operation and networking. As was the case with the NIS concept,
the analysis by Porter naturally received its inspiration from the territorially
located industrial dynamics which was part of the boom during the 1980s and
1990s to a large reflecting as well the development of the IC technologies and

85
globalization. Recent research has taken a more nuanced and critical view on
the functions – and non-functions – of clusters showing that it sometimes has
been too easy to identify and make policy and location decisions to “clusters”
(Asheim, Cooke & Martin, 2010).

Over time government has taken a more prominent position in Porter´s

research. This is especially the case in the paper Green and Competitive where
the importance of a high-end domestic demand, created by an advanced
environmental policy, for the creation of green industrial clusters and
competitiveness is analysed (Porter and van der Linde, 1998). It may be
argued that part of their argument misses the point: the transformation
pressure discussed did not have its origin in government but in the
environmental movement (Cerin, 2004). Institutions are more than
governments.

Common for these research approaches is the ambition to identify the

combination of locally unique and relatively sticky mechanisms and cultures
favouring innovativeness and creativity, i.e. the origin of competitiveness: to
have the capability to develop something which does not immediately slip over
to other actors, networks or regions.

The implicit assumption in many theories on the mechanisms of dynamic

territories where certain attractors make them grow more rapidly than others
is that people, often understood as labour, are the agents most inclined or even
forced to move. Historically this may well have been the truth. Some years ago,
however the Italian-American sociologist Richard Florida in a set of books
(2006; 2002/2014) argued that in a world where creativity and human capital
is important for competitiveness and professionals are in short supply, high
tech firms have to move where the creative class prefer to live. Although
Florida´s original conclusion had conceptual and empirical weaknesses (cf.
Nuur & Laestadius, 2009) his basic argument is of importance: in a world
where human resources are important people do not only adapt to the labour
market but create the conditions for industrial life.

More sophisticated discourses on the contextual ground for knowledge

formation (and learning) do, however, take us away from a narrow
interpretation of the territory. Introducing the concept spaces of knowledge –
reminding us of the concept economic spaces introduced by Perroux 1950
already – the geography of knowledge is no longer fettered to the territory
(Amin and Cohendet, 2004, chpt. 5). This also takes us to the approaches
common in modern sociology and management theory. Specifically, a
significant part of the knowledge management theory is focused on how to
develop and maintain learning and knowledge creation on corporate level in
global firms and thus basically independent from territorial aspects (von
Krogh et al, 2000).

86
Creative and innovative processes can nowadays be organized on global level
as well as on the regional. Multinationals that traditionally have had their
R&D units in Europe or the US (Pavitt and Patel, 1991) now relocate their R&D
activities. Within some scientific and technological areas there are also
research communities strongly connected in networks with low physical
proximity but maintaining and developing professional relations on distance.
Phenomena like these do change over time: mechanisms which contributed to
proximity some decades ago – e.g. in the industrial districts of Alfred Marshall
– may be less relevant today. And information technology development may
create potentials for communication that, at least partly, may substitute for
direct human interaction. How far this globalization of knowledge formation
has gone and whether there are limits to this process is the topic for much of
present day research. It may be assumed that innovation systems are changing
in character due to internationalization (Carlsson, 2006).

What constitutes a region or a regional system is thus far from evident. We

can imagine a set of relations which all have different geographical anchoring
– and distributed over different geographical territories – but still being
important for the system in question. The condensations of the different
relations may be interpreted as layers of different systems added to each other;
this set of layers may reveal a certain territorial structure in the form of a set
of nodes (hubs) connected through a set of network relations, together
constituting the economic space. The non-territorial systems we discussed
above may thus be included in this model.

One way to conceptually handle this variety of relations is to introduce the

concept functional regions (Hallin, 2005). Theoretically we thus get a set of
regions each of which is defined from a certain identified function and not
necessarily identical with a classical geographical or administrative structure.
A functional region from an innovation point of view must not coincide with
the regional labour market.

It may be argued whether functional regions – strictly speaking – are

territorial or not; that is of no importance for our analysis here. Of importance,
however, is that (innovation) system borders no longer – if they ever could –
can be assumed to coincide with those administrative borders created by, and
conditioning, the activities of policy makers.

Enlarging the conceptual ambition to the global level the innovation systems
approach faces a lot of problems. The meaning of concepts like global or
continental innovation systems – as introduced by Freeman (2002) – is far
from obvious. This problem was revealed already in the early analysis by
Mowery and Rosenberg (1993) relating to the U.S. innovation system. Still
more do we face the problem of analysing the industrial transformation in
China which has – with 1400 million inhabitants – during three decades
(since 1978/79) shown a unique industrial transformation; longer and more

87
rapid than any other country in history. But hitherto, most of this dynamics
has taken place in a segment of Chinese society covering approximately 300-
400 million people of which the majority live in three large regions (Pearl
River Delta, Yangtse River Region and Bo Hai Rim) and in another half a
dozen big urban areas with populations up to the magnitude of 30 million (e.g.
the Chonquing area). But in these fast-transforming regions there are large
segments of the population that are outside, or even losers in, this dynamic. It
is difficult to imagine that the NIS and RIS concepts, as used in the European
way, could have a similar meaning when applied on countries like China.
Every dynamic “region” in China is larger than the biggest European countries
and the big industrial cities have a larger population than a typical small or
medium sized European country. In addition, it should be observed that much
of this transformation – or catching-up process – more has the character of
classic growth – i.e. more roads, canals, railways, cars and houses – rather
than innovations as usually defined (Lundvall, 2006; Laestadius et al., 2008).
In short: we are here probably approaching the limits of the innovation system
concept.

Components and critiques

Within the IS families a set of core concepts are embraced. We will here
discuss and problematize some of these central aspects, and point to the need
for further clarification, specification and empirical investigation of the IS
framework. A first core feature is the explicit identification of the importance
of institutions in conditioning economic processes. Secondly, we discuss the
general understanding of the system concept, particularly focusing on
openness and intentions. Thirdly, we scrutinize the identification of systems
borders that may be related to geography – but also to other forms of
proximity (knowledge/technology). Fourthly we focus on the problem on to
what extent a “structure” is a feasible analytical concept when analysing the
dynamics of industrial and technical change. Finally, we turn to the
governance question. The aim, of course, with IS related analysis is to obtain
results that can be used for policy and management. To what extent can ISs
be governed, or even created – and how?

The role of institutions

In the definition of ISs used above there is an implicit although fundamental

criticism of the core message in standard economics: the notion of pure
atomistic equilibrium markets – i.e. characterized of all the requirements for
full competition – is not a good approximation of real economies. These are
always embedded in an institutional and cultural web conditioning activities
in the long run as well as in the short. Our focus on the role of institutions –
contributing to the creation of the system as well as being created by it –

88
follows a long tradition among economists and industrial analysts (Hodgson,
1998; North, 1990). The IS approach may be looked upon as a holistic attempt
to understand the institutional web of which the classical economic agents,
“firms” and “labour” are just two elements of many.

The role of institutions has been a core research topic for economists for a
century at least. Among the most well-known among international classics are
Thorstein Weblen, Douglas North, Ronald Coase and Gunnar Myrdal. Among
Swedish economists, not so well known in this area, we also find Johan
Åkerman and Ingvar Svennilsson (Hodgson, 1998). For some social scientists
outside the economics discipline – e.g. sociologists and political scientists –
the important role institutions take is by definition obvious. What is the book
by Putnam on the Working Democracy if not an institutional analysis of the
north and south Italian ‘innovation systems’ (Putnam, 1996)?

Understanding innovation systems as an institutional and relational

condensation is thus essential. That leads us, to the meaning of the
“institution” concept. Here we follow conventional social science theory using
a very broad definition for the concept where it is seen as “sets of common
habits, routines, established practices or rules which regulate the relations
and interactions between individuals and groups” (Edquist and Johnson,
1995, p. 9).29 According to this view, an institution is every social construct –
be it intentional or not – regulating social interaction. Thus, an institution is
not a government authority only; neither is it primarily a hospital, a home for
elderly or an opera – although these three categories may all be included in
the institution concept. Institutions may be acting on the market – patent
authorities e.g. charge applicants – but their importance is primarily not in
selling and buying but conditioning and influencing the behaviour of societies.

Accepted business behaviour, the working moral, patent systems, property

laws and registration systems, tax laws, school systems, entrepreneurial spirit
and prevailing views on gender, nature, equality and religion all belong to the
set of institutions which contributes to the embeddedness which the IS
approach intends to capture. Obviously there is variety as regards institutional
embeddedness, and thereby there is also variety between IS as regards the
impact certain institutions may have on the behaviour of actors. IS researchers
typically have the ambition to examine differences and similarities between
different innovation systems, maybe also to point to their perceived or
revealed weaknesses or strengths analysed from a certain perspective. While
such analyses could theoretically contain a very large number of institutions,
they are in practice often focused on a rather limited set. For example, one

29 Incontrast, organizations are “formal structures with an explicit purpose” (Edquist and Johnson,
1995, p. 11). This means that we do not use the term institutions when we mean the organization as
such. For example, a university may be referred to both as an institution influencing societal
interactions, and as a specific organization.

89
may scrutinize what role different countries’ institutions for property rights
have for inventiveness. However, while the analysis of cultural norms and
behaviour may be just as relevant, such institutions have more seldom been
centred in the IS literature. In part this may be due to the knowledge
background of the community of IS scholars, and also on the theories and
methodologies developed to measure such phenomena. For example, do we
really have the tools to evaluate whether the welfare system, the role of
ethnical minorities, the gender attitudes or the R&D policy is most important
for the innovation capacity of a certain region?

The systems concept

As we are dealing with a “system” we need to first of all discuss what this
implies. In general, systems are made up of components, attributes and
relationships. The components are institutions, actors and artefacts, and
attributes are the properties of the components. Relationships are the links
between these components: Market as well as non-market links. Feedback
loops and interaction provide the dynamics of the system. As Carlson et al.
(2002, p. 234) put it: ‘One result of interaction (feedback) among actors is that
capabilities shift and grow over time, and therefore, the system configuration
also changes’. Also, the system’s propensity for to be both robust and flexible,
and its ability to both induce and handle change are crucial. Another way to
define a system is to see it as “a group of components (devices, objects or
agents) serving a common purpose, i.e. working towards a common objective
or overall function” (Bergek et al., 2008, p. 408). In this context, many
researchers agree that an innovation system have the overall function of
developing, diffusing and utilizing innovations.

Already classical systems theory struggled with how to handle the

transformation of the systems approach from the natural world (e.g. physics
or biology) to the social (Emery ,1969 and Ingelstam, 2002). An important
step – especially interesting due to the evolutionary foundation of innovation
theory – taken by von Bertalanffy (1950, in Emery, 1969) is his distinction
between the equilibrium (closed) systems of physics and the steady state, on
the one hand, and open biological systems on the other. Similar to biological
systems, social systems are open which makes it possible for them to develop
and to maintain or even change their systems structure while transforming
their parts.30 Clearly, systems can be more or less open (closed) towards their
context. An extreme interpretation with a “totally open” system renders the
concept without meaning as one cannot discriminate it from its context. One
way to come to grips with the degree of openness is to specify the types of
elements that are to be included in the system. In the case of innovation
systems literature, one often includes the relevant institutions, knowledge
areas, artefacts, actors, networks and the associated knowledge flows. We will

30 Note that the biological analogues should not be extended in all directions.

90
come back to this issue of delineating the system and deciding upon relevant
elements in a more detailed way below.

Another important distinction of the system concept – relating to how the IS

concept is used by various theorists and practitioners – is whether or not the
system is seen as an analytical construct or as a system that exists in reality.
Phrased differently, one may use the IS concept to describe/analyse an
empirically observed set of interconnected actors, networks and institutions.
The IS framework thus becomes a model for us to better describe and explain
identified structures and connections. With this use of the IS concept, one may
sketch a potential system that could emerge but that doesn’t exist in practice,
or where links or functions are not in place (Bergek et al., 2008).

Assuming that we have identified the existence of innovation systems we may

make a distinction whether they are self-organized or constructed by
intention. Whether certain innovation systems are social constructs that
contain human actors and organizations that have intentions – is a typical
research question. Actors may have intentions for their activities (firms are
e.g. often assumed to maximize profit and individuals may have certain goals
for action) and these intentions may include that some functions of the system
as such needs to be strengthened. In this way, actors such as firms may actively
work towards supporting (their view of) a well-functioning system (Nilsson et
al., 2012). Rarely, however, firms take an overall responsibility for all
functions of a system, but are more likely to focus on particular sub-functions.
Moreover, policy or bridging organizations may have the intention to
construct a system with certain properties. In this quest such organizations
may strive to create a system from scratch or to strengthen an existing system.
In order to achieve that they may be engaged in the construction and
transformation of institutions which are parts of the system.

Having said this about actors’ intentions, our broad definition of innovation
systems does not as a prerequisite assume that innovation systems are the
result of intentional systems building. In fact, while in some cases collective
and coordinated action may be the case, it seems to be more common that
innovation systems emerge through the aggregated effect of uncoordinated
more or less intentional actions over long time periods. From an analytical
point of view such systems may be looked upon as self-organizing where
intentions, entrepreneurial culture, etc. may be more or less formed
endogenously. As Bergek et al. (2008, p. 408) put it: “Actors do not necessarily
share the same goal, and even if they do, they do not have to be working
together consciously towards it (although some may be). Indeed, conflicts and
tensions are part and parcel of the dynamics of innovation systems.”

When scrutinizing innovation systems, the borders may of course be defined

narrowly enough to allow for intentionally constructed systems. Not the least

91
has this “voluntaristic” approach been popular among policy makers with
intentions to create innovation systems.

System borders

In our definition of innovation systems above we have situated the IS to an

economic space, i.e. the domain where all activities of an economy take place.
Inspired by the French economist François Perroux (1950), who originally
introduced the concept economic space, we choose not to a priori restrict our
definition to a territorial – genomic – space. The possible territorial anchoring
of an innovation system thus becomes an empirical question unless we do not
explicitly restrict ourselves to geographical proximity. Consequently,
geographically defined systems – like regional ones – thus become a sub-set
of the large family of IS. Similarly, national innovation systems become one of
many possible interpretations of an IS although – as we will come back to
below – the NIS concept is the original (Lundvall, 1992, Nelson, 1993).

One may argue that geographical borders become increasingly meaningless

when speaking about innovation processes. In this epoch of globalization:
what is left of territorially defined innovation systems when not only industrial
production but also knowledge formation takes place on a global scale? Is, as
formulated by the American journalist Thomas Friedman (2005), the world
in fact flat, i.e. can everything more or less take place everywhere with zero
friction and instant connection? And does product and service innovation
require ever more knowledge domains to be integrated, rendering the analysis
of specific technologies or sectors less meaningful?

While there may be more than a grain of truth in such statements, innovation
processes are nevertheless characterized by space as well as by the specificity
and logics of knowledge domains, and the IS concept tries to capture such
features not only statically but also over time. This means that each IS analysis
needs to define the borders of the system in question. By default, each such
attempt to delineate a system must carry its inaccuracies: There are no
absolute borders to an open system but only the more or less arbitrary ones
set for the purpose of analysis. In this way an IS is an analytical construct, and
not an absolute reality. In general, there are three dimensions that are used to
delineate any specific innovation system: geography, knowledge domain and
time. For each of these dimensions’ issues of level of analysis become
important. Depending on which of these starting points the analyst chooses –
geography, knowledge domain and/or time – he or she will see different
systems. Not the same set of components (institutions, actors, artefacts) or
relations will be included in the structural description of the IS, nor will
necessarily identical attributes of the components come to focus. This implies
that the choice of perspective from which to look at the innovation processes
will be crucial for the findings. For example, being interested in the emergence
of environmental friendly alternatives to the combustion engine, one may

92
partly stress dissimilar aspects having chosen different geographical scopes
and levels of analysis (e.g. the RIS of Western Sweden, the NIS of Sweden, or
the IS of EU). Clearly, the question at hand is what will guide the choice of
borders for the system (Carlsson et al., 2002).

What may by some be seen as a weakness of the IS framework – that there are
no clear guidelines as how to discriminate the system and its parts from its
environment – may also be considered the strength of the framework giving it
its flexibility. As we do not a priori (e.g. by definition) know what phenomena
or what components to include in the various innovation systems, the choice
of delineation will be essential and has to be well informed.

While these issues of border setting are decisive – and far from easy – they
also will lead to further questions relating to the level of analysis and
measurement matters. For example, should we only count condensations of
(which?) firms, or should we include (which?) individuals (and in what of their
roles)? What level of analysis should one choose for the demarcation of a
technological innovation system? How do we count regional industrial service
offices and technical universities? And how do we include change and
transformation in the systems analysis? These questions have been dealt with
in several methodological texts related to the IS concept (Carlsson et al., 2002;
Bergek et al., 2008, Hillman et al, 2011; Magnusson and Rickne, 2012), but
more is yet to be done in this vein of research.

Dynamics

A common set of critique towards the IS approach says that in spite of claiming
to evolutionary aspects of innovation there is a risk of rather static and
structural analyses. There are several reasons for that critique. One is the
relative lack of tools to capture dynamics in a structure as well as in the
function of a system. That problem has since long, been in focus among
systems analysts (cf. texts in Emery, 1969). In short the problematique may
be described as the relation between structure and function in defining the
system. As regards the structure – functional dimension in social systems this
was early identified by e.g. Merton (1949).31 Is the system basically defined
from its structure (its components) and if so, has the system changed – or is
it another system – if components disappear and/or transform? Or, is the
system basically defined from a set of essential functions or purposes with the
implication that the system is basically the same irrespective of structural
change as long as the basic functions are there? And more fundamentally:

31 And this is particularly problematic to combine with an evolutionary perspective, a

problem with which both natural scientists (biologists) and social scientists have
struggled. The usefulness of a functions concept differs between physical systems and
biological ones as well as between natural science related systems on the one hand
and social systems on the other (cf. Nagel, 1956 in Emery, 1969).

93
What are the functions of the system, and how can these be confined? This is,
of course, also relevant for the analysis of innovation systems.

Another problem for systems researchers in general is how to measure the

performance of a specific system as well as the relative importance of it specific
attributes. A third and related problem is how to obtain relevant data for
comparative analyses between systems that largely deviate from another.

In social sciences this balance between structure and function of systems –

and the epistemological consequences of that – has since long been in focus
for many disciplines: anthropologists (Radcliffe-Brown, 1952), sociologists
(Parsons and Smelser, 1956; Merton, 1949) and political scientists (Almond,
1960). The general problem is well illustrated with a case from political
science: in his classical paper from 1960 Almond identifies the political
systems more or less completely from its functions of which he identifies four
essential input, and three essential output, functions.

During the last decade, several authors have addressed this critique by
developing a functional approach to IS dynamics (Galli and Teubal, 1997;
Johnson, 1998; 2001; Rickne 2000; Johnson and Jacobsson, 2001; Liu and
White, 2001; Bergek, 2002; Bergek and Jacobsson, 2003; Carlsson and
Jacobsson, 2004; Bergek et al., 2008; Hekkert et al., 2007). In essence, these
researchers argue that each IS – basically their focus is on technological
innovation systems, TIS – can be portrayed by a number of functions
describing what actually ‘happens’ in the system. The set of functions
identified is from the beginning empirically derived. Complemented by
findings from e.g. sociology, organizational theory and political science the
functions arrived at have a broad resonance in the literature. The approach
describes how to assess the “functional pattern” of the IS: This implies to
determine how the key processes currently work in any specific IS. Bergek et
al. (2008, p. 414) stress that the “functional pattern of a TIS is likely to differ
from that of other TISs and is also likely to change over time. Thus, the concept
should not be interpreted as implying that the pattern is either repeated or
optimal.”.

The Swedish “TIS functionalists” – suggest that the systems are based on the
following functions (Bergek et al., 2008):

• knowledge development and diffusion

• influence on the direction of research
• entrepreneurial experimentation
• market formation
• legitimation
• resource mobilization
• development of positive externalities

94
This list can be argued and the definitions questioned: do the functions
constituting TIS differ from those of the NIS, the SIS or the RIS? Or, do the
set of functions differ over countries or time? While the functional approach
by no means solves the entire dilemma of measuring dynamics or
performance it is nevertheless a valuable step along that road. The
epistemological question on how to identify the functions of a system is indeed
not trivial. It may be argued – and has in fact also been – that such
functionalistic approaches run the risk that the list of functions gets an ad hoc
character, that it can be extended or reorganized with no obvious theoretical
ground. How do we validate that the 4+3 Almond functions make the tool box
complete?, or in general terms: is my list of essential functions more reliable
than yours? 32 Nevertheless this functional approach opens for research
activities to get answers on these questions.

This dilemma can be illustrated by the following “constructed” illustrations: if

an IS is assumed to be to less (highly) entrepreneurial – how do we obtain
knowledge of whether that is a consequence of the school system, the tax
system, the (lack of) protestant ethic, the welfare state or the innovation
policy? And if a NIS produces a lot of R&D results but is not outstanding in
GDP- development – how do we know what functions – if any – to blame for
that. In short how do we obtain scientific results that differ significantly from
pure prejudices?

Governance

Moreover, the functional approach among IS analysts may be looked upon as

one answer (among others) to the “governance dilemma”. This impasse relates
to the fact that there are many actors who individually or jointly endeavour to
put various schemes into practice (let us call these governance arrangements)
with the aim to influence innovation input, its processes and its outcome: We
call this process innovation governance. The dilemma as such lies in the
difficulties in guiding or controlling such complex and ever changing
innovation processes, where so many actors and knowledge areas are
involved. The governance concept is wider than public policy: We include all
actions and actors involved in guiding or directing the process at hand. This
implies that public policy actors may or may not be crucial actors: There is
nothing in the governance approach admitting a special role for public policy.
Instead, various types of actors may take on different responsibility depending
on e.g. technological or sectorial area, geographical conditions or current
trends.

32 TheAlmond (1960) input functions are 1) political adaption and recruitment; 2) articulation of
interest; 3) aggregation of interest, and 4) political communication. The output functions are a) rule
making; b) rule application, and c) rule adjudication.

95
Not the least may a functional approach contribute in relation to the
globalization problematique: sorting out the various functions to be filled
within an IS and making them comparable over space and time. Globalization
challenges, for example, the balance between the regional, national and supra-
national (Rickne, Laestadius & Etzkowitz, 2012). Which structural
components and which functions can be expected to be served at which spatial
level? Is it reasonable to anticipate clusters to emerge within e.g. biomedicine
at a multitude of locations around Europe: And what does this imply for
national policy actors striving to support such clusters? Another governance
issue related to globalization is what types of actors take what roles in pushing
the regional, national or supra-national agenda.

All in all: governance issues are central when it comes to innovation systems.
However, even given the possibilities of the functional approach, critiques
assert that while the institutional set up is used to delimit the systems
considered, institutional changes and influences, as well as governance
aspects, have more seldom been fully captured by IS analyses. In fact, current
theoretical development has not coherently shown how to handle these issues
(Jordan, 2008). In particular, there is a lack of integration with established
governance approaches (Treib et al., 2007; Newell et al., 2008) and the IS
framework. A recent attempt along this line was presented by Hillman et al.
(2009), merging IS theories with the multi-level perspective (Geels, 2002;
2004) and governance understanding to form a comprehensive framework.

The governance issue can also be read in the more policy focused IS
documents which, in particular, have been published by the OECD (OECD,
1999; 2002; 2005a-c). Already in the early OECD work on innovation systems
it was felt among policy makers that the institutional set up for policy was not
adequate for the transformations in innovation systems that were taking
place, in particular the growing importance of innovations and the decline of
the linear model (OECD, 1999; 2002).

Later OECD reports have extended that analysis and, based on scanning state
of the art practices in several – primarily small – OECD countries identified
the tensions in the innovation systems and the potentials for policy
formulation and integration (OECD, 2005 a-c). Explicitly the growing
importance of sustainability and information society policy is identified as
new areas to consider in the reorientation of IS governance. Among the
approaches intended to handle the transformative challenges related to
sustainability, and spanning over several layers of innovation systems, we
above mentioned the multilevel analyses by Frank Geels and colleagues
(Geels, 2002; 2004 & 2015).

96
Managerial and policy implications
The aim of this paper was to enhance our understanding of the theoretical
basis on which innovation policy in several countries rests. While we did not
aspire to synthesize or develop new theory, our intention was to remind those
who use the innovation systems tool box that it is important to be aware of the
shortcomings and problems of this analytical framework as well as its
advantages.

Modern innovation theories rest on an evolutionary view where learning

processes and path dependencies are key, and where variety creation,
selection and retention mechanisms are central and where equilibrium is
imaginary. In fact, this non-existence of equilibrium creates the tensions
which constitute the engine of innovation and change. The innovation system
concepts in its various variants, is a means for framing all these processes in
their context. The system approaches have, we argue, compared to earlier
equilibrium based theories, come closer to the balance between structure and
function in the innovation landscape. It seems that one important factor in
making the frameworks useful for governance of transformation and
innovation processes has been the continuous interaction between policy,
practice and theory.

Nevertheless, there are certainly several flaws in how the frameworks manage
to depict a complex reality: openness and intentions in systems as well as
delineation of systems borders remains problematic tasks for the analyst.
Another area is to what extent innovation systems can be created intentionally
or can/should be the main framework for innovation policy. 33 Our review has
aspired to highlight such difficulties, thereby giving analysts, governing actors
and managers a better ground on which to stand.

33 ForSweden these frameworks are ways to address issues of industrial dynamics. Indeed, while
several countries and supra-national units as OECD or the EU have utilized the innovation system
approaches to varying degrees during the last decades, Sweden is perhaps at the extreme end of this
spectrum in that the IS concept has been allowed to set the agenda for national and regional policies
and program details, and even to name a governmental organization, Vinnova.

97
References
Alexander, C. (1964) Notes on the Synthesis of Form, Harvard: Harvard U.P.

Almond, G. (1960) A Functional Approach to Comparative Politics, in

Almond, G. and Coleman, J. eds (1960) The Politics of Developing Areas,
Princeton, N.J.: Princeton U.P.

Amin, A. and Cohendet, P (2004) architectures of knowledge, Oxford:

Oxford [Link].

Amin, A. and Thrift, N. eds (1994) Globalization, institutions, and regional

development in Europe, Oxford: Oxford Univ. Press.

Asheim, B-T and Coenen, L. (2006) Contextualising Regional Innovation

Systems in a Globalising Learning Economy, Journal of Technology
Transfer, 31 (1)

Asheim, B-T Cooke, P. & Martin, R., 2010, Clusters and Regional
Development – Critical Reflections and Explorations, London:
Routledge.

Asheim, B-T and Gertler, M. (2005) The Geography of Innovation: Regional

Innovation Systems, in Fagerberg, J. Mowery, D. and Nelson, R. eds
(2005) The Oxford Handbook of Innovation, Oxford: University Press.

Asheim, B-T et al. (2007) “Constructing knowledge-based regional

advantage: implications for regional innovation policy”, International
Journal of Entrepreneurship and Innovation Management.

Asheim, B. T. and Herstad, S. (2003) Regional Innovation Systems,

Varieties of Capitalism and non-Local Relations: Challenges from the
Globalising Economy, in Asheim, B. T. and Mariussen, Å. (eds)
Innovation, Regions and Projects, Stockholm: Nordregio, R [Link]
241-274.

Benner, M. ed. (2005) Innovationer, Lund: Studentlitteratur.

Benner, M. and Sandström, U. (2000) Institutionalizing the triple helix:

research funding and norms in the academic system. Research Policy, 29
(2): 291-301.

Bergek, A. (2002) Shaping and exploiting technological opportunities: the

case of renewable energy technology in Sweden. Ph.D. thesis.
Department of Industrial Dynamics, Chalmers University of Technology,
Gothenburg.

98
Bergek, A. Jacobsson, S. (2003) The emergence of a growth industry: a
comparative analysis of the German, Dutch and Swedish wind turbine
industries, in: Metcalfe, S. Cantner, U. eds. (2003) Change,
Transformation and Development. Physica-Verlag, Heidelberg, pp. 197–
227.

Bergek, A. Jacobsson, S. Carlsson, B. Lindmark, S. and Rickne, A. (2008)

Analysing the Functional Dynamics of Technological Innovation
Systems: A scheme of analysis, Research Policy, 37 (3): 407-429.

Bijker, W.E. (1995) Of Bicycles, Bakelites and Bulbs: Toward a Theory of

Sociotechnical Change. Cambrifhge: The MIT Press.

Boschma, R., 2005, Proximity and Innovation: a critical assessment,

Regional Studies, 39(1): 61-74.

Breschi, S. and Malerba, F. (1997) Sectoral Innovation Systems:

Technological Regimes, Schumpeterian Dynamics, and Spatial
Boundaries, In: Edquist, C. ed. (1997) Systems of Innovation:
Technologies, Institutions and Organizations. London: Pinter

Brulin, G. (1998) Den tredje uppgiften – Högskola och omgivning i

samverkan, Stockholm: SNS Förlag and Arbetslivsinstitutet.

Brusco, S. (1982) The Emilian Model: productive decentralization and

social integration, Cambridge Journal of Economics, 6:167-184.

Bush, V. (1945) Science the endless frontier: a report to the President July
1945, Washinton: Gvt. Printing Office.

Carlsson, B. and Stankiewicz, R. (1995) On the nature, function and

composition of technological systems in Carlsson, B. ed. (1995)
Technological Systems and Economic Performance: The Case of
Factory Automation, Dordrecht: Kluwer.

Carlsson, B. (2000) Teknologiska system som analytisk ram – en

introduktion, in: Carlsson et al. (eds) (2000) Innovationssystem, kluster
och kompetensblock – fyra essäer om innovationer, tillväxt och
sysselsättning. Stockholm: Rådet för Arbetslivsforskning.

Carlsson, B. and Stankiewicz, R. (1991) On the Nature, Function and

Composition of Technological Systems, Journal of Evolutionary
Economics 1 (2): 93-118.

Carlsson, B. (2006) Internationalization of innovation systems: A survey of

the literature, Research Policy, 35: 56-67.

99
Carlsson, B. Jacobsson, S. Holmén, M. and Rickne, A. (2002) Innovation
Systems: Analytical and Methodological Issues, Research Policy 31 (2):
233-245.

Carlsson, B. and Jacobsson, S. (2004) ‘Dynamics of Innovation Systems-

Policy-Making in a Complex and Non-deterministic World’, Paper
presented at the ‘International workshop on Functions of Innovation
Systems’ at the University of Utrecht, June 23-24.

Cerin, P. (2004) Turning Corporate Behaviour Into Sustainability, Ph.D.

thesis, Stockholm: Kungliga Tekniska Högskolan (Royal Institute of
Technology).

Coenen, L. Moodysson, J. Asheim, B.T. (2004) Nodes, Networks and

Proximities: On the Knowledge Dynamics of the Medicon Valley
Biotechnology Cluster. European Planning Studies, 12 (7): 1003-1018.

Cooke, P. (2005) “Regionally asymmetric knowledge capabilities and open

innovation: Exploring “Globalisation 2” – A new model of industry
organisation”, Research Policy 34: 1128-1149.

Cooke, P. Boekholt, P. Tödtling, F. (2000) The Governance of Innovation

in Europe. Regional Perspectives on Global Competitiveness. London,
Pinter.

Dahmén, E. (1950) Svensk industriell företagarverksamhet, band 1, (IUI),

Uppsala: Almqvist and Wiksell.

Dertouzous, M. Lester, R. and Solow, R. (1988) Made in America –

Regaining the Productive Edge, Cambridge, Massachusetts and London:
The MIT Press.

Dosi, G. et al. eds (1988) Technical Change and Economic Theory, London
and New York: Pinter Publishers.

Edquist, C. ed. (1997) Systems of innovation: technologies, institutions and

organizations. London: Pinter.

Edqvist, C. and Johnson, B. (1995): Institutions and Innovations: A

Conceptual Discussion, Presented at Söderköping Conference within the
System of Innovation Research Network 1994-95.

Emery, F. ed. (1969) Systems Thinking, Harmondsworth: Penguin.

Etzkowitz, H. Webster, A. and Healey, P. eds. (1998) Capitalizing

Knowledge – New Intersections of Industry and Academia, New York:
State Univ. of New York Pr.

100
Etzkowitz, H. (2005) Triple Helix – den nya innovationsmodellen,
Stockholm: SNS Förlag.

Etzkowitz, H. and Leydesdorff, L. eds (1997) Universities and the Global

Knowledge Economy: A Triple Helix of University – Industry –
Government Relations, London: Cassell.

Etzkowitz, H. and Leydesdorff, L. (2000) The dynamics of innovation:

From National Systems and “Mode2” to a Triple Helix of university-
industry-government relations, Research Policy, 29 (2): 109-123.

Etzkowitz, H. Webster, A. and Healy, P., eds (1998) Capitalizing

Knowledge – New Intersections of Industry and Academia, Albany:
State Univ. of New York Press.

Fagerberg, J.; Laestadius, S. & Martin, B. (2015) The Triple Challenge for
Europe – Economic Development, Climate Change, and Governance,
London: Oxford Univ. Pr.

Fagerberg, J. Mowery, D. and Nelson, R. eds (2005) The Oxford Handbook

of Innovation, Oxford: University Press.

Faulkner, W. (1994) Conceptualising knowledge used in innovation: A

second look at the science-technology distinction and industrial
innovation, Science, Technology, and Human Values, 19 (4): 425-458.

Florida, R. (2002/2014) The Rise of the Creative Class – Revisited, New

York: Basic Books.

Florida, R. (2006) The Flight of the Creative Class – The New Global
Competition for Talent, New York: Harper Business.

Freeman, C. (1979) The determinants of innovation. Market demand,

technology and the response to social problems”. Futures (June): 206-
215.

Freeman, C. (1983/2005) “Technological infrastructure and international

competitiveness”, Industrial and Corporate Change, 2005, 13 (3): 541-
569 (reprint of old OECD report).

Freeman, C. (1987) Technology policy and economic performance:

Lessons from Japan. London and New York: Pinter.

Freeman, C. (2002) “Continental, national and sub-national innovation

systems – complementarity and economic growth”. Research Policy, 31:
191-211.

101
Freeman, C. and Lundvall, B-Å, eds (1988) Small Countries Facing the
Technological Revolution, London and New York: Pinter Publishers.

Friedman, T. (2005) The World is Flat. New York: Farar, Straus and
Giroux.

Galli, R. Teubal, M. (1997) Paradigmatic shifts in national innovation

systems in: Edquist, C. ed. Systems of Innovation: Technologies,
Institutions and Organizations. Pinter Publishers, London, pp. 342–
370.

Geels, F.W. (2002) Technological transitions as evolutionary

reconfiguration processes: a multi-level perspective and a case-study,
Research Policy 31(8-9): 1257-1274.

Geels, F.W. (2004) From sectoral systems of innovation to sociotechnical

systems: insights about dynamics and change from sociology and
institutional theory. Research Policy, 33: 897–920.

Geels, F.W. (2015) The Ardours Transition to Low Carbon Energy: A

Multi-Level Analysis of Renewable Electricity Niches and Resilient
Regimes, in Fagerberg, Laestadius & Martin, 2015.

Gibbons, M. et al. eds. (1994) The New Production of Knowledge. London:

Sage Publications Ltd.

Granovetter, M. (1985) Economic Action and Social Structure: the Problem

of Embeddedness. American Journal of Sociology, 91 (3): 481-510.

Gustavsson, L. (2009) Creating Advantage: On the complexity of

industrial knowledge formation in the knowledge-based economy,
Ph.D. Thesis, Kungliga Tekniska Högskolan (Royal Institute of
Technology) Stockholm.

Hallin, G. (2005) Sektor möter tillväxt – men hur och var?, Stockholm:
ITPS (Institutet för tillväxtpolitiska studier).

Hekkert, M.P. et al. (2007) Functions of innovation systems: a new

approach for analyzing technological change Technological Forecasting
and Social Change 74: 413–432.

Hillman, K, Nilsson, M, Rickne, A and Magnusson, T. (2011): Fostering

sustainable technologies – a framework for analysing the governance of
innovation systems, Science and Public Policy, Volume 38, Number 5,
June 2011, pp. 403-415

102
Hodgson, G. (1998) The approach of institutional economics, Journal of
Economic Literature, 36 (1): 166-192.

Howells, J. (1999) Regional Systems of Innovation? in: Archibugi, D. et al.

eds (1999) Innovation Policy in a Global Economy, Cambridge:
Cambridge University Press.

Hughes, T. (1992) The Dynamics of Technological Change: Salients,

Reverse Salients and Critical Problems, in Dosi, G. Gianetti, R. and
Toninelli, P.A. eds (1992) Technology and Enterprise in Historical
Perspective, Oxford: Clarendon Press.

Ingelstam, L. (2002) System – att tänka över samhälle och teknik,

Eskilstuna: Energimyndigheten.

Isaksen, A. (2005) Regional Clusters Building on Local and Non-Local

Relations: A European Comparison. In: A. Lagendijk and P. Oinas eds
(2005) Proximity, Distance and Diversity: Issues on Economic
Interaction and Local Development, Aldershot: Ashgate, pp. 129-152.

Johnson, A. (1998) Functions in innovation system approaches.

Unpublished Working Paper. Department of Industrial Dynamics,
Chalmers University of Technology, Gothenburg.

Johnson, A. (2001) Functions in innovation system approaches. In:

Electronic Paper at the Proceedings of the Nelson and Winter
Conference, Aalborg.

Johnson, A. Jacobsson, S. (2001) Inducement and blocking mechanisms in

the development of a new industry: the case of renewable energy
technology in Sweden. In: Coombs, R. Green, K. Walsh, V. Richards, A.
eds (2001) Technology and the Market: Demand, Users and Innovation.
Edward Elgar, Cheltenham.

Jordan, A. (2008) The governance of sustainable development: taking stock

and looking forward. Environment and Planning C: Government and
Policy 26: 17-33.

Kline, J. and Rosenberg, N. (1986) An Overview of Innovation”, in, Landau,

R. and Rosenberg, N. (1986) The Positive Sum Strategy, Washington
D.C.: National Academy Press. p. 275-305.

Laestadius, S. (1980) Produktion utan gränser, Stockholm: sekretariatet

för framtidsstudier.

Laestadius, S. (1992) Arbetsdelningens dynamik, Lund: Arkiv förlag.

103
Laestadius, S. (1999) Marshallian Dynamics and New Growth Theory,
Working paper, Stockholm: Kungliga Tekniska Högskolan (Royal
Institute of Technology): TRITA-IEO R 1999:1

Laestadius, S. (2000) Biotechnology and the Potential for a Radical Shift of

Technology in Forest Industry, Technology Analysis and Strategic
Management, 12 (2): 193-212.

Laestadius, S. (2005) Det dahménska utvecklingblocket – några

reflektioner”. in Benner, M. ed. (2005) Innovationer, Lund:
Studentlitteratur.

Laestadius, S. Gustavsson, L. and Long, V. (2008) High-Tech Innovations

in Catching-Up Countries: Conditions and Perspectives, in Hisch-
Kreinsen, H. and Jacobson, D. (2008) Innovation in Low-Tech Firms
and Industries, Cheltenham: Edward Elgar.

Leydesdorff, L. Etzkowitz, H. (1998) The Triple Helix as a Model for

Innovation Studies, Science and Public Policy, Vol. 25, No 3, 195-203.

Liu, X. White, S. (2001) Comparing innovation systems: a framework and

application to China’s transitional context. Research Policy 30: 1091–
1114.

Lundvall, B-Å. ed. (1992) National Systems of Innovation – Towards A

Theory of Innovation and Interactive Learning. London: Pinter
Publishers.

Lundvall, B-Å. (1994) The Learning Economy, Journal of Industry Studies,

1 (2): 23-42.

Lundvall, B-Å. and Maskell, P. (2000) Nation States and Economic

Development, in: Clark, G.L. et al. eds. (2000) The Oxford Handbook of
Economic Geography, Oxford: Oxford University Press, pp. 353-372.

Lundvall, B-Å. Edquist, C. (1992) Comparing the Danish and Swedish

Systems of Innovation, in Lundvall, B-Å. ed. (1992) National Systems of
Innovation – Towards A Theory of Innovation and Interactive
Learning. London: Pinter Publishers.

Magnusson, A. and Rickne, A, (2013): The Role of Policy Intervention for

Technology Management: The Case of Swedish Hybrid-Electric Vehicle
Technology, in Dilek Cetindamar, Tugrul Daim, Nuri Basoglu, Berna
Beyhan (Eds.) Strategic Planning Decisions in the High Tech Industry:
Methods and Cases, Springer-Verlag.

104
Malerba, F. ed. (2004) Sectoral Systems of Innovation, Cambridge:
Cambridge U.P.

Malmberg, A. and Power, D. (2005) (How) Do (Firms in) Clusters Create

Knowledge?, Industry and Innovation, 12(4): 409-431.

Markusen, A. (1996) Sticky places in a slippery space: a typology of

industrial districts, Economic Geography, 72: 293-313.

Marshall, A. 1890/1990 Principles of Economics: an introductory volume.

8th ed. London: Macmillan.

Maskell, [Link] al. eds. (1998) Competitiveness, Localised Learning and

Regional Development. London: Routledge.

Merton, R.K. (1949/68) Social Theory and Social Structure: toward the
codification of theory and research, New York: The Free Press.

Metcalfe, S. (1995) The Economic Foundations of Technology Policy:

Equilibrium and Evolutionary Perspectives, in Stoneman, P. ed. (1995)
Handbook of the Economics of Innovation and Technological Change,
Oxford UK and Cambridge USA: Blackwell, p. 409-512.

Mowery, D.C., and Rosenberg, N. (1993) The U.S. National Innovation

System, in R.R. Nelson, ed., National Innovation Systems: A
Comparative Analysis. New York: Oxford UP.

Nelson, R. (1993) (ed.) National Innovation Systems: A Comparative

Analysis. New York: Oxford UP.

Nelson, R. (1994) The co-evolution of Technology, Industrial Structure, and

Supporting Institutions, Industrial and Corporate Change, 3: 47-63.

Nelson R. and Winter, S. (1982) An Evolutionary Theory of Economic

Change, Cambr., Mass: Belknap Press of Harvard U.P.

Newell, P. Jenner, N. and Baker, L. (2008) Governing Clean Development:

A Framework for Analysis. Norwich, University of East Anglia.

Nilsson, J-E. Uhlin, Å. (2002) Regionala Innovationssystem: en fördjupad

kunskapsöversikt. VINNOVA Rapport, VR 2002:03.

Nilsson, M., Rickne, A., Hillman, K. and Magnusson, T. (2012): The road
ahead- Conclusions and governance implications, in Nilsson, M.,
Hillman, K. Rickne, A., and Magnusson, T., Paving the Road to
Sustainable Transport, – Governance and Innovation in Low-Carbon
Vehicles, Routledge, NY.

105
North, D. (1990) Institutions, Institutional Change and Economic
Performance, Cambridge: Cambridge U.P.

Nuur, C & Laestadius, S., 2009, “Is the “Creative Class” Necessarily Urban?
Putting the Creativity Thesis in the Context of non-Urbanized Regions in
Industrialized Nations, European Journal of Spatial Development
(online 2009-02-16).

OECD (1997) National Innovation Systems, Paris: OECD.

OECD (1999) Managing Innovation Systems, Paris: OECD.

OECD (2002) Dynamising National Innovation Systems, Paris: OECD.

OECD (2005a) Governance of Innovation Systems: Vol 1: Synthesis

Report, Paris: OECD.

OECD (2005b) Governance of Innovation Systems: Vol 2: Case Studies in

Innovation Policy, Paris: OECD.

OECD (2005c) Governance of Innovation Systems: Vol 3: Case Studies in

Cross-sectoral Policy, Paris: OECD.

Parsons, T. and Smelser, N. (1956) Economy and Society, Glencoe, Ill.: The
Free Press.

Pavitt, K. and Patel, P. (1991) Large Firms in the Production of Worlds

Technology: An Important Case of Non Globalization, Journal of
International Business Studies, 22: 1-21.

Perroux, F. (1950) Economic Space: Theory and Applications, Quarterly

Journal of Economics, 64: 89-104.

Polanyi, M. (1967) The Tacit Dimension, N.Y.: Double Day Anchor.

Porter, M. (1990) The Competitive Advantage of Nations. London:

Macmillan.

Porter, M. and van der Linde, C. (1995) Green and Competitive”, Harvard
Business Review, (Sept-Oct.): 120-134.

Prop. (1996/97:5) Forskning och samhälle (regeringens

forskningspolitiska proposition, gvt. proposal of R&D policy),
Stockholm: riksdagen.

Putnam, R. D. (1996) Den fungerande demokratin – medborgarandans

rötter i Italien. Stockholm: SNS Förlag.

106
Radcliffe-Brown, A. R. (1952) Structure and Function in Primitive Society,
London: Cohen and West.

Rickne, A. (2000) New Technology-Based Firms and Industrial Dynamics

– Evidence from the Technological System of Biomaterials in Sweden,
Ohio and Massachusetts, Doctoral thesis, Dept of Industrial Dynamics,
Chalmers University of Technology, Göteborg.

Rickne, A. et al (2009) Conceptualizing and measuring operative

interaction and collective action in a region, The Dahmén Institute,
Report to VINNOVA.

Rickne, A.; Laestadius, S. & Etzkowitz, H., eds., 2012, Innovation

Governance in an Open Economy – Shaping Regional Nodes in a
Globalized World, London: Routledge.

Saxenian, A. (1994) Regional advantage: culture and competition in

Silicon Valley and Route 128. Cambridge, Mass.: Harvard Univ. Press.

Schumpeter, J. (1934/62) The Theory of Economic Development, Cambr.,

Mass.: Harvard Univ. Pr.

Schumpeter, J. (1943/2000) Capitalism, Socialism and Democracy,

London and New York: Routledge.

Sellenthin, M. (2006) Beyond the Ivory Tower – A Comparison of Patent

Rights Regimes in Sweden and Germany, Linköping: Linköping
University.

Sharp, M. (1983) Europe and the New Technologies, London: Frances

Pinter.

Shinn, T. (2002) The Triple Helix and New Production of Knowledge:

Prepackaged Thinking on Science and Technology, Social Studies of
Science, 32/4, 599-614.

Simon, H. (1969/96) The Sciences of the Artificial, 3rd ed. Cambridge, Mass:
MIT Press.

Teece, D. (1986) Profiting from Technological Innovation: Implications for

Integration, Collaboration, Licensing and Public Policy, Research Policy
15: 285-305

Torre, A. and Gilly, J.P. (2000) On the Analytical Dimension of Proximity

Dynamics, Regional Studies, 34 (2): 169-180.

107
Treib, O. Bähr, H. and Falkner, G. (2007) Modes of governance: towards a
conceptual clarification, Journal of European Public Policy 14(1): 1 – 20.

Utterback, J. et al, eds (2006) Design Inspired Innovation, Singapore:

World Scientific.

von Bertalanffy, L. (1950) An Outline of General System Theory, British

Journal for the Philosophy of Science 1: 139-164 (reprinted in Emery, F.
ed. (1969) Systems Thinking, Harmondsworth: Penguin).

von Krogh, G. Kazuo, I. and Nonaka, I. (2000) Enabling Knowledge

Creation, New York: Oxford Univ. Pr.

Wittgenstein, L. (1953/89) Philosophical Investigations, Oxford: Basil

Blackwell.

108
Diffusion of Innovations
Emrah Karakaya & Pranpreya Sriwannawit Lundberg

How do innovations diffuse in societies and organizations? It is an

interesting yet a complex question to answer. In order to shed some
lights on this fundamental question, this book chapter provides the
foundations of diffusion of innovations theory along with two
empirical cases.

The theory is discussed through a three-component model of

diffusion: innovation, sources and adopters. It mainly builds upon
the seminal work of the well-known sociologist and communication
scholar, Everett Rogers (1962).

The empirical cases are based on diffusion of solar photovoltaic (PV)

systems. The first case is about off-grid solar PV systems in
Bangladesh, while the second case is about on-grid solar PV systems
in Germany. Overall, the differences and similarities between the two
case studies in developing and developed countries let us explain how
the characteristics of innovation, sources and adopters affect the
diffusion of innovations.

Introduction
We live in a world where we get to know new innovations every day. The smart
phone applications, computer software, daily habits, household appliances
and modes of transportation are just a few to name. Perhaps, they have been
there for a while, but it is until we adopt them that they become innovations
for us. However, why does one innovation get adopted but not the other? This
is a very important question worth pondering, especially for companies,
governments and individuals that attempt to kick off the diffusion of
particular innovations.

Diffusion often refers to how innovations propagate in a specific societal

environment. Diffusion framework has been used to explain the spread of
innovations, which is perceived as new by potential adopters (e.g., by
individuals or organizations). The “newness” poses uncertainty and
underscores the importance of diffusion study. Innovation and its diffusion
often have positive effects on social, technological and economic development.
However, innovation alone without the diffusion process would not have
much effect (Hall, 2005).

151
Apart from this introduction, this book chapter is structured as follows.
Section 2 presents the definition of diffusion. Section 3 explains the three
components of diffusion: innovation, sources and adopters. Section 4
introduces two case studies (one from Germany and the other from
Bangladesh) and explains how the diffusion theory can be applied in real life
contexts. Last but not least, section 5 discusses managerial and policy
implications, section 6 suggests some further readings and section 7 discusses
the methodological considerations.

What is diffusion?
The definition of diffusion is “the process in which an innovation is
communicated through certain channels over time among the members of a
social system. It is a special type of communication, in that the messages are
concerned with new ideas” (Rogers, 2003, p. 5). The seminal book Diffusion
of Innovations, by Everett M. Rogers (1962), has been the core of a large body
of literature in diffusion research. For decades, the theory has been used by
many scholars from a variety of disciplines, such as economics, sociology,
engineering, and management.

The diffusion is not just a passive process; rather, it involves a complex process
with incremental adjustments needed to make all the parts of a system fit
together (Nathan. Rosenberg, 1994). In fact, the process involves several
related activities even before the actual diffusion takes place (N. Rosenberg,
1976). In some cases, the activities of individuals, companies or policy makers
may result in some innovations that can be easily diffused by early and later
groups of adopters. The socioeconomic cost of adoption at a later stage of the
diffusion significantly becomes lower. Because, if the innovation continues to
diffuse, technical and commercial risks will be reduced and production costs
will decline.

Historically, diffusion has been an important element of ethnological research

addressing how cultural innovations have spread among cultures. It also tried
to shed light on whether specific innovations are the results of autonomous
processes or dependent on diffusion. Academically, diffusion has been used in
various branches of social sciences and has been given various definitions.
Studies in diffusion research have been focused on explaining the variables
that influence the adoption process. Diffusion and adoption processes are
highly complicated, having several unforeseeable and context-specific factors.
In addition, diffusion processes are almost always associated with something
new, magnifying the degree of uncertainty and, hence, resulting in
unpredictable outcomes. Therefore, studying diffusion, its pattern, its model,
its factors, and its effects can help scholars and practitioners to understand

152
better the process and to formulate increasingly more efficient, fruitful, and
productive innovation diffusion schemes.

When studying diffusion, it can be important to understand the innovation at

hand. For example: Is it public or private? Is the cost high or low? Is it
standardized or dynamic?

Public vs. Private

The diffusion of innovations might result in either private or public

consequences, or sometimes both. If there are some public consequences at
hand, the diffusion process usually involves several actors and collective
actions, such as policy makers, organizations, and social movements (Wejnert,
2002). For example, environmental innovations have both public and private
consequences, as they avoid or reduce environmental harms for the whole
society. The diffusion of environmental innovations also involves policy
makers, as it was seen in the case of the Europe 2020 strategy of European
Union (EU). As a public consequence, it is widely assumed that the diffusion
of environmental innovations will ensure future employment, contribute to
economic growth in Europe, and respond to today’s major societal challenges,
especially environmental (ETAP, 2010). In general, the diffusion of
innovations that have public consequences is a more complex and lengthier
process in comparison with those that have only private consequences.

High cost vs. Low cost

The cost of adoption of an innovation could be both monetary and

nonmonetary forms, and it is directly or indirectly associated with the
innovation (Wejnert, 2002). In general, for high-cost innovations, the
economic aspects are often the most important factors for understanding the
diffusion of such innovations (Rogers, 2003). When the cost of adoption of an
innovation is perceived as high, the potential adopter needs to understand
clearly the outcomes, benefits, risks, and costs, both in short-term and long-
term, associated with the adoption. That is why high-cost innovations require
more time to be diffused and involve more actors for the diffusion process in
comparison with low-cost (non-cost) innovations.

Standardized vs. Dynamic

In some cases, innovations do change or get modified in the process of

diffusion. The change can be so continuous that the innovation gets modified
in every adoption that takes place in space and time. For instance, an
innovation can evolve during the diffusion phase when the adoption is highly
restricted by adopter’s settings. Some possible examples could be cornrows
hairstyles, as adopted by individuals; solar PV systems, as adopted by
households; and internet websites, as adopted by firms. However, some other

153
innovations can be standardized in one or a limited number of forms,
appearing in the market on an incremental basis, e.g., automobiles, Facebook,
or smart phones. In this case, the innovation does not represent a fully
heterogeneous form in space and time, instead, it is an incremental one.

Three Components of Diffusion

In the history of science, diffusion as a term was primarily used in the
chemistry and physics in order to define the process of movement of molecules
from high-concentrated regions to other regions (for example Fick's laws of
diffusion from 1855). But, can the diffusion of innovations conceptualized as
similar to the diffusion phenomena in physics and chemistry? Perhaps to
some extent, but not completely. In this book chapter, however, we refer to
some analogies between diffusion of innovations (in social world) and
diffusion of molecules (in physical world). For example, Figure 1 illustrates a
three-component model of the diffusion of innovations: source (in analogy
with high concentrated regions), innovation (in analogy with molecules) and
adopter (in analogy with low concentrated regions).

Figure 1. Simplified diffusion process (Sriwannawit & Laestadius, 2015)

Innovation

Innovation is the core element of diffusion of innovation process. The

assumption is that innovation can diffuse among potential adopters.
Innovation is a very broad term and indeed can be anything as long as it is
perceived new by adopters. The variety of innovations that are analysed in the
literature is evident in the early scholars of diffusion. For example, Griliches
(1957) analysed the diffusion of hybrid seed corn, which is a product of
controlled crossing of specially selected parental strains among American
farmers. Mansfield (1961) studied the diffusions of twelve types of
innovations, such as a trackless mobile loader and a continuous mining
machine among firms in particular industries.

The availability of a new technology or innovation does not necessarily

motivate its adoption by individuals. The perceived attributes of an

154
innovation, which is contingent upon the adopters, explain 49-87% of the
variance on the different diffusion rates of different innovations (Rogers,
2003; Tidd, 2009). These attributes are relative advantage, compatibility,
complexity, trialability, and observability.

• Relative advantage refers to the degree to which an innovation is

perceived to be better than the incumbent idea, technology, or practice
and is usually expressed as economic profitability. However, non-
economic factors (e.g., quality, satisfaction, environmental awareness
and social prestige) are also important, as it is the case of the PV
diffusion (e.g. Jäger, 2006a; Müggenburg, Tillmans, Schweizer-Ries,
Raabe, & Adelmann, 2012; Palit, 2013). Greenhalgh et al. (2004)also
emphasize that potential users will likely not consider the innovation if
they do not see relative advantages, which are foremost measured in
economic returns. Moreover, there are decisive social factors such as
user satisfaction and prestige that influence an individual’s perception
of the relative advantage of innovations.
• Compatibility is the degree to which an innovation is perceived as being
consistent with the existing values (e.g., sociocultural values and
beliefs), past experiences (e.g., previously introduced ideas), and the
needs of potential adopters. Several empirical studies points to a direct
relationship between the compatibility of an innovation and its
adoption (McEachern & Hanson, 2008; Shum, 2013).
• Complexity is the degree to which an innovation is perceived as being
relatively difficult to understand and use. Generally, there is an inverse
relationship between the perceived complexity of an innovation and its
adoption rate (Kai-ming Au & Enderwick, 1999; Labay & Kinnear, 1981;
Völlink, Meertens, & Midden, 2002).
• Trialability is the degree with which an innovation may be
experimented on a limited basis. Innovations with high trialability often
have a higher diffusion rate (Makse & Volden, 2011; Rogers, 2003),
although some other studies (Labay & Kinnear, 1981; Völlink et al.,
2002) indicate an absence of a relationship between trialability and the
adoption of innovations in the energy sector.
• Finally, observability is the degree to which the results of an innovation
are visible to others. According to Tidd (2009), the rate of adoption of
an innovation increases when it is easier to see the benefits of this
innovation.

Even though these attributes have been widely studied and used in scholarly
research, they have sometimes been considered as abstract and irrelevant for
practitioners. These attributes are, in fact, simple to understand and can be
operationalized in practical cases but it is crucial that the attributes be
analysed as ‘perceived’ not as ‘intrinsic’ (Weiss & Dale, 1998).

155
Sources

Sources of diffusion could be the individuals, organizations or institutions at

which the diffusion process originates. They can be innovators, government
agencies, companies or any individuals. Sources are usually important change
agents – that influences the decisions of potential adopters in a desirable
direction- in the diffusion process. This means that potential adopters can be
influenced to adopt an innovation by the pressure of the social system
generated via adopters, public policies, shareholders and organizations (Bass,
1969; Frondel, Horbach, & Rennings, 2008). Nevertheless, change agents’
efforts and the adoption rate of a technology do not always correlate. More
payoffs of change agents’ efforts occur at specific stages. The highest response
to the efforts is when an opinion leader – a relatively well-known individual
or organization that has the ability to influence the opinions of others – adopts
the change.

In the literature, the common sources of diffusion are often conceptualized as

policy measures, previous adopters and firms. In the following bullet points,
we exemplify these.

• Policy measures: Policy makers can foster specific regulations to guide

suppliers and adopters to choose an innovation or a specific design,
which in return may increase the economical relative advantage of the
innovation. For example, in the case of environmental innovations, the
policy can have the capacity to create expectations for an innovation
that is essential to foster the diffusion. In this context, until 2012 in
Germany, policy makers provided clear signals regarding the growth
potential of solar PV systems through the implementation of feed-in
tariff (Dewald & Truffer, 2012; Hoppmann, Peters, Schneider, &
Hoffmann, 2013; Staffan Jacobsson & Lauber, 2006). Such tariffs
influence the perceived economic relative advantage of solar PV and,
therefore, the diffusion (Jäger, 2006b).
• Previous adopters: In general, potential adopters are often influenced
by what they see and hear from their peers. The peer effects has
received a lot of attention from scholars analysing the academic and
health practices (e.g. Trogdon, Nonnemaker, & Pais, 2008;
Zimmerman, 2003). In the case diffusion of PV systems, several
scholars (e.g. Bollinger & Gillingham, 2012; Graziano & Gillingham,
2014; Müller & Rode, 2013) have also asserted that the peers, who have
already adopted the PV systems in the same neighbourhood, increase
the diffusion rate among potential adopters. This resonates quite well
with the recent work of Pentland (2014) on social physics. He
emphasizes that innovations and good ideas spread faster if the
potential adopters have regular physical, e.g. face-to-face, interactions.
In a well-connected global world, although the potential PV adopters

156
 might have tight virtual connections with their peers from other
regions, the most of physical interactions takes place in the local
neighbourhood. As Graziano and Gillingham (2014) also argue, such
spatial dimension is especially critical for innovations that have both
private and public good characteristics, e.g. solar PV systems.
• Firms: The primary economic function of an industrial firm is “making
use of productive resources for the purpose of supplying goods and
services to the economy in accordance with plans developed and put
into effect within the firm” (Penrose, 2009, p. 12). In the renewable
energy sector, the activities of firms not only result in supplying goods
and services but also in the creation of new knowledge and the
development of different types of designs (S. Jacobsson & Bergek,
2004). The firms are important stakeholders in the development of
renewable energy policies, and therefore, they try to influence the
political decisions about the design of financial support systems and the
grid access (Wüstenhagen, Wolsink, & Bürer, 2007, p. 2686). For
example, in Germany, the energy utility firms are well- known for their
effective lobbying strategies for renewable energy, such as, regular and
personal maintenance of contact to politicians and forming a policy
network through associations (Sühlsen & Hisschemöller, 2014). Firms
could be conceptualized as change agents, which influence the opinion
of potential adopters. For the meso/macro level analysis, firms could
be conceptualized as actors or complementary inputs. Whether the
analysis is at micro or meso/macro level, local solar firms are often
identified as important drivers of the diffusion of solar PV systems. For
example, in the case of PV systems in Germany, Dewald and Truffer
(2012) identified that local solar firms stimulated rapid diffusion and
market formation. They argued that the successful diffusion is not only
driven by the strong policy support and favourable geophysical
conditions, but also by the market formation activities of local solar
firms. This is in line with the study of Fabrizio and Hawn (2013) which
analysed the role of local solar companies in the USA. Conceptualizing
the local firms as complementary inputs, they identified that local firms
do function as important drivers of diffusion of PV systems.
Adopters

Adopters are the individuals, organizations or institutions who take up or

embrace the innovations. This means that adopters are the individuals or
other units of adoption that make the decision to adopt the innovation
(Rogers, 2003). Any individual can be a potential adopter, however, not all
potential adopters necessarily decide to embrace the innovation at any certain
point of time. 2 shows the points in time at which different groups adopt
innovations. The distinction between the different adopter groups is based on
behavioural characteristics regarding innovativeness and corresponding to

157
normal distribution statistic percentages. Innovators are those who want a
certain product as soon as it becomes available, are willing to take risks, and
have financial capabilities. They are cosmopolites who act and have contacts
regionally and globally. Early adopters are a larger group who also seek new
products but are less sensitive to “hype” (e.g., they look more into
functionality). They are often local and generally have the highest degrees of
opinion leadership (Rogers, 2003). The early majority is the first mass of
people to adopt a product, and this is where the curve reaches maturity. The
late majority adopts when the majority of the market is already familiar with
the product. Sales tend to slow during this phase. Finally, the laggards adopt
when the product is soon to be removed from the market. These adopters are
more price-sensitive and sceptical (Rogers, 2003).

Figure 2. Adoption curve and the chasm. Adapted from Rogers (2003) and Moore (2002)

Typically, the adoption process begins relatively slowly, but once a critical
mass is reached, it becomes an automatic mechanism that forms an S-shaped
curve. This critical mass is one reason that, after a relatively slow start, the
rate of adoption can form an S-shaped curve. During the diffusion process,
influence can be exerted between adopter groups. Individuals who are
influential within the social system and who spread information about an
innovation are defined as opinion leaders and are generally to be found among
the early adopters (Rogers, 2003).

According to Moore (2002), there can be gaps between different adopter

groups. One of the most important gaps is between the early adopters and the
early majority, defined as the chasm (see Figure ). This occurs when a new
product or service cannot be translated into a significant benefit (Moore,
2002). Sometimes, early adopters can create bad references for the early
majority. For example, StudiVZ of Germany (founded in 2005, a pioneer
social networking platform – similar to Facebook) has failed to cross the
chasm between its early adopters (in Germany, Switzerland and Austria) to
the early majority (worldwide individuals elsewhere). However, on contrary,

158
Facebook (founded in 2004) has successfully crossed the chasm between its
early adopters (the students in the US colleges) to the early majority
(worldwide individuals elsewhere).

The diffusion rate of innovations, which is the ratio of current number of

adoptions to the total number of potential adoptions, is mainly based on the
attributes of an innovation as perceived by potential adopters (as discussed
previously). Besides the attributes of innovation, there are some other factors
that can affect the adopters. Such as:

• Types of innovation-decision: affect whether an innovation should be

adopted or rejected. They can be optional decision made by individuals
independently from other members; collective decision reached by
consensus with others; and authority decision made by relatively few
members who possess status, power or expertise. These three types are
arranged on a continuum indicating the power that each individual
possesses in decision-making – from complete power via partial power
to no power at all. The rate of adoption decreases when there are more
people involved in the decision process. However, the diffusion process
may involve a mix of all of these decision‐making types, depending on
the type of technology, regulations and adopters, as is sometimes the
case of the renewable energy technologies in different countries (Bodas-
Freitas et al., 2010; Reardon, 2009). For example, if there is an
innovation to build on an apartment, multiple adopters (who live in the
same apartment) may have a say on the final decision.
• Communication channels: affect the innovation-decision process.
Communication is the process through which information is created,
received, and shared. A main goal is to achieve mutual understanding
among the participants. Interpersonal communications (including non-
verbal observations) and mass media channels (television and internet)
are important influences on the diffusion rate of the innovations in a
social system. The former is more powerful in convincing a social
system to accept a new innovation (Mahajan, Muller, & Bass, 1990;
Rogers, 2003). Communication between adopters and the observability
of the adoptions can induce peer-effects, whereby the decision of
potential adopters may be influenced by the previous adopters
(Bollinger & Gillingham, 2012). Recent literature has paid much
attention to how peer-effects influence the diffusion of PV technology
(Müller & Rode, 2012; Rode & Weber, 2012).
• Nature of the social system: can affect the adoption. A social system is
a group of interrelated actors. Rogers (Rogers, 2003, p. 24)defines a
social system as “a set of interrelated units that are engaged in joint
problem-solving to accomplish a common goal”. The members of a
social system may be individuals, informal groups, organizations
and/or subsystems. Units in a system can differ in their behaviour by

159
 means of homo- and heterophily: homophily refers to the similarity
between individuals, e.g., regarding education level, beliefs and social
status, whereas heterophily is when individuals differ on these
attributes. One distinctive challenge is that individual adopters are
often heterophilous. This makes it difficult for an actor attempting to
diffuse an innovation to choose only one single approach (Rogers,
2003).

Potential adopters can be influenced to adopt an innovation by the pressure

of the social system generated via adopters, public policies, shareholders and
organizations (Bass, 1969; Frondel et al., 2008). Some recent research have
identified the effects of network externalities as being significantly important
for the diffusion rate of innovations (Goldenberg, Libai, & Muller, 2010; Van
den Bergh, 2013).

The adopters are crucial in shaping and diffusing technologies. For example,
collaboration of users in energy technology can lead to greater diffusion,
technological development, and possibly new product innovation (Ornetzeder
and Rohracher 2006). Nevertheless, since the 1980s, when there was an
attempt to implement renewable energy technology, social acceptance has
been neglected. Initial surveys of public acceptance of renewable energy, in
fact, showed very high support of renewable energy. However, in reality, a
large difference exists between general public support and support for specific
projects or implementations. Policy makers have been misled by this
discrepancy between general positive support and low support for specific
contexts directly affecting individuals. Moving from general to specific and
from the global or national to the local level, direct investigation of public
support of local sites should be investigated. Acceptance of general projects
and resistance to specific implementation may occur because people tend to
support renewable energy as long as it is not in their backyards. Therefore,
specific cases must be investigated to provide understanding of local contexts
(Wustenhagen et al. 2007). The inclusion of users in technology diffusion
research can provide better understanding of environmentally friendly
technologies and enhance dissemination. Recently, there has been an interest
in academia in investigating social acceptance and users’ perspectives through
the case-study approach (see e.g. Rohracher 2003 in Austria; Mallett 2007 in
Mexico; Muggenburg et al. 2012 in Ethiopia; Shyu 2013 in China). For small-
scale renewable energy technologies, like household PV systems,
Wustenhagen et al. (2007) suggested diffusion of innovation literature
(Rogers 1995) can help in understanding social acceptance because it can be
regarded as market acceptance or a market adoption process. Nevertheless,
because of the nature of energy technologies, which are constrained to other
infrastructures, their diffusion can be more complicated than that of other
products.

160
The roles of potential users are of high importance for all four elements of the
diffusion: innovation, communication channel, time, and social system
(Rogers, 2003). First, in terms of the innovation itself, understanding users’
needs is a vital determinant of its success. Users’ involvement in the
innovation process leads to development of a more valuable innovation,
enhancing its adoption. Therefore, integrating users from an early stage and
actively engaging them throughout the process are necessary. Second, an
appropriate communication channel must be selected for effective facilitation
of the diffusion process. Along with the communication channel, the cognitive
distance between the source and the recipient and users’ absorptive capacity
should be taken into account. Third, the time and its three sub-aspects are
strongly connected and highly dependent on users. The first sub-aspect, the
decision-making process, involves users’ deciding whether to adopt or reject
the innovation. The second aspect is the relative earliness or lateness of its
adoption by other members in a social system. The last aspect is the rate of
adoption, measured directly by counting the number of users who choose to
adopt the innovation within a certain time frame. Finally, users are obviously
the main part of the social system. Without users, the system would not be
complete because the function of the system would not be fulfilled.

Empirical cases: Solar Photovoltaic Systems

In today´s world, we are facing several environmental and societal challenges-
at which understanding diffusion of innovations could be a part of the
solutions. For example, concerns about climate change and limited resources
of fossil fuels have prompted governments to support the emergence and
diffusion of renewable energy systems. At the same time, there are still 1.3
billion people living without electricity (IEA, 2012). Generating electricity
from solar energy, an abundant and renewable source, using photovoltaic
(PV) systems is one means to tackle these challenges.

A new technology’s being superior to the preceding technology does not

guarantee its successful dissemination, as is the case for PV systems (Miller
2009). Photovoltaics (PV) is an electricity generation method using solar
radiation. Based on grid connectivity, PV systems can be separated into two
categories: on-grid and off-grid. An on-grid PV system is integrated into the
utility grid. During the day, electricity is generated via the PV system. The
excess amount is transferred to the national grid for general use at any grid-
connected location. When the PV system cannot generate sufficient electricity,
the electricity can be taken from the utility grid; thus, a battery is not a
necessary component. On the other hand, an off-grid PV system is used in
remote areas where the grid is unavailable; hence, a battery is an important
part of the off-grid system. During the day, solar energy is collected and
transformed into electricity. The excess amount of electricity is stored in

161
batteries for later use. Despite the use of PV technology for rural electrification
since the 1960s (Lorenzo 1997) together with its rapid technological
development and price decline during the last few years, its use is still low
compared to conventional energy sources (REN21 2014). The rapid price drop
has made PV systems an attractive energy technology, even among low-
income inhabitants. However, the significant price decline has not been fully
appreciated by relevant actors, such as policy makers and users who still
perceive a PV system as an expensive option (Bazilian et al. 2013).

In the following two sub-sections, we present two case studies. One of them is
about off-grid solar PV Systems in Bangladesh, while the other is about on-
grid solar PV systems in Germany (see Table 1).

Table 1. Comparison of two case studies

Case Innovation Main initial Adopters

Study source

A Off-grid solar Grameen Shakti Some local

PV Systems Bangladeshis

B On-grid solar Hartmann Some local

PV systems Energitechnik Germans
GmbH

These two cases from two different continents have been selected to represent
the diffusion process of PV systems because of three main reasons. Firstly, the
underlying technology are similar (solar PV systems) but the applications are
different (on-grid vs. off-grid). Secondly, the sources in both cases are of the
same type (i.e., private firms) but they undertake different firm strategies.
Thirdly, the adopters in both cases are householders, however with different
socioeconomic characteristics. Overall, all these differences and similarities
between two cases let us explore how vital the characteristics of innovation,
source and adopters are to the diffusion of innovations.

Case study from Bangladesh

This case study is mainly based on Grameen Shakti, a local energy company
operating in rural Bangladesh. Bangladesh is an overcrowded country with
approximately 160 million inhabitants. One-third of the population lives
below the poverty line (CIA, 2013). More than half of the rural population do
not have access to electricity (IEA, 2014). In addition, power services in rural
areas are of poor quality. There is a great need and a large market for reliable

162
electricity at an affordable price. This is a potential for a stand-alone off-grid
energy technology like Solar Home System (SHS), which is a set of
photovoltaic systems installed on the rooftop to convert sunlight into
electricity. It is a self-contained generation and distribution system that
generates much lower maintenance and operating costs than fossil fuel
alternatives (Cabraal, Cosgrove-Davies, & Schaeffer, 1998).

Case study in Bangladesh analyses the diffusion of energy technology for rural
electrification. It is a highly relevant and timely challenge for many
communities in developing countries. We focus on the diffusion of SHS in
particular among the population with low income or those belonging to the
“bottom of the pyramid” (BOP). Strong focus in the BOP discourse has been
placed on international business or multinational companies (Prahalad, Di
Benedetto, & Nakata, 2012; Prahalad & Hammond, 2002)but it tends to
neglect the role of government (Bawakyillenuo, 2009) and local entrepreneurs
or industrial activities within developing countries (Linna, 2012).

Grameen Shakti was founded by Muhammad Yunus, also the founder of

Grameen Bank and recipient of the Nobel Peace Prize for his microfinance
concept. In Yunus’s opinion, development must be based on a positive
economic change in the bottom half of the population (Muhammad, 2003).
Consequently, Grameen Bank as well as Grameen Shakti focus their activities
on the BOP (Prahalad, 2010) . GS was established to disseminate affordable
and renewable energy technologies to rural and remote areas in Bangladesh.
The company has become one of the largest of its kind in the world (Asif &
Barua, 2011; Sovacool & Drupady, 2011) with approximately 1,500 offices and
12,000 employees covering all 64 districts of Bangladesh. The diffusion of
GS’s first SHS began in 1996. By the end of 2013, GS had provided SHS to
more than 1.3 million rural households across the country (GS, 2010). SHS
has successfully penetrated the rural Bangladeshi market with no direct
subsidies (Kolk & van den Buuse, 2012). On the contrary, SHS is still an
emerging technology with low dissemination in most countries (Rebane &
Barham, 2011). This motivates us to learn more regarding the factors that
affect the high adoption rate in Bangladesh, which we discuss in the following.

Miller (2009) identified two types of capital that are barriers to entrepreneurs
dealing with solar energy. First, market infrastructure capital—such as local
branches, sales people, and technicians—is necessary. Second, customer
finance is important in order to make the products affordable for the target
group. Grameen Shakti has overcome these two barriers, and its diffusion is
at a large nationwide scale. The analysis here focuses on the characteristics of
the source that are specific to the BOP market, here identified as a social
entrepreneur. These source characteristics that equip Grameen Shakti with
competitive advantage comprise four items: (a) unique company background
in development work related to the source’s history, leader, and values; (b)
good qualifications and strong effort of change agents highly involved with

163
local communities, both before and after the adoption; (c) affordable and
manageable economic schemes specifically designed for the low-income
adopters; and (d) a distribution system quick to respond to adopters’ financial
situations. The source characteristics are assumed to be the most important
factor affecting the high adoption rate of PV systems in Bangladesh because
they are unique and they specifically address the BOP market. Understanding
the source characteristics can equip various types of actors with the
capabilities to begin a diffusion process in the low-income segment. The
Grameen Shakti case reveals that a local social entrepreneur can drive the
diffusion process by building a previously unattractive market to become a
market with large potential adopters. The social characteristics of Grameen
Shakti may be largely because of its founder, Muhammad Yunus, and his
‘innovation’ of the microfinance concept. As opposed to Prahalad’s BOP
concept, which may be considered a top-down approach, Grameen business
may be considered a complementary perspective, focusing on capability
building in local firms rather than waiting for investment from multi-national
companies.

In summary, this case study shows that the source could be conceptualized as
a social entrepreneur. The source characteristics may be the most important
factor that affect GS’s adoption rate because they are unique and specifically
address the BOP market. This market has great potential but also contains
vulnerable consumers. Social entrepreneurship is one manner in which to
approach this market. GS’s strategies have overcome a common barrier of
high initial cost. The GS case reveals that a local social entrepreneur can drive
the diffusion process by building up from a previously unattractive market to
become a market with large potential adopters.

Case study from Germany

Germany is a leading producer of solar photovoltaic-power, accounting for

one-third of the global photovoltaic (PV) systems’ installed capacity (GTAI,
2013). Although Germany has a relatively limited solar radiation potential, it
has outperformed countries that have a larger land area and higher solar
radiation, such as France, Spain, Turkey, Brazil and Japan. The mechanisms
behind the German story of PV systems’ diffusion have been discussed at
length in the literature (Dewald & Truffer, 2011; e.g. Staffan Jacobsson &
Lauber, 2006; Karakaya, Nuur, Breitschopf, & Hidalgo, 2014). One of the
most important drivers of the diffusion process has been the activities of local
solar companies (Dewald & Truffer, 2012). According to some studies (Lettner
& Auer, 2012; Pérez, Cervantes, Báez, & Domínguez, 2013; Spertino, Leo, &
Cocina, 2014), solar PV energy in Germany has already achieved grid parity
by 2012, i.e., solar PV energy can directly compete with conventional
electricity sources in terms of the levelized cost of electricity generation.

164
The German feed-in tariff scheme is widely accepted as the strongest driver
for the diffusion of PV since 2000 (Dewald & Truffer, 2011). This scheme
ensures that solar PV adopters (when they supply electricity to a grid) are paid
with fixed feed-in tariffs over 20 years, beginning from the time the solar PV
systems are connected. However, the feed-in tariff for solar PV systems has
decreased more rapidly than that for any of the other renewable energy
technologies (Wirth, 2013). Although solar PV systems in Germany are often
assumed to be at grid parity, the PV market has recently faced uncertainties
related to the cuts in the feed-in tariff.

The focus of the case study is the source of innovation, Hartmann

Energitechnik GmBH (HET) in Germany. This company has been chosen for
two reasons. Firstly, it is located in the spatial lead market of PV systems’
diffusion in Germany (Karakaya et al., 2014). Secondly, it has witnessed the
diffusion of PV systems since 1990s as being one of the pioneer local solar
companies in Germany (Karakaya, Hidalgo, & Nuur, 2015). It has been a part
of solar initiative movement (Drücke, 2004) and featured in several energy
magazines (SB, 2007; e.g. SWW, 2004).

The company offers not only PV systems, but also some other technologies,
including solar thermal systems, pellet ovens etc. As shown in it is located in
a rural area, near a village, called Oberndorf in Rottenburg (Am Neckar). It is
a local solar company founded in 1995 by a local entrepreneur, Thomas
Hartmann, a native of Oberndorf. He is also the co-founder of two solar
initiatives: Solar-Partner e.V and Sonnenhaus-Institut e.V. The former is a
network of companies, freelance solar consultants and partner companies.
The latter is an association of architects, engineers and managers of the solar
industry, focusing on solar-heated and solar-electrified buildings.

This case study demonstrates several important motivators for adoption.

Achieving grid parity does not necessarily motivate potential adopters for PV
systems. As discussed by Rogers (2003), for wide adoption, innovations
should not only have high compatibility, high trialability, high observability,
high relative advantage and low complexity but also be communicated and
driven by change agents, e.g. adopters and other actors.

Most of the adopters in our case have been motivated for PV systems in order
to be self-sufficient and independent from electricity supply. Such desire is
often complemented with environmental awareness, peer effects and financial
stability. This is in line with previous research, which investigated similar
phenomenon in other countries before grid parity (Balcombe, Rigby, &
Azapagic, 2013, 2014; Jäger, 2006a). In our case, the increasing electricity
retail prices, as influenced by policy measures, have motivated potential
adopters to be less dependent on electricity supply. In this context, PV systems
have been often perceived to be compatible with the desire of potential
adopters towards being self-sufficient and independent.

165
Moreover, complementing the previous research (Dewald & Truffer, 2012;
Fabrizio & Hawn, 2013), the local solar company is identified to be an
important motivator for adoptions. The vision of the company and its local
entrepreneur has reduced the perceived complexity of PV systems. Several
activities organized by the local entrepreneur, eg., solar-walks and open-door
days, have also increased perceived trialability and the observability of PV
systems. Because PV systems require some level of knowledge about the
technology and its operation, a high level of communication between local
solar companies and adopters is a key factor to minimize the perceived
complexity and to facilitate the decision made by the adopters.

Managerial and policy implications

The diffusion of innovations in general and solar photovoltaics in particular
have several managerial and policy implications. This book chapter provides
two examples focusing on the source of innovation. The implications are
drawn from previous studies and two cases: one from a developing country
and the other from a developed country. The case studies discussed in this
chapter suggests that the involvement of all stakeholders—adopters, local
communities, firms, international organizations, financial institutions, and
government— are crucial when it comes to high-cost dynamic innovations
with public consequences. Without proper collaboration, effective marketing,
and dedicated government support, the barriers to adoption cannot be easily
overcome. However, technology diffusion is context specific and so is the
adoption of PV systems. Therefore, in general, the local conditions of a
particular case of an innovation should be well understood by companies,
governments and individuals who are in favour of wide diffusion.

Suggested further readings

Diffusion research originated in and has been conducted largely by scholars
in the advanced parts of the world. Various topics related to the diffusion of
innovation have been studied since the 19th century (Boas 1896; Tarde 1903)
and continue to be studied in academia. The framework is highly
interdisciplinary in character. To date, this field has been addressed in a large
amount of literature spanning such fields as folklore, cultural change,
economics, technological change, and so on (Sriwannawit and Sandström
2015). Such depth and breadth of interest in diffusion research may be
because of its wide application, the existence of many issues needing
resolution, the specificities across cases, and the degree of generalization,
among others.

166
A number of other studies can be considered complementary, filling in gaps
in the diffusion framework. Therefore, we suggest the following studies as
further readings:

• Cohen and Levinthal’s work on absorptive capacity (1990) addressed

the importance of users’ capability in the exploitation of new knowledge
based on related prior background. This work implies that adoption
requires appropriate knowledge and some form of training. Their
concept has been widely applied at the firm and organizational level.
• Nonaka and Takeuchi (1995), together with Jaffe, Trajtenberg, and
Henderson (1993), help to fill in this complex understanding of the
creation, utilization, and diffusion of knowledge, which is an intangible
matter. This intangible aspect of diffusion may be lacking in Rogers’
work.
• The work of Wejnert (2002) who categorizes the variables that
influence diffusion in three major groups: environmental settings (such
as geographical and sociopolitical context), the characteristics of
innovations, and the characteristic of adopters.
• Beise (2001, 2004)’s lead market hypothesis on the international
diffusion of innovations.

On methodological considerations and the use of

sources in this chapter
It should be noted that this book chapter (i.e., 6 sub-chapters) are mainly
based on the following research papers and dissertations:

• Sriwannawit, P., & Sandström, U. (2015). Large-scale bibliometric

review of diffusion research. Scientometrics, 102(2), 1615-1645.
• Sriwannawit, P., & Laestadius, S. (2015). Determinants of the Diffusion
of Solar Home Systems: Case Study among Low-Income Inhabitants in
Bangladesh. Energy & Environment, 26(5), 803-816.
• Sriwannawit, P. (2015). Power to the people: Diffusion of renewable
electricity in rural areas of developing countries. Ph.D. Thesis.
• Karakaya, E., Hidalgo, A., & Nuur, C. (2015). Motivators for adoption of
photovoltaic systems at grid parity: A case study from Southern
Germany. Renewable and Sustainable Energy Reviews, 43, 1090-
1098.
• Karakaya, E., Nuur, C., & Hidalgo, A. (2016). Business model challenge:
Lessons from a local solar company. Renewable Energy, 85, 1026-
1035.
• Karakaya, E. (2015). Diffusion of dynamic innovations: A case study of
residential solar PV systems. Ph.D. Thesis.

167
Therefore, the content of this chapter may contain identical parts/paragraphs
with the sources mentioned above.

References
Asif, M., & Barua, D. (2011). Salient features of the Grameen Shakti
renewable energy program. Renewable and Sustainable Energy Reviews,
15(9), 5063–5067. [Link]

Balcombe, P., Rigby, D., & Azapagic, A. (2013). Motivations and barriers
associated with adopting microgeneration energy technologies in the UK.
Renewable and Sustainable Energy Reviews, 22, 655–666.
[Link]

Balcombe, P., Rigby, D., & Azapagic, A. (2014). Investigating the importance
of motivations and barriers related to microgeneration uptake in the UK.
Applied Energy, 130, 403–418.
[Link]

Bass, F. M. (1969). A new product growth for model consumer durables.

Management Science, 15(5), 215–227.

Bawakyillenuo, S. (2009). Policy and Institutional Failures: Photovoltaic

Solar Household System (PV/SHS) Dissemination in Ghana. Energy &
Environment, 20(6), 927–947.
[Link]

Beise, M. (2001). Lead Markets, Country-Specific Succes Factors of Global

Diffuison of Innovations. ZEW Economic Studies.

Beise, M. (2004). Lead markets: country-specific drivers of the global

diffusion of innovations. Research Policy, 33(6-7), 997–1018.
[Link]

Bollinger, B., & Gillingham, K. (2012). Peer Effects in the Diffusion of Solar
Photovoltaic Panels. Marketing Science, 31(6), 900–912. Retrieved from
[Link]
12/eco-labels2011/[Link]

Cabraal, A., Cosgrove-Davies, M., & Schaeffer, L. (1998). Accelerating

sustainable photovoltaic market development. Progress in Photovoltaics:
Research and Applications, 6(5), 297–306.
[Link]
pip226>[Link];2-c

168
CIA. (2013). Central Intelligence Agency, The World Factbook: Bangladesh,
Accessed on 5 September 2013
[Link]
factbook/geos/[Link].

Dewald, U., & Truffer, B. (2011). Market Formation in Technological

Innovation Systems—Diffusion of Photovoltaic Applications in Germany.
Industry & Innovation, 18(3), 285–300.
[Link]

Dewald, U., & Truffer, B. (2012). The local sources of market formation:
explaining regional growth differentials in German photovoltaic markets.
European Planning Studies, (December 2012), 37–41.

Drücke, O. (2004). RegioSolar : Kurzstudie Solarinitiativen in

Deutschland. Bundesverband Solarindustrie e.V. (BSi).

ETAP. (2010). Clean , Clever , Competitive Eco-innovation for a sustainable

future, Newsletter Issue 18. Brussels, Belgium: Environmental
Technologies Action Plan (ETAP), European Commision.

Fabrizio, K. R., & Hawn, O. (2013). Enabling diffusion: How complementary

inputs moderate the response to environmental policy. Research Policy,
42(5), 1099–1111. [Link]

Frondel, M., Horbach, J., & Rennings, K. (2008). What triggers

environmental management and innovation? Empirical evidence for
Germany. Ecological Economics, 66(1), 153–160.
[Link]

Goldenberg, J., Libai, B., & Muller, E. (2010). The chilling effects of network
externalities. International Journal of Research in Marketing, 27(1), 4–
15. [Link]

Graziano, M., & Gillingham, K. (2014). Spatial Patterns of Solar Photovoltaic

System Adoption: The Influence of Neighbors and the Built Environment
ǂ . Journal of Economic Geography, 1–25.
[Link]

Greenhalgh, T., Robert, G., Macfarlane, F., Bate, P., & Kyriakidou, O. (2004).
Diffusion of innovations in service organizations: systematic review and
recommendations.” Milbank Quarterly 82.4 581-629.

Griliches, Z. (1957). Hybrid corn: An exploration in the economics of

technological change. Econometrica, Journal of the Econometric Society,
25(4), 501–522.

169
GS. (2010). Grameen Shakti, Grameen Shakti in Figures, Accessed on 22
January 2014
[Link]
&layout=blog&i d=54&Itemid=78.

GTAI. (2013). Industry Overview- The photvoltaic Market in Germany.

Germany Trade & Invest.

Hall, B. H. (2005). Innovation and Diffusion. In J. Fagerberg, D. Mowery, &

R. Nelson (Eds.), The Oxford handbook of innovation (pp. 459–484). New
York, NY: Oxford University Press. [Link]

Hoppmann, J., Peters, M., Schneider, M., & Hoffmann, V. H. (2013). The
two faces of market support—How deployment policies affect
technological exploration and exploitation in the solar photovoltaic
industry. Research Policy, 42(4), 989–1003.
[Link]

IEA. (2012). World Energy Outlook 2012. International Energy Agency.

International Energy Agency (IEA).

IEA. (2014). World Energy Outlook 2014 Factsheet, 1–5. Retrieved from
[Link]
EO_FactSheets.pdf

Jacobsson, S., & Bergek, A. (2004). Transforming the energy sector: the
evolution of technological systems in renewable energy technology.
Industrial and Corporate Change, 13(5), 815–849.
[Link]

Jacobsson, S., & Lauber, V. (2006). The politics and policy of energy system
transformation- explaining the German diffusion of renewable energy
technology. Energy Policy, 34(3), 256–276.
[Link]

Jäger, W. (2006a). Stimulating the diffusion of photovoltaic systems: A

behavioural perspective. Energy Policy, 34(14), 1935–1943.
[Link]

Jäger, W. (2006b). Stimulating the diffusion of photovoltaic systems: A

behavioural perspective. Energy Policy, 34(14), 1935–1943.
[Link]

Kai-ming Au, A., & Enderwick, P. (1999). A cognitive model on attitude

towards technology adoption. Journal of Managerial Psychology, (15
(4)), 266–282.

170
Karakaya, E., Hidalgo, A., & Nuur, C. (2015). Motivators for adoption of
photovoltaic systems at grid parity: A case study from Southern Germany.
Renewable and Sustainable Energy Reviews, 43, 1090–1098.
[Link]

Karakaya, E., Nuur, C., Breitschopf, B., & Hidalgo, A. (2014). Spatial
Dimension of Lead Markets: Evidences from Diffusion of Photovoltaic
Systems in Germany. In DRUID Society Conference 2014, CBS,
Copenhagen, June 16-18. Retrieved from
[Link]
[Link]

Karakaya, E., Nuur, C., & Hidalgo, A. (2016). Business model challenge:
Lessons from a local solar company. Renewable Energy, 85, 1026–1035.
[Link]

Kolk, A., & van den Buuse, D. (2012). In Search of Viable Business Models
for Development: Sustainable Energy in Developing Countries. Corporate
Governance: International Journal of Business in Society, 1–20.
[Link]

Labay, D. G., & Kinnear, T. C. (1981). Exploring in the the Consumer of

Decision Energy Process Adoption Solar Systems. Journal of Consumer
Research, 8(3), 271–278.

Lettner, G., & Auer, H. (2012). Realistic roadmap to PV grid parity for all
target countries. Vienna University of Technology – PV Parity Project,
(October).

Linna, P. (2012). Base of the Pyramid (BOP) as a Source of Innovation:

Experiences of Companies in the Kenyan Mobile Sector,. International
Journal of Technology Management and Sustainable Development.

Mahajan, V., Muller, E., & Bass, F. M. (1990). New Product Diffusion Models
in Marketing : A Review and for Research. Journal of Marketing, 54, 1–
26.

Makse, T., & Volden, C. (2011). The Role of Policy Attributes in the Diffusion
of Innovations. The Journal of Politics, 73(01), 108–124.
[Link]

Mansfield, E. (1961). Technical change and the rate of imitation.

Econometrica: Journal of the Econometric Society, 29(4), 741–766.
Retrieved from [Link]

171
McEachern, M., & Hanson, S. (2008). Socio-geographic perception in the
diffusion of innovation: Solar energy technology in Sri Lanka. Energy
Policy, 36(7), 2578–2590. [Link]

Miller, D. (2009). Selling solar: the diffusion of renewable energy in

emerging markets.

Moore, G. A. (2002). Crossing the chasm.

Muhammad, Y. (2003). Banker to the poor: The story of the Grameen Bank.
Aurum Press Limited,.

Müggenburg, H., Tillmans, A., Schweizer-Ries, P., Raabe, T., & Adelmann, P.
(2012). Social acceptance of PicoPV systems as a means of rural
electrification — A socio-technical case study in Ethiopia. Energy for
Sustainable Development, 16(1), 90–97.
[Link]

Müller, S., & Rode, J. (2012). The Diffusion of Photovoltaic Systems in

Wiesbaden, 1–17.

Müller, S., & Rode, J. (2013). The adoption of photovoltaic systems in

Wiesbaden, Germany. Economics of Innovation and New Technology,
22(5), 519–535. [Link]

Palit, D. (2013). Solar energy programs for rural electrification: Experiences

and lessons from South Asia. Energy for Sustainable Development, 17(3),
270–279. [Link]

Penrose, E. (2009). The theory of the growth of the firm (4th Editio). Oxford
University Press. Retrieved from
[Link]

Pentland, A. (2014). Social Physics, How Good Ideas Spread- The Lessons
from a New Science. New York: The Penguin Press.

Pérez, D., Cervantes, V., Báez, M. J., & Domínguez, F. T. (2013). PV Grid
Parity Monitor Residential Sector. Spain eclareon. Spain Eclareon.
Retrieved from [Link]
energy/files/documents-and-links/pv_gpm_2_residential_2013.pdf

Prahalad, C. K. (2010). The fortune at the bottom of the pyramid:

Eradicating poverty through profits (revised and ppdated 5th
anniversary edition). [Link]. Upper Saddle River,
New Jersey: Pearson Education, Inc.

172
Prahalad, C. K., Di Benedetto, A., & Nakata, C. (2012). Bottom of the
pyramid as a source of breakthrough innovations. Journal of Product
Innovation Management, 29(1), 6–12.

Prahalad, C. K., & Hammond, A. (2002). Serving the world’s poor,

profitably. Harvard Business Review, 80(9), 48.
[Link]

Rebane, K. L., & Barham, B. L. (2011). Knowledge and adoption of solar

home systems in rural Nicaragua. Energy Policy, 39(6), 3064–3075.
[Link]

Rode, J., & Weber, A. (2012). Does Localized Imitation Drive Technology
Adoption? A Case Study on Solar Cells in Germany. Retrieved from
[Link]
[Link]/media/loewe_eigenlogik_der_staedte/dokumente_downlo
ad/artikel/Rode_Weber_2012.pdf

Rogers, E. M. (1962). Diffusion of Innovations. New York: The Free Press.

Rogers, E. M. (2003). Diffusion of innovations (5th ed.). New York: The Free
Press.

Rosenberg, N. (1976). Perspectives on Technology. New York: Cambridge

University Press.

Rosenberg, N. (1994). Exploring the black box: Technology, economics, and

history. Cambridge University Press.

SB. (2007). Ein Zentrum fürs Leben. Solar Boulevard (SB), (1). Retrieved
from [Link]
energietechnik/unternehmen/presse/Solarboulevard_1_07_Hartmann.p df

Shum, K. L. (2013). A general technology adoption model of solar

photovoltaic under feed-in tariff policy incorporating learning and
network effects. Journal of Renewable and Sustainable Energy, 5(4),
043128. [Link]

Sovacool, B. K., & Drupady, I. M. (2011). Summoning earth and fire: The
energy development implications of Grameen Shakti (GS) in Bangladesh.
Energy, 36(7), 4445–4459. [Link]

Spertino, F., Leo, P. Di, & Cocina, V. (2014). Which are the constraints to the
photovoltaic grid-parity in the main European markets? Solar Energy,
105, 390–400. [Link]

173
Sriwannawit, P., & Sandström, U. (2015). Large-scale bibliometric review of
diffusion research. Scientometrics, 102(2), 1615–1645.
[Link]

Sriwannawit, P., & Laestadius, S. (2015). Determinants of the Diffusion of

Solar Home Systems : Case Study Among Low-Income Inhabitants in
banglades. Energy & Environment, 803–816.

SWW. (2004, August). Flexibilität hört auf dem Dach nicth auf. Sonne, Wind
& Wärme (SWW), 1–4. Retrieved from [Link]
[Link]/images/hartmann-
energietechnik/unternehmen/presse/sonderdruck_sonne_wind_waerme.
pdf

Sühlsen, K., & Hisschemöller, M. (2014). Lobbying the “Energiewende”.

Assessing the effectiveness of strategies to promote the renewable energy
business in Germany. Energy Policy, 69, 316–325.
[Link]

Tidd, J. (2009). From Models to the Management of Diffusion. In Gaining

Momentum, Managing the Diffusion of Innovations. Imperial Colloge
Press.

Trogdon, J. G., Nonnemaker, J., & Pais, J. (2008). Peer effects in adolescent
overweight. Journal of Health Economics, 27(5), 1388–99.
[Link]

Van den Bergh, J. (2013). Policies to enhance economic feasibility of a

sustainable energy transition. Proceedings of the National Academy of
Sciences of the United States of America, 110(7), 2436–2437.
[Link]

Weiss, J. A., & Dale, B. C. (1998). Diffusing Against Mature Technology.

Industrial Marketing Management, 27(4), 293–304.
[Link]

Wejnert, B. (2002). Integrating Models of Diffusion of Innovations : A

Conceptual Framework. Annual Review of Sociology, 28(1), 297–326.
[Link]

Wirth, H. (2013). Aktuelle Fakten zur Photovoltaik in Deutschland.

Fraunhofer ISE, (21.3.2013).

Wüstenhagen, R., Wolsink, M., & Bürer, M. J. (2007). Social acceptance of

renewable energy innovation: An introduction to the concept. Energy
Policy, 35(5), 2683–2691. [Link]

174
Völlink, T., Meertens, R., & Midden, C. J. H. (2002). Innovating “Diffusion
of Innovation” Theory: Innovation Characteristics and the Intention of
Utility Companies To Adopt Energy Conservation Interventions. Journal
of Environmental Psychology, 22(4), 333–344.

Zimmerman, D. (2003). Peer effects in academic outcomes: Evidence from a

natural experiment. Review of Economics and Statistics, 85(1), 9–23.

175
 Location as a Matrix of Competition
Cali Nuur

Why is Microsoft located in Seattle, Washington in the USA? Why have

places like Silicon Valley, Northern Italy, and Baden Wurttemberg in
Germany and closer to home, Gnosjö become synonymous with industrial
competiveness? What are the mechanisms that influence industrial
location? Why are some regions rich while others poor? Why are some
industries attracted to locate in a specific area? Why are there regional
variations in terms of competitiveness? What are the factors that
influence of the concentration and dispersion of production? Or to put it
simply what is in location/place?

For many decades, if not centuries, the importance of location has been
the subject of academic, regional development and industrial policies’
discussions. In the academic debate, a number of concepts have been
developed which have also become pervasive in policy circles to enhance
regional competitiveness. These include, but not limited to, concepts such
as industrial districts, clusters Growth poles, Technopolis, science parks,
creative cities, regional innovation systems, just to name a few. With
slight variations, the underlying theme of these concepts is that what takes
place in a location affects the competitiveness of firms, people and have
policy implications. The aim of this chapter is to provide an overview of
some of the key concepts relating to location1.

The earlier literature

Long before the dawn of modern-day means of transportations and
communications, e.g. cars, jets and Information and Communication
technology (ICT), the German economist Johan Heinrich von Thünen 34
(1780-1850) provided a rational for the location of economic activities. In
what has become known as Land Use Theory, he used a circular model and
illustrated the relationship between supply and demand. Others such as Lösch
(1954) and Christaller (1966) also developed the so called Central Place
Theory which describes urban patterns of location. Also Hotelling (1929; see
Hover, 1948, for an overview) has provided theoretical frameworks in regards

34 See
Beckmann, M [1972], ”Von Thünen Revisited: A Neo-Classical Land Use Model,” Swedish Journal of
Economics 1-7.

110
to spatial competition. In many of these models, contemporary economies of
location such as knowledge formation, externalities and relationship building
were ignored as the models were based on assumptions and postulations that
were based on achieving equilibria.

For example, Christaller (1966) assumed similar spatial distribution of

households, while Hotelling (1929) assumed spatial distribution of price and
elasticity (see Parr; 2002; McCann; 2002 and Laestadius; 1992 for further
discussions). An important scholar in this context is Alfred Weber (1909) who
provided a linear model – a least cost approach- of industrial location that
considered transportation costs as an important element in the particular
firm’s choice of location. A key conjecture of Weber’s model was that the
particular firm pursued minimisation of production, labour and
transportation costs when deciding the location of the firm. In addition, the
assumption was that the raw materials that the particular firm needed to turn
into finished goods were produced in the same vicinity, while lighter materials
were produced near the source of raw materials. Furthermore, the assumption
was that demand was local. Because production inputs, including labour and
raw materials, were assumed to be available in abundance and ubiquitous, and
the demand market was assumed to be geographically concentrated nearby
the single most important determinant of location was the minimisation of
transportation costs. Walker (1989) argues that since the model was based on
the conditions of a single plant choosing an optimum location for its operation
it did not provide the reasons for industrial agglomeration and the external
linkages between the particular firm and actors that provided auxiliary
resources.

Location and development

In the period following the second world war, there emerged a literature on
the relationship between economic development and the location of
economic activities. This literature was in response to criticism regarding
the shortcomings of the models described earlier and analysed the impact
of the location of industrial activities on the region/localities/countries.

In addition, there emerged another set of literature which discussed the

mechanisms behind the emergence of the core and periphery of economic
development. Some of this literature moved the level of analysis to
developing as well as nearly independent countries and discussed the
underlying reason for un(der)development 35 (see, e.g. Furtado, 1964).
Influenced and adopting a Marxist approach the key message was that
spatial development – mainly at the country level- is a question of the
relationship between centre (old colonial powers) and the peripheral (newly
independent countries) where the core dictated the terms and the periphery

35 This is worth a chapter of its own

109
conformed (see, e.g. Amin, 1972,1974, 1977; Arighi 1977, Emmanuel, 1977
and Furtado, 1964, 1973; Frank, A., 1969)36.

At the same time, even in developed countries new concepts and

perspectives that offered new insights and perspectives on spatial
competition begun to emerge. For instance, the French economist, Pérroux
(1950) objected to the static and abstract models that ostensibly explained
spatial competition and proposed the notion of an “economic space” that
was based on linkages and relationship. Lausén (1969, p 21) captures the
dissatisfaction with previous scholars who he claimed treated the
geographic space as “a passive rigid container that conditions the dynamic
evolution of economic forces”.

Perroux approach was conceptualised as growth poles (póle de croissance)

and became the point of departure for regional development economists
including Thomas (1969), Lausén (1969) and Hansen (1972) to name but a
few. There also emerged another stream of literature on the regional
dimension of economic development. For instance, Walker (1989) asserts
that the presence of a lead industry in a given territory is likely to make the
whole region/nation prosper and develop since the successful industry
depends on local inputs. Likewise, the same development elsewhere could
indeed diminish production in another region (Pletsch 1982).

Another scholar worth mentioning in this context is Walter Isard (1956,

1959 and 1960) who pioneered the field of regional science and made a case
for multidisciplinary approach, mainly the inclusion of all aspects of the
social sciences, in analysing regional competitiveness. By and large, the
birth of regional science has its origin in the criticism on the abstract
models, which assumed space as homogenous and available to every actor.
These failed to explain why different locations favoured or disfavoured the
growth and development of a particular industry. Soon studies were
commissioned and concepts developed which viewed the territory a
dynamic entity has the ability to influence the decline and growth of
industry.

In summary, the conceptualisation of space and subsequent introduction of

growth poles theory attempted to answer the questions of: What factors play
a role in the economic development of a region; what are the features of
successful industries; what mechanisms contribute to the evolution of

36 There is a myriad of literature on the impact of colonialism in ensuring the evolution

and sustention of the periphery. Some of these have taken a Marxist approach where
independicia was key theoretical point of departure (see, Frank, 1969; Furtado, 1964;
Emmanuel, 1977 and Arighi, 1977) while others took a somewhat structural approach
(see, e.g. Prebrish 1956). Admittedly, I am name dropping as there is more to this
literature stream than merely a critique of orthodox economic theory.

111
peripheral regions? And how do clusters of innovative firms emerge? Pérroux
asserted that territory had the ability to become a stimulus for the general
economic condition of a region or a nation by inducing propulsive
industries3738 and lead firms (Lever 1980, Tomas 1969, Lasuén 1969). For
Pérroux and other development oriented economists, the notion of
development is about the birth and death of industries at a particular place i.e.
industrial dynamics. Old industries die because product and process
innovations make previous products redundant and open up new areas of
production.

It might be argued that Pérroux’ s ideas were related to the creative

destruction approach adopted by Schumpeter (1943), who contended that
economic development is affected by factors such as technology and the
diffusion of technology as well as endogenous combinations such as the
introduction of new products that the consumer has not tried before or a
qualitative extension of an available product, the introduction of new methods
of production or marketing method, the exploitation of new markets and the
organisation of a new industry. However, it is worth noting that Schumpeter
did not have the geographical dimension as his point of departure. Perroux’s
growth poles theory recognizes the ability of technology to shape territorial
dynamics and become a driver of economic development and thus an
important factor that is exogenous to the traditional comparative-based
advantages of land, labour and capital as the basic prerequisites to growth.
Lasuén (1969) suggests that growth poles theory relates to Schumpeter’s
theory of innovation in that is also discusses new process and new products
that affect the structure and composition of industries in a given location. The
geographic location that housed old industries become peripheral while the
location of new industries become core regions and start to develop. In other
words, innovations contribute to the territory undergoing the same kind of
dynamics as that of the individual firm. To argue the similarity of product and
geographic space in economic development, Pérroux (1950), who obviously
was influenced by the Schumpeterian approach, discusses the features of
leading industry, industrial complexes, sectoral and spatial relationships
between dominant industrial location and the industry, and the relationship
between industrial clustering and the general development of the nation.

As indicated earlier, Pérroux’s approach came at a time when several other

social scientists discussed the notions of underdevelopment. Apart from
sociologists and political scientists discussed earlier, development economists
were also conceptually engaged in the discourse. For example, Albert

37 See Amin S. (1976), 'Unequal Development: An Essay on the Social Formations of

Peripheral Capitalism' New York: Monthly Review Press.
38 Lever (1980) defines propulsive industry as firms which are “significantly linked by

flows of goods with a large number of other industries and which are therefore likely
to transmit growth more widely” Lever (1980, p.501)

112
Hirschman (1958, 1967) in studies done in peripheral regions in the south of
Europe and in South America, developed the concepts, unbalanced growth as
well polarisation and trickling down39 to describe how growth in a region is
affected by the evolution of new growth poles as a result of a dominant/lead
industry. According to Hirschman (1967), the development of a lead industry
induces a process of innovations, which positively affect the general progress
of the region. The emergence of growth poles as a result of a propulsive
industry could lead to the positive effects of polarisation, but it could also have
a negative impact in terms of a trickling-down effect on other areas leading to
uneven development as core areas attract manpower, capital and public
infrastructure from surrounding areas while peripheral areas undergo a
process of decay (Amos Jr 1988).

The growth poles as a strategy of regional development gained prominence in

the 1960s and early 1970s as governments saw them as an opportunity to
influence regional development (Amos Jr. 1988, Polése 1999). These post-war
decades were characterised by an economic reconstruction of industry and the
subsequent boom of the economies of Europe and North America. In
academia, factors such as the availability of statistics, a process of statistically
significant urbanisation/migration in some regions and differences in
population distributions contributed to the usage of terms such as
“disparities” and “lagging”. To reduce regional economic polarisations and to
ensure the development of peripheral regions, many governments established
departments with the task of pursuing and stimulating regional development
policies that could contribute to even development. Through diverse forms of
incentives, such as subsidies, investment in infrastructure as well as
channelling public funds into lagging regions, governments hoped to
encourage regional parities (Polése 1999).

Industrial districts

The political economist Alfred Marshall who also wrote the influential book
“principles of economics” and largely accredited with bringing the field
economics in the realm of science is the sage of the literature on industrial
districts. In addition to describing the propulsive ability of the territory,
Marshall (1890/1920) laid the foundation for the reasons as to why there are
industrial agglomerations. In the chapter Industrial organization – the
concentration of specialized industries in particular localities, Marshall
discusses the advantages that geography could offer to economic actors
following empirical observations made on the districts surrounding
Lancashire and Sheffield (Laestadius 1999).

39 This
should not be confused with the politically laden notion of ”Trickle-down
economics”, also known as ”trickle-down theory” which is basically the notion that
as wealth increases for the rich, it will result in spillover to the poor.

113
Marshall (1890/1920) argued that concentration of economic activities
resulted from historical circumstances, including factor endowments, local
demand and the geographic conditions of the area. Nevertheless, perhaps his
chief contribution was in revealing the relationship between the internal
capabilities of individual firms and external mechanisms that proximity
facilitated. According to Marshall, firms develop internal economies, which
allow them to either cut production costs or increase sales. To a certain degree,
accessing and sharing resources with other economic players in the same
geographic proximity is the result of what takes place outside the firms’
borders. In this respect, Marshall played a central role in the theory of
economic organisation (Laestadius 1992).

Marshall described economies of location and in so doing laid the foundation

for understanding the externalities that firms can derive outside its borders.
For the sake of simplicity, economic geographers group agglomeration
economies into two economies: urbanisation economies and localisation
economies. The former emerge when people and economic activities locate
near to each other in order to benefit from proximity. These are often external
to the area, but internal to the firms in the area as they gain economies of scale
in production (Malmberg 1998). On the other hand, localisation economies
are firm specific and emerge because of geographic proximity, which allows
the firms to have collective benefits that are unavailable to firms in other
areas.

According to Malmberg (1998), both urbanisation and location economies are

largely influenced by the same set of factors and develop from shared
infrastructure. Physical proximity allows the companies to produce at a lower
cost as transportation, thus transaction costs are reduced, and they can benefit
from cheaper provisions that arise from shared accessibility to harbours,
roads, etc. Furthermore, it facilitates the flow of information and allows the
companies to share technical, education and labour market infrastructure.

It is worth noting that in addition to the studies on the relationship between

territory and competition, Marshall also discussed various aspects of
economic development, including the role of politics and the division of labour
(Laestadius 1992). As regards location, Marshall (1920) offered three major
advantages that geographic proximity provides to people and firms: creation
of a labour market, external economies of scale, and learning opportunities.

First, geographic proximity facilitates the creation of a specialised labour

market as it attracts both employees and employers. On the one hand, a
person seeking employment is inclined to move to a place where he can get a
job and, on the other hand, employers are inclined to settle anywhere where
it is easy to select and recruit personnel. This approach is often referred to as
localisation economies (Malmberg 1998).

114
The second important advantage of geographic proximity is the ability to offer
external economies of scale. Concentration of business activities creates
external economies through vertical relationships, such as sub-contracting
agreements between firms in the value chain, and vertical co-operation for the
firms that produce the same kind of goods (Marshall 1920). At the same time,
it gives the concentrated firms advantages in terms of lower costs over firms
in other territories.

According to Marshall, the third major advantage of geographic proximity is

the ability of geography to enhance face-to-face interactions between
economic actors. Asheim (2000) suggests Marshall’s main contribution to
social science is in revealing proximity as a factor that enhances the creation
of learning opportunities. Through relationship ties and the facilitations of
information exchanges between different actors, geographic proximity fosters
learning and innovation that provide the necessary resources for
competitiveness (Malmberg et al, 1996).

According to Malmberg et al (1996, p 90) the geographic proximity allows for

tacit knowledge exchange because this kind of knowledge “does not reside in
blue prints and formulae, but is based on personal skills and operational
procedures which do not lend themselves to be presented and defined in either
language or writing”. This aspect of learning activities is not only developing
the competitive strengths of the existing firms, but also of enhancing the
growth of new firms was a key aspect of location.

Marshallian dynamics of location, especially with respect to relationship and

knowledge spillovers as an important dimension in understanding spatial
competition – in particular in regional development policy circles- were
confined to bookshelves until Piore and Sabel entered the scene in the mid-
1980s. In their much-acclaimed book “The industrial divide” (1984), they
discuss the emergence of two major paradigms supposed to have shaped
economic development in post-war Europe and the United States. First, they
analysed how mass production technology at the turn of the twentieth century
ended the era of handicraft production and gave rise to the Fordist era in the
USA and parts of Western Europe. Secondly, they describe the stagnation of
the world economy in the 1970s and the arrival of new flexible production
methods, in which manufacturers shifted one product or process to another
depending on the market needs.

Amin and Thrift (1995) have also described this period as the post-Fordist or
post-industrial era, which is characterised by among other things, the
diffusion of technology, volatile markets, and the growth of a large number of
small- and medium-sized firms in parts of Europe. As the unstable market
conditions curtailed returns on investments, e.g. machines often became
redundant before the investment costs were recovered; many firms
experimented with flexible working conditions and production methods. The

115
resulting high level of efficiency, which previously had been reserved for
multinational large corporations that thrived on mass production, as well as
active support from institutions, enhanced the growth of small- and medium-
sized firms.

Following on this, “Marshallian districts” became the symbol and point of

departure for understanding and analysing the dynamics of the “third Italy40
(Cossention 1996, Brusco 1986, Becattini 1995, Baker 1995, Brusco 1995, Dei
Ottati 1994). This name is used to differentiate industrial districts in the south
of Italy and the triangle of Genoa, Turin and Milan in the northeast and
northwest of Italy from the rest of the country. These studies revealed the
presence of many industrial districts that were home to small- and medium-
sized firms with flexible production methods. The studies found that, for
example, in the knitted goods industry in Capri, the ceramic tiles industry in
Sassuolo, and the textile industry in Prato. Taken altogether, these districts
numbered around 29 (Cossention 1996).

In accordance with Marshallian dynamics i.e. specialised labour markets,

learning etc., the Italian case studies showed that they had the capacity to
attract skilled workers, which was one of the major advantages that Marshall
(1920) had written about. Accordingly, the availability of jobs for people who
want to take them and the level of entrepreneurship indicated that the districts
were growth areas (see, for instance, Cossention 1996, Brusco 1986, Becattini
1995, Baker 1995, Brusco 1995, Dei Ottati 1994). The division of labour
between the Italian industrial districts allowed each district to have a unique
form of specialisation. Workers and technicians move between the different
firms, for example between suppliers and buyers, but they tend to stay within
the district. In this way, tacit knowledge among the workers could be
accumulated (Berggren et al, 1996, Becattini 1995, Brusco 1995).

In Sweden, the Gnosjö region located in the province of Småland in the

southern part of the country showed the same pattern in terms of generating
small enterprises. The small companies in this region were argued to combine
competition and collaboration. Just like the Third Italy, the presence of socio-
economic networks that created a breeding ground for entrepreneurship was
observed (Wigren 2003, Brulin 2000, Johanisson 1984).

The ability of Gnosjö region to foster entrepreneurial activities and nurture a

climate of trust between the economic actors through embedded relations that
span through the community is referred to as “the Gnosjö spirit”41. According
to Gummesson (1997), the foundations of this spirit started in the sixteenth

40 The Marshallian district concept has also been used to analyse dynamics regions in
many European, Asian as well as North America. In addition, in the 1980s and 1990s it
was the dominating concept in regional development.
41 In Swedish “Gnosjöandan”.

116
century when a weapon factory was established in the scarcely populated,
small-farming area. The iron industry and trade rose during the sixteenth and
seventeenth centuries when foreign smiths taught new methods. After the
death of the Swedish warrior king Charles XII and Sweden’s defeat in the
Great Northern War, local blacksmiths witnessed a sharp fall in the demand
for their metal products. Following the decrease in demand, many of the now
self-employed blacksmiths started to expand their production lines. However,
from the mid-nineteenth century, the wood industry and glassworks took the
place of the iron industry.

In summary, the characteristics of the industrial districts of the Third Italy

and the Gnosjö region42 can be summarised as follows (Baker 1995, Becattini
1995, Brusco 1995, Asheim 2000):

Geographic concentrations of many small- and medium-sized firms.

• Dense socio-economic networks that encourage inter-firm

relationships and enhance competitiveness.
• Institutional arrangements that facilitate the flow of ideas, which
enhances development of production methods and market innovations.
• Willingness to adapt to new circumstances and flexible production
systems.
• Division of labour between the districts allowing for high levels of
competence and accumulation of tacit knowledge.

Meanwhile across the Atlantic, territorial competitiveness and the

significance of the region as an important unit of development gained further
importance when Anna Lee Saxenian published “Regional Advantage: Culture
and Competition in Silicon Valley and Route 128” (1994). In this book based
on a study of the two areas of route 128 in Massachusetts and Silicon Valley in
California, she concluded that the success of the latter depended on the
dynamic horizontal networks of managers, companies and supporting public
institutions. In contrast to the static vertical integration of inter-firm
relationships in Route 128, she found that the dynamic networks that existed
in Silicon Valley provided a receptive atmosphere, which encouraged co-
operation while still maintaining a degree of competition.

At the same time, several social scientists (For instance, Staber 1998,
Markusen 1996, Mascanzone et al 2000, Baker 2000) began to develop

42 It is worth noting that although the Third Italy and the Gnosjö region are similar in the
spirit of encouraging entrepreneurship and the many informal networks, there are also
differences between the regions. For example, in the Gnosjö region, the industrial
specialization is horizontal, while in the Third Italy it is vertical.

117
concepts that were largely based on the work of Marshall to explain spatial
competition. They examined key features of special areas in terms of economic
development and began to categorise the different kinds of industrial districts.

Staber (1998) argues that there are two distinct kinds of industrial districts.
First, there are the “Marshallian” type industrial districts with competitive
inter-firm relationships. These districts have, of course, firms that come and
go because of forces of competition. New firms are born out of older firms,
new entrants emerge as a result of needs within the districts, and firms are
closed as a result of increased costs, lack of customers, etc. Secondly, there are
industrial districts, mostly like the ones in the Third Italy, which have inter-
firm relationships and are socially integrated. The dynamics arise as a result
of institutional arrangements for collective learning and social relationships.
Staber (1998) asserts that the backbone of industrial districts is the
institutional setting with common, but varying, mechanisms of integration
based on kinship, religion, and relationships. Actors are involved in dense
social networks that facilitate not only competition due to peer pressure, but
also co-operation through drawing on network resources.

In the USA, Markusen (1996) found the presence of industrial districts that
were quite different in characteristics from those in Europe. In Sticky Places
in Slippery Space: A Typology of Industrial Districts, she gives an in-depth
analysis of neo-Marshallian industrial districts. Markusen (1996) contends
that the research findings from studies of the Italian districts, which are based
on the socio-economic networks of small- and medium-sized enterprises, are
far from representative for other industrial districts throughout the globe. To
capture differences in terms of networking, regional development and
institutional context, she proposes two alternative models of industrial
districts.

The first kind of industrial district is similar to Pérroux’s growth poles where
a dominant firm pushes the region forward, which is an important element in
growth pole theory. Markusen (1996) calls it the hub-and-spoke. These kind
of industrial districts consist of one or several major employers with supplier
networks and service providers. These are often industrial areas that house
single enterprises like the Boeing complex in Seattle, or larger areas that house
firms in the same industry like the motor industry in Detroit.

The dynamics of hub-and-spoke industrial districts is centred on the ability of

small- and medium-sized firms (SMEs) to interact in networks, and the
capacity of a large firm (an engine enterprise) to accommodate the smaller
firms. The SMEs are linked through transactional relationships to the engine
enterprise, to which they sell to or learn from (Barkley 2001). The co-
operation and competition between the companies are usually dictated by the
engine enterprise. The small and medium sized companies in the hub-and-
spoke gain external economies from each other by virtue of their number. For

118
example, at the Boeing complex in Seattle, several SMEs supply Boeing with
various inputs (Dunning 2000, Markusen 1996). Mascanzoni et al, (2000)
report the presence of sixty hub-and-spoke districts in the Third Italy. These
are characterized by vertical and horizontal networking between the firms,
which are mainly small suppliers and an engine enterprise. Small local
networks of firms import, store and distribute raw materials necessary for the
engine enterprise to produce finished goods. These drive the pace of product
innovations in the district. These locomotive companies in the Italian sense
are global players with international reputations and they are characterised by
their excellent professional knowledge of craftsmanship, technical and social
know-how, and innovation ability. Their success is a result of technological,
organizational and marketing abilities that have been accumulated during
many years of conducting business activities in the districts. Close proximity
between the engine enterprise and their suppliers and customers facilitates
the transfer of strategic information, which small- and medium-sized
suppliers are dependent upon.

There is a two-way linkage of relationship between the firms in a hub-and-

spoke district. First, there is a vertical relationship between the engine
enterprise and the SMEs that are clustered, and, secondly, a horizontal
relationship between the SMEs. Competition is limited because most of the
SMEs supply different services or products, or they are all involved in the
value chain of the engine enterprise (Barkley 2001).

The second category according to Markusen (1996) is the satellite platform

and state anchored industrial districts, which often are induced by policy
measures to enhance regional development. Defence installations,
universities and concentrations of government agencies belong to this
category. Policy measures play a vital role not only in their evolution, but also
in their successful development. In these kind of industrial districts, there is
no collaboration between the companies as each and every one of the tenants
is engaged in different kinds of activities. Most of them are large players that
determine the pace of development.

Although Marshall (1920) discussed the knowledge aspect of economic

concentration, most of the studies on “Marshallian districts” from the eighties
onward had their analysis in the institutional setting with a focus on cluster
dynamics in explaining their evolution and functioning. The case studies of
the Italian districts and other areas around the globe described the
relationship dynamics, including the collaborative arrangements between
economic actors and public bodies in achieving development. This aspect of
cluster based relationships as a facilitator of regional development was to a
certain extent discarded by mainstream economists.

119
Post globalization concepts
In the early 1990s, spatial competition models including Marshall’s were
aligned with new growth theories 43 , mainly due to the importance of
knowledge which came to be viewed as a factor of competition. These theories
hold that steady economic growth in a nation or a region depends on
technological progress and knowledge utilisation (Glaeser 2000, Ekstedt
2001). Accordingly, the development of new technologies and their diffusion
by way of knowledge accumulation are critical components in a company’s
productivity and henceforth on the context. Much of the literature builds on
the works of Romer (1990), in which he models the micro-economic
environment from a knowledge accumulation perspective and the underlying
theme is that the development of the individual firm depends on knowledge
accumulation and, consequently, its productivity and innovative capabilities
depend on the extent of knowledge form creation and proliferation.

At the same time, the new growth theories brought forth the impact of
research, development, and the spin-off effects of knowledge-based firms in
the context of location. Accordingly, knowledge accrual is often the result of
education, experience and training that is contextually generated and
diffused. Therefore, much of this research stream attempts to explore the
geographic dimension of economic development by looking at mechanisms
that lie outside the borders of the particular firm, including the extent of
knowledge spillovers between actors. These are mainly the mechanisms that
were earlier discussed by Marshall (Laestadius 1999).

The recognition that knowledge creation and diffusion is an important

mechanism to achieve economic competition and territorial development
coincided with a period that was marked by political and economic changes.
The collapse of the Berlin Wall and the subsequent opening up of new markets
in Eastern Europe and the expansion of the European community (later on,
the European Union) are just a few examples of the political developments.
On the economic front, globalisation, improvements in transportation and
information flows mainly in the telecommunications industry and the
disintegration of the larger manufacturing industry in the developed world
(Castells 1999) led some to predict the “death of distance” (see Cairncross
1997). While others, e.g. Amin and Thrift (1995, p.2), argued that
globalisation” does not represent the end of territorial distinctions and
distinctiveness, but an added set of influences on local economic identities and
development capabilities”.

The developments mentioned above gave an added significance to knowledge

as a catalyst of economic development and heralded the evolution of new

43 This is sometimes dubbed as endogenous growth models.

120
concepts such as “the knowledge economy” (Castells 1999), “the learning
economy” (Lundvall 1992) to capture the knowledge aspect of development.
The spatial dimensions of these changes were captured by concepts such as
the advent of a “new economic geography”. In the early 1990s, Paul Krugman
for instance developed his trade theory of “the new economic geography”, in
which he constructed an equilibrium-based model of trade to explain the
concentration of business activities in a given place. His model of interplaying
factors contained: increased returns to scale, transport costs and factor
mobility. According to this model, increased return to scale motivates
producers to limit production costs, transport costs are reduced because of the
geographic proximity, and factor mobility is achieved because of the
expansion of the labour market (Krugman 1996). Tson Söderström, et al,
(2001) in a much publicised report in Sweden, for example employ the
metaphors of “forces of concentration and forces of dispersion” to describe the
features of the new economic geography. On the one hand, the new economic
geography is manifested in the replacement of traditional factors such as
economies of scale by relationship-based factors (forces of concentrations).
On the other hand, according to Söderström et al, (2001) the new economic
geography is characterised by the forces of dispersion. These include the
presence of infrastructure, which attracts skilled labour, research and
development, customers and suppliers.

In addition, since the “new” concepts reveal a positive relationship between

innovation, knowledge and economic development (Lundvall 2001,
Archiburgi et al 2001, Storper 2000a, Audretsch 2000a), they had an
influence on industrial policy. Thus they contributed to a rethinking of
regional and national development strategies, especially at a time when
previous policy measures appeared to be unsustainable. The relative successes
of industrial districts and the digital boom orchestrated by the advent of the
Internet have also certainly enhanced the development of new frameworks of
competition; a paradigm that is by and large based conceptual-based
development (Raines 2001). In the period leading up to the new millennium,
there was a widespread debate on the “knowledge economy” as a mechanism
that would contribute to regional and national development. In the report
A New Economy? (2000), the OECD described knowledge creation as the
ultimate panacea to overcome the challenges of the global economy. In terms
of regional development, the debate stressed the importance of codified forms
of knowledge that could be accessed and generated through institutions of
higher education.

The new regional development policy, which evolved from these discussions
is referred to as endogenous regional policy, which simplified means a
departure from the Keynesian inspired convergence model aimed at achieving
regional parity. There was also an emphasis on a systems-based strategy to
enhance regional competitiveness (see, Nuur et al, 2009 and Nuur and

121
Laestadius, 2010). By adopting a regional/national innovation systems’
approach the government hoped Swedish industry would become
internationally competitive. The theoretical undercurrents of the concept of,
“National Innovation Systems” (NIS) related to the work of Christopher
Freeman (1988) who described the success of Japan. Freeman (1988)
discusses how industry and public policy interplayed to contribute to
innovative methods, process and products. The NIS concept has been
developed by, among others, the economist Bengt Åke Lundvall. Central to the
notion of NIS is the idea that differences between nations in terms of economic
development depend on a myriad of factors including the degree of knowledge
utilisation, institutions and infrastructure rather than on basic factor
endowment:

“National innovation systems are, by definition, localized and

immobile, and thus able to provide firms with valuable capabilities
and framework conditions not available to competitors located
abroad, even under the most open market conditions imaginable.”
(Lundvall 2000, p.364).

The relative success of Silicon Valley and Route 128 in Massachusetts in

fostering entrepreneurship and knowledge-based dynamics (Saxenian 1994)
and the political as well as economic changes mentioned earlier increased the
importance of innovation systems as a developmental tool. In Sweden, the
central government has with the establishment of Vinnova (the Swedish
Agency for innovation systems) embraced an industrial policy with the goal of
enhancing the competitiveness of Sweden as a nation by promoting the
innovative capabilities of industry. Following this, the parameters of the
national innovations systems approach were expanded to include actors at the
regional level with the advent of regional innovation systems (RIS) (Bager-
Sjögren & Rosenberg 2004).

In Sweden, the seeds of the regional innovation systems approach which are
discussed by Rickne and Laestadius (2016 – in this volume) were sown in the
1970’s when higher education was decentralised (Andersson et al, 2004). The
establishment of universities and university colleges in almost every political
province (län) and the development of science and technology parks in the
1990s is a testimony of this approach. The underlying implication of this
policy approach is a belief that knowledge spillovers between geographically
proximate actors or functionally interacting actors would have a positive effect
on regional development and contribute to the emergence of knowledge-based
firms that would replace the decline of traditional sectors.

Industrial Clusters

If the national and regional innovation systems approach is one that answers
the “how’s” of knowledge generation, one concept that has its point of

122
departure in the “whereas” of development is “Clusters”. In the 1990s and the
earlier part of 2000, this concept had become part and parcel of the regional
and national economic development debate (see, for example, Stymme 2004).
In the economic, geographic and business literature, the phenomenon of
cluster was too often used interchangeably with industrial districts and
science and technology parks. In the past, scholars have used the perspective
of geographic agglomeration, which Marshall pioneered to describe clustering
of economic activities. This description denotes the tendency of firms in the
same field to locate in the same geographical area. In most circumstances,
however, the physical location of firm agglomeration is based on factor
endowment and does not necessarily mean the presence of interdependencies
(Malmberg 1998).

Cluster dynamics

As indicated earlier, clustering of economic activities is not a new

phenomenon, but has rather occurred for centuries. 44 . However, it gained
potency in 1990 when Michael E. Porter, Professor of Business at Harvard
University, authored a book entitled “The Competitive Advantages of
Nations”. The findings in this book are based on research carried out in several
European countries, including Sweden. The Swedish analysis is contained in
a book entitled “Advantage Sweden”, which was co-authored by Örjan Sölvell
and Ivo Zander at the Stockholm School of Economics. For the purpose of this
chapter, I intend to embrace Porter’s (1990) definition of a cluster, which is:

” Geographically proximate groups of interconnected companies and

associated institutions in a particular field, linked by commonalities
and complementarities. The geographic scope of a cluster can range
from a single city or state or a country or even a group of neighbouring
countries.” (Porter 1990, p.254)

The origins of clusters are the same as those of agglomerations and include
historical circumstances, proximity to sources of endowment, local demand
conditions and random elements (Porter 1990). However, not many of these
factors act independently, but rather they interact with one another.
Technology is one such factor. Krugman (1996) argues that technology is a
vital factor that induces clustering, but not the main one. Instead, technology
helps the pace of the cluster development, as it takes some time to discover
and diffuse it. Technology has been cited as one of the factors that contributed
to the growth of the well-known cluster of Silicon Valley. The discovery and
subsequent diffusion of the semiconductor and the computer processor
certainly played a leading role in the development of Silicon Valley (Saxenian
1994).

44 See for example the discussion earlier Marshallian districts.

123
Tson Söderström et al (2001) assert that clusters go through various stages
that range from “the heroic” to the “renaissance” phase. The heroic phase is
the initial period of clustering and is characterised by the presence of a few
actors, mainly entrepreneurs or a flagship enterprise. During this phase, there
is almost no vertical relationship between the actors. The second phase of
clustering is marked by the presence of dense networks of relationships that
are vertically integrated and a critical mass of firms that are concentrated in
the same area. The final phase is characterised by “diversity” and
specialisation, with many sub-clusters and firms that are vertically integrated.

Rosenfeld (1997) also argues that economic clusters have a life cycle that is
marked by embryonic, growth, maturity and decay stages. The embryonic
stage of a cluster is the result of technological innovations and/or inventions
and internal direct investments made either by an individual or a firm. The
growth stage is marked by the developments of markets, the spin-offs of firms,
the attraction of imitators that develop unique products, competition that
allows for benchmarking and the facilities to attract entrepreneurship. The
mature stage is marked by new entrants, including service providers, shared
infrastructure that leads to cost efficiency for the companies and the ability to
combine competition and cooperation. The decay stage takes place when the
products that the cluster produced are replaced by cheaper production
elsewhere or substituted for as a result of new innovations.

When introduced in the early 1990’s by Porter, the cluster approach defied
traditional models of the location of economic activities, which were mainly
based on comparative advantage, and were static in their nature (Malmberg
et al 1996). In contrast to agglomeration theories which are mainly embraced
by economic geographers, Porter (1990) took a business science perspective
and contended that it is the interplay between geography, institutional
linkages and inputs from formal and informal organisations that forms the
basis for competitive advantage of nations. In addition, this perspective
recognised the importance of interdependencies between firms in terms of
relationships and knowledge building in developing production competencies
(Cooke 2002). Porter never described how large a geographical area ought to
be, or how many firms ought to be concentrated in one specific area to be
called a cluster (Berggren & Laestadius 2003). However, one vital element of
his thesis was that firms do not live in isolation, but rather depend on other
firms and organisations for various kinds of resources that enhance
productivity, innovation and new business formations.

124
The diamond model45

Having previously written about firm-based competitive strategies, including

value-chain management and firm-level competitive strategy through cost
reductions, Porter (1990) developed a diamond-shaped model to describe the
main elements that contribute to the competitiveness of nations. The model
consists of four major interplaying factors that determine territorial
competition: demand conditions, factor conditions, firm strategy, and
related and supporting industries. It also included two exogenous variables
that could be accessed by any nation or region: chance and government policy.
Two of the components of the model, demand and factor conditions, are
external mechanisms. These lie outside the borders of the firm and relate to
the availability of a demand market and the production factors necessary to
meet the demand. Accordingly, the basis for competitive strength of a nation
is the presence of factor conditions, such as provision of upgradeable natural
resources, capital and labour, combined with market demand conditions, such
as local and national demands.

Porter (1990) relates the demand and factor conditions of nations to two
micro-economic-based elements: individual firm strategies and the presence
of related, and supporting industries. These two elements are firm- and
branch-specific factors and form the basis of the cluster concept. The idea
behind this is simple; Rival firms and other firms in the same field tend to
appear in a cluster where they have access to common services which reduces
transactions costs and allow for the provision of human resources. The
presence of related and supporting industries facilitates a process of
innovation and value-creating mechanisms, without the individual firm
needing to make all the investments on its own. The interdependencies that
arise in terms of inputs, such as training and specialised labour, are managed
collectively. In this way, the firms can significantly reduce their production
costs as compared to rivals that are located in different places.

An analytical instrument and a policy tool

The concept of economic cluster gives an analytical opportunity to scholars,

as well as to policy makers, to capture the mechanisms that lead to industrial
development in a nation or a region (Malmberg 2002; Brown 2000). Through
the dynamic linkages of relationships and transactional activities, the
argument went that the cluster concept helps to identify competitive sectors
of the economy (Malmberg 2002). In addition, it provides scholars with
instruments to study how firms perform in an institutionalised system rather
than seeing the firms as static units usually described by statistic variables of

45 For
an in-depth analysis of this model on Swedish industrial dynamics, see
Advantage Sweden (Sölvell et al, p. 24)

125
size, number of employees, etc. According to Malmberg (2002), using the
cluster approach provides insights into functional linkages such as:

• Vertical transactional links, i.e., exchange of goods, know-how, etc.

between firms, customers and suppliers.
• Horizontal competitive links, i.e., competition between firms in
acquiring factors of production such as labour.
• Knowledge spillovers through interaction between actors.
• Horizontal co-operation through formal and informal contacts such as
strategic alliances, networking, etc.
• Commonalities such as technologies, complementary activities,
infrastructure and labour pools.

As illustrated by Malmberg (2002) above, using the cluster approach to

analyse regional or national development poses measurement challenges
since functional linkages are subjective and difficult to quantify. According to
Berggren and Laestadius (2003), the adoption of a quantitative cluster
approach to study industrial development is a complex one, since the extent
of relationship connections between proximate economic actors is hardly
accessible in industrial statistics. Instead, they propose a qualitative
evaluation, which considers horizontal and backward linkages in the form of
“development pairs”. Berggren and Laestadius (2003) exemplify this with the
Finnish-Swedish cluster of the telecom companies Nokia and Ericsson, which
they argue developed parallel with a distinctive feature of competitive rivalry.

In addition to being an academic analytical instrument, the cluster concept

had also become a popular tool among policy makers to enhance regional and
national development (Sadler 2004; Gordon &McCann 2000, Raines 2001,
Stymme 2004). From a policy perspective, an important factor of the concept
is the degree of co-operation that is needed to manage interdependencies
between public institutions and organisations and more importantly policy
measures to provide auxiliary inputs to companies to raise their
competitiveness. As Waits (2000) contends, adopting a cluster-based
approach allows policy makers to have insights into the factors that drive the
economies of particular regions and the challenges that face them.

O’Mally and Van Egeraat (2000) examined how a cluster-based approach

helps us understand regional and industrial development in Ireland. Despite
an active policy strategy to induce and promote the growth of new industry
using a cluster-based approach, they found that the performance of Irish
traditional firms (Food processing, manufacturing, service, etc.) was not
weaker than the performance of the designed clusters à la Porter.

Raines (2000) puts forward insights into how the cluster concept is embraced
as a developmental approach in seven regions across the European Union (Art

126
Vally in France, East Sweden46, Limburg in the Netherlands, North Rhine
Westphalia in Germany, Pais Vasco in Spain, Scotland and Tampere in
Finland). In general, he found that a cluster-based approach is influencing
regional and national development policy either by complementing existing
development policies or through using it as a new framework to achieve
competitiveness.

Brown (2000) examined the literature on clusters and cluster policy and
argues that there is a lack of coherency in both how the concept is defined
scholarly and how policy makers apply it in practice. While some European
countries such as the Netherlands have adopted a “top-down” approach to
implementing clusters, other countries including Scandinavia have opted to
have a bottom up strategy where cluster development is a local/regional issue.

Brandt (2001) has also studied cluster-based approaches in Sweden and the
other Scandinavian countries47 and he contends that Porter’s cluster approach
has shaped industrial dynamics in Sweden. He identified two basic measures
to develop cluster dynamics: top-down and bottom-up approaches. These two
approaches are based on providing direct and indirect support to devising
cluster-based strategies. The direct approach is marked by a strategy of
“picking winners” and “implicitly” targeting available competitive sectors. The
indirect approach strategy involves devising programmes, such as
“partnership” programmes, and endogenous development initiatives to
induce cluster dynamics.

Despite the popularity the cluster concept has gained in both policy and
academic circles, it is laden with ambiguities in terms of scope, contents and
definitions (Malmberg 2002). For instance, in an article entitled
“Deconstructing Clusters: Chaotic Concept or Policy Panacea?” Martin and
Sunley (2003) discuss how “Porterian” clusters are shaping economic
development policies. The authors assert that the concept is “vague” as to both
definition and scope and that it contributes to “conceptual and empirical
confusion” (2003, p.10). According to Martin and Sunley (2003), much of the
cluster debate centres on policy discourse to find “fit” strategies to enhance
development and that from a scholarly perspective it does not inject new blood
into location as a matrix of competition. Their critique is based on how the
concept is defined (they cite 10 different definitions) and that the political
discourse surrounding it, as a tool to encourage innovation, clouds the
scholarly value and the lack/difficulty of empirical foundations (Sadler 2004).
In respect to the policy dimension, Martin et al (2003) suggest the role of

46 InSwedish: Östergötland.
47 Isaksen (1997) identified 143 clusters in Norway and Nielsen (2001) identified 149
clusters in Denmark.

127
policy should be to encourage productivity in all firms and caution against a
policy that is committed to a “mind-set” of creating clusters.

Also, Feser (1998) points to the inconsistency of cluster definitions and the
lack of clarity in understanding the dynamics that lead to the development of
clusters. He argues that this results in an inconsistency in policy measures on
cluster developmental approaches.

In a study commissioned by ITPS, the Swedish Institute for Growth Policy

Studies, Malmberg (2002) follows this path of criticism to a certain extent. He
asserts that it is unclear whether the term denotes the functional dynamics of
geographically proximate organisations or the geographical scope. He argues
that the ambiguity of the concept is strengthened by the policy discourse
designed to enhance regional development. In policy terms, a cluster is
“formed when an actor (often with a base in a public organisation, or private
supporting institution) identifies a cluster, existing or potential, gives it a
name (often with the suffix of “valley”) and starts to act to strengthen, develop
and market it” (Malmberg 2002, p.16).

Apart from the companies which form the core, a cluster also encompasses
public or non-governmental organisations that provide inputs such as
competence development, education facilities, research and development,
marketing and other business development mechanisms that make the cluster
competitive (Porter 1998, 2000). However, although public policy towards
encouraging cluster development is vital, according to Enright (2000) the role
of policy is limited to improving rather than initiating and implementing.
Apart from providing public goods and competence generation mechanisms,
active policy participation aimed at “creating” a cluster could impede cluster
development, as noted by Enright and Ffwocs-Williams (2000, p.4):

“Guidelines on policy towards clusters must be premised on a view of

government as supporting existing and emerging clusters rather than
trying to create them from scratch. A policy aimed at developing
entirely new groups of firms in selected sectors can entail high costs,
high risks, serve as a screen for outmoded forms of industrial
targeting.”

McDonald & Vertova (2001) also maintain that policy arrangements are
beneficial in identifying potential clusters, to facilitate co-operation and
competition, provide linkages between firms and support institutions as well
as develop infrastructure. However, active policy participation in creating
clusters from scratch and imitations could restrain their development and
lead to lock-in effects.

In Sweden, in a much-publicised report entitled “[Link]”, Tson Söderström

et al (2000) caution against the proliferation of strategies that aim at creating

128
clusters in all regions of Sweden. While the authors maintain that Sweden has
the capability to generate dynamic clusters outside the Greater Stockholm
area”, they caution against “spreading resources too thinly over Sweden”
(2000, p.23).

On the other hand, Rosenfeld (1995) takes a policy perspective on the

development of clusters and contends that there is a need for active policy
intervention to promote cluster-based strategies to achieve economic
development. He argues that public policy should not only focus on building
the necessary infrastructure, but also on “building” competitiveness and
innovation by, for example, investing in social capital and learning.
Furthermore, he asserts that cluster definitions have been complicated by the
political discourse about networking and other inter-firm dynamics.

Some classifications of clusters48

As indicated earlier in this section, Porter (1990) did not explicitly discuss the
geographical scope of a cluster. This creates a size problem, especially since
the concept has become embedded in the industrial development debate. The
business economist Michael Enright (1996), however, is one of several
researchers who have tried to describe the geographical dimension of clusters.
He has suggested the following four varieties based on the works of Porter
(1990):

• Industrial clusters. In this case he uses Porter’s definition of a

concentration of companies in the same field and proximate
institutions. This implies that basically the geographical scope of this
cluster could be more or less anything. The reason why Enright uses the
word “industrial” in this respect can be understood given that Porter’s
studies are based on nations rather than regions or small geographical
areas.
• Regional clusters. Concentration of firms in the same or related branch
and regionally based proximate institutions. Paradoxically, regional
clusters allow firms to have a measure of both co-operation and
competition, as they are outward-oriented, such as the firms in the
Gnosjö region. Isaksen (2001, p 104) argues that the term “regional
clusters” encompasses a broad description of “industrial districts,
innovative milieus, local industrial complexes and new industrial
spaces”

48 There are several ways to classify clusters. Different researchers have provided
different typologies of clusters and my ambition with this study is not to give a
thorough typology. Instead this classification merely attempts to shed some light on
the topic and is not meant to offer a solid typology.

129
 • Industrial districts. Firms that are geographically bound and engaged
in complementary and common production processes, such as the firms
in the Third Italy.
• Business networks. Firms that are not necessarily located in the same
vicinity, but that have continuous communication and interaction with
one another. Although the actors have relational bonds and conduct
transactions with one another, this kind of cluster is functional rather
than geographical, and it does not necessarily mean that they are in the
same or related branch.

Park (2000) identified six kinds of clusters based on scope, technology and
division of labour. These are “three basic types of Marshallian districts, hub-
and-spoke, mature satellite, and pioneering high-tech industrial districts”
(Park 2000:329). Dunning (2000) uses a classification also with six varieties,
similar to that of Park:

• One category is the hub-and-spoke cluster, which is induced by the

presence of an engine enterprise. The dynamics of this cluster evolve
around gaining external economies through supplier and customer
networks.
• Another category is industrial districts, such as the districts of the Third
Italy. This cluster is characterised by factors of production, such as
access to labour and natural resources. The dynamics of the industrial
districts are achieved through external economies in terms of reduction
of transportation, transaction and labour costs. Three of the categories
are clusters induced by policy measures to create learning and
innovation opportunities. These are:
• Institutional clusters that are formed as a result of their proximity to
institutions of higher education and evolve around universities.
• Public bodies that provide public infrastructure. Examples are
Farnborough and Aldershot in the United Kingdom, which are centres
for aerospace and atmospheric research.
• Generic clusters that mostly are started as a science or technology park.
Their success depends on sophistication, upgrading of infrastructure
and an innovative milieu that encourages growth through knowledge
fluidity and learning externalities.
• The last category is export zones, which developing countries have
established in a bid to improve competitive strengths and to obtain
foreign currency. Multinational companies are invited to make a
presence in these clusters.

Rosenfeld (1997) takes a relationship perspective and identified three kinds of

clusters based on relationship densities.

130
 • Working clusters. This is a mature agglomeration of firms that
recognise their interdependencies. The firms and the supporting
institutions of the cluster are involved in interactive processes of
learning and innovation. The social structure allows for exchange of
information and ideas, business start-ups and networking, and it seeks
to strengthen the systematic and economic value and image of the home
region. This type of cluster, which is sometimes called “overachieving”
clusters, is found in the Silicon Valley, the ceramic tiles industry in Italy,
and the horticulture industry in Holland.
• Latent clusters. This kind of cluster is referred to as “underachieving”
clusters. These clusters have the potential to develop and become a
working cluster, but so far there is no social structure that allows for a
free flow of information. Despite the presence of firms and supporting
institutions, interaction between actors is weak. Although the actors in
the clusters are sharing much, they are not aware of the
interdependencies that exist between them. They also lack supporting
organisations that actively seek to create an arena for networking and
interaction. One such cluster is the biotech cluster in the research
triangle in North Carolina.
• Potential clusters. This kind of cluster is also known as “wannabees”.
They have many characteristics of a cluster, but lack competitive
advantages. These clusters are often found in the technology sector and
they are often initiated, designed and selected by policy makers.
Another type of potential cluster is an agglomeration of a group of firms
in the same or related industry that lack political power, skilled workers
or exposure. Examples of such clusters are the rural cluster of wood
products in eastern Oklahoma and microbreweries in Oregon.
Creativity and location49

In the last few decades, the notion of creativity has emerged as concept in
industrial location and in the policy regarding regional and national
development. If Michael E. Porter was the sage of competitive advantage by
devising the concept of clusters and its dynamics at the beginning of the 1990s,
Richard Florida has given potency to the old notion that creativity is the
precursor of competition and in – locating creativity to large urban areas –
also become “the cool-cities guru” of the new millennium (see Peck, 2005 for
an overview).

49 This text is largely based on Nuur, C., & Laestadius, S. (2009). Is the ‘creative class’ necessarily
urban? Putting the creativity thesis in the context of non-urbanised regions in industrialised
nations. European Journal of Spatial Development, 6, 10-22.

131
The point of departure of Florida’s creative class thesis is that the global
economic integration witnessed in the last decades makes obsolete the
conventional wisdom that differences between nations, regions and cities
depend on classical comparative advantages. Competitiveness – and here, we
are close to the Porter (1990; 1998) position – is a social and cultural
construction. Florida focuses on creativity, which according to him differs and
develops differently between cities that are assumed to be its loci. For
example, in the “Rise of the creative class: And How It’s Transforming Work,
Leisure, Community and Everyday Life” (2002c) he argues that in the era of
globalisation US cities with diversified economic activities and diversified
human capital formation have experienced the advent of a creative class while
others have witnessed the erosion of creativity. The creative class uses
technology as a channel to produce/sell products, penetrate markets and
communicate (Florida, 200b).

The members of the creative class are assumed to be attracted to milieus that
provide them with amenities that nurture their creativity and conform to the
phases in their life projects (Florida, 2004). Their mobility is thus conditional
and influenced by the erosion of the demarcation lines between work and
leisure. The creative class, according to Florida, constitutes a significant part
of the population of cities that are characterised by diversity and tolerance in
terms of ethnic, cultural, religious and sexual orientation (Florida, 2002c).

In short, the creativity thesis put forward by Florida (2002a; 2002b; 2002c;
2004 and 2005) could be summarised as follows:

• Competitiveness depends on success in recruiting talent – and this is

true more than ever in the knowledge based economy.
• There is a new and intensive global competition on talent – due to
globalization as well as to the rise of new Tiger economies.
• The highest probability to get into contact with these talented workers
is in highly urbanised regions.
• Talented people have strong expectations based on lifestyles, which
have a significant impact on their willingness to follow job
opportunities; they optimize their whole life project; closeness to the
opera may be as important as career options. Consequently, firms have
to follow their staff to the triple T –regions (Technology, Talent and
Tolerance), if they want the best ones, rather than the other way around.
• Triple T regions may develop a certain self-propelling dynamic
incorporating a lifestyle oriented economy (galleries, theatres and
meeting points) highly integrated with the lifestyles of the creative class.

Thus – following the above- it could be concluded that regions and cities that
cater to the needs of the creative class win while those that do not face a
melancholy demise. Overall, to policy makers in industrialised nations the

132
notion of creativity as an instrument of development is appealing since it is
not necessarily footed in certain regions, nations or industries. In addition –
superficially at least – it opens the way for creativity enhancing policies
primarily in areas in decline. Is this analysis reasonably plausible and is the
process unidirectional, or is there a potential variety as regards the
development paths towards the knowledge-based economy? Will non-
urbanised regions in the old industrialized countries die while larger
urbanised ones prosper? Is the role and behaviour of the creative class a
general or universal phenomenon independent of time and location or is this
a process related to specific historical or geographic conditions or relevant for
certain life cycles of industries or technologies? And more importantly what
do we mean by “creativity” and “the creative class”?

Creativity- a Contested Concept?

The characterisation of Florida´s creative class is loosely based on occupations

(engineers, architects, Scientists, economists etc.) (Florida, 2002c, 2004). It
also contains “bohemian” individuals who pursue lifestyles that add to the
variety of elements in the creative class concept (Florida, 2002a). In the end,
however, it might be the case that the creative class is both wider than the set
of citizens that have a Bachelor’s or Master’s degree (many bohemians do not)
and more narrow: teachers, social workers and pilots are e.g. not included in
the creative class (see e.g. Markusen, 2006 for a critical review of this).
However, close to a study by Janik and Toulmin (1973) on the creativity of
Wittgenstein’s Vienna (although not referring to it) Florida (2002a) has
examined how the geography of bohemians corresponded to the geography of
diversity (immigrants, gays), talent (defined as those with at least a Bachelor’s
degree) and high-tech industry location. This conclusion has been questioned
by other scholars, however (cf. e.g. Markusen, 2006 & Wojan et al, 2007).

In a wider sense, creativity as a concept is not only solely based on the level of
education attained by individuals. It concerns idea generation (cf, Franken,
1994; Stenberg and Lubart, 1999; Clark and James, 1999; Boden, 2004). It is,
as defined by Robert E. Franken, (1994, p396) “the tendency to generate or
recognize ideas, alternatives, or possibilities that may be useful in solving
problems, communicating with others, and entertaining ourselves and others”
or as Boden (2004, p1) puts it “the ability to come up with ideas or artefacts
that are new, surprising and valuable”. “Ideas” here include concepts, poems,
musical, compositions, scientific theories, cookery recipes, choreography,
jokes- and so on. “Artefacts” include paintings, sculptures, steam engines,
vacuum cleaners, pottery, origami, penny whistles – and many other things
you can name.

From the above approaches, it is clear that the notion of creativity relates to
the generation and appropriation of novel ideas by individuals (Franken,
1994; Stenberg and Lubart, 1999; Boden, 2004). Hence, it is not characterised

133
by the level of education of the individual but by a process of spontaneity in
its inception and sustainable outcomes although Clark and James (1999)
assert that it can also be oriented endeavour and as such depends on previous
knowledge. Even at this stage, Rehn and De Cock (2007) argue that the
literature on creativity focuses on how it evolves without paying much
attention to the potentials. Creativity may be limited to certain areas of
knowledge as well as spilloverspillover between such areas. It is, however, far
from easy to show that creativity within certain aspects of the arts has any
relation to creativity within e.g. high-tech industries (cf. Markusen, 2006 and
Wojan et al, 2007).

Limiting our perspectives to only Florida’s approach might lead to a failure to

consider the importance of context/community related tacit knowledge and
the social capital that underlies much of the generation and dissemination of
creative ideas that is relevant for industrial transformation. Industrial
creativity does not happen in solitude but instead takes place through
interplays with the result of cumulative learned actions and existing industrial
practices and technical solutions developed among people that do not
necessarily have academic degrees. Knowledge creation and diffusion often
takes place through vast formal and informal networks that interact and
provide ancillary inputs which contribute to nurturing existing capabilities
and the fostering of new ideas. This entrepreneurial creativity was formulated
already a century ago by Marshall in the famous line “(...) If one man starts a
new idea, it is taken up by others and combined with suggestions of their own;
and thus it becomes the source of further new ideas” (Marshall 1890/1920,
p.271). If this kind of – maybe traditional or old fashioned – creativity is losing
its importance then phenomena related to such a historical transformation
should be included in the analysis, not simply neglected. From the context of
regional development, creativity should be put in the social context where it is
induced, generated and diffused. As such the term involves the capacity of
actors to individually or collectively breed ideas (novel or not) which
eventually promote competitiveness in the environment or context where they
are introduced.

Creativity as a concept – not least as formulated by Boden (2004; see above)

– has a clear connection with innovativeness. Readers familiar with
innovation theory may recall the Schumpeterian definition (1934/1968) of
innovation as creative combinations. Already for Schumpeter (1943) it was
clear that the “new” could be a surprising combination of something “old”.
And he used the term creative destruction to capture the renewal capabilities
of capitalistic societies (Schumpeter 1943). Inspired by Schumpeter, students
of industrial and technological transformation have since then viewed
creativity as an important pillar of the mechanisms that contribute to renewal
(e.g. Dosi et al., 2000, Nelson, 1994).

134
There is also, following Boden (2004) an obvious connection between
creativity and the worlds of design and engineering. This has, among others
already been formulated by Simon (1962, 1996) who relates these activities to
the “sciences of the artificial”. What engineers (and designers) do is creating
– through design processes – artificial worlds which are largely based on
existing knowledge which is reorganized and synthesized into new forms.
Putting aside, the conceptual discussion on the notion of creativity and
limiting ourselves to spatial development, Florida’s perspective relates to the
abilities of technology, talent and tolerance in creating the conditions for
development. This is vital and worth analysing. While Florida’s “tolerance”
might not have been on the agenda half a century ago when the economic
thought governing regional differences was coined, technological diffusion
making industrial production redundant and the erosion of talent in the form
of brain drain were two issues that were even then seen as crucial in regional
economic transformations (Myrdal, 1957, Hirschman 1958, Hansen
1966/1972, Lausén 1969, Clark and Whiteman 1983). Also discussed by
Gunnar Törnqvist (1983) and Andersson (1981) the core theme of this
literature was that an innovative environment facilitates knowledge
transmission, competence development, regional learning and as such
provides synergies to innovate product, processes and methods that could be
commercialised. Evidently, Florida’s suggestion of the magnet effect of cities
discussed in “The rise of the creative class: And How It’s Transforming Work,
Leisure, Community and Everyday Life” and the repulsive effects discussed in
“The flight of the creative class:

The New Global Competition for Talent” although devised from the US
context, could be applicable to regions and as such match Myrdal’s spread and
backlash repercussions on peripheral regions. Thus, the rise of the creative
class in relatively urbanised regions is matched by the fall of the creative class
in peripheral regions since in absence of employment opportunities young
people in particular tend to migrate to urban regions in search of better living
conditions as stipulated by Myrdal (1957). At the same time, we are aware that
Florida’s creativity notion centres on an age-old question of the mechanisms
which promote and foster knowledge formation. Students of economic
geography have for decades discussed whether it is regional specialisation or
diversity that nurtures knowledge creation and diffusion. The specialisation
argument takes its point of departure from the contributions of Marshall
(1920), Arrow (1962) and Romer (1986). Known as the MAR perspective, the
underlying premise is that economic growth depends on the extent of
knowledge utilisation and technological progress. Scholars using this
perspective model the micro-economic environment from a knowledge
accumulation perspective and argue that firm competitiveness, productivity
and innovative capabilities depend on the extent of knowledge transfer within
organisations and knowledge spillover to other actors (e.g. Romer, 1986;
Feldman and Audtretsch, 1999). It is important here to note that the MAR

135
supposition also captures the importance of location, as it reveals the
economies associated with R&D and as such explores the geographic
dimension of economic development by looking at mechanisms that lie
outside the borders of the particular firm, including knowledge spillovers
between actors.

The creativity thesis put forward by Florida as an element of development, on

the other hand -manifested in the three T: s that make up Florida’s thesis –
Technology, Tolerance and Talent – relate strongly to Jane Jacob’s (1969)
study in which she argued that cities have magnetic abilities to attract people
and industries that perform diverse operations i.e. the opposite of the MAR
thesis. But it also has a family resemblance not only with the previously
mentioned study by Janik and Toulmin (1973) but with a Swedish study
entitled K-samhällets Framtid (Andersson and Strömquist 1988) in which it
is argued that the four Ks; Kunskap (Knowledge), Konst (Art), Kreativitet
(Creativity) and Kommunikationer (Communication) are the cornerstones of
the knowledge society. Andersson and Strömquist (1988) argued the future
growth of that society was based on developing webs, the hubs of which are
found in the rapidly growing K-regions with strong similarities to Florida’s T-
regions.

Jacobs (1969) argues that urbanised environments provide the nexus of

economic development. In The Economy of Cities, Jacobs (1969, p1) poses the
question “why some cities grow and why others decay”. In contrast to the
suggestions of the MAR arguments, such as knowledge spillovers due to
specialisation, Jacobs argues that the varieties in skills emanating from the
creative urban context are the decisive elements. The Jacobs’ diversity
supposition gained approval following several studies contrasting it with the
MAR perspective; For instance, Glaeser et al., (1992) compared the MAR and
Jacobs hypothesis by using data from 170 cities in the United States and found
it consistent with the Jacobs position that it is the variety that cities offer and
not the specialisation of labour/industry that explained knowledge spillovers.
Feldman and Audtretsch (1999) also tested the diversity versus specialisation
thesis by asking the question “does the specific type of economic activity
undertaken within any particular region matter?” and found that industrial
diversity is conductive to knowledge spillovers. Desrochers (2001) adopting a
Schumpeterian perspective also found support for Jacobs’ externalities
concerning the creative combinations of existing unrelated things promotes
human creativity and innovations in regions. From a policy perspective
Desrochers (2001) cautioned against policy strategies to promote the
emergence of regional specialisations.

In the Swedish context, several studies placing the creativity thesis in a

Swedish context have emerged; for instance; Andersson et al., (2005)
confirmed Jacobs’ hypothesis with regard to commercial patenting in Sweden
(1994-2001). According to this study, the propensity to engage in commercial

136
patenting activities was larger in urbanised regions that have diversified
economic activities. Also Melander and Florida (2007) used a path analyses
and measured traditional variables of human capital formation (wages,
education level etc.), technology and talent and found that the three relatively
urbanised cities of Stockholm, Gothenburg and Malmö accounted for 50
percent of the Swedish creative class. Stockholm scored highest with 30
percent, followed by Gothenburg 11.6 percent and Malmö 6 percent.

Tinagli et al., (2007) defining the creative class as the “the share of workforce
engaged in conceptual and creative types of occupation, like managers,
scientists, architects, engineers, artists, entrepreneurs, and many others”
argue that overall there is a north-south divide in Sweden. While urbanised
counties of Stockholm, Gothenburg and Malmö account for 60 percent of the
Swedish creative class by scoring high in the tolerance (64 percent of the
immigrant population are domiciled here).

Conclusions and implications

There is a myriad of literature on the underlying mechanisms of industrial
location and the policy implications relating to that. This chapter has provided
a brief overview of some of the key concepts. In summary. while the old
literature sought to discuss location from an equilibrium approach, followed
by the literature on the evolution of the centre as well as the periphery which
put location in the context of regional development. In recent years, we have
witnessed the emergence of concepts such as industrial districts, clusters,
creative cities, etc. which view the economies of location and the policy
implications could be summarised as embarking on a concept driven regional
development. The underlying conjecture is that, there is a “path dependency”
and territorial “lock-in” which means that history matters in understanding
industrial dynamics and the development of an industry in a particular
location. These relate to the notion that regions and locations that have similar
historical backgrounds would experience similar patterns of economic
development in the future (Malmberg & Maskell 1997). Accordingly, a
territory’s ability to develop and compete with other territories is conditioned
by history and comes because of cumulative decisions made by companies and
policy makers.

Not least the onset of globalisation has brought a shift in regional

development. While earlier, Keynes dictated the goals of regional
development i.e. regional convergence with the goal to offset regional
disparities, globalisation brought challenges and resulted in an endogenous
approach. This approach is focused on regional growth and centres on
inducing human capital formation, entrepreneurship and knowledge creation
and diffusion. To many old industrialised regions what this meant was a

137
challenge to acquire and put in motion the prerequisite of modern location
theories. This included overhauling of previous convergence measures and
promote mechanisms the educational infrastructure as well as the way of
thinking in terms of business development. To implement these theories into
practice poses both threats and opportunities that have far-reaching
implications on the role of human capital formation in general, and on higher
education and research development in particular. In addition, regional
development programmes focused on mechanisms to promote the growth of
small- and medium-sized firms (SMEs). Encouraged by the debate on the new
economy and the arrival of the information communications technology
(ICT), regions focused on measures to induce knowledge creation and
relationship building mechanisms in a hope to regenerate their local
economies. Some areas are concentrating on accessing codified knowledge,
while others are benchmarking successful concepts such as innovations
system (Lundvall 1990), Clusters (Porter 1990) to improve the competitive
strengths of their regions. The aim of these developmental initiatives is to
overcome potential disadvantages of firms in terms of size and location and to
facilitate learning, innovation, knowledge sharing, and creation of territory-
specific types of knowledge that are crucial to competitiveness.

In this context, in the 1990s, the regional development debate in Sweden, for
instance, was the three areas of northern Italy, Silicon Valley and, closer to
home, the Gnosjö region, have inspired local and regional development
policy initiatives. The clusters of the Third Italy (Becattini 1990) are spread
over several districts of Italy (Johanisson 1995; Brulin 2000), Silicon Valley
(Saxenian 1995) is one geographically large area, and the Gnosjö region of
Sweden consists of four municipalities. Silicon Valley is argued to have all
the necessary characteristics of the new economy. Its success was partly
attributed to the birth of the micro-electronics industry and the abundance
of venture capitalists, professional networks and institutions of higher
education, which are characteristics that the other two regions lack. The
industrial dynamics of the other two regions are instead characterised by the
presence of small- and medium-sized enterprises and their ability to
generate entrepreneurial spin-offs, to manage interdependencies and to
develop horizontal relationships that involve both formal and informal
institutions.

In addition, compared to Silicon Valley, the regions in Italy and Sweden are
not known for being home to knowledge-based firms. Instead, these regions
are known for the ability of the actors to engage in horizontal development
dynamics in which relational assets play a vital role. In terms of regional
development, the only similarity these three regions have in common is that
they were situated “in the periphery” before they were discovered.

138
Also, the onset and proliferation of the digital technology was hailed as an
important regional development tool that would compensate municipalities
in peripheral regions for the lack of higher education institutions. By all
means, ICT application has improved the competitiveness of firms and it has
brought with it the emergence of new kinds of firms, such as call centre firms.
These firms arrived at a time when it was observed that more and more
economic activities could be outsourced or carried out in temporary projects
by other companies. In a hearing at the end of 1990, on regional
development, it was alleged that four out of five companies that had received
state or regional subsidies to establish a presence in remote areas were firms
in the ICT industry (SOU 1999:138).

In a nutshell, what we have witnessed the last two decades could be described
as a concept driven regional development in which modern location theories
with a point of departure in knowledge building and relational exchanges
between different actors became the Mädchen für alles. According to
modern location theory, product and process innovations take place through
interaction between geographically proximate actors in systems like clusters
(Porter 1990), innovation systems (Lundvall 1990) and competence blocs
(Eliasson 1996). Through relational linkages facilitated by physical
proximity, the diffusion and transfer of contextual knowledge has become an
important mechanism to determine the competitiveness of not only
individual firms, but also the location they function from. Consequently,
collective knowledge spillover between different geographically proximate
actors results in new products and/or services and the improvement of the
existing ones (Audretsch, 2000; Maskell et al 1998; Porter 1990).

The concept of clusters (Porter, 1990) in particular, was on the lips of policy
makers. As Porter (1998) argued the assumption was that the “global
economy” does not diminish the role of the local environment but links
production facilities to a local milieu with global strategic capabilities. Global
firms need local capabilities to reach customers and they are dependent on
locally produced resources to be able to compete globally. According to
Porter (1998), this creates a paradox. Maskell et al (1998) argue that the
carriers of knowledge are accessed locally and by developing their
competences and operational capabilities the firms become able to innovate
products and processes.

Also the concept of creative cities (Florida, 2002) should be seen from the
above context where knowledge creation, innovations and creativity have
come to be viewed as mechanisms that impact location of economic activities
and the policy implications related to that.

139
References
Amin, S. (1972) Underdevelopment and dependence in Black-Africa- Origins
and contemporary forms. The Journal of Modern African Studies, 10(4), pp.
503-24.

Amin, S. (1974) Accumulation on a world scale: a critique of the theory of

underdevelopment, Hassocks: Harvester Press.

Amin, S. (1977) Unequal Development: An Essay on the Social Formations of

Peripheral Capitalism. Monthly Review Press.

Arrighi, G. (1978) The Geometry of Imperialism – the Limits of Hobson’s

Paradigm, London: NLB.

Emmanuel, A. (1969) Unequal Exchange: A Study of the Imperialism of Trade.

Alonso, W. (1968). Urban and regional imbalances in economic development.

Economic development and cultural change, 17(1), 1-14.

Amin, A. & Thrift, N. (1995) “Globalisation, Institutional “thickness” And

Local Economy”. In Healey, P., Cameron, S., Davoudi, S., Graham, S. and
Madani-pour A, (eds.) Managing cities, the new urban context, Chichester:
John Wiley.

Amos Jr O. M. (1988) Unbalanced Regional Growth and Regional Income

Inequality in the latter stages of Development. Regional science and Urban
Economics 18.

Andersson, Å. E and Strämquist, U. (1988) K-Samhällets Framtid (Eng: The

future of the K-society), prisma.

Andersson, Å. E. (1981). Structural change and technological development.

Regional Science and Urban Economics, 11(3), 351-361.

Arrow, K. J. (1962). The economic implications of learning by doing. The

review of economic studies, 155-173.

Arthur, B., (1989), “Competing Technologies, Increasing returns, and Lock-

in by historical small events”, Economic Journal 99.

Asheim, B., (1996), “Industrial Districts as Learning Regions: a Condition

for Prosperity”; European Planning Studies, Vol. 4, No. 4.

140
Asheim, B., (2000), “Industrial Districts: The Contributions of Marshall and
Beyond” (In Clark, Gordon L.; Feldmann, Maryann P.; Gertler, Meric S.
(Ed.): (2000): Oxford Handbook of Economic Geography. Oxford
University Press).

Asheim, B., (2002), “Temporary Organisations and Spatial Embeddedness

of Learning and Knowledge creation”. Geografiska Annaler, Vol. 84 B, No.
2.

Audretsch, D., (1993), “Industrial policy and International Competition” in

Industrial policy in the European community” edited by Nikolaides, P.

Audretsch, D., (2000), “Knowledge, Globalisation, and Regions: An

Economist’s Perspective” In Regions, Globalization, and the Knowledge-
based Economy, Edited by Dunning, H.J, Oxford University Press.

Baker P., (1995), “Small Firms, Industrial Districts And Power Asymmetries”,
International Journal of entrepreneurial behaviour and research, MCB
University.

Becattini, G., (1992), “The Marshallian Industrial District As a socio-

economic notion”, in Pyke, F. / Becattini, G. / Sengenberger, W. (1990)
(eds.), Industrial districts and inter-firm co-operation in Italy,
International Institute for Labour Studies, Geneva.

Beckmann, M. (1972) “Von Thünen Revisited: A Neo-Classical Land Use

Model”, Swedish Journal of Economics, 1-7.

Berggren, C., Brulin, G., & Gustafsson, L. (1998) “Från Italian to Gnosjö. Om
det sociala kapitalets betydelse för livskraftig industriella regioner”.,
Stockholm: the council for working life research.

Bergström, F. (1998) “Regional Policy and Convergence of Real per capital

income among Swedish regions” SSE/EFI working paper series in
Economics and finance no.8

Boden, M. A. (2004) The creative mind, Myths and Mechanics. 2nd edition
Routledge, Taylor

Brant, M. (2001) “Nordic Clusters and Cluster Policies”, in Cluster Policies-

Cluster Development? Edited by Mariussen, Å- Nordregio R2001:2.

Brown, R. (2000) “Cluster Dynamics in Theory and Practice with

Application to Scotland”, Regional and Industrial Policy Research paper,
Number 38: European Policies Research Centre, niversity of Strathclyde.

141
Brulin, G. (2002) “Faktor X, Arbete och Kapital i en lokal värld “, Stockholm,
Atlas.

Brusco, S. (1986) “Small Firms and Industrial Districts; The experience of

Italy”, In New Firms and Regional Development in Europé; Edited by D.
Keeble and E. Wever, London- Croom Helm´.

Cairncross, F. (1997) “The Death of Distance: How The Communications

Revolution will change our lives”. Boston, Massachusetts. Harvard
Business School Press.

Castell, M. (1999) “Informationsåldern: Ekonomi, samhälle och Kultur.

Nätverkssamhällets framväxt,” Diadolos, Göteborg.

Clark, G, L and Whiteman J (1983) Why Poor People Do Not Move: Job Search
Behavior And Community and Everyday Life, Basic Books, New York, NY.

Cooke, P. (2002) “Knowledge Economies: Clusters, Learning and Co-

Operative Advantage” Routledge.

Christaller, W. (1966) Central places in southern Germany. Prentice-Hall.

Cossentino, F., Pyke, E., & Sengenberger, E. (1996) “Local and Regional
Response to Global pressures; The case of Italy and its Industrial
districts”- Research Series; International Institute of Labour studies-
Geneve.

Cox, R. (1995) “Lessons From 30 Years Of Science Parks in the USA”, In

J. Gibb, (ed.) Science parks and innovation centers: Their economic and
social impact, conference proceedings, Berlin, Feb. 13-15. Amsterdam:
Elsevier.

Ottati, G. D. (1994). Cooperation and competition in the industrial district as

an organization model. European Planning Studies, 2(4), 463-483.

Nelson R.R (Ed). Princeton University Press, Disequilibrium Amongst Local

Labor Markets. Environment and Planning A 15(1) 85 – 104.

Dosi, G., Nelson, R., & Winter, S. (Eds.). (2001). The nature and dynamics of
organizational capabilities. OUP Oxford.

Dunning, J. H. (2000) “Regions, Globalization, and the Knowledge

Economy: The issues Stated”- Oxford University Press.

Ejermo, O. (2005) Technological diversity and Jacobs’ externality hypothesis

revisited. Growth and Change, 36(2), 167-195.

142
Engstrand, Å-K. (2003) “The Road Once Taken-Transformation of labour
markets, politics and place promotion in two Swedish cities, Karlskrona
and Uddevala 1930-2000” Working Life In Transition.

Enright, M. & Ffowcs-Williams, I. (2000) “Enhancing the competitiveness of

SMEs in the Global Economy: Strategies and Policies”, Paper presented at
the conference of ministers responsible for SMEs and Industry ministers,
Bologna, Italy 14-15 June 2000.

Enright, M. (1996) “Regional Clusters and Economic Development- A

Research Agenda” in Business Networks: Prospects for Regional
Development- Edited by Staber, U. Schaefer, N. and Sharma, B. de Gruyter,
Berlin.

Epstein, G. (2003) “The Role and Control of Multinational Corporations in

the World Economy”, in: The handbook of Globalisation, Edited by J.
Michie, Edward Edger publishing house, Cheltenham, UK.

Feser, E.J. (1998) “Old and New Theories of Industry Clusters” In Clusters
and Regional specilization; On Geography Technology Networks; London,
Pion Limited.

Florida, R. (2004) Cities and the Creative Class. Routledge, New York, NY.

Florida, R. (2002a) Bohemia and economic geography. Journal of Economic

Geography 2 (1) 55-71.

Florida, R. (2002b) The economic geography of Talent. Annals of the

association of American.

Florida, R. (2002c) The Rise of the Creative Class: And How It's Transforming
Work, Leisure.

Florida, R. (2005) The Flight of the Creative Class: The New Global
Competition for Talent. Harper Francis.

Frank, A. (1969) Capitalism and Underdevelopment in Latin America:

historical studies of Chile and Brazil, New York: Monthly Review Press.

Franken, R, E. (1994) Human Motivation. Brooks/Cole publishing, CA.

Freeman, C., (1988) “Japan: A new National System of Innovation?”, in

Technical and Economic Theory, edited by Dosa, G, Freeman, C, Nelson R,
Silverberg & G, Soete L: UK printership.

Furtado, C. (1964) Development and Underdevelopment. University of

California Press.

143
Furtado, C. (1973) The Concept of External Dependence in the Study of
Underdevelopment. In The Political Economy of Development and
Underdevelopment, edited by C. K. Wilber. New York: Random House.

Glaeser E.L., Kallal H.D, Scheinkman J.A and Shleifer A. (1992) Growth in
Cities. Journal of

Glaeser, E, L. (2000) “The new economic of urban and regional growth” in

(Clark, Gordon L. Feldmann, Maryann P.; Gertler, Meric S. (Ed.): (2000):
Oxford Handbook of Economic Geography. Oxford University Press)

Glasmeier, A. K. (2000) “Economic geography in practice: local economic

development policy” in (Clark, Gordon L.; Feldmann, Maryann P.; Gertler,
Meric S. (Ed.): (2000): Oxford Handbook of Economic Geography. Oxford
University Press)

Granovetter, Mark. (1985) “Economic Action and Social Structure: The

Problem of Embeddedness”, American Journal of Sociology. Vol. 91 No. 3.

Gummesson, O. (1997), “Därför lyckas Gnosjö: bygden som har blivit ett
begrepp “, Stockholm, Ekerlids.

Hall, P. (1974) The containment of Urban England. The geographical Journal,

140(3) 386-408.

Hirschman, A. O. (1958) “The Strategy of Economic Development”, New

Haven, Yale University press.

Hirschman, A. O. (1967) “Development projects observed”, The Brookling

institutet Washington.

Hoover, E. M. (1948) The location of economic activity.

Isaksen, A. (1997) “Regional Clusters and Competitiveness: The Norwegian

Case” European Planning studies, Vol 5 Number 1.

Isaksen, A. (2001) “Building Regional Innovations System: Is Endogenous

Industrial Development possible in the Global Economy?”, Canadian
Journal of Regional Science, XXIV:1.

Isard, W. (1951) Interregional and regional input-output analysis: a model of

a space-economy. The review of Economics and Statistics, 318-328.

Isard, W. (1954) Location theory and trade theory: short-run analysis. The
Quarterly Journal of Economics, 305-320.

Jacobs, J. (1969) The Economy of Cities. Random House, New York, NY.

144
Johanisson, B. (1984) “A cultural perspective on small business-local
business climate” in the International small business journal, vol 2.

Johanisson, B. (2002) “Regional Policies and Comparative Advantage””,

Cheltenham, Edward Elgar.

Kebble, D. (1997) “Small firms, Innovation and Regional development in

Britain in the 1990s”, Regional studies.

Krugman, P. (1996) “The self-organizing economy”- Blackwell Publishers.

Krugman, P. (2000) “Where in the world is New Economic Geography?” (In

Clark, Gordon L.; Feldmann, Maryann P.; Gertler, Meric S. (Eds) Oxford
Handbook of Economic Geography, Oxford University Press)

Laestadius, S. (1992) “Arbetsdelningens dynamik” , Arkiv, Lund.

Laestadius, S. (1999) “Marshallian dynamics and new growth theory,”

TRITA IEO-R 1999:1 KTH.

Laestadius, S. (2005) “Det Dahménska utvecklingsblocket – några

reflektioner” i Benner, Mats, ed., Innovationer – omvandling och dynamik,
Studentlitteratur.

Lasuén J.R. (1969) On growth poles Urban studies, Vol 6, Number 2.

Lever, W.F. (1980) Manufacturing Linkages, Industrial Dynamics and The

Transmission of Growth”. Regional science and Urban Economics.

Lindberg, H. (1999) “Industriell Omvandling I Söderhamn och Hudiksvall

under trävaruepoken 1850-1995”, Licentiattuppsats ekonomiska-
historiska institution Uppsala Universitet , Uppsala.

Losch, A. (1954). The economics of location. Translated by WH Woglom and

WF Stolper. New Haven: Yale UP.

Lundequist, P & Power, D. (2002) Putting Porter into Practice? Practices of

Regional Cluster Building: Evidence from Sweden. European Planning
Studies, V. 10. 6.

Lundvall, B-Å. (1992), “National Systems of Innovations- Toward a theory

of innovation and interactive learning”, Printer Publishers, London.

Malmberg, A. (1998),”Den gäckande agglomerationsfördelen: teoretiska

principer och empiriska paradoxer” Svensk Geografisk Årsbok, Årg. 74.

145
Malmberg, A., & Maskell, P. (1997), “Towards an explanation of regional
specialisation and industry agglomeration” in Regional Specialisation and
Local environment: Learning and Competitiveness. NordREFO 3.

Malmberg, A. (2002) “Klusterdynamik och regional näringslivsutveckling,-

begrepp diskussioner och forskningsöversikt” – a report for ITPS.

Malmberg, A., Solvell O., Zander, I. (1996) “Spatial clustering, Local

Accumulation of Knowledge and Firm Competitiveness”, Geografiska
Annaler. Series B, Human Geography, Vol 78.

Markusen, A. (1996) “Sticky Places in Slippery Space- A typology of

Industrial districts”, Journal of Economic Geography, vol.72, No.3.

Marshall, A. (1890/1920/1916/ 1920,) “Principles of Economics. An

Introductory Volume.”

Martin, R., & Sunley, B. (2003), “Deconstructing Clusters: Chaotic Concept

or Policy Panacea?”, Journal of Economic Geography 3:5.

Mascanzoni, D. & Novotny, M. (2000), “Lokomotivföretaget I Italien-Motor

för sysselsättning och regional utveckling “, Stockholm.

Mascanzoni, D. (2003) “Snabbväxande Kluster I Italien-från fattigdom till

spridd välfärd” – Institutet för Tillväxtpolitiska studier, Elanders Gotab,
Stockholm.

Maskell, P., & Malmberg, A. (1999) “Localized learning and industrial

competitiveness”, Cambridge Journal of Economics, 23:

McCann, P. (2002) “Classical and Neoclassical Location- Production

Models” in: Industrial Location Economics; McCann (editor), Edvard
Edgar, Cheltenham, UK.

McDonald F. Vertova G. (2001) “Geographical concentration and

competitiveness in the European Union” European Business review, MCB
University Press.

McGranahan, D., & Wojan, T. (2007). Recasting the creative class to examine
growth processes in rural and urban counties. Regional studies, 41(2), 197-
216.

Mellander, C and Florida, R. (2007) The creative class or the human class”,
Cesis, electronic working

Morgan D. Thomas. (1969) “Growth pole theory: An examination of the some

of its basic concepts”. In “Hansen; Growth centres in regional development.

146
Morrison, P.A and Wheeler, J.P (1976) Rural Resistance in America?
Population Bulletin 31, (3-26).

Myrdal, G. (1957) “Economic theory and the underdeveloped regions”,

London: Ducksworth.

Nelson, R (1994) The Co-evolution of Technology, Industrial Structure, and

supporting institutions.

O`malley, E.,& Van Egeraat, C.(2000), “Industry Clusters and Irish

Indigenous manufacturers: Limits of the Porter view” , in the Economic
and Social Review Vol 31, Number 1.

Park, S-O. (2000) “Innovation Systems, Networks, and the Knowledge-

Based Economy in Korea” in Globalization, and the Knowledge-based
Economy, Edited by Dunning, H.-Oxford University Press.

Parr, J.B. (2002) “The Location of economic activity: Central Place Theory
and the Wider Urbanisation” in Industrial Location Economics; McCann
(editor), Edward Edgar, Cheltenham, UK.

Peck, J. (2005) Struggling with the Creative Class. International Journal of

Urban and Regional

Perraton, J. (2003) “The scope and Implications of globalisation”, in The

handbook of Globalisation, Edited by J. Michie, Edward Edger publishing
house, Cheltenham, UK.

Pérroux, F. (1950) “Economic space; Theory and applications;” Quarterly

Journal of Economics, Vol. 46.

Pettersson, J-E. (1983) Avesta Jernverks AB 1883-1983. (Eng: The Avesta iron
Works 1883-1983)

Phelps, N. A., Fallon, R. J., & Williams, C. L. (2001). Small firms, borrowed
size and the urban‐rural shift. Regional Studies, 35(7), 613-624.

Piore, M. Sabel, C. (1984), “The second industrial divide.” New York

Polése, M.E. (1999) “From Regional Development to Local Development: On

the Life, Death and Rebirth(?) of Regional Science as a Policy Relevant
Science”, Canadian Journal of Regional science XXII:3 (Autumn 1999).

Porter, M. (1990) “The competitive advantages of Nations” London and

Basingstoke, MacMillan.

Porter, M. (1998) “Clusters and the new economics of Competition” Havard

Business Review Nov-Dec.

147
Porter, M. (2000) “Locations, Clusters and Company Strategy”, (In Clark,
Gordon L.; Feldmann, Maryann P.; Gertler, Meric S. (Ed.): (2000): Oxford
Handbook of Economic Geography. Oxford University Press)

Putnam, R. D. (1993) The prosperous community;: Social Capital and Public

Life. in the American Prospect, Vol 3 No.13.

Raines, P. (2001) “The cluster Approach and the Dynamics of regional policy
making”, The European policy Research centre: Glasgow. Regional and
Industrial policy research paper Number: 47.

Andersson, R., Quigley, J. M., & Wilhelmsson, M. (2005). Agglomeration and

the spatial distribution of creativity. Papers in Regional Science, 84(3),
445-464.

Romer, P. M. (1986). Increasing returns and long-run growth. The journal of

political economy, 1002-1037.

Romer, P. M. (1990). “Endogenous Technological Change”, Journal of

political economy, Vol. 98.

Rosenfeld, S. (1995), ”Overachievers: Business Clusters that work, Prospects

for regional development”, Chapell Hill NC.

Rosenfeld, S. (1997) “Bringing Business Clusters Into the Mainstream of

Economic Development”, European Planning Studies, Volume 5, Number
1.

Sadler, D. (2004) “Cluster Evolution, the Transformation of Old Industrial

and the Steel Industrial Supply Chain in North East England” Regional
studies, Vol 38.1.

Sandefur, G.D, and Scott, W.J. (1981) A Dynamic Analysis of Migration: An

assessment of the

Sandell S. H. (1977) Women and the Economics of Family Migration. The

Review of Economics and

Saxenian, A.L. (1994) “Regional Advantage: Culture and Competition in

Silicon Valley and Route 128 Cambridge”, MA: Harvard University Press.

Schumpeter, J. (1934), “The Theory of Economic Development”, Cambridge,

Harvard University Press.

Scott, A J. (2000) “Economic Geography: the great half century.” (In Clark,
Gordon L.; Feldmann, Maryann P.; Gertler, Meric S. (Ed.): (2000): Oxford
Handbook of Economic Geography. Oxford University Press)

148
Simon, H (1962, 1996) The Sciences of the Artificial, 3rd ed., MIT Press,
Cambridge, Mass.

Söderström, H-T, Braunerhjelm, P, Friberg R, Norman, V, Sölvell (2001),

“[Link]-Sverige I den nya ekonomiska geografin” SNS- Stockholm.

Sölvell, Ö, & Birkinshaw., J (2000), “Multinational Enterprises and The

Knowledge Economy: Leveraging Global Practices,” in Regions,
Globalization, and the Knowledge-based Economy, Edited by Dunning, H.

Sölvell, Ö., Zander, I., and Porter, M., (1991), “Advantage Sweden,”
Stockholm: Norstedts.

Staber, U. (1998), “Inter-firm cooperation and competition in industrial

districts” Organisation studies 19.

Storper, M. (2000a), “Globalization, localization and trade” In: Clark G.L.,

Feldman M.P., Gertler M.S., (eds.). The Oxford Handbook of Economic
Geography. Oxford, Oxford University Press.

Storper, M. (1995), “The resurgence of regional economies. Ten years later:

the region as a nexus of untraded interdependencies” European Urban and
Regional Studies, 2.

Storper, M. (1997), “The regional world: Territorial Development in a global

economy”. N.Y London, Guilford.

Storper, M. (2000b), “Globalization and Knowledge Flows: An Industrial

Geographer’s Perspective” In Globalization, and the Knowledge-based
Economy, Edited by Dunning, H.J, Oxford University Press.

Sweden. Population and Environment, 3(2) (125-134).

Tacoli, C. (1998) Rural-Urban Interactions: a guide to the literature.

Environment and Urbanization

Tignali, I, Florida, R, Ström, P, Wahlqvist E. (2007) Sweden in the Creative

Age. Research report;

Törnqvist, G. (1983). Creativity and the renewal of regional life. In Creativity

and context: A seminar report (Vol. 50, pp. 91-112). Lund: Gleerup.

Waits, M. (2000), “The Added Value of the Industry Cluster Approach to

Economic Analysis, Strategy Development, and Service Delivery,”
Economic Development Quarterly, Vol. 14, No.1.

149
 Walker, R. (1989), “A Requiem for Corporate Geography: New Directions in
Industrial Organization, the Production of Place and Uneven
Development”, Georgrafiska Annaler. Series B, Human Geography Vol. 71.
No.1.

Weber, A. (1909) Theory of the location of industries. Chicago: The University

of Chicago Press.

Wigren, C. (2003), “The spirit of Gnosjö- The Grand Narrative and beyond,”
JIBS Dissertation Series No.017.

Wojan, T. R., Lambert, D. M., & McGranahan, D. A. (2007). Emoting with

their feet: Bohemian attraction to creative milieu. Journal of Economic
Geography, 7(6), 711-736.

150
Dynamics in Swedish Industrial and
Political history

This article by Eric Giertz was first published in November 2016 by

the Division of Sustainability and Industrial Dynamics, Department
of Industrial Economics and Management (INDEK)), KTH in the
anthology: A Dynamic Mind. Perspectives on Industrial Dynamics
in Honour of Staffan Laestadius. (Editors Pär Blomkvist & Petter
Johansson)

Eric Giertz, Ph.D., is a Professor emeritus in Industrial Economics and

Management at KTH. He is specialized in Business Development linked to
structural changes in industry. For 20 years, Giertz worked as a strategy
consultant in the field of business development. His clients included companies
in manufacturing and service industries as well as ministries and authorities.
He has experience as member of executive boards in companies in various
industries and as chair and part owner of successful new ventures. Giertz is the
Founder of KTH Executive School, which develops the strategic leadership and
business skills of corporate executives. He was Vice President for Business &
Community Liaison at KTH 2006-2011. He is also a member of the IVA, the
Royal Swedish Academy of Engineering Sciences, and the former Chairman of its
Management Division.

1
Dynamics in Swedish Industrial and
Political history
Eric Giertz

Rules and regulations for industry are of great importance for the
wellbeing of a country. In this article, I discuss the complex interplay
between governmental actions and industrial development in
Sweden during one and a half century. I divide the period into three
phases.
The first phase initiated in the mid-1800s when a number of decisions
speeded up the process of industrialization in Sweden – giving
Sweden many new ventures founded by entrepreneurs. The second
phase initiated by the depression after World War I. The ownership
of Swedish industry was concentrated to a few financial spheres,
which closely cooperated with the Social Democratic government and
the reformist labor movement. The Social Democrats lost the election
in 1976 and the crises in the late 1970s initiated a new political era
which was followed by the third phase in the late 1980s.
Globalization has fundamentally changed possibilities for small
nations, like Sweden, to form national strategies for growth. Still the
“Swedish model” from last century evokes a nostalgic response from
many Swedes. Some concepts and institutions, as well as rules and
regulations, also still survive with some strength, even though they
originated in a different era and are less relevant today. I hope that
the reader of this article shall be aware that yesterday’s successful
principles of setting rules for industry, as well as organizing and
managing different businesses might no longer be appropriate.

Introduction
In this article the development of the Swedish industry, and the industrial
dynamics in Sweden, is linked to political and societal changes. One of my
ambitions is to demonstrate how the industrialization and the development of
a competitive industry in Sweden was formed and has been maintained by a
complex interplay between government and business. An interplay between

2
political movements in society, governmental decisions and regulations on the
one hand and business and industrial development on the other. The rules and
regulations, as well as the interactions and cooperation between different
actors, has changed considerable over time. Shorter periods of great turmoil
were followed by longer periods of a relatively stable growth. The development
during one phase has of course also built the foundation and the starting point
for the next. In this article I have divided the development 1850-2015 into
three different growth periods, each initiated by a period of political and
industrial turmoil and great changes in society (This whole article, and
especially the beginning, draws on 37 articles published in Giertz (Ed) 2008).
A number of liberal reforms initiated the first phase during two decades in
the mid-1800s. Freedom of business was introduced, an act introducing the
principle of limited company liability was passed and the guild system was
abolished. Sweden also introduced free trade, a new banking law and the right
for any citizen to establish firms. In parallel, a political new order brought new
men with liberal ideas to the representative bodies. Government initiated
investments in new infrastructure, which were combined with local, private
investments. The liberal reforms and the new infrastructure set up the
foundations for a belated but rapid industrialization in Sweden. Most new
firms completely focused on the domestic market but some also became
successful export companies in the wood-, pulp and paper-, steel- and
engineering industries. During the second half of the 19th century a new
entrepreneurial capitalism flourished, which hardly resembled the traditional
and stable social structure in the beginning of the 19th century.
The turmoil following World War I initiated the second phase. There was
full democracy in Sweden from 1921, but a deep depression and a tough
deflationary monetary policy also followed the war. Between 1920 and 1922,
liquidations and bankruptcies were very common in Sweden. Many company
founders and their families lost their businesses. A few financial spheres took
over the ownership. The spheres became even fewer after a new depression a
decade later. The consolidation of the export industry was accompanied by a
massive consolidation of the infra service industries when government
increased its grip on the infrastructure, through the acquisition of all local
phone associations and telephone companies, the nationalization of all private
railroads and the establishment of Vattenfall for regional power generation
and later the deployment of a nationwide transmission grid.
It was not only the control over the Swedish industry and infrastructure that
was centralized in the 1930s. After the 1936 election, the political power in
Sweden was gathered in the reformist labor movement's hands for many
decades to come. Leading Social Democrats replaced the socialist dogmatism
for a pragmatic cooperation with industry and built close personal contacts
with the most important owners and CEOs in Swedish industry. They saw a
competitive export industry as the foundation of a future welfare society.
During the 1950s and 1960s Swedish industry harvested major export

3
 successes. Sweden's share of world trade rose almost dramatically; Sweden
took the lead in terms of economic growth and a welfare society emerged
through the close cooperation between central representatives of
government, industry and trade unions in a very centrally but informally
managed truly corporative society.
In the late 1960s, the consensus and close cooperation between government
and private industry was put to an end. A young generation of leading Social
Democrats, that formed a new government in 1969, took a distance from the
former close cooperation with big business. In the beginning of the 1970s,
thoughts of socialization, governmental enterprises, wage-earner funds,
new labor laws and socialistic national economic planning were on top
of the political agenda among Social Democrats in Sweden. Thus, towards
the end of the second phase the close and informal cooperation between
government and industry ended and was replaced by more confrontation,
legislation and formal bureaucracy. However, the belief in the strong state
with a powerful government responsible for developing, controlling and
distributing welfare to all citizens in the country rather increased than
decreased in the beginning of the 1970s.
The third phase was initiated in the mid-1970s, when the development
took a fast new turn with industrial crises and heavily underfunded welfare
systems. The crisis was manifested in decreased production, reduced
market shares for Swedish industry on the global market, increased
unemployment, negative trade balance and low profitability in many
industries. The Social Democrats lost the election in 1976 and a new
political era, with pending majorities, was born. Focus was once again
transferred to economic growth. But this time – not only incumbents – but
also new ventures and higher efficiency in the public sector were
important targets. Deregulation, which increased the pressure for change
in domestic and public activities were carried out.
Since the late 1980s, thus, a new pattern emerged – a globalization phase
had begun. Trade barriers were taken away and new trade blocs were
formed. Meanwhile, the information and communications technology
shrunk the world further. Many markets became global. The financial
market is global, both manufacturing and service companies are operating
on global markets. The companies that provide infra services are nowadays
competing on international markets. Questions about corporate locations
of activities are constantly under discussion, as well as the exchange of labor
in increasingly open economies. Major structural changes permeate the
business world. Companies are acquired and merged. Many companies
concentrate on their core business and choose to procure services that were
previously made in- house. Globalization has also fundamentally changed
possibilities for small individual nations, like Sweden, to form national
strategies for growth.

4
Liberal reforms, industrialization and
entrepreneurship
Until the mid-1800s Sweden was relatively untouched by the incipient
industrialization in other parts of Europe. Liberal ideas that had taken root
elsewhere had not left any deep traces in Swedish society, but in the mid-
1800s a series of reforms were implemented.

Liberal reforms, a new political order and new

communications are introduced
A number of liberal reforms were initiated in 1844 when king Karl XIV Johan
was succeeded by his more liberal minded son Oscar I and grandson Karl XV.
During two decades a lot of fundamental changes were introduced. Along with
a general liberalization Sweden introduced the right for any citizen to establish
firms and free trade was introduced; import and export bans were lifted, and
both foodstuffs and raw materials were completely duty free. A new banking
law was added, which led to the establishment of new commercial banks. The
banks attracted capital which was loaned to an expanding industry.
Predecessor was A.O. Wallenberg, who founded Stockholms Enskilda Bank
already in 1856. The number of commercial banks peaked in 1908. Then there
were 84 companies (Lindgren 2008, p. 87). The liberal reform period in the
mid-1800s ended with a reform of the political life in 1865. The political new
order brought new men to the representative bodies. Conservative groups
(nobility, clergy, bourgeois and peasants) lost their former privileges. They
were replaced by more liberal forces, in city councils as well as in the new two
chamber parliament, which advocated industrial development.
The state also took on new commitments, which were financed by large
borrowings abroad. These included development of communications, which
improved dramatically from the mid-1800s (Kaijser 2008:1). A novelty was
the electric telegraph. A telegraph agency (Telegrafverket) was established in
1853 to build a state-owned national telegraph system. That same year the
Stockholm-Uppsala telegraph line was in use for public traffic. Another
novelty was the railroad. In 1853 parliament decided to build a nationwide
railway. From the start the railway network was planned as a nationwide
transport system. Government built a network of trunk lines but left it to
private railroad companies to build the ramified connections. In 1880 there
was a total of 6,000 kilometers railway in the country, of which 2,000 were
owned by the state and controlled by Swedish Rail (SJ). So in constructing the
nationwide railway system governmental initiatives were combined with local
private entrepreneurship. For telephony, as well as electricity, on the other
hand the exploitation started almost invariably by local initiatives.

5
 In 1880 the Stockholm Bell Telephone Company established the first
telephone network in Sweden with just over one hundred subscribers
(Kaijser 2008:1). The Bell Company sold telephone networks to public
institutions and companies. Bell applied the same concept in many other
cities in the US and in Europe. In some countries the Bell Company had
a patent, but not in Sweden. In Stockholm the Bell Company was
challenged. The engineer and entrepreneur Henrik Tore Cedergren felt that
the phone should also be aimed at households. He established a competing
telephone company in 1883, Stockholms Allmänna Telefonaktiebolag (the
Stockholm Public Telephone Company), which purchased technical input
from a newly founded engineering company, Telefonaktiebolaget LM
Ericsson, which also supplied some spare parts to the telegraph agency.
Stockholms Allmänna challenged the Bell Company by offering
significantly lower subscription fees. Bell company responded by lowering
their fees. The intense competition between the companies led to a rapid
increase in the number of subscribers. In 1885, there were 5,000 telephone
sets in Stockholm, which was more than in any other city in the world at
that time! There also were around 400 other local networks all over
Sweden at that time.

A belated but rapid and entrepreneurial industrialization

The liberal reforms, the freedom to start new companies, a political new
order and free trade – which was introduced in the mid-1800s – set, along
with investments in infrastructure and a new banking law, the foundations
for a belated but rapid industrialization in Sweden. New manufacturers
produced food, agricultural implements and textiles for the domestic
market. The town Borås in the west of Sweden converted from a marketplace
to a textile city and the town Norrköping became the center for the wool
industry (Schön 2008). The first steam sawmill began operating in Ådalen in
northern Sweden in 1849 and was quickly followed by others. A huge
international demand, especially from Britain, paved the way for a rapid
expansion of the sawmill industry along the coast of Norrland.
Workers, without experience from their new profession, sought out from
around the country to job opportunities in the new towns. They often lived
in poor housing conditions, while new entrepreneurs could make huge
profits and expose their new wealth in an extravagant way of living. But the
market was characterized by rapid fluctuations in both demand and
prices, which gave the whole industry a speculative character. Big profits
and shortage of labor suddenly was replaced by bankruptcies and
unemployment. In parallel many new successful export companies in the
pulp and paper, steel and engineering industries were born during the
second half of the 1800s. A new entrepreneurial capitalism flourished,
which hardly resembled the traditional and stable social structure in
Sweden.
6
The industrialization meant that people moved from rural areas to towns and
cities as well as from agriculture to industry. But it was not a relocation of a
constant number of citizens. During the 19th century Sweden's population
increased from just over 2 million to over 5 million despite a massive
emigration of more than one million Swedes, primarily to America. In 1850,
there were still only 20,000 industrial workers across Sweden. But
liberalization and communications speeded up the industrialization and
created new jobs. It was among poor people on the countryside that the
emerging industry recruited its workforce. Industrial work was, like
emigration, an attractive alternative to a hopeless existence in poorest misery.
It was mainly younger people who broke up and sought their livelihood in the
new industries. They moved from place to place, and from company to
company. They often ended up in the big cities that contained many
companies and thus many alternative workplaces. To a large extent, it was also
firms in cities who paid the highest salaries.
The emerging industrialization reformed society and created new social
classes. A new working class emerged, which was free from the immediate and
patriarchal control of their former landlords and masters who had
compensated them in kind. Now they got their compensations by cash
payment and they could sell their labor to the highest bidder on a free labor
market. But the industrial freedom was limited in other ways. Work in the
industry was regulated, disciplined, mechanized and monotonous. Working
conditions were often poor, accidents common and working hours late. The
financial and social security had also disappeared with the increased freedom.
Shareholders, company directors and engineers, who ruled the industrial
development, were often more anonymous, remote and inaccessible than the
former masters.
Towards the end of the 1800s internationalism and free trade ideals did
indeed give way to the rise of nationalism, but industrial manufacturing still
pushed further momentum and increased prosperity. Exports tripled around
the turn of the century, thanks mainly to the success of the engineering and
the pulp and paper industries. But the increase in trade came to an abrupt end
during World War I.
On August 1 1914 World War I broke out. The war had a great impact on
Swedish industry even though Sweden stayed neutral and kept out of the war.
Imports were strangled already in the beginning of the war, while exports
continued for a few years. Sweden was transformed from a capital importing
to a capital exporting country. Eventually, however, the Swedish neutrality
policy also came to include the trade policy. As a result, Sweden’s foreign trade
almost entirely ceased. Also trade with non-belligerent countries was
complicated after 1917, when submarines started to attack trade vessels.
The loss of export markets hit many Swedish businesses hard. Most
engineering companies however were richly compensated by an increasing

7
 domestic demand in an expanding war economy. They got large orders from
government and some of them were extremely profitable during the war.
Many profitable engineering companies also integrated backwards during
the war and increased their leverage through large investments in
acquisitions, new machinery and the electrification of their fabrication. The
investments were partly financed by loans. But a new Swedish bank act of
1911 also gave commercial banks the right to trade with shares. Many
banks did that by starting new subsidiary investment companies, e.g.
Investor. The banks combined their lending with buying new issued
preference shares in borrowing companies. The new banking law also
favored a concentration of the banking industry in Sweden.

Consolidation, centralization and

corporatism
An almost non-existing import during World War I created great problems
for the Swedish population. There was a shortage of fuel, machinery, spare
parts and consumables, but also of food. When the war broke out Sweden
was far from self-sufficient in grain and feed. Both animal and plant foods
became scarce. Rationing was introduced and ration coupons were
handed out to Swedish citizens for sugar, coffee, butter, bread, potatoes,
milk and so on. A continued population growth and urbanization also led
to a significant shortage of housing in the cities.
The war in Europe also had great impact on domestic policies. The war
became a source of inspiration for revolutionary forces all over Europe. In
war-torn countries it created fertile ground for subversive revolutions. In
Russia, the Tsar abdicated after the March revolution in 1917 and in August
the same year the Bolsheviks took over power. Also in Germany revolts broke
out and in 1918 Wilhelm II renounced his crown. The German Republic was
proclaimed. The revolutionary currents also gained ground in Sweden. A
general unrest in the country and numerous hunger demonstrations and
riots supported the revolutionary agenda. The threat of a revolution
appeared increasingly more real and paved the way for rapid and radical
democratic reforms.

Full democracy is introduced in Sweden

Left winds were rash in the 1917 general election in Sweden. A liberal Prime
Minister formed a new government and the Social Democrats participated
for the very first time. Together they initiated a revision of the
constitution, especially the introduction of universal suffrage for both men
and women. In 1919 and 1921 the Swedish Parliament adopted the
constitutional amendment which meant the democratic breakthrough in
10
Sweden. In 1919 the Parliament also adopted a law on working hours. This
meant that working hours were limited to 8 hours per day or 48 hours
per week. The Swedish trade union movement had succeeded, with the
help of government, and despite employers' unwillingness, to enforce the
law. So the reformist worker’s union celebrated great triumphs. On the
parliamentary road two of its main demands, universal suffrage and the
eight-hour day, had successfully been implemented. It was decisive victories
and the revolutionary currents among workers in Sweden gradually
declined.
During World War I and the years immediately following inflation rates had
gone up. The prices quadrupled in 1913 to 1920. But the inflationary period
was followed by a deep depression and a tough deflationary monetary policy,
which led to prices again halved in two years. Companies that during the war
had borrowed money from the banks and issued new preference shares ended
up in difficulties. They still had their big debts, but the assets no longer
corresponded to purchase values. They were also confronted with an
extremely tough market. Many engineering companies, which had been
favored by the war economy, lacked customers when they would switch from
military to civilian markets. Some Swedish companies also had assets in
Russia, which were confiscated after the revolution.

Ownership in industry is concentrated to a few financial

spheres
During 1920 to 1922 liquidations and bankruptcies succeeded each other in
Swedish industry, which led to large transfers of ownership. Also many
individuals were driven into bankruptcy and livelihoods deteriorated in the
short term for a large majority of the population. Unemployment was
widespread. In early 1922, a third of the members of LO (Sweden’s combined
blue collar union) were unemployed. But from the rubble grew a new
competitive export industry.
Many company founders; entrepreneurs, engineers and their families lost
their businesses. But in came new owners; commercial banks and financiers.
The number of commercial banks had peaked in 1908 and through mergers
they were much fewer after the war. It was therefore a relatively small number
of dominant ownership spheres within the Swedish financial circles who came
to answer for the reconstruction. The control of the iron and steel industry,
the pulp and paper industry and the engineering industry gathered at a small
number of hands, which became even less after the Kreuger crash and a new
depression in the beginning of the 1930s. In 1934 a very large part of the
Swedish export industries was controlled by less than a dozen financial
spheres.
The new principal owners of Swedish export companies started to implement
consolidation and structural changes in the half-century old export industries.
They were also eager to expand the markets and start rationalization and
9
 streamlining of operations in the companies that now belonged to them. They
recruited rationalization oriented engineers as CEOs and other leading
positions in engineering companies. The new executives were more industry
strategists than former inventors and entrepreneurs. After World War I the
profile of Swedish business leaders changed. You no longer became a CEO in
a large company by starting or inheriting a business, but by educating yourself
and making a white collar career to the profession of a business leader.
Under the new leaders, many of the existing companies would develop very
positively. However, after the mid-1930s industrial development was
mostly about rationalization, restructuring and internationalization of
existing businesses. More rarely, about creating or establishing new
ventures. It was mainly within already established Swedish companies that
one would come to pick up innovations and establish new product areas.

Government increases its grip on infrastructure

While the export industry consolidated and rationalized there was an
apparent consolidation also in the infrastructure area. Government
increased its grip on the infrastructure (Kaijser 2008:2). Government took,
through the establishment of Vattenfall in 1909, first responsibility for
regional power generation and later also the deployment of a nationwide
transmission grid. Government took an even greater responsibility for the
railways. In 1938 a decision was made on the nationalization of all private
railways in Sweden. Swedish Rail (SJ) was then a national railway monopoly
in the whole country. Telegrafverket successively acquired all local phone
associations and telephone companies. In 1918, the acquisition was finally
made of Stockholms Allmänna Telefonaktiebolag. Telegrafverket (later
Televerket) thus created a nationwide de facto monopoly in the telephony
field and turned to a Swedish PTT. From the 1930s the government did
chose to direct and control the expansion and tariffs in a way that
guaranteed companies and people in all parts of Sweden access to infra
services on almost equal terms.
It was not only the control over the Swedish industry and infrastructure that
was centralized in the 1930s. After the 1936 election the political power in
Sweden was gathered in the reformist labor movement's hands for many
decades to come. The Social Democrats began, under the leadership of
Per Albin Hansson, to build the Swedish welfare society. Already before
1936 Per Albin had developed very close, but discrete, personal contacts
with some of the most important owners and CEOs in Swedish industry
(Isaksson 2008). The Social Democratic vision included the creation of an
efficient and streamlined Swedish export industry. The government also
desired a peaceful labor market and the avoidance of blockades, boycotts
and strikes. Efficient manufacturing processes and a competitive export
industry was seen as the foundation of the Social Democratic vision for a
future welfare society. It was

10
the foundation that would allow for increased living standards for a large
majority of the Swedish people.

A corporatist Swedish society is born

The interests of the new industrial owners and industry leaders' coincided
largely with the reformist labor movement. After the election in 1936 the
leading representatives of the large Swedish corporations chose to abandon
their previous negative attitude towards the Social Democrats. Instead they
decided to cooperate with the two branches of the reformist labor movement
in Sweden; the worker’s union (LO) and the government. They gathered
around the common interest of rational and efficient export industries. After
the 1936 LO Congress the reformist trade unions also took an active
responsibility for the development of a more competitive Swedish industry
and promoted rationalization; consolidation as well as automation and
implementation of work study methods. Negotiations were started with the
Employers' Confederation (SAF). A master treaty, the so-called "Saltsjöbaden
Agreement", was signed in 1938. The agreement gave the central
organizations, SAF and LO, increased influence and acted in a clearly
centralist direction.
The rapprochement between the central organizations on the labor market
was followed by cooperation between government and industry. The outbreak
of World War II in the fall of 1939 gave an extra boost to the corporate efforts
in Sweden. Closures and refurbishment of defense also contributed to the
close collaboration between government and parts of industry. Government
introduced special military taxes and issued war loans, which allowed very
large and rapid increases in public spending, largely to the Swedish
engineering industry, whose advanced products gained an increasingly
important role in the military defense. During the war, there was a
pronounced spirit of cooperation both at the political level and between the
parties on the labor market. Laws, including regulations about working hours
and holidays, were breached and overtime and shift work became much more
common. At the same time, new regulations about official duty made strikes
and conflicts impossible. During the war, there also was in essence a wage
freeze even in the expanding and lucrative engineering industry.
The good relations persisted after the war. The Prime Minister, Per Albin
Hansson, who died in 1946, had replaced the socialist dogmatism for a role as
pragmatic father of the nation. This was also true for his predecessor, Tage
Erlander, who served as Prime Minister of Sweden and the leader for the
Swedish Social Democratic Party from 1946 to 1969. The Social Democrats
disassociated themselves from further nationalizations of private industry.
World War II was, in contrast to World War I, followed by increased economic
activity. Swedish industry was blessed with increased orders. During the

11
 1950s and 1960s Swedish industry harvested major export successes.
Sweden's share of world trade rose almost dramatically; Sweden took the
lead in terms of economic growth. The export industry increased its share
of employment and was the driving force in the whole of the Swedish society.
The flourishing Swedish export industry laid the foundations of the welfare
society that emerged in the decades after World War II. It created the
preconditions in terms of economic growth and employment and
contributed, both directly and indirectly, to the financing of the growing
public sector. The success of the Swedish export industry was therefore not
only a concern for business owners and leaders, but equally for the Social
Democratic government and the labor movement. In the late 1960s,
Sweden was one of the world's richest countries and the Swedish welfare
society attracted international attention.
In a relatively short time, a majority of the Swedish population had raised
their standard of living from a precarious life of poverty to a life of full
employment, relatively good income and seemingly great security. In less
than half a century the average industrial worker had tripled their real
wages. In addition, he was covered by pension insurance, medical
insurance, unemployment insurance and statutory paid holiday. In the
welfare society, government also took responsibility for the welfare of
citizens in terms of health care, education and housing. Government
reallocated resources, through taxes and charges, between different stages
of life, between rich and poor and between those who worked and those
unemployed or unable to work.
There was a national consensus that laid the foundations of the welfare
society. The development was driven systematically by central
representatives of government, industry and trade unions in the context of a
very centrally and informally managed truly corporative social society. The
corporatist society was build on a clear division of roles between
government, industry and trade unions. I present some characteristics of
the corporatist Swedish society below.

Efficiency in details and in overall structure

The export success of Swedish industry after World War II laid the
foundation for full employment. There was a shortage of labor – not a lack
of work. In consequence, there was almost total agreement in Sweden that
it is desirable to streamline human labor in factories. The Swedish labor
movement also wanted to ensure that objective piecework salaries should
be guaranteed throughout the industry in Sweden. Consequently, in 1944
the labor court in Sweden decided that the use of work and motion studies
should be compulsory throughout the Swedish engineering industry (Giertz
1981).

12
The employers, the union movement and Social Democratic politicians all
wanted to introduce a corporative version of Scientific Management. The most
rational production methods were to be determined on objective grounds. A
comprehensive change in the wage formation process for industrial workers
was one of the cornerstones underlying the carefully planned introduction of
Scientific Management throughout the Swedish industry after World War II.
In 1944, a verdict in the Labor Court (Arbetsdomstolen) stated that all piece
rates in the engineering industry must be decided objectively. All engineering
companies were thus compelled to employ time-study engineers, to determine
– on objective grounds – which working methods were to be applied for a
certain task and the time required performing that task.
Because of the verdict, the Employers Confederation and The Labor Union
reached an agreement on work-study in 1948. Swedish companies thus not
only had the right but a contractual obligation to employ motion and time
study engineers and to perform time and motion studies on all manual work
on the shop floors. Starting with the 1955 agreement the parties also agreed
upon introducing a performance standard based upon Methods-Time
Measurement (MTM). MTM is a predetermined motion time system to
analyze the methods used to perform any manual operation or task and, as a
product of that analysis, set the standard time in which an average worker
should complete the task. The US consultancy company Maynard´s launched
MTM in 1948 and MTM soon became widely spread in the Swedish
engineering industry. Thus, from 1955 the parties could check the
performance appraisals and piecework times made by any time and motion
study engineer against hindsight. This unique arrangement is often over-
looked when describing the corporatist Swedish society, which created the
post-war economic miracle in Sweden.
The Labor Court´s verdict was entirely in line with the implications of the
concept of the “policy of wage solidarity” and the principle of “equal pay for
equal work” accepted at that time. Workers who performed the same volume
of work should get the same pay, regardless of what company or industry they
were employed by. The ability of the individual company to pay, or the local
supply of labor, was not to influence the earnings of the workers.
One outcome of the application of the principle of “equal pay for equal work”
was – quite intentionally – that export companies in industries which
internationally paid high wages, such as the automotive industry, were
subsidized, while the rug was systematically pulled from under companies in
industries that internationally payed low wages, such as the clothing industry.
This type of guided structural change had been outlined in detail already in
the 1930s by one of the leading ideologists of the Social Democratic Party,
Sigfrid Hansson (brother of Per Albin Hansson). Labor was to be transferred
from inefficient companies to efficient ones and from low-wage to high-wage
industries. By means of retraining and “relocation policies” labor was to be
transferred to those regions and sectors of industry that could afford to pay

13
high wages. In this way, the value of the nation´s total production would
increase, and people would generally enjoy higher living standards.
Not only had the labor productivity come into focus, but also the
mechanization and automation of production. The larger engineering
corporations in Sweden generally coped very well with this transformation
and they took the lead when digital electronics was introduced on the shop
floors. Corporations like ASEA, Volvo and Electrolux even started to develop
industrial robots and auto carriers for internal use during the 1960s and later
introduced them to the market. The Swedish machine tool industry also was
very competitive and, until the 1970s, it was one of the key Swedish industries.
It achieved international recognition, and as an equipment supplier to
Swedish industry it played an important part in the success of Swedish
manufacturing industry in general. You could definitely argue that production
engineering was an important developing block (Dahmén 1989) in Swedish
industry after WW II.

Government facilitates mobility on the labor market

In their deliberations in Saltsjöbaden in 1938, the parties on the labor market
had declined governmental interference. To avoid governmental interference
also became the main track for three decades in the postwar labor market
policies. Government legislated working hours and holidays, but refrained in
general from the regulation of employment conditions. In the agreement from
1938, the parties stated that it was the employers’ right to organize, manage
and freely hire and fire employees (Giertz 1981). The agreement supported the
structural changes and efficiency measures that the Social Democrats also
sought. The government on the other hand, took an active role to facilitate
labor mobility. It built up the resources to re-educate people, to move people
and set up employment agencies. The relatively widely accepted Social
Democratic "moving van policy" explicitly aimed at facilitating and supporting
the restructuring and rationalization of the Swedish industry.
In the corporatist society, the Social Democrats went arm-in-arm with the
owners of major companies in Sweden. They pursued a policy that clearly
favored the accumulation of capital within established companies and
disadvantaged new ventures (Jakobsson 2008). The list of listed companies
at the Stockholm stock exchange also stayed more or less unchanged from the
mid-1930s to the mid-1970s. At the same time, the Social Democrats gradually
tightened taxation of wealthy individuals through the introduction of
progressive income tax and wealth tax and they increased gift and inheritance
taxes. One consequence was that the dominance of the larger well established
Swedish export companies increased heavily in Swedish industry.
Ideologically the Social Democrats justified their close cooperation with the
Swedish capitalists by pointing out that large established companies became

14
increasingly dominant. They argued that the individual entrepreneurship
belonged to history. Consolidations and economy of scale would eventually
lead to capitalism undermining its institutional and political importance.
Eventually it would peacefully pave the way for socialism in democratic
countries.

Demand-driven engineering sciences are introduced

The corporative consensus also included an increased governmental interest
in academic engineering research and education. During World War II
government started to look upon engineering sciences as a national
investment (Deiaco & Reitberger 2008). Professor JD Bernal at the University
of Cambridge inspired that thinking. He argued that engineering sciences
should be more demand-driven for the benefit of citizens, society and
businesses. Therefore, government decided that specific sectorial research
institutes should perform applied research, linked to common needs in
different industries. Government along with incumbents in a specific industry
was co-funding each institute. The research institutes would function as
autonomous units but they were to be located next to corresponding
institutions at the technical universities, especially KTH.
That the large Swedish incumbents in one industry jointly sanctioned a
research program was sufficient justification for government to put in
corresponding funding. Thus, government completely relied on market forces
– that of incumbents – to control the direction, expansion and contraction of
applied research in various engineering fields. The set up created a major
expansion and upsurge of the higher education and research in engineering
sciences.
During the 1950s, there was a very common optimistic belief in a better future
that relied on technological development. This belief was one reason to further
increase governmental investments in PhD studies and basic research at the
technical universities. In parallel incumbents initiated applied research
programs next door. The incumbents decided on large research projects, co-
funded by government, in different sectorial research institutes. Sometimes
the researchers almost didn´t know who of the colleagues were employed by
the university and who belonged to an institute. This resulted in a Swedish
model for technical universities that was unique. The institute sector stayed
small but there were strong links between basic research, applied research and
industry. Education at technical universities also became closely connected to
research and to actual problems in industry. This created a Swedish doctrine
that was valid almost until the end of the 20th century.
The close cooperation and interplay between technical universities, research
institutes and incumbents were accompanied with an extremely close
cooperation between individual public customers and individual private

15
 corporations in Sweden. In those so-called developing couples, engineers
and technicians from public customers and private companies worked side
by side with research, product development and implementation of new
products and systems. This gave way for special conditions for the
commercialization of research and new technology in the corporatist
Swedish society.

Developing couples paves the way for export industries

As mentioned earlier, after World War I government assumed greater
responsibility for the development and operation of the infra service
systems in Sweden. The extension, expansion and modernization of the
infra service systems involved major technical challenges and led to a very
close relationship between public clients and the engineering corporations
that supplied the technical input.
When power generation and the nationwide transmission grid was built it
was the public company Vattenfall and the private company ASEA – the
country's leading manufacturer of electrical equipment – which jointly
took on the challenge. They developed a close cooperation in high-voltage
engineering. This form of close and long-term cooperation between a
private Swedish company and a public customer has been named a
developing couple (Fridlund 1999). ASEA obviously found it very valuable
to have a close cooperation with a demanding pilot customer. ASEA relied a
lot on engineers in Vattenfall, who had great expertise in the operation of
large power systems. Additionally, ASEA could perform pilot tests with new
equipment in Vattenfall’s existing plants. It contributed greatly to ASEA in
the 1950s becoming a world leader and a successful export company in
high-voltage technology. Even in the nuclear sector, ASEA got a leading
position through public orders.
Developing couples played important roles in post-war Sweden. Each
couple consisted of a public client and a private engineering company,
which developed, manufactured and supplied necessary products and
systems. Developing couples also arose within the railroad sector. SJ
developed a close and long-term technical cooperation both with ASEA in
the electrification of railroads and with LM Ericsson in developing signaling
systems and equipment used for control and security of railway traffic.
During the 1950s, an indigenous development of locomotives also started
with ASEA that first resulted in thyristor converters for DC motor drives
and Rc locomotives and later in asynchronous motors and X2000 trains
(Kaijser 2008:2).
The most pronounced development couple, however, was the cooperation
between the Swedish PTT, Televerket, and LM Ericsson. Televerket
acquired Stockholms Allmänna in 1918. Many leading engineers and
technicians in the company were transferred to their former supplier, LM
Ericsson. Televerket and LM Ericsson soon initiated a close cooperation
16
in the development of switchboards and other important components. The
collaboration between Ericsson and Televerket however differed somewhat
from that in other developing couples. Televerket had since 1891 developed
and manufactured their own products and systems in Teli, an industrial
division of their own. Teli's first factory was located in Stockholm but was
through a parliament decision moved to Nynäshamn in 1913. Later Teli
established new factories all over Sweden in Vänersborg, Gothenburg,
Sundsvall, Kristinehamn and Skellefteå. Since Televerket manufactured
their own products they almost didn´t buy anything from Ericsson.
However, Televerket and Ericsson joined forces to develop new products
and systems.
The cooperation between the Swedish PTT (Televerket), with its own
industrial division (Teli), and L M Ericsson got an even firmer shape when
they established a joint research and development company, Ellemtel
Utvecklings AB. Ericsson and Televerket owned the new company equally.
Ellemtel’s primary task was to develop, on its owners’ behalf, what would later
become the AXE system. AXE was an electronic and automated telephone
switching system for telephone stations Ellemtel was purely a development
company without its own production. Production would instead be the
responsibility of the company’s two owners using their respective production
facilities. Televerket and Ericsson manufactured identical products for the
domestic respectively the export markets. The close cooperation with
Televerket gave LM Ericsson a great advantage in being able both to share
development costs and to quickly gain experience from the operation of new
products and components before they were launched abroad.
There were also other developing couples of great significance, such as The Air
Force and FMV (The Materiel Administration of Swedish Armed Services) as
clients and engineering companies like Saab as suppliers.

More orthodox socialistic ideas rules government

In the early 1970s, the well-organized Swedish welfare state, with its high
standard of living, still brought much attention and admiration abroad.
However, new forces were in motion under the idyllic and polished surface of
the Swedish welfare society. The once docile inhabitants of the Swedish
community started, in the late 1960s, to oppose of those in power in the
corporatist Swedish welfare society. They expressed a series of reactions
against the established society. Protests, unrest and discontent also spread to
the workplaces. In the early 1970s, both contractual and unlawful strikes
became increasingly common. The commotion also contributed to another
change. The previously cherished work and motion studies and piece rate
systems were thrown out of the Swedish manufacturing industry. Socio-
technical systems and autonomous groups were instead the new mantra
regarding work organization in Swedish factories. These changes occurred at

17
 the same time as demands for real socialization received backing from a
new generation of leading Social Democrats.
Orthodox socialistic ideas took rote in government in 1969 when Olof Palme
succeeded Tage Erlander as Prime Minister in Sweden. Palme rejuvenated
the government, paving the way for more radical ideas, which had taken
root among the young Social Democrats. A few years earlier, in 1967, the
Social Democratic Party Youth League (SSU) held a Congress that really
reflected the left winds (Ljunggren 2008). The Congress adopted a very
radical action plan. The resigning President, the future Prime Minister
Ingvar Carlsson, and the new President elected by the Congress, the future
Minister of Finance Bo Ringholm, had endorsed the congressional
opinions. The Congress demanded, among other things, the nationalization
of the banking industry, pharmaceutical industry, pharmacy services,
construction materials industry and petrol and oil trading. SSU also
demanded an end to the cooperation with capitalists and demanded that
citizens would have real power and control over the means of production.
Government wanted to end the close collaboration with the capitalists and
instead implement orthodox socialistic ideas.
Krister Wickman took up a new position as Minister of Industry in
Olof Palme's cabinet. Already May 29, 1969, he presented Bill 121, which
became the basis for the formation of Statsföretag AB; an industrial
conglomerate with extravagant expansion plans (Fredriksson 2008). The
idea was that government, through Statsföretag, would be responsible for
the production of basic goods in the Swedish society, such as schoolbooks,
computers for pupils, plastic bicycles, postal cars, steel, food and medicines.
Government also introduced a more active industrial policy in other
respects. A new governmental agency, the Board for Technical Development
(STU) was established in 1969. Government argued that the innovation
process – the transfer of ideas and research results into commercially
successful products – must be seen as a single, cohesive chain of activities.
When STU was established, government also started a new investment
company. One mission was to exploit innovations made by employees in
government authorities and agencies and state-owned companies.
Formally, the state's passive role in the industrial development was replaced
by an active, proactive involvement at government level. It meant a sharp
departure from the previously established division of roles between
government and industry, where government completely handed over the
commercialization to private incumbents.
Government awarded STU a role in the development and
commercialization of new technology. However, in parallel, a more
hostile approach towards rationalization and new technology also gained a
foothold among social democrats and in the trade unions. In particularly
the increasing computerization, automation and use of industrial robots
was in focus. Politicians raised demands to stop or stem the use of new
technology, so that people had enough time to adapt to new conditions.
Government started new

18
delegations (Dataeffektutredningen, Datadelegationen and Data- och
elektronikkommittén). This signaled a very radical change of perspective in
the Swedish labor movement that spread very fast.
Even if few people noticed, already in the early 1970s you could see the first
signs of an upcoming crisis. Therefore, subsidizes became a part of the newly
established concept of industrial policy. Selective support measures, targeted
at industries dominated by small enterprises, like textile and clothing,
footwear, glass, furniture and foundry industry, were the result. A new
government agency, The National Industrial Works (SIND), was established
in 1973. It was given the responsibility of handling the programs of selective
subsidizes. During the period 1973-75 pay-outs doubled, mainly due to
increased support operations in crisis-affected companies and sectors, but
also through increased R&D funding and increased payments for export
promotion.

Wage-earner funds initiates great tensions

Despite the "oil crisis", and the increased costs of industrial policy, there was
not a widespread "sense of crisis" in the country before 1976 (this section
draws on Giertz (1991) p. 253-269 and 387-407). In 1974, the major export
companies made very high profits overall. The political debate was dominated
by discussions about the "excess profits" – not on "industrial crisis". Growth
and increased standards of living were seen rather as governed by the laws of
nature than as a result of innovations, rationalization and hard work of the
population. The older social democrats’ missionary on these matters was more
or less forgotten, which gave significant consequences for the actions taken by
government and trade unions.
Sweden’s confederation of blue-collar unions (LO) pushed the issue of
economic democracy at the beginning of the 1970s. The list of companies on
the Swedish stock exchange had remained almost intact since the mid-1930s.
To own shares therefore appeared as safe as putting money in the bank, and
there had almost been no need for issuing new shares or for venture capital.
The established incumbents had intercepted new technology, new products
and new markets. In 1971, the convention of LO proposed a study to
investigate the possibility of introducing collective capital formation. In 1975,
the study, under the leadership of LO economist Rudolf Meidner, was
presented. “The Meidner Plan” recommended the introduction of wage-
earner funds. The plan proposed that 20 percent of a company’s profits should
be used to buy newly issued equity shares. This proposal entailed a gradual
transfer of ownership from private individuals and institutions to collective
entities governed by union-appointed directors. The funds would eventually
own 50 percent of companies in the long term. The LO convention supported
and approved the Meidner Plan in the summer of 1976.

19
 The discussions about wage-earner funds were partly ditched when the
Social Democrats lost the election in the autumn of 1976. However, in
1982 Olof Palme was forced, partly against his will, to make wage-earner
funds one of the election’s key issues. LO demanded that wage-earner
funds should be given a prominent position on the Social Democrats’
agenda. However, a considerably modified proposal was put on the table by
that time. A group led by the future Minister of Finance Kjell-Olof Feldt,
had revised the original plan. The Social Democrats won the election and
they introduced a modified version of wage earner funds in 1983. However,
in 1991 a right winged government, under the Conservative Prime Minister
Carl Bildt's leadership, immediately cancelled the taxation. A few years
later, the funds were closed down.
When the ideas of wage-earner funds started to take shape in the beginning
of the 1970s, it was the death knell for the government's earlier cooperation
with industry. The tension between industrial owners and CEOs on the one
hand and leading Social Democrats and labor union leaders on the other
increased considerably at the beginning of the 1970s. Thus, the former
foundation of the corporate Swedish society was destroyed.

Informal cooperation is replaced by formal bureaucracy

The tension between government and industry increased even more when
government abandoned its former non-interventionist role on the labor
market. Investment funds, which encouraged companies to reduce tax
by locating manufacturing plants to regions with unemployment, were
introduced. The union movement also relinquished its tradition of
negotiating and agreeing with employers on numerous common issues, and
instead turned to government with a request for legislation. Various labor
law reforms, including the Act concerning Security of Employment, Board
Representation for the Employees, Position of Union Representatives and
Co- determination at Work, were implemented by the Minister of Justice,
Carl Lidbom, during the period 1972-77. The new labor laws resulted in an
almost complete turnaround compared to the previous "Spirit of
Saltsjöbaden". There is surprisingly little attention paid to this turnaround
in more recent discussions about the content of the so called Swedish
model. There might even be some people who believe that the 1970s labor
law reforms are part of the original Swedish model and a reason for the
Swedish economic miracle in the 1960s.
Behind the visible changes, also a subtler but important development
lurked. The relationship between ministers in the government and the
business community's top representative changed radically. In previous
Social Democratic post-war governments, the Prime Minister as well as
other important ministers – such as the Ministers of Trade and Finance –
had close

20
personal contacts with top representatives in important financial spheres and
with employed CEOs. In the 1970s however, confidentiality, personal contacts
and informal consultations between government and business leaders were
replaced by bureaucratic contacts between officials, officers and ombudsmen
of various kinds. The direct contacts between government and industry
became extremely frosty. The earlier consensus had been replaced by powerful
confrontation at crucial points. In parallel, the public sector developed its own
staff organization in the 1970s. Government agencies, public authorities,
ministers and organizations representing various groups in society took over
the handling of an increasing number of tasks.
The personal contacts between politicians and industry leaders were replaced
by interaction between civil servants from the large public authorities, leading
trade unions and large employer associations. Oddly enough, these stiff
structures were put in place when Swedish industry was just about to enter
into a new phase characterized by great industrial dynamics.

From crises to an entrepreneurial global

economy
In 1976, the Social Democrats lost the general election. After 44 years of
uninterrupted government a liberal-conservative coalition, led by the Centre
Party leader Thorbjörn Fälldin took over. His predecessor said that when
entering the cabinet, the new government could just start picking presents off
a Christmas tree arranged by the Social Democrats. However, the new
Minister of Industry, Nils G Åsling, barely had time to take up the post until
his office became an Emergency Room for failing firms. Sweden's economy
was greatly unbalanced. A shrinking export industry could no longer finance
government´s galloping spending.
Shortly after the seemingly good "excess profit year" in 1974 – which also
brought greatly increased wage costs for industry – structural problems
appeared. Productivity decreased and real earnings were reduced.
Unemployment rose, particularly among young people. Special support
measures were directed at small enterprises and companies who arranged
jobs for young people received special grants. Above all, there was a dramatic
increase in direct efforts to sustain employment in the country's ailing basic
industries and incumbents. Concealed behind the increase was direct aid to
shipbuilding, steel, forestry, mining and textile industry. During the period
1977-79 over nine billion crowns was paid in direct subsidies only to the
shipbuilding industry, which represented approximately 280,000 crowns per
employee, or 120 percent of the shipbuilding industry's total payroll. The crisis
also made, ironically, the new liberal-conservative government responsible for
the largest socializations in the country's history. With an ambition to pursue
structural rationalization, government entered as the sole owner of several

21
 companies; for example, Svenskt Stål AB (SSAB) and Svenska Varv AB
(Swedish Shipyards).

SME:s and Universities in focus in regional policy

When the incumbents in basic industries went into trouble, the country's
total production decreased greatly. The industrial crisis also spoke quickly
in reduced market shares on the world market, rising unemployment,
negative trade balance and low profitability in many industries. In the
wake of the crisis, the liberal-conservative government's interest to
encourage the growth of small and medium sized enterprises increased. This
also initiated a major change in the institutional conditions that set the
scene for the Swedish innovation system.
As a result, the business associations were converted to regional
Development Funds in 1978. SIND and the county council in each county
became principals. The Development Funds, which received increased
resources and access to their own loan funds, would primarily focus their
support to start-ups and existing businesses with up to 200 employees.
The election in 1982 reiterated the Social Democrats to power. The crisis
had drained government finances. A widespread sense of crisis had helped
to keep the Swedish wage increases and inflation in line with the main
competitor countries. Consolidations and eliminations had also cleaned up
the structure in industry. The remaining companies did well, and the
industry's competitiveness on foreign markets further improved by a
devaluation of sixteen percent, which the Minister of Finance Kjell-Olof
Feldt pushed through immediately when the new government was installed
in 1982.
Swedish industry in general started to recover in the 1980s but there were
significant regional imbalances. The eliminations of plants, enterprises and
certain industries had resulted in a redistribution of employment.
Employment increased in those cities where a dozen successful engineering
corporations or pharmaceutical groups decided to expand. However, it
declined sharply in regions, which had previous employment in industries
such as shipbuilding and steel works. The regional imbalances had great
impact on industrial policy in the mid-1980s, when Ingvar Carlsson
assumed the position of Prime Minister after the assassination of Olof
Palme.
In the mid-1980s, the Social Democratic Minister of Industry, Thage G.
Peterson, no longer had the possibility to direct an increasing number of
blue- collar jobs to regions with unemployment. Now when successful
Swedish industrial corporations gained market share, invested in research
and development, introduced new technology and improved their
profitability, they did not increase the number of blue-collar jobs in Swedish
factories. On the contrary, manufacturing industry's share of
employment in Sweden
22
decreased throughout the 1980s. Nearly 100,000 jobs disappeared from the
export industry. At the end of the decade, all companies producing any kind
of goods accounted for less than 20 percent of employment on the Swedish
labor market.
While routine factory jobs vanished, the need for skilled technicians and
engineers increased in the export industry. Overall, the industry doubled its
spending on research and development during the 1980s. The need for MSc´s
in engineering took off. However, the investments were concentrated to a little
more than a dozen corporations in the engineering, pharmaceutical and
chemical industries. They concentrated largely on the metropolitan areas and
the vicinity of universities. A lot of investigations, as well as the media, also
highlighted research as a vehicle for growth in the emerging knowledge
society. Cities with a university were supposed to be the winners in the future
society (Andersson 1988). They argued that there was strong links between
research and higher education, the creation of new competence, the growth of
new knowledge-based businesses and thus new jobs (Deiaco 2002). Cities
with universities were pinpointed as the growth centers and the labor markets
of the future, where highly skilled, highly paid and creative people flocked.
Media made out the universities to be the future engines for the growth of new
knowledge-based companies. For regional policy-makers it seemed obvious
that there were strong links between research, new knowledge-based
companies and future employment. International studies also supported a
positive correlation between the percentage of highly educated and the
economic growth of a country. This made higher mass-education get its
definitive breakthrough in Sweden. The expansion also enabled the dispersion
of both higher education and research.
In 1988, Sweden received four new universities in Luleå, Karlstad, Växjö and
Örebro. At the same time there were investments made in regional colleges,
which almost invariably came to engage themselves in some kind of research.
Thus from the mid-1980s research and higher education became an
instrument of regional policy. New universities and regional colleges were the
base for regional business policies. They were not only designed to satisfy the
need for skilled labor on the regional market. They would also conduct their
own research and hopefully become the breeding ground for comprehensive
new enterprises with great employment potential.

Internationalization and decreased productivity hits

manufacturing
Heightened global trade characterized the end of the 1980s. Countries that
previously had very protectionist trade policies joined forces in large trading
blocs that positioned themselves in the global arena. This change was most
evident in Western Europe, which strived to resuscitate its former role as a
great industrial power. In 1986, after Spain and Portugal joined the European

23
 Economic Community (EEC), the organization comprised 12 member
nations and the process had begun to convert it into a type of European
federation. Everything would be eliminated that hindered free movement
between member countries for goods, services, capital and people. By the
end of the 1980s, this vision of creating a single internal market had started
to take concrete shape.
When the decade shifted in 1989-90, developments in Europe took yet
another turn. The walls between East and West were torn down at an
astonishing speed. On October 3, 1990, the two German states reunited. The
Soviet Union was dissolved and Eastern European countries became
independent nations that embarked on a difficult journey toward
democracy and a market economy.
For the nation of Sweden, many of its curves started to bend in the wrong
direction as of 1987. On an international scale, Sweden still had an
extremely low unemployment rate, which was very much due to the
expansion of the public sector and the growth of the automotive industry.
Wages and salaries soared to new heights as a consequence. In combination
with price increases, for which insufficient capacity utilization within
industry was another cause, the competitiveness of Swedish industry
declined even more. Market shares for Sweden’s engineering industry
decreased 10 percent in three years and Sweden’s balance of payments
plunged radically. Companies were forced to borrow more money outside
the country and interest rates skyrocketed. The trade surplus also started
to decline, despite a strong economy, low oil prices and a falling exchange
rate for the U.S. dollar, which had a devaluation effect on the Swedish krona
in the most important European export markets.
The new economic reality caught the eye of the political establishment
towards the end of the 1980s. The Social Democratic government in Sweden,
under the leadership of Ingvar Carlsson, carried out major reforms to
Swedish economic policy. The Minister of Finance, Kjell-Olof Feldt, was the
driving force behind many important changes that enabled Sweden to
adapt better to the rapid internationalization of its economy. In a short
time, a number of measures were carried out, including deregulation of the
credit market and abolishment of foreign exchange controls.
The engineering industry in Sweden was also becoming internationalized.
In the beginning of the 1980s, it was a Swedish export industry with the
majority of its employees in Sweden. By the end of the decade, it was an
international industry with most personnel abroad. Thus, at the close of the
1980s, major Swedish engineering companies had expanded their foreign
production base, due in part to acquisitions of international competitors,
but also due to substantial direct investments in other countries at the
end of the decade. They rationalized their international expansion with
arguments that they needed to get closer to the market or have local
production in countries, but in actual fact, Swedish production units had
become less profitable in the

24
1970s and the 1980s, and the corporate leaders of major Swedish corporations
like Ericsson, Electrolux, Atlas Copco and Sandvik explained, quite frankly,
that Sweden was less and less of an economically viable alternative, compared
to the possibility of setting up operations locally in other countries, which they
were contemplating. The reasons were the country’s high wage costs, shortage
of labor, high sick leave rates and low productivity levels. At the end of the
1980s, many Swedish engineering companies invested directly within the
tariff walls of the EEC in countries like Germany, France, the Netherlands,
Italy and Spain. In the early 1990s, labor-intensive production was moved out
of Sweden also to low-cost countries in Eastern Europe.

New crises set the scene for a more competitive future

In the beginning of the 1990s, a recession was waiting. It was magnified by
Iraq's invasion of Kuwait in the summer of 1990. The industry had to adjust
production capacity to a declining demand. Cutbacks arose in virtually all
industries. Questions concerning the location of production no longer dealt
with the location of new plants but rather decisions on where to do necessary
closures. Confidence that incumbents could guarantee full employment in the
future faltered in almost all political parties and on all levels. In the industrial
and regional policy, more and more weight was given to new ventures and to
small and midsize enterprises. This also enforced a shift in the governmental
regional policy, from measures that primarily reallocated employment, to
efforts that supported regional or local development programs. By now, the
Social Democrats were no longer very keen on owning the Swedish industry.
The industrial dynamics during the 1970s and 1980s, and the failures in state-
owned companies, made politicians in almost all parties hesitate towards
governmental ownership in industry.
In the wake of the recession prices on commercial real estate plummeted. In
two years, they fell by more than 50 percent. New construction ceased, which
hit hard on the construction and building industries. The downturn on the
property market first hit the real estate companies, but also had serious
repercussions on investment companies and banks. The Social Democratic
government presented a "crisis package", which included savings in the public
sector and reduced health insurance. The Social Democrats also made changes
related to industrial policy. Three governmental agencies – SIND, STU and
Statens energiverk – merged into a new governmental agency, The Swedish
National Board for Industrial and Technical Development (Närings- och
teknikutvecklingsverket, NUTEK).
In 1991 a very radical tax reform, agreed upon two years earlier both by the
Social Democrats and the liberal-conservative parties, was implemented. It
lowered the previous extreme progressivity of the Swedish income tax to more
normal levels. It was still a progressive income tax, but in principle, the Swede

25
 would in the future not pay more than half of a wage increase or an extra
income in tax. The progressive income tax was combined with a flat tax, of
30 percent, on capital gains.
In the chaotic situation, the Social Democrats once again left government.
After the 1991 election conservative leader, Carl Bildt, became Prime
Minister of a liberal-conservative government. The financial crisis forced
a series of costly commitments. Alongside dealing with the unemployment,
government had to guarantee bank liabilities to third parties, give
subsidizes to support some banks and nationalize the private bank
Nordbanken. In May 1993, government set up a special governmental
agency, Bankstödsnämnden, to handle further support to banks in trouble.
The crisis weakened public finances and unemployment was widespread.
Nobody put trust in Swedish economy any longer, which imposed new
problems. When the credit market was deregulated, the Swedish krona had
been linked to the European currency unit ECU at a fixed exchange rate. The
Riksbank – Sweden's central bank – tried the longest to stave off the
attacks against the krona. The overnight interest rate even rose to
improbable 500 percent on September 16, 1992, before the Governor, Bengt
Dennis, gave up. The krona was allowed to float, which in reality meant that
it sank like a stone in November 1992.

Joining EU facilitates the consolidation of Swedish economy

From the beginning of the 1990s, Sweden's relationship to the European
Community was an important issue in the growth debate in Sweden. As the
vision of a common internal market took concrete form, it became
increasingly clear that Sweden would join the European Union (EU). EU
was realized in November 1993, and in January 1995 it was extended to
include Finland, Sweden and Austria. EU's so-called convergence criteria
facilitated the consolidation of the Swedish economy during the first half of
1990's.
When Sweden joined the European Union, the Social Democrats had
returned to government. In 1994, Ingvar Carlsson put together his third
Cabinet. Göran Persson was appointed Minister of Finance. He
implemented welfare cuts and tax increases to place Sweden in a position to
qualify for the European Economic and Monetary Union. The job was done
and the budget deficit went down but unemployment rose considerably.
International confidence in the Swedish economy was restored. The krona
was appreciated and interest rates were reduced to historically low levels.
Inflation ceased almost completely, but a larger decline in the
unemployment rate was slow in coming. The development in Europe also
took a different turn after the walls between East and the West were torn
down. Over time, it created a greatly enlarged EU.

26
Deregulation, privatization and productivity on the political
agenda
In Great Britain, the Conservative government of Margaret Thatcher started a
program of deregulation and privatization in the financial sector, transport
sector and the telecom sector in the beginning of the 1980s. This liberalization
movement spread to other countries.
Already in the end of the 1980s, the so-called right wing in in the Swedish
Social Democratic government, under the leadership of Kjell-Olof Feldt, had
secured that productivity issues once again were high up on the social
democratic agenda. Feldt spoke as former social democrats did, when he
stressed that growth requires a more efficient production throughout society
– of goods as well as services. This could be the result of introducing new
technologies, but also from introducing new and more efficient regulations,
management, structures and work methods in working life.
A productivity delegation was appointed in 1989. It stressed that real growth
is based on efficiency improvements, which ensure that we get better
productivity and quality and better service in all parts of working life. The
delegation's report, which was delivered in 1991 (SOU 1991:82), focused on
weak productivity improvements in the Swedish working life. The delegation
looked at the nation's total production capacity and stressed that the
transformation pressure, especially in service sectors, must increase.
A wave of deregulations in sectors such as transport, electricity, media, postal
services and telecommunications were rolled out. Sweden was to play a
leading role in the transformation when firms and agencies in domestic
monopoly or oligopoly industries faced competition from new competitors
and international players who expanded their geographic markets.
The taxi sector was deregulated in 1990. The reform aimed to improve access
to taxis at all hours and across the country, by making it easier for taxi drivers
to set up a business. By increasing competition, it was also expected that taxi
fares would stay at a relatively low level. As a result, new franchising networks
challenged traditional federative taxi networks. In order to increase
competition and lower prices, domestic air travel market regulations changed
in 1992. The reform meant that former privileged airlines, like SAS and
Linjeflyg, met new competition from budget airlines. The railroad monopoly
however stayed untouched for long. In 1988, a cut was made through the
nationalized railway monopoly (SJ); since then, infrastructure is handled by a
public-sector agency (Banverket). Trains were until 2011 only run by the
government-owned monopolist (SJ), but since 2011, SJ competes with other
railway companies.
A deregulation of the Swedish electricity industry was started in the 1990s. A
first step in the process was the corporatization of the Swedish state-owned
utility Vattenfall in 1992. The deregulatory process culminated with a new
Electricity Law, which entered into force in 1996 (Högselius and Kaijser

27
 2007). The new regulation introduced competition on the electricity market
in the generation and distribution of electricity. The aim was to provide
consumers with greater freedom of choice and better opportunities for
putting pressure on costs and prices in the supply of electricity. When
charging customers, the total cost of electricity should be separated into two
components, the transmission price and the price of the electricity.
Deregulations and privatizations of different state-owned monopolies have
continued over time. In 2009 Sweden’s state-owned pharmacy monopoly,
which was originally formed in 1969, ended. A huge part of about 1,000
individual pharmacies was sold to large and medium-sized companies in
eight different clusters. But the deregulation also paved the way for new
entrants into the market for drug sales. A similar procedure followed in 2010
when the market for annual, compulsory vehicle control was opened for
competition. Previously all vehicle controls had been provided by a
national monopolist. The government has been partitioning the incumbent
and gradually selling it on commercial terms.
In addition, Swedish governments have, since the 1990s, experimented
with boosting public-service efficiency in different sectors by allowing
private actors to compete with publicly owned vendors on tax-financed
markets. Many schools are now independently run, and in health care as
well as child and elderly care private management is a growing – and
frequently debated
– trend.

Outsourcing shapes new specialized industries

The deregulation and privatization in the public sector was accompanied by
a similar trend in many industries. Already in the mid-1980s, many Swedish
incumbents started to concentrate on their core activities. They spun off
numerous businesses and got other companies to supply everything that
was not regarded as their own core competence.
Outsourcing, the hiving off of activities that are considered peripheral,
did affect simple activities just as much as advanced or complex ones. The
activities that an incumbent decided not to keep in-house were sourced
from suppliers, with whom they often have long-standing relationships.
Their own employees were in that way replaced by a more flexible and
specialized external work force, with both highly standardized and highly
specialized services. The external work force may be self-employed
consultants, employees of large consultancy corporations, call centers,
facility management providers or agencies specializing in supplying
anything from doctors to blue-collar workers. In addition, the
manufacturing of components was outsourced in many engineering
companies. An example was Ericsson, who in 1996 decided to sell
production plants in Sweden – manufacturing components they still
needed – to contract manufacturers.

28
 For the outsourcing companies, concentrating on their own core activities
and selling off other activities caused major changes in the relationships they
had with those involved in their business systems. Spinning off numerous
employees into different independent companies of course resulted in fewer
employment contracts and more business contracts for the outsourcing
company. Moreover, the new business contracts sometimes involve a stronger
tie than the former employment contracts, since the employee could cancel
the latter, while a company that has a contractual obligation to deliver must
fulfil its undertaking in order to retain its credibility on the market.

Figure 1: In the mid-1980s large incumbents started to spin off different activities, from rather
simple local services to professional services, manufacturing, computer operations to sales
activities. In the beginning, they sometimes formed new companies, called for instance
Support AB, Assist AB, Partner AB, Consultancy AB, Education AB or Data AB, within their own
corporation. Eventually specialized service companies in different service sectors later
acquired many of the new companies.
The outsourcing of different staff and service activities in private as well as
public sector created expanding B2B industries of different kind. Some
expanding industries offer customers services on a local market, i.e. repair
and maintenance, janitorial services, installations, painting services, security
services, facility management, local transportations and canteen services.
Others offer more knowledge-intensive professional services, i.e.
accountants, technical consultants, IT consultants, architects, management

29
 consultants and marketing agencies. Due to technical development, some
support could also be provided on a distance. New work places providing
distance support, i.e. computer operations monitoring, call centers and
help desks, were in the 1990s, by new vendors, concentrated to locations
where suitable labor was available. Finally, many large companies made
changes in their staffing of sales activities in the beginning of the 1990s.
Employed sales forces appointed for an indefinite period were
complemented by a new kind of sales forces, i.e. franchisers, telemarketing
and sales agencies. For the new sales force, sales commission accounted for
the lion´s share of revenues.
The outsourcing of different services challenged the established system for
wage formations in Sweden. The traditional system was built on wage
formation at central level between the parties on the labor market. It
also included “the industrial union principle”, meaning that all blue-collar
or white-collar employees in an industry were covered by the same
agreement, but the agreements in another industry could differ.
Consequently, a driver in a manufacturing company could earn more,
especially when driving at night, than a driver employed by a transportation
company. These differences accelerated the outsourcing of some activities.
When a former staff activity was divested from an outsourcing company,
it found itself in a commercial environment where it was exposed for
competition. This also involved a change in conditions for the employees,
whose wages were influenced to a greater degree by the company´s profit
level and the competition on the local labor market. It also meant that what
used to be a peripheral and perhaps neglected part of a large incumbent
suddenly was transformed into a core activity in a spin-off company.
Wage formations and earnings are nowadays very different in different
parts of the service industries in Sweden. For large groups of employees, for
example in the manufacturing industry and the public sector, wages are still
controlled in a more traditional manner. Wage differences between
different professions and positions are relatively small, and wages are
almost completely independent of individual performance or results. Other
industries, i.e. janitors and call centers, are still controlled by central
agreements but wage floor has been lowered compared to manufacturing
industry and different kinds of incentives have been introduced. In other
industries and activities, i.e. consultancy and sales, market forces outside
the central parties’ control control wage formation. Wages are relatively
high but wage differences are also great. In many of those industries, there
is a clear link between the earnings and the individual employee's job
performance and the revenues of the individual work place or company.

30
Televerket and Ericsson are divorced
The deregulation, internationalization and privatization broke up most
developing couples in Sweden. However, the very close cooperation between
the monopoly Televerket and Ericsson continued until the mid-1990s (Giertz
2015:1). In fact, a gigantic development project was started in 1987 in the
jointly owned development company Ellemtel, which had previously
developed the very successful AXE system. The new project was named AXE-
N, where N stood for Network. However, the benefits from a successful project
would be substituted by internet that simultaneously began to take shape in
the early 1990s. The AXE-N project was stalled in 1994 and completely closed
down on December 8, 1995. This was a chock not only for 700 employees but
also for a number of R&D related ICT consultancy companies who almost
overnight were out of job.
Towards the end of the project, the marriage between Televerket and Ericsson
ended up in a divorce. They had been a developing couple for about 70 years
but the time to separate had come. Already in 1990, government had taken
some steps on a liberalization journey by transforming Televerket into a public
company. In 1993, the new company Telia AB was formed and a new
Telecommunications Act was launched. This initiated a first significant
liberalization effect. When Telia was to become a competitor in the
telecommunication industry and not a state-owned domestic monopolist
many things changed. In 1994, Ericsson bought Teli, Telia’s equipment-
manufacturing units, which meant that the domestic AXE manufacturing was
transferred to Ericsson. In 1995, Ericsson also bought Telia’s holding in
Ellemtel.
On June 13, 2000, a partial privatization of Telia AB, through an initial public
offering (IPO) on the Stockholm stock exchange, was realized. It was a huge
interest among Swedes in general. Government decided that all investors who
applied for 100 shares or 400 shares would get them, while those who applied
for more only got a fraction. The Minister of Industry and Infrastructure,
Björn Rosengren, made adds about the ''people's share'' and 954,000 Swedes
actually became shareholders in Telia AB. The IPO was done just before the
third generation mobile standard, 3G or UMTS (Universal Mobile
Telecommunications System), was about to be installed in Europe and Japan.
Licenses for the new system were issued in the European countries in the late
1990s and early 2000s. Different countries chose different approaches.
The Swedish Post and Telecom Authority (PTS) chose to allocate the licenses
through a "beauty contest". Telia applied but on December 16, 2000 PTS
announced that Telia did not receive a license. Telia was the only former PTT
in the world finding itself in this position. The licenses were instead allocated
to Europolitan (which later became Vodafone and then Telenor), Hi3G (later
3), Orange (later acquired by France Telecom) and Tele2. Telia was in shock.
The former monopolist not only met new competitors. Telia was not even

31
 allowed to compete on the mobile market. However, Telia found a solution.
Telia and its competitor Tele2 soon announced their intention to create a
joint network company, Swedish UMTS Network Company, to utilize the 3G
license that Tele2 had been awarded. It was a joint venture on a 50-50 basis,
which was approved by the Competition Authority in the spring of 2002.

Ericsson survived when the bubble bursted

In the year 2000, Ericsson was riding high (Giertz 2015:1). Ericsson had
more than 100,000 employees globally, despite the fact that Ericsson had
closed down their own production of processors and other components and
outsourced manufacturing plants to Flextronic in the late 1990s. Still
Ericsson had more than 43, 000 employees in Sweden. The skills and
competence of Swedish employees however, had change over time.
Engineers and technicians of various kinds had replaced blue-collar
workers.
In the year 2000, the turnover in Ericsson was close to 300 Bsek and
Ericsson’s share of Swedish GDP was two percent. Ericsson was in the
very center of a worldwide hype around the potential of the internet.
Ericsson had around 200 million AXE lines in place in 120 countries.
However, the success was very much linked to a fast penetration of GSM
systems in the late 1990s. Ericsson held an estimated 40 percent share of
the world’s mobile market. At the turn of the millennium, Ericsson was on
top and expecting to deliver many brand new 3G-systems, especially to tele
operators in Europe. However, the market failed!
One important reason why the market failed was that most European
countries chose to allocate the 3G licenses through auctions. European
tele operators spent around 130 billion Euro to buy licenses for 3G-
networks that were supposed to give people the freedom to use their
mobile phones for reading e-mail, browsing the Internet, placing video
calls, enjoying music, buying products and services, making reservations
and so on. The nature of the auctions was designed to increase competitive
pressure on bidders by offering fewer licenses than the number of
operators likely to bid. The tele operators were put in a difficult position,
because if they lost the auction they were out of business in the next phase.
They took risks, made high bids, and ended up with large debts. However,
when they got the licenses they could not afford to buy and install the
systems.
The tele operators went into serious trouble. When the trouble leaked out
to the public the hype on the stock market was put to an end. On March 6,
2000 the index on the Stockholm stock exchange had been all time high.
The total value was 4,800 Bsek and Ericsson alone was valued to 1,709
BSek. The next day the bubble started to burst, but Ericsson still showed
good figures in 2000 and coped rather well initially. The share value of tele
operators on the other hand fell quickly all over Europe. They had no
choice but to reduce their
32
investments sharply. This of course also hit Ericsson and on October 20,
2000, the share price went down drastically. Still the figures for the full fiscal
year 2000 were quit ok.
In Ericsson, the crisis was a fact in March 2001 (Karlsson & Lugn 2009). More
than 700,000 employees in the tele vendor industry around the world lost
their jobs in a very short time. There also were many consolidations
throughout the industry. However, people still wanted new mobile phones, so
the market for handsets was almost not affected at all, which partly explains
why Nokia was not hit that hard. Ericsson on the other hand had great
problems with their handsets. Former, when the demand exceeded the supply,
Ericsson had done well because their phones were reliable. Ericsson still
shared the top position with Nokia and Motorola in the late 1990s. However,
in 2000 people on the consumer market started to ask for more. Ericsson
decided to concentrate on core business, that is tele system – not consumer
products. Despite cost cuttings and massive layoffs, the fiscal year 2001 ended
with a very big loss. It was Ericsson’s first loss in more than half a century.
More than 8,000 consultants also lost their customer in 2001. But it was not
only a question of profit and loss, but also about the cash flow situation. The
lack of orders from the tele operators was one thing but customers also paid
late due to bad liquidity. Ericsson was forced to borrow money to pay salaries.
The situation was alarming in March 2002. The customers were afraid to put
orders because they did not believe Ericsson would survive. The board decided
to issue new shares. Ericsson was only weeks from bankruptcy when the new
money turned up.
Ericsson survived and took off again in 2004. By then global employment was
down to a little more than 50,000 employees globally and 21,000 in Sweden.
Thus, Ericsson’s global and Swedish employment fell to less than half of their
previous levels in only four years. In Sweden, the ICT sector outside Ericsson
was also hit. Via its external networking and collaborating, Ericsson had
contracted tens of thousands of Swedish companies in ICT-related
development, consulting, manufacturing and maintenance. In the successful
attempt to return to profitability, Ericsson slashed this spending. With this
multiplicative effect and Ericsson’s own layoffs, a huge volume of ICT-
expertise was re-deployed in Sweden. That is still the case. Ericsson’s
employment worldwide has increased since 2004, but employment in Sweden
is continuously decreasing. Ericsson is no longer only a manufacturing
company exporting goods, but also a global service provider that operates
telecom system worldwide on a contractual basis.

A diversified and globally competitive ICT sector is born in

Sweden
Not only Ericsson made layoffs in the turn of the millennium. In parallel a lot
of .com companies and internet consultancy companies, which had been

33
 valued to fantasy figures during the hype, were in deep trouble. They lay off
people or went into bankruptcy. The new millennium started with a
complete turmoil, but out of the ashes raised a dynamic new ICT sector.
When Ericsson no longer absorbed most of the ICT competence in the
country people had to turn somewhere else. Ericsson’s former employees
looked for job elsewhere, but thousands of ICT consultants also had to look
for new customers. That development actually started on a massive scale
already in the mid-1990s when Ellemtel decided to close down the AXE-N
project, one of the largest projects in Swedish industry so far. Many
competent engineers working with R&D related software development
turned to other employers or customers in different sectors, for example the
engineering industry and the banking industry. Information and
Communication Technologies are of great importance for industrial
transformations.
Other more application oriented developers and entrepreneurs benefitted
from the fact that Sweden, and particularly Stockholm, much thanks to
Telia and Ericsson, had a well-functioning infrastructure and citizens who
were early adopters of new internet services. Many very successful new
ventures, such as Skype, Spotify, MySQL, Unibet and Klarna, have been
born in the segment Software and Net Services. Some of them are spun off
from universities like KTH, but many are also spin offs from industry. In
some sub- segments, such as the gaming industry, new ventures, like King,
Mojang, Jice and Stardoll, seems to have a strong foothold among students.
Most new ventures in the segment Software and Net Services seem to have
one thing in common. They have not been reliant on contacts with Swedish
incumbents or Swedish agencies (Giertz 2015:2).
When the bubble busted, in March 2000, it seems as if the explosion spread
many seeds, which found good soil for new businesses and new ventures.
However, at the same time it changed the conditions for governmental
efforts to support demand-driven research and innovations in Swedish
industry. Before 2000, incumbents like Ericsson, Televerket and ABB could
set the agenda for both higher educations in engineering sciences and
applied research in technical universities as well as in sectorial research
institutes. However, when companies like ABB and Ericsson decided to
concentrate their business and close down their factories the existing
research in fields like microelectronics and optics lost their former
demanding “customers”, who set the agenda for demand-driven applied
research. A sectorial research institute like Acreo than became more of an
incubator than an institute partly financed by Swedish incumbents with a
common research interest. Consequently, governmental funding partly
transferred to help the establishment of new ventures, mainly in the
hardware sector.

34
EU influences innovation policy and regional growth policy
When Ingvar Carlsson announced his retirement in 1996, Göran Persson
became the new Prime Minister. One important policy issue was to lower the
unemployment rate. By then research and technology-based new ventures
were regarded as important potential growth engines in Swedish policy.
Around the turn of the millennium scientists and politicians pointed out that
in Sweden we invested more in R&D than most other countries. Sweden was
also ranking in top in terms of new patents. Yet we lay in the bottom of the
OECD list when it came to new ventures and entrepreneurship. Many heavy
commentators claimed it was due to an inability to commercialize research to
reap the benefits of our outstanding research in the form of new enterprises
in the country. Consequently, great policy interest was concentrated to
research – both in universities and in research institutes – in fields like ICT,
biotech and energy. Seed capital, venture capital, incubators, patent
counselling and the like were included in a new policy agenda.
The new agenda made the Social Democratic government form a new Swedish
governmental innovation agency, VINNOVA (Verket för innovationssystem),
on January 1, 2001. The new agency, which is sorting under the Ministry of
Enterprise and Innovation, replaced part of the former governmental agency
Nutek. VINNOVA got the assignment to administer state funding for research
and development. The agency's mission as defined by the government is to
promote development of efficient and innovative Swedish systems within the
areas of technology, transportation, communication and labor. The agency
should accomplish this by giving financial aid to research-focused companies
for research, development and legal costs. VINNOVA have contacts with and
supports universities, research institutes and public sector organizations as
well. VINNOVA also acts as a National contact agency for the EU framework
program for research and innovation, which has in a way increased the
funding.
You can argue that the public funding of research and innovation increased
when Sweden joined EU. An active policy to support the development in all
regions in Sweden was even more boosted when Sweden joined the EU
(Nilsson 2008). A broad political consensus came to prevail that Sweden
should regain its share of the money paid to the common funds. Politicians
and researchers, and representatives of business and trade unions, both at
national and regional level, joined forces to make sure Sweden would be able
to gain a rightful share of the common cake. Sometimes the discussions on
development started rather in the availability of EU funds, than in discussions
about actual needs and possibilities in the region. On the national level, Nutek
initially played a central role in helping the regions in this context. However,
when Nutek was closed down in April 2009, the responsibilities were
transferred to a new government agency, Tillväxtverket (The Swedish Agency
for Economic and Regional Growth).

35
 The Swedish Agency for Economic and Regional Growth is sorting under
the Ministry of Enterprise and Innovation. The main task is to distribute
EU funding to promote entrepreneurship and regional growth. The mission
is to strengthen the competitiveness of Swedish small and medium-sized
enterprises or future entrepreneurs directly and work to improve the
general framework for doing business. The agency also has the ambition
to build networks for cooperation and investment initiatives that they hope
will strengthen the business sector in the region.
Ambitions on the regional level have to some extent removed national
policy focus from productivity issues. Former right-wingers among the
Social Democrats, such as Kjell-Olof Feldt, had left. They made room for
other ideas. A similar reversal took place within the liberal-conservative
block, which was to some extent manifested already when they regained
power in 1991. Within both the two political blocks, the key policy areas have
lately been formed on other considerations than economic growth on the
national level. The taxation emphasizes justice not incentives, labor law
stresses security not mobility, and regional policy stresses equality in all
parts of the country and so on. Growth policy does no longer permeate all
policies on the national level. Instead, growth policy has been regarded as
a separate policy area. Economic growth is no longer regarded a concern of
general policy but the concern of a separate innovation policy and a regional
growth policy.

Spiders – commercial network-builders in modern industry

The industrial dynamics in modern industry differs from the dynamics
a century ago. In the old days, Henry Ford was the role model for the recipe
for success, to merge companies, to acquire suppliers and distributers in
order to cover the entire chain from beginning to end in one single
corporation. Today many companies concentrate on their core business
and choose to procure services that they previously ran in-house. This
leads to a more diversified business world and to companies that are more
specialized. Specialized companies, which must interact with one another
in different business systems. The interaction is organized by spiders, who
build different kinds of networks between more or less autonomous actors
in the modern business world. While the spiders themselves account for
an insignificant fraction of total employment, they are becoming
increasingly important for large parts of modern business life. I differentiate
between spiders of three different kinds; contractors, replicators and
brokers (Giertz 2000).
When companies concentrate on their core business, they partly turn from
being employers to being contractors. As contractors, they engage different
service companies, contract manufactures, consultants, freelancers,
suppliers and subcontractors to produce specific tasks. Contractors bring
together various specialists in temporary projects. Typical contractors
are project
36
managers; turnkey construction contractors book publishers, festival
organizers and film producers. Contractors engage many people but have very
few people employed. Researchers in Sweden started to observe contractors
during the 1990s and called them imaginary organizations (Hedberg 2000).
Today Contractors with an extreme business concept, who don´t have an
office and are interacting with subcontractors mainly through telecommuting,
sometimes called virtual organizations. Contracting is spreading because of
outsourcing and spin-offs as well as increased specialization and
professionalization of services. Relations between contractors and suppliers
are often stable and long-term, even though each supplier is normally part of
many different networks.
Outsourcing of different services of course also leads to the formation of new
service companies and new service industries in the B2B sector. Many of these
services, like machine repairs, installations and janitorial services, just as
many B2C-services, like car repairs, hairdressing or body care, must be
produced and delivered locally to the customer. However, since companies
and people in different locations have similar needs, it is possible to replicate
one successful business concept from one location to another – and this is
exactly what the second kind of spiders, replicators, do. The replicator is the
central node in a replicating organization.
Replicating organizations, which are replicating a business model to several
geographical locations, have during the last decades become a dominant type
of organization in many different industries. One of the origins of replicating
organizations can be found in the retailing sector and today domestic,
international or global replicating organizations are dominating the retailing
sector in developed countries. Replicating organizations are also since long
very common in the hotel and restaurant industry, with McDonald's as the
global role model, and also dominating part of the transportation industry,
like Taxi, and rental industry, like car rental. Domestic and international
replicating organizations are spreading very rapidly and becoming more
common in a great variety of local B2C as well as B2B services. This means
that different local entrepreneurs more or less simultaneously can exploit
different geographic markets with the same standardized business concepts.
The building blocks of a replicating organization include brands, corporate
identity programs, logotypes, designs, manuals and marketing and sales
promotion.
Replicating organizations can be organized in many different ways. In some
cases, the replicator at the top – the central node in a replicating organization
– owns both the business concept and all local branches. The US retail chain
Walmart, with more than two million employees, is such an in-house retail
chain. In franchise organizations, the replicator at the top is a franchiser that
owns the business concept but gives local entrepreneurs, franchisees, the
rights to exploit the concept on a certain local market. In federative
organizations similar local companies voluntarily get together to set up a

37
 replicator as common central node. Together they can decide upon a
common brand and set up common purchasing and marketing
organizations. In recent years a forth kind of replicating organizations,
using internet and sharing economy, have been set up. Thus, organizations
like Uber and Airbnb compete with other more traditional replicating
organizations. This forth kind of replicating organizations work more like
brokers than traditional replicating organizations.
Some Swedish and Nordic replicating organization have been very
successful on the international and global market during the last decades.
Some are found among retail chains, i.e. H&M and Ikea. Both companies
have built up their own product range. On the supply side, they have worked
as contractors buying manufacturing from many different suppliers. On the
sales side, on the other hand, both companies own almost all their stores
in-house in their global replicating organizations. Other successful
replicating organizations are delivering local manual services on the global
market. They started on the Nordic Market and expanded through
acquisitions when the outsourcing trend went global. One example is the
ISS, which offers facility management services and cleaning. In 2010, the
group had more than 520,000 employees in more than 50 countries.
Another example is the Swedish security group Securitas, which had over
280,000 employees in 45 countries in 2010.
The third kind of spiders, brokers, serves as intermediaries in transactions
involving capital, goods or services, by matching a defined supply from
sellers with a corresponding demand from potential buyers. Traditional
brokers, such as insurance brokerage or real estate brokerage and
telemarketing can be rather labor-intensive brokering activities and sales
commissions usually account for the lion´s parts of revenues among those
brokers. Other forms of traditional brokerage activities, such as freight
brokerage, travel agencies and mail order firms are not as labor-intensive.
Traditionally they were based more on mechanical promotional efforts and
passive order taking. However, today traditional brokers in many different
segments are outcompeted by new web-shops, web-portals and virtual
stores. Internet-based brokers of various kinds increase the transparency
and competition in many different industries. In Sweden Blocket, an online
buying/selling website has completely changed brokerage on the Swedish
second hand market. You select a location, item category, and then post
your ad. It works well since almost everyone in Sweden uses it.

A Service society with many different business logics

Where people work and what they do has, due to the industrial dynamics in
Sweden, changed dramatically over time. Traditional categorizations of
companies say essentially nothing about the unique features of different
operations from the standpoint of organization, management, skill profiles
or
38
markets. This is especially true for different kinds of services – which becomes
increasingly bad when the service sector completely dominates working life.
Therefore, I have developed a catalogue of business logics, which complement
the traditional classifications into economic sectors or industries (Giertz
2000). My six main categories, which are broken down into 21 different
business logics, focus on different organizational and management
prerequisites.
It is obvious to anyone that management principles and success factors will
vary from one business logic to another. There is a difference between
organizing and managing the work of a steel mill or a manual assembly line, a
tax office, a consulting firm, a retail store or a call center. However, despite
these obvious differences many people do not specify what business logic they
are referring to when they describe the evolution of management roles
organizational development or leadership. In 2000, I, together with Staffan
Larsson, analyzed how employment in different business logics had developed
over time in Sweden between 1970 and 1997 (Giertz & Larsson 2000). In 2011,
I made a follow up until 2009.
At the beginning of the 20th century, more than half the Swedish population
made a living directly from what the soil, the forests, the mountains and the
sea produced. Nowadays the conditions under which these fundamental
livelihoods are pursued have changed dramatically. Agricultural yields have
increased thanks to better farming methods and the work has been
mechanized. Mechanization has also revolutionized other raw materials
extraction. Thanks to combine harvesters, fishing trawlers, timber processors
boomer rigs and other machines only 1.6 percent of the Swedish working
population was needed to do all raw materials extraction in Sweden in the
beginning of the 21th century.
Since the mid-1960s, the manufacturing sector in Sweden has undergone a
similar transition process as the sector of raw materials extraction.
Mechanization, automation and restructuring have changed the need for labor
– both quantitatively and qualitatively. The price of labor is highly important
in industries that still employ numerous unskilled workers. As a result, many
companies have located their labor-intensive manufacturing operations in
low-wage countries. In parallel, mechanization and automation has reduced
the work forces in remaining factories in Sweden as well. Put simply, semi-
skilled repetitive tasks have been replaced by positions that require a broader
understanding of operations, heavier responsibility and greater technical
skills. The statistics clearly show the development. In 1970, companies
producing some kind of goods still employed 29 percent of the Swedish
working population. In 2009 no more than 13.8 percent of the work force in
Sweden – both white collar and blue collar – were employed by companies
producing any kind of goods, including everything from steel and pulp- and
paper industries to engineering companies as well as breweries and bakeries.

39
 Quite a lot of people are involved in the distribution of goods from factories
to final customers. Originally, many factories were concentrated in coastal
cities that had large ports for inbound and outbound shipments. Starting in
the mid-19th century, the growing railway network gave rise to additional
transport links. Long-distance trucking and intensive air cargo services
later supplemented these. Eventually the shipments led to the
establishment of shops and department stores that could offer a growing
variety of merchandise from all corners of the earth. Today, as the
production of goods has become increasingly more concentrated, parts and
components are being shipped back and forth between factories in different
parts of the world. The volume of goods shipments is thus constantly
growing. Meanwhile many companies are trying to reduce their inventories
and buffer stocks. They would prefer short cuts where goods are shipped
directly from manufacturers to retailers – or even better, directly to the
customers. Despite the heavy increase in transportations, the number of
people employed in the transportation industries, e.g. trucking firms and
sea lines, the transshipping industries, e.g. seaports, airports, terminals,
warehouses and logistic companies, and the retailing industries, e.g. shops,
stores, pharmacies and kiosks, has been slowly declining in Sweden, from
16.7 percent in 1970 to 13.8 percent in 2009.
In a modern society, there are also basic common services that cannot
be controlled only by the wishes of individual users or their willingness to
pay. Basic common services are shared resources. In almost every nation,
basic common services include public administration, the justice system,
the national defense system, the police system, the customs service, fire-
fighting services and the prison system. Basic common service also includes
the operation of infra services of different kinds, such as roads, railroads,
water works, telecom operators and distribution systems for electricity.
Basic common services also include organizations that safeguard common
interests,
e.g. the interests of all citizens, all municipal inhabitants or all members in
for example a church, a confederation or a union. The share of the
working population in Sweden that are employed by organizations
delivering basic common services has been surprisingly stable over time,
increasing from 8.1 percent in 1970 to 9.2 percent in 2009.
Besides distribution of goods and basic common services, there is a very fast
growing service sector. Services of various kinds – health care, child and
elder care, consultancy work, restaurants, handicrafts, entertainment and
so on – are providing gainful employment to more and more people in
Sweden. The share of Sweden’s working population engaged in delivering
operational service production of various kinds increased from 36.2
percent in 1970 to
54.9 percent in 2009. One factor behind this increase is that today Swedish
women work outside the home to as great an extent as men. Before WW
II, children and elderly were cared for by women working in the home but
nowadays this work has become professionalized and salaried.

40
A considerable proportion of the service sector is directly aimed at us in our
capacity as consumers, e.g. restaurants, hotels, day care centers, theaters,
gyms, dentists, hairdressers, taxis, fast-food stands. The share of working
people in Sweden that deliver direct services to consumers at local
establishments increased from 12.8 percent in 1970 to 25.1 percent in 2009.
Services aimed at companies and organizations (B2B) had roughly half as
many employees in Sweden as consumer services (B2C) in the beginning of
the 21th century but they were growing even faster. The reason for this was of
course that companies and organizations increasingly choose to outsource
professional services instead of supplying them in-house. One very fast-
growing sub-sector was knowledge intensive services, e.g. consultancy firms
and accountants, which increased its share of employment in Sweden from 2.9
percent in 1970 to 8.5 percent in 2009. Another large sub-sector was local
manual services, such as various kinds of maintenance and repair work,
electrical installations, painting, plumbing, janitorial and security services. In
2009 11.3 percent of the working population in Sweden delivered local manual
services of some kind to customers on the local market.

Figure 2:

41
 Entrepreneurship is back on the policy agenda
A few years into the 2000s, a consensus grew on the desirability of a broad
widespread entrepreneurship (Braunerhielm 2012). Not only an
increased number of super entrepreneurs that would drive new technology-
based companies. Entrepreneurship and business creation is an equally
important ingredient to increase efficiency in different parts of working life
– including domestic and local services. Fragmentation of organizations and
companies into smaller units, and a subsequent streamlining of operations,
could have significant impact on the pressure for change and growth. The
emergence of a rich flora market funded, entrepreneurial enterprises can be
of equal importance for economic growth as a single new company with
high growth potential.
The wish to support new ventures and entrepreneurship made government
introduce some changes in regulatory systems, for example in the tax
systems. Swedish laws and regulations were in the beginning of 2000s still
very much inspired by Harvard economist John Kenneth Galbraith, who
advocated policies and regulations that favored existing incumbents. He
argued that individual efforts and individual incentives had become less
important. That thinking was still reflected in the Swedish regulatory
system in the beginning of the millennium. Something had to be done.
Sweden, compared to other countries, had a very high inheritance tax and
gift tax, which were taxed progressively. The inheritance tax and gift tax in
Sweden caused many serious problems in connection with generation shifts
in private companies, because private assets were often locked up in the
companies. When assets were shifted, the heirs were often forced to either
sell the company or take out a substantial private loan to cover the taxes.
For several decades, the inheritance and gift taxes in Sweden, combined with
a wealth tax
– which also imposed taxes upon working capital in companies – forced many
business owners either to sell their firm before the transfer to the next
generation, or to move outside Sweden. Many financial advisors in Sweden
had since the mid-1970s recommended successful Swedish business owners
to take up residence in another country. All those taxes were repealed in the
beginning of the new millennium. The inheritance and gift taxes were repealed
on January 1, 2005 and the wealth tax on January 1, 2007. Taxes on residential
property were repealed in 2008. This not only pawed the way for new
entrepreneurs but also made successful Swedish entrepreneurs move back
home to Sweden.

42
Managerial and policy implications
During the last fifteen years globalization has fundamentally changed
the rules and action opportunities for politicians. The financial market is
global, both manufacturing and service companies operate in a global
market. The companies that supply services to the infrastructure are
international. Questions on corporate localizations of the development,
production and back-office businesses are constantly up for discussion, as
well as the exchange of labor in an increasingly open economy.
Globalization has also opened for efficiency improvements and structural
changes in almost all industries.
The industrial dynamics in manufacturing industries can partly be conceived
as conflicting with an increased local entrepreneurship. Economies of scale
are utilized and small companies are acquired and merged when industries
are consolidating. In some mature manufacturing industries, there are only a
few companies left, which compete with one another in the global market.
Alongside companies that expand on the global market are becoming more
specialized.
Of course, there are also new ventures born in manufacturing industries.
Some of them origin from research and new technology and they might very
well grow into successful global companies of importance for the export and
trade balance in Sweden. However, today only about 13 % of the people
working in Sweden are employed by manufacturing industry – and the
percentage is getting lower every year.
The Swedish work force has already moved from manufacturing industry to
service industries of many different kinds. Thus a growth policy of tomorrow
must not only – or even primarily – be about individual innovations or export
companies with a high growth potential. Equally important – perhaps – is a
policy, which improves flexibility, efficiency and productivity in different
service sectors that today accounts for 85 percent of all jobs in Sweden. Better
quality and service with the same resources in all different service sectors is
almost certainly of vital importance to grow the economy in Sweden. From a
societal perspective, the main measure in this context might not be to get
existing enterprises to employ more people. Perhaps it is even better to
support the establishment of more market-funded enterprises that continue
to finance their operations through revenues from services delivered on a
competitive market.
One important question is if the business world´s ancillary systems in Sweden
have kept up with the changes in industry. Are the parties on the labor market,
authorities, ministries, legislative systems, insurance schemes, welfare
systems, working hours, wage formation and so on adjusted to a global,
flexible and dynamic business life with different contracts and incentives in
different industries? Or is a great part of the ancillary systems that were
cemented in the 1970s – just when the steady growth of Swedish economy
begun to turn back down – still controlling the agenda when government,
authorities, employer associations, unions and others discuss future policy
43
 formulations in Sweden?

Another interesting question is if working life research and management

theory in Swedish universities is adjusting to actual trends in today’s
industry. Traditional management theory was formulated in close contact
with manufacturing industry, which nowadays constitutes a very small
part of working life. Some focus is definitely devoted to large incumbents in
the engineering industry who integrate forward and become more of service
organizations. Some focus is also devoted to services in the public sector, i.e.
health care, and to the retailing industry. However, the question remains: Is
enough attention devoted to new management trends in different service
sectors? Where can we in Sweden find courses on how to manage
replicating organizations of various kinds or how to set up different net
services on a domestic or global market or how to organize and manage
successful knowledge intensive consultancy companies?

On methodological considerations and

the use of sources in this article
All content in this article was published before, mostly in Swedish, in
various thesis, reports, books and articles. The main source is an anthology
in Swedish that I edited in 2008. The anthology (Då förändras Sverige)
contains 37 articles, with individual references, written by 25 Swedish
researchers. Other important sources are reports in Swedish, written by
myself, together with various co-authors. These reports are the result of
assignments for public authorities, ministries, industries, employer
confederations, trade unions, research funding organizations and
individual companies in Sweden during more than 40 years. A few of are
mentioned in the reference list below. Thus, this article is to a very large
extent a very short conclusion and translation of my previous work and
previous publications.

44
Suggested further readings
• For those who understand Swedish and want to get more into detail I
strongly recommend to read the anthology Då förändras Sverige (Giertz
(Ed) 2008).
• A classic book of great importance on industrial development in Sweden
during two centuries is written by Lennart Schön: Sweden´s road to
modernity: An economic history. (Schön 2011).
• For those interested in my catalogue of business logics I recommend
Measuring Success. (Giertz 2000).
• Later development – especially in the ICT-sector – is presented in the
anthology Small and beautiful – The ICT success of Finland and
Sweden. Vinnova VA 2015:06.

45
 References
Andersson, M., Braunerhjelm, P., Thulin, P. (2012) Creative destruction and
productivity: entrepreneurship by type, sector and sequence, Journal of
Entrepreneurship and Public Policy 1, 125-146.
Bodin, M., Bredbacka, C, Giertz E. (1996) In (and out of) pace – on business
of today and its ancillary systems – of yesterday? Stockholm: Swedish
Employers Confederation.
Braunerhielm, P., Eklund, K., Henrekson, M. (2012). Ett ramverk för
innovationspolitiken, Stockholm: Samhällsförlaget.
Carlsson, B and Jacobsson, S. (1997). The Technological System for Factory
Automation. Pages 37-59, Technological Systems and Industrial Dynamics.
Boston: Kluwer Academic Publishers.
Chesbrough, H. (2003). Open Innovation: The New Imperative for Creating
and Profiting from Technology. Boston: Harvard Business School Press.
Christensen, C., Anthony, S., & Roth, E. (2004). Seeing What's Next - Using
the Theories of Innovation to Predict Industry Change. Boston: Harvard
Business School Publishing.
Christensen, C.M. (1997). The Innovator’s Dilemma. Boston: Harvard
Business School Press.
Dahmén, E. (1989). Development Blocks in Industrial Economics. Page 136-
148. Stockholm.
Deiaco, E., Giertz, E., Reitberger, G. (2002). Teknikparkens roll i det svenska
innovationssystemet – historien om kommersialisering av
forskningsresultat, Stockholm: Vinnova VFI 2002:3.
Deiaco, E., Reitberger, G. (2008). Teknisk och kollektiv forskning i
samförståndets land. In E. Giertz (Ed) Då förändras Sverige, Lund:
Studentlitteratur.
Ehrnberg, E and Jacobsson, S. (1995). Technological discontinuities and
Company Strategies – Machine Tools and Flexible Manufacturing Systems,
ch. 10 in Carlsson, op. cit., 273-322.
Fagerfjäll, R (2008). Entreprenörer i exportindustrins barndom. In E. Giertz
(Ed) Då förändras Sverige, Lund: Studentlitteratur.
Fredriksson C. (2008). Statsföretagsambitionerna. In E. Giertz (Ed) Då
förändras Sverige, Lund: Studentlitteratur.
Gessler, F. (2002). The development of wireless infrastructure standards.
Doctoral dissertation, Dept. of Industrial Economic and Management at KTH,
Stockholm: Royal Institute of Technology (KTH).

46
Giertz, E. (1981). Om Arbetsstudieutbildningens Institutionalisering i
Sverige. Doctoral dissertation, Dept. of Industrial Economic and
Management at KTH, Stockholm: Royal Institute of Technology (KTH).
Giertz, E. (1987). A New Era for Technical and Clerical Staff in Industry,
Stockholm: The Swedish Work Environment Fund.
Giertz, E (1991). Människor i Scania under 100 år – industri, arbetsliv och
samhälle i förändring, Stockholm: Norstedts.
Giertz, E., Larsson S (2000). Tjänstesamhällets arbetsmarknader – Analys
av verksamhetstyper och kompetenskrav 1970-1997. Stockholm: NUTEK och
Rådet för arbetslivsforskning.
Giertz, E (2000). Measuring success – Identifying Performance Indicators.
Malmö: Celemi International AB.
Giertz, E., Blomberg, H. (2000). Tillväxt och lönebildning – Om löne- och
anställningsvillkor på tjänstesamhällets nya arbetsmarknader. Stockholm:
Kungl. Ingenjörsvetenskapsakademien.
Giertz, E. (Ed) (2008). Då förändras Sverige. 25 experter beskriver
drivkrafter bakom utvecklingen, Lund: Studentlitteratur.
Giertz, E. Levén, P., Sörensson, R. and Logui, M. (2013). Företag inom
informations- och kommunikationsteknik i Sverige 2007-2011. Stockholm:
Vinnova VA 2013:07.
Giertz, E. (2015:1). Understanding the Swedish saga through an historical
exposé. In E. Giertz, A. Rickne & P. Rouvinen (Eds.) Small and beautiful –
The ICT success of Finland and Sweden, Stockholm: Vinnova VA 2015:06.
Giertz, E. (2015:2). The Swedish ICT sector today. In E. Giertz, A. Rickne & P.
Rouvinen (Eds.) Small and beautiful – The ICT success of Finland and
Sweden, Stockholm: Vinnova VA 2015:06.
Isaksson, A. (2008). När socialdemokratin blev statsbärande. In E. Giertz
(Ed) Då förändras Sverige, Lund: Studentlitteratur.
Jakobsson, U. (2008). Storföretagen och den svenska modellens guldålder.
In E. Giertz (Ed) Då förändras Sverige, Lund: Studentlitteratur.
Kaijser, A. (2008:1). Den hjälpande handen – Staten och infrastrukturen. In
E. Giertz (Ed) Då förändras Sverige, Lund: Studentlitteratur.
Kaijser, A. (2008:2). Staten stärker greppet om infrastrukturen. In E. Giertz
(Ed) Då förändras Sverige, Lund: Studentlitteratur.
Karlsson, S., Lugn, A. (2009). Att förändra världen: en berättelse om Lars
Magnus Ericsson och hans efterföljare, Stockholm: Sellin & Partner.
Laestadius, S and Rickne, A. (2012). The theoretical foundation for Swedish
innovation policy. Pages 18-50, Innovation and Governance in an Open
Economy. Routledge, Oxon.

47
 Lindgren, H. (2008). Liberalisering och reglering. Finansmarknadernas
etablering 1850-1920. In E. Giertz (Ed) Då förändras Sverige, Lund:
Studentlitteratur.
Lindmark, S. Andersson, E., Johansson, M. and Bohlin, E. (2004). Telecom
Dynamics - History and state of the Swedish telecom sector 1970-2003,
VINNOVA Analysis VA 2004:04.
Ljunggren, S-B. (2008). Socialdemokratin och Bosse Ringholm år 1967. In
E. Giertz (Ed) Då förändras Sverige, Lund: Studentlitteratur.
Nilsson, J-E. (2008). Regional tillväxtpolitik – det politiskt möjliga. In E.
Giertz (Ed) Då förändras Sverige, Lund: Studentlitteratur.
Saabs Pressinformation (1962). Saab lancerar ny svensk datamaskin,
Linköping: MP-19-62, maj 1962.
Schumpeter, J. (1942). Capitalism, Socialism and Democracy. New York:
Harper & Row.
Schön, L. (2008). Entreprenörer i den tidiga industrialiseringen. In E. Giertz
(Ed) Då förändras Sverige, Lund: Studentlitteratur.
Schön L. (2010) Vår världs ekonomiska historia – den industriella tiden.
Stockholm: SNS Förlag.
Schön, L. (2011) Sweden´s road to modernity: An economic history. Lund:
Studentlitteratur.
SOU 1991:82. Drivkrafter för produktivitet och välstånd, Stockholm.
Sörlin, S. (2008). Universitetsutbildning – från privilegium till medborgerlig
möjlighet. In E. Giertz (Ed) Då förändras Sverige, Lund: Studentlitteratur.
Utterback, J M (1994). Mastering the Dynamics of Innovation. Harvard
Business School

48
49