Content uploaded by James Simmie
Author content
All content in this area was uploaded by James Simmie on Jan 28, 2015
Content may be subject to copyright.
This article was downloaded by: [James Simmie]
On: 26 April 2012, At: 04:25
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
European Planning Studies
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/ceps20
Path Dependence and New
Technological Path Creation in the
Danish Wind Power Industry
James Simmie
a
a
Department of Planning, Oxford Brookes University, Headington
Campus, Gipsy Lane, Oxford, OX3 0BP, UK
Available online: 26 Apr 2012
To cite this article: James Simmie (2012): Path Dependence and New Technological Path Creation
in the Danish Wind Power Industry, European Planning Studies, 20:5, 753-772
To link to this article: http://dx.doi.org/10.1080/09654313.2012.667924
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-
conditions
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.
Path Dependence and New
Technological Path Creation
in the Danish Wind Power Industry
JAMES SIMMIE
Department of Planning, Oxford Brookes University, Headington Campus, Gipsy Lane, Oxford OX3 0BP, UK
(Received March 2011; accepted September 2011)
A
BSTRACT In this paper, it is argued that in order to overcome the limitations of canonical path
dependence and new path creation theories that arise from the different ontological approaches
of economics and sociology, a hybrid socio-economic theory is required that examines the
contributions of intelligent agents to the creation of new technological pathways in conditions of
path dependence. The main aim of such a theory is to explain the processes by which agents may
collectively contribute to the emergence of new technological pathways and overcome the
barriers confronting them as a result of the evolution of historical forces that establish the path-
dependent trajectories of contemporary technologies. The arguments are illustrated empirically
by analysing the roles of inventors and innovation pioneers located in economic niches together
with the diffusion of new technologies to the attainment of critical mass and the creation of new
pathways. The roles and processes engaged in by such actors, and the path-dependent barriers
confronting them, are illustrated using the historical creation the renewable energy technological
pathway in wind power in Denmark.
Introduction
Over the last two decades or so, a growing dissatisfaction with some of the underlying
assumptions of neo-classical economics, such as those of the hypothetical existence of
rational average actors whose actions drive aggregate economies towards equilibrium,
has led to an evolutionary turn in economic theory. Inspired by the work of the Austrian
economist Joseph Schumpeter (Schumpeter, 1939, 1942), a number of distinguished econ-
omists such as Nelson and Winter (1982), Dosi et al. (1988), Hodgson (1993), Arthur et al.
(1997), Foster (1997), Metcalfe (1998), Potts (2000), Fagerberg (2003), Dopfer (2004),
Metcalfe and Foster (2004) and Witt (2003) have sought to understand how real economies
evolve through time. Neo-Schumpeterian theories concerning the role of technological
innovation in driving economic evolution have been prominent in explanations of such
Correspondence Address: James Simmie, Oxford Brookes University, UK. Email: jsimmie@brookes.ac.uk
European Planning Studies Vol. 20, No. 5, May 2012
ISSN 0965-4313 Print/ISSN 1469-5944 Online/12/050753–20 # 2012 Taylor & Francis
http://dx.doi.org/10.1080/09654313.2012.667924
Downloaded by [James Simmie] at 04:25 26 April 2012
evolution (Witt, 1993; Dosi et al., 1998; Metcalfe, 1998). Between them they have studied
technological innovation at the company level, competition, diffusion and structural
change at the levels of markets and sectors and growth, long waves and international
trade at the macro-level. Studies such as these have established the value of adopting an
evolutionary perspective to understand the historical growth and change of real as
opposed to hypothetical economies.
As this is still an emerging approach in economic theory, there is, as yet, no settled view
on which elements of evolutionary theory should be the standard analytical tools of
analysis. So far, at least four main approaches to analysing economic change and inno-
vation have been developed. These are the Darwinian inspired biological analogy and
the notion of the co-evolution of institutions with economic change, complex adaptive
systems theory, panarchy and path dependence theory.
This paper focuses on the last of these four alternatives. In particular, it seeks to address
the question “In conditions of path dependence how are new technological and industrial
pathways created in the first instance?” It is argued that, in addition to the standard
analyses of path dependence and lock-in, we need better explanations of how new techno-
logical and industrial pathways are created in the first instance. How are they related to
previous historical conditions, past path-dependent trajectories and current circumstances?
What are the sources of novelty and innovation? How is novelty and innovation created
and introduced into the ongoing evolution of history? Why are some new technologies
selected in preference to others?
Following this introduction, these questions will be addressed in the remainder of the
paper. First, the original canonical economic explanations for path dependence are laid
out as the theoretical departure point for the paper. Secondly, sociological theories of
new path creation are reviewed. It is argued that these provide a theory of knowledgeable
agents to explain the initial creation of new technologies within the historical context in
which they are conceived. Thirdly, a hybrid socio-economic theory is proposed
that seeks to explain the processes of new technological path creation in conditions
of path dependence. Fourthly, some empirical evidence of new path creation is
analysed. This draws on the creation of the new renewable energy technological
pathwayinwindpowerinDenmarktoillustrate the theoretical arguments developed
in the paper.
A final section draws these arguments and the historical evidence together. It evaluates
the validity of the theoretical arguments as illustrated by the history of new path creation in
the Danish wind power industry.
The Economics of Path Dependence
The seminal economic work of David (1985, 1986) on “QWERTYnomics” provides the
initial inspiration for studies of technological path dependence. According to the David
model, new technological pathways start as a result of small “historical accidents”,
“chance events” or “random” actions. Subsequently, one or more of these chance
events is contingently selected for reasons not immediately connected with the original
event. When that happens path dependence occurs as the original accidental events
become progressively “locked-in” as the development pathway through the operation of
various autocatalytic “network externalities” (David, 1985, 1986) or “increasing returns
effects” (Arthur, 1989, 1994). The final feature of this canonical model is that, once
754 J. Simmie
Downloaded by [James Simmie] at 04:25 26 April 2012
lock-in has occurred it is assumed that a technology, industry, institution or industrial
location pattern will persist until such time as it is disrupted by an “external shock”.
This standard model was recently re-stated by Vergne and Durand (2010). They define
path dependence narrowly as “a property of a stochastic process which obtains under two
conditions (contingency and self-reinforcement) and causes lock-in in the absence of
exogenous shock” (Vergne & Durand, 2010, p. 737). The main characteristics of this cano-
nical model are summarized in Figure 1.
The original David model was developed to challenge two central assumptions in neo-
classical, equilibrium based micro-economic theory. These are, first, that market forces
will ensure that the most efficient technological solutions will finally prevail. Second,
that decisions are, in principle, reversible and will be reversed if and when better technol-
ogies become available. David’s (1985) example of the QWERTY keyboard and Arthur’s
(1990, 1992) analysis of the adoption of VHS in preference to Betamax as the standard
VCR format challenged both of these assumptions.
In the first place, they argued that inefficient or sub-optimal technologies can become
locked-in as industry standards and may persist for considerable periods of time. This is
despite the fact that alternative and more efficient technologies are available. Thus,
lock-in is one of the core concepts in the path dependence model. It expresses the idea
that historical contingency combined with the emergence of self-reinforcing effects
drives a technology or industry along one path rather than another and that this pathway
may be based on inefficient or sub-optimal technologies, institutional or organizational
arrangements.
Secondly, in place of the possible reversibility of decision-making, a key defining
characteristic of path-dependent processes is “non-ergodicity”. This is an inability to
shake free of their history. From this perspective, path-dependent processes arrive at out-
comes that evolve as a consequence of the process’s own history. This history cannot be
reversed. As a result, early chance events and decisions can reverberate through history
closing off alternative paths and confirming a particular path with the implication that
outcomes need not be based on the most efficient technologies available.
David (2005) has argued that the main point of path dependence is to restore the
importance of causal, historical economic explanation involving sequential actions. This
laudable aim, however, currently lacks historical explanation of how and why new
technological pathways are created in the first instance. Martin and Sunley (2006) rightly
Figure 1. Basic economic model of path dependence (after David, 1985, 1986; Arthur, 1989, 1994;
Vergne & Durand, 2010)
Note: Arrows indicate uni-directional historical development of pathway towards lock-in.
Path Dependence and New Technological Path Creation 755
Downloaded by [James Simmie] at 04:25 26 April 2012
point out that “placing too much emphasis on random accidents as the sources of path
creation constrains and undermines such causal explanation” (p. 428). Despite this much
of the path dependence, literature claims that the emergence of novelty and the consequential
creation of new technological pathways is essentially a serendipitous process.
The inadequate discussion of path creation in path dependence gives rise to a curious
paradox in the theory. On the one hand, it is argued that historical processes are important
in shaping the development of a new technology or industry once it has emerged, while on
the other hand history plays no part in shaping its initial creation. But, as Witt (2003)
points out new technologies do not emerge in “virgin markets”. Perez (2010) also
argues that “Studies of innovation (the emergence of novelty) have shown that the
introduction of technical change is not random but (itself) path dependent and interdepen-
dent with other innovations clustered in systems ...” (p. 199). Thus, in reality, new
technologies often have to emerge into a complex landscape of historical path-dependent
developments which themselves may possibly provide significant barriers to that
emergence. New path creation in such historical contexts is not explained by recourse
to the notion of serendipity as suggested in the basic model of David (1985, 1986),
Arthur (1989, 1994) and Vergne and Durand (2010).
The Sociology of New Path Creation
The lack of an adequate economic explanation of how new pathways are created in the first
instance has led a number of scholars to turn to sociological approaches that focus on the
activities of reflexive agents in new path creation. Prominent among them are Garud and
Karnøe (2001). They offer an alternative theory to economic versions of path dependence.
They adopt a sociological ontology and argue that any theory of new path creation should
attach a significant role to the importance of knowledgeable agents and the considered
“mindful deviation” of entrepreneurs from established paths. They suggest that entrepre-
neurs of various kinds create new pathways as they navigate the current flow of events in
“real time” and seek to set new processes in motion by “mindful deviation” (Garud &
Karnøe, 2001, p. 2).
The early work of Garud and Karnøe (2001) takes its inspiration from the “Social con-
struction of technology systems” literature (e.g. Bijker et al., 1987). This suggests that
human agency is distributed across a multiplicity of actors who are embedded in networks
and emerging technological pathways. Thus, actors both contribute to the creation and
emergence of new pathways and are embedded in them as a result of their involvements
in their previous historical development. But at any point in real time, these pathways
provide the platforms for potential departures from them. This is summed up in the
phrase that “agency associated with technological entrepreneurship is distributed,
embedded, and can vary by paths” (Garud & Karnøe, 2001, p. 281).
In later work, Garud et al. (2010) take issue with the first four stages of the standard path
dependence theory shown in Figure 1. They argue that, with respect to path creation,
“initial conditions are not given but rather constructed by actors who mobilize specific
sets of events from the past in pursuit of their initiatives ... what is exogenous and what
is endogenous is not given but depends on how actors draw and redraw their boundaries.
Emergent situations are not “contingencies” but instead afford embedded actors the pos-
sibilities to pursue certain courses of action while making others more difficult to pursue.
Self-reinforcing mechanisms do not just exist but are cultivated. Rather than lock-in there
756 J. Simmie
Downloaded by [James Simmie] at 04:25 26 April 2012
is ever the possibility of creative destruction ... with those with the most to lose proac-
tively making their own creations in order to survive” (Garud et al., 2010, p. 769).
It is evident from these arguments that what is being proposed here is not simply a more
detailed reworking of the early stages of path dependence theory, as outlined in Figure 1,
but an alternative theory based on a major ontological shift to theorizing the emergence
and development of a technology in terms of the interactions between actors embedded
in networks. The main elements of this alternative are summarized in Figure 2.
Garud et al. (2010) state that they “do not embrace a path dependence perspective”
(p. 762). Nevertheless, the “Social construction of technology systems” literature in
some interpretations sees path dependence and path creation as continual inter-linked
processes. This position is put succinctly by Martin and Sunley (2006) with respect to
economic evolution which they argue may “be understood as an ongoing, never ending
interplay of path dependence, path creation and path destruction that occurs as actors
in different arenas reproduce, mindfully deviate from, and transform existing socio-
economic-technological structures, socio-economic practices and development paths”
(p. 408). The theoretical aim of this paper is therefore to develop this insight by linking
evolutionary theories of path dependence with a sociological explanation of how new tech-
nological pathways are created by knowledgeable agents in the first instance in a hybrid
socio-economic theory of new path creation. Before developing this new hybrid theory,
it is necessary to point out that since the original narrow economic explanation of path
dependence, based on contingency and self-reinforcing processes leading inevitably to
lock-in that can only be broken by external shocks, additional causes of path dependence
have been identified. The paper now turns to an evaluation of these arguments because
they form the bases of our understanding of the evolution of the historical conditions
into which new technologies emerge, and the contemporary path-dependent barriers that
may possibly block their development and diffusion to a critical mass that would mark the
creation of a new technological pathway.
Beyond Contingency and Self-reinforcement
The standard economic model of path dependence provided by David (1985, 1986), Arthur
(1989, 1994) and Vergne and Durand (2010) is designed to show how contingency and
Figure 2. The sociology of new path creation (after Garud & Karnøe 2001; Garud et al., 2010)
Note: Arrows indicate direction of historical development of new pathway and iterative reflexive
feedback loops.
Path Dependence and New Technological Path Creation 757
Downloaded by [James Simmie] at 04:25 26 April 2012
self-reinforcing process can lead to the outcome of stability and lock-in of (inefficient)
technologies. In this argument, the introduction of new technologies is blocked for
considerable periods of time until being replaced as a result of some major external
shock. But other theorists have identified a number of additional forces that can lead to
path-dependent outcomes. These historical forces influence strongly the evolution of the
initial conditions in which new technologies are created and the resulting contemporary
barriers to new path creation.
The forces that may lead to the evolution of path-dependent development pathways
include cognitive parameters in the form of technological paradigms, existing rules and
norms exemplified by institutional hysteresis and established technological regimes.
These barriers are all formed and maintained by path-dependent processes and are in
addition to the economic forces leading to path-dependent development.
Cognitive barriers to new path creation were first explored by Dosi (1982). He
developed the Kuhnian concept of scientific paradigms (Kuhn, 1962) and argued that
technologies develop and diffuse along trajectories that are path-dependent because of
the influence of the technological paradigms within which they are developed. Such para-
digms are the prevailing models for the solution of techno-economic problems (Markard &
Truffer, 2006). They may be defined as “a collectively shared logic at the convergence of
technological potential, relative costs, market acceptance and functional coherence”
(Perez, 2010, p. 186).
For Dosi, a technological paradigm involves a specific set of heuristics and visions on
how to do things and how to improve them. They represent the conventional wisdom of the
relevant community of practioners and their collectively shared cognitive frames (Dosi &
Grazzi, 2010). The development pathways of new technologies proceed along trajectories
that are shaped and constrained by the paradigms within which they have been
inaugurated. This is a path-dependent process because the rate and direction of technical
change are limited by the currently acceptable cognitive frameworks within any given
technological paradigm.
Geels (2002) argues that the effects of technological paradigms are not only limited to
the search heuristics adopted by engineers that strongly influence the direction of travel
along a technological trajectory but also become embodied in the socio-technical land-
scape. This represents “a set of deep structural trends ... (expressed) in the material
context of society for example the material and spatial arrangements of cities, factories,
highways and electricity infrastructure” (Geels, 2002, p. 1260). In the face of such
powerful cognitive and material barriers, the creation of new technical search heuristics
may take years because it proceeds in small incremental steps and often involves
experiments and failures (Geels, 2004).
The historical and cumulative development of knowledge along a particular techno-
logical trajectory can turn a once competitive technology into the basis of a declining
industry. In this instance, learning that once functioned as “barriers to entry” for outsiders
may be transformed into “barriers to exit” for insiders (Bianchi, 1986).
In a second important development of the canonical path dependence model North
(1990) and Setterfield (1993, 1995, 1997) argue that institutions, in the form of socially
constructed rules of behaviour, can also become locked-in. In such cases, they may
provide barriers to the creation of new technological pathways. North (1990), for
example, argues that the increasing returns discussed by Arthur that generate economic
path-dependent development, also apply in institutions. This is because institutions
758 J. Simmie
Downloaded by [James Simmie] at 04:25 26 April 2012
develop as a web of interdependent formal and informal rules that work together to favour
and reinforce some kinds of actions and exclude others.
Setterfield (1993, 1995, 1997) also developed the idea that institutions can become bar-
riers to change. He introduced the concept of institutional hysteresis in which institutions
and economic pathways are argued to co-evolve in an interdependent way. In the short run,
institutions are said to be “exogenous” to the economy and to provide a stable framework
for economic activity. This initial stability may not be conducive to the emergence of new
technologies. As path dependence sets in, however, it is also argued that institutional struc-
tures become “endogenous” to the economy and open to feedback effects from changes in
the economy. This leads to the co-evolution of path-dependent institutions and economic
pathways. Echoing the canonical theory of path dependence North and Setterfield both
argue that, as with technologies, some institutional structures that emerge may not be the
most efficient. As a result of their path-dependent development, this means that some
inefficient and sub-optimal institutions may become locked-in for long periods of time
(Martin & Sunley, 2006). Thus, in conditions of institutional hysteresis where inefficient
and sub-optimal economic pathways and institutions co-evolve together and become
locked-in to these forms for considerable periods of time, they can form hostile historical
conditions and powerful contemporary barriers to new technological path creation.
Technological regimes provide a third type of barrier to new path creation. This concept
was first introduced by Nelson and Winter (1977). As part of their development of an
evolutionary approach to economic change, they argued that firm behaviour is governed
by a set of technological routines. The combined set of such routines they defined as a
technological regime. They noted that the problem-solving activities of engineers were
not oriented so much to changes in costs or demand conditions, as predicted by neo-
classical micro-economic theory, but remained remarkably stable over time. They were
focused on particular problems and informed by certain ideas of how these problems
might be solved. Nelson and Winter (1977) illustrate these ideas using the long-term
and incremental development of the DC3 aircraft programme during the 1930s.
This original, narrow definition of a technological regime was broadened by the sociol-
ogists Rip and Kemp (1998) to incorporate much more of the social context of technology.
In contrast to Nelson and Winter’s (1977) narrow definition, they defined a technological
regime as “the whole complex of scientific knowledges, engineering practices, production
process technologies, product characteristics, skills and procedures, and institutions and
infrastructures that make up the totality of a technology” (Rip & Kemp, 1998, p. 340).
In this conception, technological regimes represent an intermediate level between specific
innovations as they are conceived and developed by particular agents and what Rip and
Kemp (1998) call “social landscapes”. They argue that technological regimes represent
on the one hand the boundary conditions within which technologies are developed and
on the other links to the more concrete and slow to change social landscape.
The idea that there are different degrees of technological path dependence at different
analytical levels has been influential in subsequent studies of the relationships between
individual technologies and whole technological systems and their stability and change.
Geels (2004), for example, takes up the sociological approach of Rip and Kemp (1998)
to analyse the barriers to the adoption of new technological systems. He analyses
transitions from one technological system to another in terms of three path-dependent
analytical dimensions. Technological regimes together with other types of regime form
the meso-level of this analysis. Above them in a hierarchy of relative stability and hardness
Path Dependence and New Technological Path Creation 759
Downloaded by [James Simmie] at 04:25 26 April 2012
lies the socio-technical landscape that includes the material aspects of society and is
therefore difficult and slow to change. Below the meso-regime level, a multiplicity of
technological niches come and go.
This idea of multiple levels of technological path dependence may also apply to
institutional path dependence described above. Thus, at the landscape level, institutions
may take the form of legal rules enacted by governments. At the grass roots level, they
may be the historical ways of doing things. And at the meso-level, they may be informal
but conventional rules that incur social sanctions if they are not followed.
Overall, the processes of new path creation are confronted by three major types of path-
dependent barriers in addition to basic economic barriers such as how competitiveness and
costs are defined, risk aversion and network externalities. These barriers arise because of
the processes leading to path dependence and different degrees of lock-in. They include
cognitive barriers in the form of technological paradigms, institutional hysteresis and
social barriers emanating from the groups who are the bearers of existing technological
and other types of regimes.
Hybrid Theory
Given the historical evolution of path-dependent development trajectories that form the
bases of the initial conditions and barriers, outlined above, that confront the creation of
new technological pathways, the evolutionary hybrid theory proposed in this paper
seeks to explain how new technological pathways are created in conditions of historical
path-dependent development. The theory starts with the assumption that new pathways
are not created by disembodied economic forces but by knowledgeable agents. In order
to explain the reflexive actions of these agents, it is necessary to add a sociological
dimension to the analyses of the causes of path dependence.
In their sociological alternative to traditional economic path dependence theory, Garud
and Karnøe (2001) and Garud et al. (2010) argue that path dependence is a process that is
influenced by history but that is also made out of that history as a result of the interactions
between embedded agents and the structures in which they are embedded. In this case,
technological deviations and departures may be introduced, mostly incrementally, over
time by agents of change. In their earlier work, Garud and Karnøe (2001) argue that the
key agents are entrepreneurs who mindfully deviate from contemporary flows of events
to create new pathways. Later they broaden this view to include “also those who
develop complementary assets .. . those in institutional forums ... and customers who
offer critical inputs that shape emerging paths” (Garud & Karnøe, 2003, p. 279). This
list, however, does not exhaust the potential actors who may contribute to the creation
of new technological knowledge and economic novelty which may be regarded as the
key drivers of new path creation.
Upstream from entrepreneurs who may be responsible for the commercialization of new
knowledge in the form of innovations, individual and collective inventors play key roles.
In the past, individual private sector inventors have been the pioneers of new technologies,
for example, in successive Kondratieff waves of technological development. But, the
increasing scientific basis of new technologies has also been accompanied by a growing
role for large R&D intensive companies (e.g. pharmaceuticals) and public sector actors
such as universities (e.g. genetic technologies) and government research establishments
(e.g. nuclear technologies). Unlike entrepreneurs selling into competitive markets,
760 J. Simmie
Downloaded by [James Simmie] at 04:25 26 April 2012
organizations such as these are often supplying markets for such services as health and
power that have been established or regulated by governments. Thus, a free market
model in which private sector entrepreneurs are the main agents of new technological
path creation has limitations in its potential applicability in contemporary Western
economies, particularly those characterized by “co-ordinated capitalism”.
Garud and Karnøe (2003) also ask what mindfully deviating actors actually do to create
new pathways. They suggest two possible strategies and illustrate them with an analysis of
the early path creation of wind power technologies and industries in Denmark and the US.
They argue that the early pioneers in Denmark adopted a strategy of “bricolage”. This is
defined as “a process of moving ahead on the basis of inputs from actors who possess
local knowledge, but through their interactions, are able to gradually transform emerging
paths to higher degrees of functionality” (Garud & Karnøe, 2003, p. 296). In other words,
actors in this case sought to improve the existing technologies by incremental innovations.
This created a new wind power pathway by modest adaptations over a long period of time.
In contrast, actors in the NASA-led US public wind power research programme adopted
a strategy of “breakthrough”. This is defined as “actors attempting to generate dramatic
outcomes rather than adaptiveness, an unyielding vision to leap-frog the Danish initiative”
(Garud & Karnøe, 2003, p. 279). In this case, the actors involved were trying to generate
new technologies by radical innovations. In the event, the Danish actors’ strategy
succeeded while that of the US actors was mostly an expensive failure.
There is also a substantial literature from other fields such as historical sociology and
political science on the activities of knowledgeable agents that may lead to the creation
of new institutional arrangements and new technological pathways. Of particular note
here is the introduction to their book by Streek and Thelen (2005). In it they identify
three ways in which the “gradual transformation” (p. 9) of institutional arrangements
may take place. These are displacement, layering and conversion. Such discontinuities
and transformations may equally apply to the actions of knowledgeable agents in new
path creation. They represent three empirically substantiated ways in which the socio-
logical interactions of actors may “transform emerging paths to higher degrees of function-
ality” (Garud & Karnøe, 2003, p. 296) and thereby create new technological pathways.
Such transformation may also involve purposive planning (Sydow et al., 2010).
Displacement is defined as the “slowly rising salience of subordinate relative to
dominant institutions” (Streek & Thelen, 2005, p. 31) and, in the argument advanced
here, technologies. Defection from one to the other is the mechanism by which change
takes place. This is made possible by the existence of some incoherence within existing
arrangements that opens up possibilities for “mindful deviation”, the cultivation of
alternative technological paradigms, the rediscovery of dormant technologies and/or the
assimilation of external practices.
The early history of the motor car provides one example of this process at work. During
the late nineteenth century, the previous long-term development of steam power led to this
emerging as an important early propulsion technology for the new “horseless carriage”.
Even Henry Ford spent many of his early years trying to build a steam road carriage
and a farm locomotive. But, after he moved to Detroit, he learned that a German engineer,
Nicholas Otto, had built an internal combustion engine driven by petrol. From then on,
Ford concentrated on developing a petrol-driven motor car. Once he went into mass
production with his Model T, steam was no longer developed as a serious propulsion
technology for motor cars.
Path Dependence and New Technological Path Creation 761
Downloaded by [James Simmie] at 04:25 26 April 2012
Layering involves adding new rules, procedures, structures or technologies to what
already exists. Geels (2002) provides an example of this process at work with his analysis
of the rise of steam ships. He argues that, initially, steam engines were added to sailing
ships as an auxiliary device. In this way, the new technological pathway was created by
adding the new form of propulsion to what already existed. The subsequent differential
growth of steam technology eventually led to the dominance of this new technology
over the old.
Conversion is the reorientation of an institution or technology in terms of its form or
function. An example of this process at work is that, in the early days of radio its main
function was to facilitate the communication of information concerning phenomena
such as the weather and maritime movements. It was not until the 1920s that its main
function began to shift and convert towards broadcast entertainment. This created the
new production pathway for the production of content for this new function of the medium.
Taken together these different types of agent activities suggest that transformative
new technological pathways can be created in ways other than those involving external
shocks, “gales of creative destruction” and terminal path breaking. These are relatively
rare events. Breakthrough innovations and their rapid diffusion to create new pathways
are also the exception rather than the rule.
The hybrid theory advanced in this paper is therefore primarily concerned with explain-
ing new path creation by knowledgeable agents in the context of transformations of the
path-dependent characteristics of the economic environment, technological paradigms,
institutions and technological regimes. These provide powerful historical barriers to
change. Rather than confronting them directly, it is often easier for agents to initiate
new technologies in niches that are not subject to the full force of the tides of history.
Niches may be regarded as important minority selection environments in which the
process of new path creation may be started. “They provide locations for learning
processes, e.g. about technical specifications, user preferences, public policies, symbolic
meanings. Niches are locations where it is possible to deviate from the rules of the existing
regime” (Geels, 2004, p. 912). While not all new technologies start in such niches, they
often provide important potential locations for the process of mindful deviation that is
required for new path creation (Garud & Karnøe, 2001).
A “niche” may be defined as an application context in which the new product or tech-
nology is temporarily protected from the standards and selection rules of the prevailing
paradigm (Kemp et al., 1998; Hoogma et al., 2002; Markard & Truffer, 2006). Niches
provide space for novelties to incubate without being subjected to prevailing competitive
market pressures or the normal selection criteria that accompany the dominant techno-
logical trajectories. The informal rules of niche environments are less articulated and
subject to higher degrees of uncertainty than those of the established trajectories (Geels,
2004). In niche conditions, it is also possible to draw on new or deviant local or
international knowledge in order to develop new business networks, value chains and
user–producer relationships.
While niche environments provide opportunities for technological inventions to be
introduced, they do not of themselves create inventions. Those are introduced as a
result of interactions between knowledgeable economic actors, policy-makers and
the potential users of inventions or their results. Following Garud and Karnøe (2001),
it is argued that strategic agency is significant in new path creation. This involves
the “mindful deviation” from established paths not only of individual private sector
762 J. Simmie
Downloaded by [James Simmie] at 04:25 26 April 2012
entrepreneurs but also, among others, of researchers in universities, large companies and
government research establishments.
The commercialization of inventions and the creation of new pathways in the form of
innovations is more often than not a long and incremental process. Frequently, older path-
ways continue for some time while newer ones are initially introduced as additions to
existing technologies rather than immediate replacements for them. The nature of these
processes is captured by the concepts of “displacement”, “layering” and “conversion”
(Streek & Thelen, 2005).
Niches may shelter initially technological inventions from the path-dependent tides
of the economic selection environment, technological paradigms, institutions and techno-
logical regimes. Nevertheless, eventually these barriers have to be overcome if they are
to be able to diffuse to a critical mass and the creation of a new pathway in the overall
technological landscape.
A critical mass is reached where, in the face of existing barriers, sufficient economic
agents are prepared to switch to the new alternatives (Witt, 1997). Critical mass is a
well-known phenomenon in non-linear dynamic systems (Lorenz, 1993). It may be
defined as a point of discontinuity that induces a dramatic turn away from an existing
system (Witt, 1997). The current economic selection environments, technological
paradigms, institutions and regimes always favour the existing and widely used product
variants. For this reason, success in the creation of new technological pathways comes
down to the prerequisite to pass a critical mass (Witt, 1997).
The diffusion of innovations is a key process that spreads their use from their original
invention in niche conditions to the point at which critical mass is achieved and a new
economic pathway is created that represents a significant discontinuity in the technological
landscape. In contemporary economies, as with the introduction of an innovation,
diffusion also requires knowledgeable agents. The roles of diffusion agents have a long
pedigree in the diffusion literature (Brown, 1981; Rogers, 1995). Thus, when a major
innovation is introduced, commercial marketing agencies or in-house people are often
given the task of promoting them and triggering a diffusion process (Witt, 1997).
Innovation diffusion may also need to be organized by collective action. This is
because setting the diffusion process in motion is like providing a public good. Without
collective action, the early adopters would have to bear the initial network diseconomies
while later adopters would profit from the investments of the early adopters (Witt,
1997). Thus, public action may be needed to overcome some of the barriers to new
path creation. The introduction of electric cars is one example where such action is
necessary.
The theoretical arguments proposed in this paper are summarized in a simplified form in
Figure 3. For the purposes of exposition, the figure is divided into five main elements. In
the first, it is argued that the initial conditions confronting the introduction of a new tech-
nology are determined by previous rounds of the historical evolution of path-dependent
technological development trajectories. Given these initial conditions, the new path cre-
ation process starts with the mindful deviation of knowledgeable agents. In the case of
new technologies, these agents are defined as “inventors”. Such deviation is easier in
niches that are not subject to the full force of the initial conditions. In a second stage of
the new path creation process, inventions are taken up by innovators who introduce trans-
formations and discontinuities into the contemporary mix of technologies. In order to
reach critical mass and to become a new technological pathway, these introductions
Path Dependence and New Technological Path Creation 763
Downloaded by [James Simmie] at 04:25 26 April 2012
have to overcome one or more of four main path-dependent barriers to new path creation.
These are the economic selection environment, the relevant existing technological para-
digms, institutional hysteresis and the relevant contemporary technological regimes.
The outcome of the tensions and conflicts between innovation and the barriers to new
path creation may be either the diffusion of a new technology to critical mass and the
creation of a new pathway, the de-locking of an existing pathway or the continued
lock-in of path-dependent development trajectories. It is not possible to predict in
advance which of these three outcomes will prevail in the economic landscape.
This hybrid approach to understanding new path creation is illustrated in this paper by
the introduction of new wind power technologies in Denmark following WWII. This
new pathway was created in the context of the established path-dependent technological
trajectories for the generation and supply of electricity.
This hybrid is an evolutionary theory in the terms proposed by Martin and Sunley
(2006). It seeks to establish an empirically grounded basis for understanding new
technological path creation in the context of the “ongoing, never ending interplay of
path dependence, path creation and path destruction that occurs as actors in different
arenas reproduce, mindfully deviate from, and transform existing socio-economic-
technological structures, socio-economic practices and development paths” (Martin and
Sunley, 2006, p. 408).
Figure 3. Hybrid socio-economic theory of new path creation
Note: Arrows indicate direction of trajectory of new path creation and reflexive feedback loops.
764 J. Simmie
Downloaded by [James Simmie] at 04:25 26 April 2012
New Path Creation: The Introduction of Wind Power Technologies in Denmark
Initial Conditions
At the beginning of the twentieth century, the established technological pathways for the
generation and supply of electricity did not apply much beyond urban areas. This provided
significant niches in rural areas that were not connected to grid supplied AC electricity. As
a result, inventions and innovations in the generation of electricity using wind power were
first developed and used in rural areas. There they were used to produce direct current
(DC) electricity powering batteries. This type of current cannot be transmitted over
long distances via a grid system.
After the War restrictions were imposed on Danish diesel fuel imports. The supply of
diesel for rural electricity generators was uncertain. This added to the niche conditions
for the adoption of wind generators in rural areas.
Confidence in the reliability and price of oil supplies was also shaken by the external
shock of the Yom Kippur War in 1973 and the Iran/Iraq war of 1980 – 1988. They
sparked widespread political debates. These led to a switch to coal-based energy gener-
ation and the emergence of a strong anti-nuclear movement that rejected nuclear power
as a long-term solution to Denmark’s energy needs.
Path Creation Processes
The first Danish wind turbines were invented in the rural niche beyond the contemporary
reach of urban based grid connected electricity suppliers. The most influential pioneering
inventor was Poul la Cour. He wanted to bring the benefits of electricity to people living in
the countryside. In order to achieve this, he managed to mobilize resources from the
Danish government to build his first DC windmill in Askov east of Esbjerg in 1891.
This was a smock type mill with a fantail to point the sails into the wind and self-
reefing shuttered sails. In the same year, he also invented a device called a “kratostat”
to smooth out the power fluctuations caused by variable wind speeds.
In Denmark, early innovations in AC wind turbines were inspired by deviation from the
dominant technological paradigm and the development of an alternative in a local minority
culture of enthusiasm for wind power. Johannes Juul was one of the key pioneer inventors.
He was a student of Poul la Cour and managed to mobilize resources from the Danish
utility SEAS to fund a niche research and development programme to produce electricity
using wind power. As a result of this programme, the first Danish AC wind turbine was
built in 1950 at Vester Egesborg on the south-west coast of Zealand. It was grid-connected
using an induction generator and was left in service for a decade. The successful
demonstration of Juul’s experimental turbines enabled him mobilize further resources
by securing public niche funding from the Danish government to construct a larger
model in 1958 at Gedser about 130 km south of Copenhagen.
These early incremental inventions gradually increased the power output of wind turbines.
But none of them could generate more than 200 kW of electricity. At this level of pro-
ductivity, they were not competitive in the markets outside their various niches. It
became clear that power outputs in excess of 1 MW would be needed to make wind turbines
economically competitive with other sources of energy at “normal” prices.
The first reliable 1 MW wind turbine was constructed by the Teacher Group in Tvind.
Like Juul, they started by deviating from the dominant fossil fuel-based technological
Path Dependence and New Technological Path Creation 765
Downloaded by [James Simmie] at 04:25 26 April 2012
paradigm for the generation of electricity. They did not accept that nuclear power was an
acceptable alternative to this paradigm. As a result, they set about demonstrating that wind
power could provide electricity on a scale and at a price acceptable to electricity utility
companies. They were led by pioneer innovator Amdi Peterson. To keep costs down,
they purchased a second-hand generator and gearbox originally used on a mine hoist. A
low-speed shaft was bought second-hand from a shipyard. Volunteers built the concrete
tower. They went to Stuttgart to learn from Ulrich Hutter how to make variable pitch
fibreglass blades. The grid was supplied via a rectifier converter that was purpose built
for the turbine by students from the technical University of Denmark supervised by
Ulrik Krabbe. The wind turbine was completed in 1978 and remains in operation in
2010. It laid the foundation by which Danish wind turbines have diffused to become
commercially competitive in world markets.
In 1979, the Danish parliament passed legislation that extended the niche market for
wind turbines by providing wind turbine purchasers with a 30% subsidy as long as the
turbine was of an approved type. In addition, in the late 1970s, the connection of small
AC turbines to the national grid in Denmark was formalized by Danske Elvaerkers
Forening (Danish Association of Electricity Supply Undertakings). This established a
collective ability for anyone with an approved wind turbine to connect to the national
grid and to receive payments for any surplus electricity they supplied to the network.
These niche market conditions were further extended with legislation that allowed
individuals who had shares in co-operatively owned turbines in or near where they
lived, to pay no tax on the income from electricity sales up to about 9000 kWh. This
together with the 30% capital subsidy made well-sited and maintained wind turbines an
attractive financial investment for local individuals and co-operatives.
Path Establishment Processes
The diffusion of wind-powered DC generators developed slowly between the wars. During
this period, the manufacturing company Lykkegaard was an important diffusion agent.
They made and installed around 20 machines before the Second World War. As a
result, the Lykkegaard Company made and installed around 60 of their four-bladed wind-
mills in the early 1940s. Even so, despite strong niches and weak barriers to creation of this
new technological pathway, the total number of installed wind generators in Denmark
remained small at this time.
The diffusion of the pioneer inventions by la Cour and Juul to create a new industrial
pathway in Denmark required a combination of both niche home market users and knowl-
edgeable manufacturing diffusion agents. The main users were individual farmers and
local rural co-operatives. The main diffusion agents were agricultural machinery manufac-
turers. The incremental innovations made by the early pioneers and subsequently by wind
enthusiasts demonstrated the practicality of generating grid-connected AC electricity
using wind turbines. Danish agricultural machinery manufacturers saw a market opportu-
nity to diversify their product ranges. Their long experience of engineering robust and
reliable agricultural equipment made Danish wind turbines as reliable as was possible
for an entirely new product.
One of the first diffusion agents to diversify in this way was Vestas. In the 1970s, it was
a leading Danish manufacturer of mobile cranes and agricultural equipment employing
about 100 people. It diversified by buying the manufacturing rights to the successful
766 J. Simmie
Downloaded by [James Simmie] at 04:25 26 April 2012
HVK wind turbine developed by Karl Erik Jorgensen and Henrik Stiesdal. Production
started in 1979. The firm is now the largest producer of wind turbines in the world with
over 13,000 employees in 2007 (Musgrove, 2010, p. 107).
Path-dependent Barriers to New Path Creati on
Although Denmark was the first country in which a significant new technological pathway was
created in wind turbines, this was achieved in the face of a number of path-dependent barriers.
The first of these was the economic hostility of existing electricity generators. They had long-
term investments in fossil fuel-based generation technologies. Their arguments regarding the
relative competitiveness of these technologies, when compared with renewable alternatives,
werealsoframedinsuchawayastoexcludetheir consequential costs such as pollution and
waste disposal. As a result, the diffusion of wind turbine technology was slowed until its
purely financial costs could converge with those of conventional technologies. This could
only be achieved by the development of reliable 1 MW and larger turbines.
The diffusion of wind turbine technology also had to overcome the prevailing technologi-
cal paradigm of centrally generated, AC, grid-connected electricity generation. The conven-
tional wisdom of this paradigm was that matching the supply with the demand for electricity
required reliable base load generation, some flexible capacity for meeting peaks in demand
and technological systems for matching both peaks and troughs in demand with the available
supply. In the context of this paradigm, wind power was seen as an intermittent source of
energy not able to supply either base load requirements or peak demand. It was also seen
as requiring additional technological systems in order to make use of its energy in the first
instance and to accommodate both local consumers’ exports and imports of electricity at
different times. Overcoming these barriers required the development of complementary tech-
nological systems that could cope with the intermittent supply of energy and institutional
changes requiring existing electricity generators to make some use of wind energy.
The diffusion of wind turbine technology was also confronted not so much by institutional
hysteresis as by institutional churn. At some points in time such as 1985, 1990 and 1996, the
Danish Government specified that Danish energy companies should make use of wind power.
But at others such as 1996 and 1999 this was placed in doubt as a result of the EU seeking
to open up national electricity markets to international suppliers over which the Danish
government had little control. Thus, the development of the new pathway did not follow a
continuous sequence of positive stages. There were also setbacks in its development. Such
institutional churn brought into conflict the long-term investment strategies required to
introduce the new technology with short-term political and institutional changes. Overcoming
this barrier required long-term certainty in the market for renewable electricity in order to
stimulate the investment required to diffuse widely the new technology.
Landscape Change
By 1980, the new wind turbine manufacturing pathway had been created in Denmark. At
that time, there were a dozen manufacturers producing grid-connected AC wind turbines
with rotors ranging from 5 to 15 m producing 5 –55 kW of electricity. Further incremental
innovations increased their efficiency throughout the 1980s. This increase in productivity
was sufficient to compensate for the removal of the Danish government’s 30% capital
purchase subsidy in 1989.
Path Dependence and New Technological Path Creation 767
Downloaded by [James Simmie] at 04:25 26 April 2012
During the 1980s and 1990s, this new pathway developed to supply some 6.65% of
Danish electricity by 2004. Figure 4 shows the take-off and growth of the contribution
of the new technological pathway to the supply of electricity in Denmark. The graph
illustrates the argument in this paper that the Danish inventions and innovations in wind
energy technologies led to the creation of a new indigenous economic pathway that
makes a significant new contribution to the supply of electricity in Denmark.
The Danish wind turbine pathway was created over a long period of time starting with
the inventions of Poul la Cour at the end of the nineteenth century and culminating with
the successful diffusion of the technology by firms such as Vestas, Bonus, Micon and
Nordtank in the 1980s. The initial conditions in which the new technology developed
were set by previous rounds of path-dependent development. In the early days, these
were the limited reach of electricity grid systems in rural areas. Later on, they also included
the initial competitive market advantages of fossil fuel-based centrally generated electricity.
It was started by the mindful deviation of knowledgeable actors in rural and publicly created
niches. It was developed primarily through incremental innovations that added a new layer
of electricity generation technology to the existing fossil fuel-based technologies.
This layering process involved adding new rules in the form of legalizing grid connec-
tion of approved wind turbines; procedures in terms of government subsidies, tax relief
and feed in tariffs; structures in the form of government regulations and technologies
in the form of wind turbines and complementary technological systems to the existing
electricity supply system in Denmark.
The early pioneers were inspired by alternative technological paradigms either in terms of
the supply of electricity in rural areas or by an anti-nuclear animus. It affected and was affected
by the existing technological landscape for the generation of electricity. Thus, on the one
hand, it added a new competitive technology to that generation landscape and on the other
its diffusion was made easier by oil price shocks to the conventional fossil fuel-based
electricity supply landscape. These iterative interactions are summarized in Figure 5.
Figure 4. Denmark: new path creation, electricity generation by wind power 1980– 1994
Source: Adapted from Gipe (1997).
768 J. Simmie
Downloaded by [James Simmie] at 04:25 26 April 2012
Conclusions
It has been argued in this paper that the canonical model of path dependence does not
attempt to provide an explanation of the processes associated with the intentional creation
of new technological pathways. It is primarily concerned with the emergence of path
dependence. Nevertheless, new pathways are not created by disembodied economic
forces. They require social action by knowledgeable pioneering individuals, universities,
companies and/or governments. The aim of the hybrid theory proposed in this paper is
to provide a potential empirically grounded basis for understanding the interactions
between the evolution of path-dependent economic development and the social actions
of pioneering agents who contribute to new path creation.
In this context, it is argued that new path creation is an iterative process in which the
initial conditions are set by previous rounds of path-dependent development. In these con-
ditions, knowledgeable agents who are identified as inventors and innovators may set in
motion the creation of new technological pathways by consciously deviating from past
Figure 5. New technological pathway created in wind power (Denmark 1950–1980s)
Path Dependence and New Technological Path Creation 769
Downloaded by [James Simmie] at 04:25 26 April 2012
practices and introducing and attempting to diffuse new technologies. The general ways in
which new technologies are diffused by agents include the adoption of incremental strat-
egies such as displacement where the salience of subordinate technologies increases to
overtake that of the dominant technology, layering, where new technologies are added
to those already present, and conversion where old technologies are changed.
These strategies do not require that existing pathways are completely broken in order to
facilitate the creation of new ones. In the case of displacement, a dominant technology is
superseded by a subordinate one but may continue for some time afterwards. Layering
involves the addition of a new technology while previous technologies also continue to
be employed. Conversion makes use of previous knowledge to develop a new technology.
This is the closest to a path-breaking scenario, although remnants of the old may continue
in the new.
Given that the canonical version of path dependence argues that lock-in is the outcome
of the process and that new pathways are created in initial conditions of previously locked-
in development pathways, path-dependent processes themselves establish major barriers
to new path creation. Foremost among these barriers are cognitive lock-ins in the form
of existing technological paradigms, institutional lock-ins in the form of institutional
hysteresis and technological regime lock-ins in the form of the entrenched power of the
social groups who are the bearers and principle protagonists of those regimes.
The outcomes of the iterations between knowledgeable agents and forces involved in
the different stages of new path creation are unpredictable. They may lead to changes
in the existing technological landscape as a new technology diffuses to a critical mass
at which point it could be described as constituting a new pathway. Alternatively, the
barriers to diffusion may be so great as to prevent new path creation and maintain previous
path-dependent development trajectories.
A final element in this iterative model is that the technological landscape itself can be either
a force for continuing lock-in or a force for new path creation. Contemporary concerns about
climate change, for example, if translated into effective collective action can change
dramatically the technological landscape in which power is generated and used.
These arguments have been illustrated by case studies of the initial introduction of
wind power technologies for the generation of electricity in Denmark. This is only one
example and further historical case studies and empirical analyses would be necessary
to demonstrate the degree of general applicability of the theoretical approach outlined
in this paper.
The Danish case study suggests that new technological pathways are created initially in
niches. In the case of wind power, these consisted of rural areas where the dominant urban,
centralized and grid-connected electricity generation and supply system did not exist or
were markets created by government fiscal policies that protected the new technology
from “normal” market competition. New niches were also created by changes at the
level of the overall technological landscape. They included rapid changes in the relative
competitiveness of fossil fuels as a result of war and revolution in the Middle East and
climate change.
The knowledgeable agents of new path creation included individual inventors and
innovators. Sometimes, their inventions were not taken up at all and at others it required
considerable periods of time for them to be developed into innovations. The diffusion
agents for these innovations were either private production companies or developers
exploiting niche conditions.
770 J. Simmie
Downloaded by [James Simmie] at 04:25 26 April 2012
The creation of the new technological pathway in wind power was not marked by
radical breakthrough innovations. Indeed, in the US, for example, government attempts
to sponsor radical innovation by contracting different types of firms and organizations
to attempt to leapfrog existing levels of technological development proved unsuccessful.
Instead, new pathways were created by incremental innovations that either added the new
technology to existing ways of generating electricity or recombined older technologies
in new ways.
The path-dependent barriers to new path creation were formidable. In some cases,
they were consciously challenged by innovators. This was the case where the prevailing
technological paradigm was deliberately rejected either to bring electricity to rural areas
or to demonstrate a viable alternative to nuclear power. In other instances, the power of
the social groups supporting existing technological regimes was weakened by landscape
developments such as rapid rises in oil prices, climate change and nuclear disasters.
As a result of these interactions, the technological landscape for electricity generation
has changed slowly over a period of half a century or more. The process of new path cre-
ation and the use of wind power have been largely incremental. Nevertheless, the EU as a
whole now has a target of 20% of energy generation to be provided by renewables by 2020.
The majority of this will be provided by wind power. Thus, this source of electricity has
now diffused to the point at which it can be described as a new technological pathway.
References
Arthur, W. B. D. (1989) Competing technologies, increasing returns, and lock-in by historical events, Economic
Journal, 99 (394), pp. 116–131.
Arthur, W. B. D. (1990) Positive feedbacks in the economy, Scientific American, 262 (February), pp. 92–99.
Arthur, W. B. D. (1992) Learning and adaptation in the economy. Santa Fe Institute Working Paper 92-07-038,
Stanford University, Stanford.
Arthur, W. B. D. (1994) Path dependence, self-reinforcement and human learning, in: W. B. D. Arthur (Ed.)
Increasing Returns and Path Dependence in the Economy, pp. 133–158 (Michigan, MI: Michigan University
Press).
Arthur, W. B., Durlauf, S. N. & Lane, David A. (Eds) (1997) The Economy as a Complex Evolving System II
(Reading, MA: Perseus Books).
Bianchi, P. (1986) Industrial restructuring within an Italian perspective. Working Paper No. 2, Nomisha,
September.
Bijker, W. E., Hughes, T. P. & Pinch, T. J. (1987) The Social Construction of Technological Systems (Cambridge,
MA: MIT Press).
Brown, L. A. (1981) Innovations Diffusion—A New Perspective (London: Methuen).
David, P. A. (1985) Clio and the economics of QWERTY, American Economic Review, 75(2), pp. 332–337.
David, P. A. (1986) Understanding the economics of QWERTY, in: W. N. Parket (Ed.) Economic History and the
Modern Economist, pp. 30–49 (Oxford: Blackwell).
David, P. A. (Ed.) (2005) Path Dependence in Economic Processes: Implications for Policy Analysis in
Dynamical Systems Contexts. The Evolutionary Foundations of Economics (Cambridge: University Press).
Dopfer, K. (Ed.) (2004) The Evolutionary Foundations of Economics (Cambridge: Cambridge University Press).
Dosi, G. (1982) Technological paradigms and technological trajectories: A suggested interpretation, Research
Policy, 11(3), pp. 147–162.
Dosi, G. & Grazzi, M. (2010) On the nature of technological knowledge, procedures, artefacts and production
inputs, Cambridge Journal of Economics, 34(1), pp. 173–184.
Dosi, G., Freeman, C., Nelson, R., Silverberg, G. & Soete, L. (Eds) (1988) Technical Change and Economic
Theory (London: Frances Pinter).
Fagerberg, J. (2003) Schumpeter and the revival of evolutionary economics: An appraisal, Journal of
Evolutionary Economics, 13, pp. 125–159.
Path Dependence and New Technological Path Creation 771
Downloaded by [James Simmie] at 04:25 26 April 2012
Foster, J. (1997) The analytical foundations of evolutionary economics: From biological analogy to economic
self-organisation, Structural Change and Economic Dynamics, 8(4), pp. 427–451.
Garud, R. & Karnøe, P. (2001) Path Dependence and Creation (London: Lawrence Erlbaum Associates).
Garud, R. & Karnøe, P. (2003) Bricolage versus breakthrough: Distributed and embedded agency in technology
entrepreneurship, Research Policy, 32, pp. 277–300.
Garud, R., Kumaraswamy, A. & Karnøe, P. (2010) Path dependency or path creation? Journal of Management
Studies, 47(4), pp. 760–774.
Geels, F. W. (2002) Technological transitions as evolutionary reconfiguration processes: A multi-level perspec-
tive and a case study, Research Policy, 31, pp. 1257–1274.
Geels, F. W. (2004) From sectoral systems of innovation to socio-technical systems: Insights about dynamics and
change from sociology and institutional theory, Research Policy, 33, pp. 897–920.
Gipe, P. (1997) Wind Energy Comes of Age (New York, NY: John Wiley).
Hodgson, G. (1993) Economics and Evolution. Bringing Life Back into Economics (Oxford: Polity Press).
Hoogma, R., Kemp, R., Schot, J. & Truffer, B. (2002) Experimenting for Sustainable Transport. The Approach of
Strategic Niche Management (London: Spon).
Kemp, R., Schot, J. & Hoogma, R. (1998) Regime shifts to sustainability through processes of niche formation: The
approach of strategic niche management, Technology Analysis & Strategic Management, 10(2), pp. 175–195.
Kuhn, T. (1962) The Structure of Scientific Revolutions (Chicago, IL: University of Chicago Press).
Lorenz, H. W. (1993) Nonlinear Dynamical Economics and Chaotic Motion (Berlin: Springer-Verlag).
Markard, J. & Truffer, B. (2006) Innovation processes in large technical systems: Market liberalization as a
driver for radical change? Research Policy, 35, pp. 609–625.
Martin, R. & Sunley, P. (2006) Path dependence and regional economic evolution, Journal of Economic
Geography, 6(4), pp. 395–437.
Metcalfe, S. (1998) Evolutionary Economics and Creative Destruction (London: Routledge).
Metcalfe, J. S. & Foster, J. (Eds) (2004) Evolution and Economic Complexity (Cheltenham: Edward Elgar).
Musgrove, P. (2010) Wind Power (Cambridge: Cambridge Uuniversity Press).
Nelson, R. R. & Winter, S. G. (1977) In search of a useful theory of innovation, Research Policy, 6(1), pp. 36–76.
Nelson, R. R. & Winter, S. G. (1982) An Evolutionary Theory of Economic Change (Cambridge: Belknap Press).
North, D. C. (1990) Institutions, Institutional Change and Economic Performance (Cambridge: Cambridge
University Press).
Perez, C. (2010) Technological revolutions and technological paradigms, Cambridge Journal of Economics,
34(1), pp. 185–202.
Potts, J. (2000) The New Evolutionary Micro-economics: Complexity, Competence and Adaptive Behaviour
(Cheltenham: Edward Elgar).
Rip, A. & Kemp, R. (1998) Technological change, in: S. Rayner & E. L. Malone (Eds) Human Choice and
Climate Change, vol. 2, pp. 327–399 (Columbus, OH: Battelle Press).
Rogers, E. M. (Ed.) (1995) Diffusion of Innovations (New York, NY: Free Press).
Schumpeter, J. A. (1939) Business Cycles: A Theoretical, Historical and Statistical Analysis of the Capitalist
Process (New York, NY: McGraw-Hill).
Schumpeter, J. A. (1942) Capitalism, Socialism and Democracy (New York, NY: McGraw-Hill).
Setterfield, M. (1993) A model of institutional hysteresis, Journal of Economic Issues, 27(3), pp. 755–774.
Setterfield, M. (1995) Historical time and economic theory, Review of Political Economy, 7(1), pp. 1–27.
Setterfield, M. (1997) Rapid Growth and Relative Decline: Modelling Macroeconomic Dynamics with Hysteresis
(London: Macmillan).
Streek, W. & Thelen, K. (2005)
Beyond Continuity: Institutional Change in Advanced Political Economies
(Oxford: Oxford University Press).
Sydow, J., Lerch, F. & Staber, U. (2010) Planning for path dependence? The case of a network in the Berlin–
Brandenburgh optics cluster, Economic Geography, 86(2), pp. 173–195.
Vergne, J.-P. & Durand, R. (2010) The missing link between the empirics of path dependence: Conceptual
clarification, testability issue, and methodological implications, Journal of Management Studies, 47(4),
pp. 736–759.
Witt, U. (Ed.) (1993) Evolutionary Economics. The International Library of Critical Writings in Economics
(London: Edward Elgar).
Witt, U. (1997) “Lock-in vs. critical masses—industrial change under network externalities”, International
Journal of Industrial Organization, 15(6), pp. 753–773.
Witt, U. (2003) The Evolving Economy (Cheltenham: Edward Elgar).
772 J. Simmie
Downloaded by [James Simmie] at 04:25 26 April 2012
