![Evolution of national patent applications in alternative propulsion technologies (source: INPI-OPI, [59]).](https://www.researchgate.net/profile/Vanessa-Oltra/publication/248498051/figure/fig1/AS:791731637321728@1565774940643/Evolution-of-national-patent-applications-in-alternative-propulsion-technologies-source_Q320.jpg)
![Renault and Peugeot patent portfolios (source: [66]).](https://www.researchgate.net/profile/Vanessa-Oltra/publication/248498051/figure/fig2/AS:791731637338114@1565774940675/Renault-and-Peugeot-patent-portfolios-source-66_Q320.jpg)
Content uploaded by Vanessa Oltra
Author content
All content in this area was uploaded by Vanessa Oltra on Aug 14, 2019
Content may be subject to copyright.
Sectoral systems of environmental innovation: An application to the French
automotive industry
Vanessa Oltra ⁎, Maïder Saint Jean
GREThA (UMR CNRS 5113), University Montesquieu Bordeaux IV, Avenue Léon Duguit F-33608 Pessac, France
article info abstract
Article history:
Received 9 December 2007
Received in revised form 18 March 2008
Accepted 18 March 2008
Available online xxxx
This article seeks to show how a sectoral system approach may contribute to the analysis of the
determinants of environmental innovations. By using Malerba's [F. Malerba: Sectoral systems of
innovation and production, Res. Policy Vol. 102, 845-859, (2002)] concept of sectoral system of
innovation and production, we develop a sectoral framework based on three building blocks:
technological regimes, demand conditions and environmental and innovation policy. Within this
framework, the sectoral patterns of environmental innovation result from the interplay between
these three blocks. The conceptual framework is applied to the case of the French automotive
industry, with a specific focus on the development of low emission vehicles. The analysis shows
how technological regime and demand conditions lead to technological inertia, and so to a strong
persistence of the dominant design.Finally, environmentaland innovativepolicy are considered in
an integrated way, so that we can study how they influence technological regime and demand
conditions, and in the meantime how they are conditioned by these two blocks.
© 2008 Elsevier Inc. All rights reserved.
JEL classification:
B52
Q55
Q58
L62
Keywords:
Environmental innovation
Sectoral systems of innovation
Automotive industry
1. Introduction
In a broad sense, environmental innovations can be defined as innovations that consist of new or modified processes, practices, systems
and products which benefit the environment and so contribute to environmental sustainability. The literature on the determinants of
environmental innovation generally focuses on the role of regulation (for example, [1–3]). Since the controversial paper by Porter and van
der Linde [4], it is acknowledged that environmental regulation may be a driver of technological change depending on the type of
instruments used and on the context in which they are applied. Nevertheless environmental innovations cannot be considered to be a
simple and systematic response to regulatory pressure. Apart from the regulatory stimulus, many other factors, such as knowledge bases,
technological opportunities and appropriability conditions, as well as demand conditions, influence the technological responses of firms.
These various determinants can be integrated into a sectoral approach of environmental innovations.
Indeed sectors provide a key level of analysis for economists in the examination of innovative and production activities. A huge
empirical literature on sectoral case studies provides a rich (and heterogeneous) set of empirical evidence on the features of
sectors, on their technologies, production, innovation and demand conditions. In order to provide a multidimensional, integrated
and dynamic view of sectors, Malerba [5] develops the concept of “sectoral system of innovation and production”, which is defined
as “a set of new and established products for specific uses and the set of agents carrying out market and non-market interactions
for the creation, production and sales of those products”([5], p. 250). This notion, which draws from basic concepts of evolutionary
Technological Forecasting & Social Change xxx (2008) xxx–xxx
⁎Corresponding author.
E-mail addresses: vanessa.oltra@u-bordeaux4.fr (V. Oltra), saintjea@u-bordeaux4.fr (M. Saint Jean).
TFS-17063; No of Pages 17
0040-1625/$ –see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.techfore.2008.03.025
Contents lists available at ScienceDirect
Technological Forecasting & Social Change
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
theory and from innovation system approaches focuses on the processes of transformation of the system.
1
It provides a useful tool
for a descriptive analysis of the structure and boundaries of sectors, for the identification of the factors affecting innovation and
competitiveness of firms, and for the development of new public policy indications.
This paper seeks to show how such a sectoral framework can contribute to the analysis of environmental innovations by
integrating their various determinants on the supply, technological and demand sides. This may help to understand the
microeconomic dynamics of environmental innovation and to draw policy implications on the type of instrument to be used
according to the stages of the innovation process and the type of targeted industrial sector. We propose to apply our conceptual
framework to the case of the French automotive industry, with a specific focus on the development of low emission vehicles. The
recent development of hybrid and fuel cell vehicles illustrates the efforts completed by the automotive industry to comply with
environmental regulations. Up to now, technological change in the automotive industry is anchored in the internal combustion
engine paradigm, while forced to explore alternative technological paths to meet long run environmental objectives. This trade-off
between the exploitation of the dominant design and the exploration of alternative engine technologies shapes the environmental
innovative activities of automotive firms.
The organization of this article is as follows. Section 1 presents the framework of analysis and discusses the role of the three
building blocks: technological regimes, market demand conditions and environmental and innovation policy. In Section 2, the
framework is applied to the case of the French automotive industry, illustrating how the various technological trajectories followed
by the automotive industry are driven by the interplay between technological regime characteristics, demand constraints and
policy instruments.
2. A sectoral analysis of environmental innovations: framework of analysis
Our analysis of environmental innovations is based on an adapted version of Malerba's “sectoral system of innovation and
production”, in particular with a more restricted conception of institutions limited to the instruments of environmental and
innovation policy. Such a sectoral framework enables to thoroughly analyse the process of environmental innovation and its
determinants. In this section, we present the three building blocks of our framework: technological regimes (Section 2.1), market
demand conditions (Section 2.2) and environmental and innovation policy (Section 2.3).
2.1. Technological regimes
Research on technological regimes can shed light on the microeconomic dynamics of environmental innovation. The literature
on environmental innovations mainly focuses on the role of regulation as a stimulus for technological innovations (cf. Section 2.3),
but not much attention is paid on the innovation process itself and on its features and determinants at the industryand firm levels.
A better understanding of the knowledge bases and of the learning processes that underlie the development of environmental
innovations may bring new insights on the sources of innovation and on the directions of the resulting technological trajectories.
The use of the evolutionary literature
2
on technological regimes enables to put the study of environmental innovations in an
industrial dynamics perspective. Such a perspective allows for an integrated and dynamic analysis of environmental performances
and competitiveness of firms. So the innovative strategy of firms, and in particular their way of combining competitiveness
objectives with environmental criteria,
3
as well as path dependency and technological lock-in phenomena, can be better grasped.
The literature on technological regimes provides a useful framework for empirical analyses of the microeconomic dynamics of
innovation. The concept of technological regime, initially developed by Nelson and Winter [9], corresponds to a description of the
technological environment in which industrial firms operate. It identifies the properties of learning processes, sources of
knowledge and nature of knowledge bases that are associated with the innovation processes of firms active in distinct sets of
production activities [10]. Malerba and Orsenigo [11] define a technological regime as the combination of four factors: knowledge
bases, technological opportunities, appropriability conditions and cumulativeness of innovation. They empirically study how these
factors shape the innovative patterns of firms and so the properties of the industrial dynamics. Using patent data the authors show
that the patterns of innovative activities differ systematically across technological fields, while remarkable similarities emerge
across countries for each technological field. These results strongly suggest that “technological imperatives”, and so the
characteristics of technological regimes, play a major role in determining the patterns of innovative activities.
In the Schumpeterian tradition, the authors distinguish between two types of technological regime:
4
the “entrepreneurial
regime”is characterized by an innovative base which is continuously enlarging through the entry of new innovators and the
erosion of the competitive and technological advantages of the established firms in the industry; and the “routinised regime”is
based on the dominance of a few established firms which are continuously innovative through the accumulation over time of
technological capabilities. These patterns of innovation are seen as the results of specific combinations of technological regime
conditions. In Malerba and Orsenigo [11,13] the ‘motor vehicles industry’typically corresponds to a “routinised regime”
1
“It departs from the traditional concept of sector used in industrial economics because it examines other agents in addition to firms, it places a lot of emphasis
on non-market as well as on market interactions, and focuses on the processes of the transformation of the system it does not consider sectoral boundaries as
given and static”([5], p. 250).
2
For a general discussion on the contributions of the evolutionary theory to environmental economics, see Van den Bergh [6] and [7].
3
For a discussion and a modelling work on the combination of environmental criteria with productive efficiency and quality of products in the innovative
strategy of firms, see for example [8].
4
This distinction has been introduced in [12].
2V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
characterized by high appropriability and cumulativeness conditions which allow innovators to accumulate technological
knowledge and to build up innovative advantages over potential entrants. This progressively creates a technological leadership
from the established firms which explains the concentration of innovation, the stability in the hierarchy of innovators and the low
rate of entry, typically observed in this type of technological regime.
By extending the previous taxonomic exercises ([13,14]), and by focusing more on the role of technological barriers to entry,
Marsili [15] proposes a new typology of regimes which distinguishes five industrial technological regimes: science-based,
fundamental processes, complex systems, product engineering and continuous processes. This typology provides a more detailed
framework especially regarding the characteristics of the knowledge bases and the sources of barriers to entry. Each regime is
defined by a specific combination of technological opportunities, technological entry barriers, inter-firm diversity in the rate and
directions of innovation, diversification of the knowledge base, external sources of knowledge, links with academic research and
nature of innovation. By collecting empirical data, Marsili [15] identifies and characterizes the industries composing each regime.
Within this framework, the ‘motor vehicles industry’corresponds to a ‘complex (knowledge) system regime’. The detailed
description of this regime gives new insights on the patterns of innovative activities in the automotive industry. According to
[15,16], the complex (knowledge) system regime is characterized by medium-high levels of technological opportunity, high entry
barriers in knowledge and scale, and high persistence of innovation. The distinctive feature of this regime is the high degree of
differentiation of the knowledge base of firms, especially in upstream technologies, but also the role of external sources of
knowledge. The author emphasizes that, in a complex system regime, firms are active in a wide range of technical fields along
similar search trajectories, but with a certain variety in their ability to exploit technological opportunities strongly related to R&D
activities. This pattern of knowledge diversification is an important feature of the technological regime of the ‘motor vehicle
industry’which may contribute to explain the innovative strategy of automotive firms. Marsili also argues that the complexity of
the knowledge base is the main source of technological entry barrier. Even if suppliers represent important sources of external
knowledge, and therefore potential sources of innovative entry, their contribution has to be integrated within a complex system of
external sources in which other actors, such as public institutions, users and competitors are of considerable relevance. As a
consequence, although suppliers belonging to other technological regimes and sectors –particularly to the mechanical and
electrical-electronic area –do acquire competencies in transportation technologies, they are not likely to enter the transportation
sector. This complex set of relationships between suppliers and producers in the production of knowledge and technology should
be taken into account when analysing the environmental innovations developed by the automotive industry.
2.2. Market demand conditions
A large proportion of work on technological change and innovation is concentrated on the “supply side”dynamics. In a
Schumpeterian perspective, evolutionary economics tends to assign a passive role to demand, and so to consumers in the
innovation process. Neither the debates about technology push and demand pull effects, nor did the treatment of demand by
Schmookler [17] develop a real account of demand conditions and consumers' preferences. In the 1980s, some scholars insist on
the role of the interactions between users and suppliers as a key element in the innovation process ([18,19]). In the same period, a
considerable body of research focuses attention on the role of demand conditions and heterogeneity of consumers' preferences in
the process of technology diffusion and adoption.
5
This research area can be split up into two types of literature: on the one hand,
research on technological competition and network externalities, on the other hand, industry life cycle analyses.
Research on network externalities considers contemporaneous competitions between rival variants of the same technology
([23–26]). Arthur [23] identifies five sources of increasing returns to adoption which may lead to a monopoly of one technology,
and so to a technological lock-in: learning by using, network externalities, scale economies in production, informational increasing
returns and technological interrelatedness. This latter source of increasing returns to adoption is linked to the development of sub-
technologies and infrastructure which go together with the adoption of a given technology. It tends to favour technological lock-in
in the sense that the other technologies, if less adopted, may lack the requisite infrastructure or may requirea partial dismantling of
the more widespread technology's infrastructure [23]. Arthur shows that, while unbounded increasing returns to adoption lead to
an inevitable monopoly by a single technology, bounded returns to adoption and heterogeneity of adopters may lead to a shared
market.
In the industry life cycle literature, the focus is on the evolution of industry structure and on the emergence of a dominant
design: through time the technology stabilizes, and so does the industry structure, which generally becomes quite concentrated
(see for example [27–29]). Once the dominant design has emerged, an era of incremental change takes place in which
organizations focus on incremental improvements of the dominant design. More recently, there have been versions of this theory
that stress demand side dynamics. A dominant design emerges and industry stabilization occurs not so much because a particular
satisfactory technological pattern is found, but rather because there are network economies due to bandwagon effects on the
demand side [30].
One of the most influential expressions of the role of demand in technology competition has been Christensen's [31]
examination of disruptive technologies. Disruptive technologies are new technologies that introduce a different performance
5
According to Metcalfe [20], adoption analysis considers the decision taken by agents to incorporate a new technology into their activities, while diffusion
analysis is concerned with how the economic significance of a new technology changes over time. Moreover it is important to precise that in the evolutionary
literature the diffusion process is characterized by post-innovative improvements of the diffusing technology. These post-innovative improvements play a vital
role in increasing the rate of diffusion within existing applications and in extending the technology to new applications [21,22].
3V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
package from mainstream technologies and are inferior to mainstream technologies along the dimensions of performance that are
the most important to mainstream consumers. Technology disruption occurs when, despite its inferior performance on focal
attributes, the new technology displaces the mainstream technology from the mainstream market [32]. Christensen introduces the
idea of ‘performance oversupply’to explain thatmainstream consumers adopt the disruptive technology in spite of the superiority
of the incumbent technology. The principles of performance oversupply state that, once consumers' requirements for a specific
functional attribute are met, evaluation shifts to place greater emphasis on attributes that were initially considered secondary.
Adner [32] further develops this demand-based view by formally modelling the role of the demand environment in shaping
competitive dynamics. By building a model characterizing the utility functions of different market segments, Adner [32] explores
the influence of the demand structure which is defined in terms of heterogeneity in consumers' requirements and preferences. The
author shows that the essential aspect of technology disruption is consumers' decreasing marginal utility (when performance
improvements go beyond their requirements), leading to a decreasing willingness to pay for improvements. This argument
complements the notion of performance oversupply, by suggesting that technology disruption is likely to occur when consumers
are willing to accept a worse price/performance ratio because the absolute price of the new option is sufficiently low. So the price at
which the new technology or product is offered becomes critical to a disruptive outcome. Within the same line of inquiry, Malerba
and al. [30] show that the successful introduction of a radically new technology in an industry, where a dominant design and a
small collection of dominant firms have emerged, may be dependent upon the presence of a group of experimental users, or of a
niche of consumers, willing to experiment the new inferior product or technology.
The various contributions on technology competition depart onewith each other regarding their respective assumptions on the
characteristics of the competing technologies. Do the competing technologies offer exactly the same characteristics in the sense of
Lancaster [33]? Windrum and Birchenhall [34] use the distinction between two cases of displacementof an established technology
by a new one: technological substitution and technological succession. In the latter case, the new technology offers one or more
new service characteristics
6
that were previously unavailable when using the old technology. Using a simulation model, the
authors show that successions are more likely to occur when the gain in direct utility of consumers from the new technology is
high. They also identify a “sailing ship effect”in the sense that the entrance of new firms competing on a new technology
stimulates old technology firms to innovate in order to improve the quality of their products.
7
Consequently, the probability of a
succession depends on the relative rates at which new and old technology firms successfully innovate.
These findings will be of great relevance when studying environmental innovations and their diffusion. In the literature on the
determinants of environmental innovations, it is generally assumed that market forces alone would provide insufficient innovation
incentives and thatconsumers' willingness to pay for environmental improvements will be too low[2]. Nevertheless, severalempirical
studies tryto identify and to evaluatethe incentive effects linked to environmental pressure coming from consumers.
8
Research works
on non-regulatory pressures often cite consumers as playing a role in the environmental performances of firms, with particular
emphasis on the emergence of “green consumers.”
9
But, even if the ecological concerns of consumers and the expected increase in
future demand for environmental products is assumed to trigger environmental product innovations, a formal analysis of the demand
structure and of the prevailing purchase criteria is still lacking.
10
The same criticism can be made on the analyses of environmental
technology diffusion.
11
A critical issue lies in the way consumers do value and take into account the environmental characteristics of
products. This evaluation strongly determines the adoption and the diffusion of environmental product innovations.
2.3. The role of environmental and innovation policy
Since the controversial paper by Porter and van der Linde showing that ‘properly designed environmental standards can trigger
innovation that may partially or more than offset the costs of complying with them’([4], p.98), the debate has been strong on
whether environmental regulations act as a constraintor a stimulus to technological innovation, and thus to economic growth and
competitiveness. The literature on environmental innovations agrees that regulation may be a driver of technological change
depending on the type of instruments used (command and control versus market-based instruments) and on the context in which
they are applied. According to Jaffe and al. [45], economic instruments (taxes and tradable permits) tend to be more cost effective
than regulation and provide ongoing incentives for firms to adopt new technology.
12
However empirical works have shown that
regulatory design is a key factor that may influence firms' innovative response, in particular when considering its stringency,
flexibility and limiting uncertainty. Stringency relates both to the absolute reduction in environmental impacts and to the fact that
compliance using an existing technology is either not possible or costly. Therefore by being technologically challenging, stringent
regulations may provide a spur for innovation. Ashford and al. [46] argue that, while stringency is the most important factor for
6
For a presentation of the concepts of technical and service characteristics of technology and product, see for example [35].
7
Significant improvements are often induced in technologies under competitive threat from the new technology, so that the diffusion curve is shaped by the
evolving pattern of competitive advantage between rival technologies —a phenomenon which is called “sailing ship effect”[36].
8
For example, [37,38] and more recently [3] and [39].
9
See for example, [40,41] and [42].
10
We can quote as an exception [43].
11
For a survey on the diffusion of environmental technologies, see Kemp [44] and [1].
12
In theoretical models, the superiority of economic instruments is generally derived under the assumption that environmental innovations are developed by
the polluting firms and not sold to other firms and that there are no information problems. “A clear weakness of the innovation models is that they do not
consider the policy-making process in the context of technological uncertainty and do not distinguish between different technologies for achieving
environmental improvements”([44], page 314).
4V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
eliciting an innovative response, flexibility towards the means of compliance, variation in the requirements imposed on different
sectors and compliance time periods contribute to stimulate alternative technologies.
In spite of the incentive effects and the increasing stringency of regulation, we can observe that the technological progress
induced by environmental regulation is most often incremental. In this respect, [47] and [48] show that environmental policy
instruments rarely lead to radical innovation,
13
but rather support incremental innovation and technological diffusion. In sum, it
seems that there is no single best instrument to foster environmental innovation and that the most common response of firms to
environmental policy is incremental innovation and adoption of technologies with short-term paybacks. Market-based
instruments and standards are not substitutes and not sufficient to induce innovation so that other policy instruments are needed.
Industry covenants (or voluntary agreements) in which a group of firms belonging to a particular industry makes a voluntary
commitment to achieve environmental goals can give more flexibility to firms.
14
But the effectiveness of such voluntary
agreements is conditioned by the existence of credible threats (product boycotts, stricter regulation in the future) and by additional
incentives regarding monitoring, penalties and inspections. More demand-oriented measures also need to be considered in order
to shape market conditions in favour of alternative technologies. Information and knowledge-sharing measures such as eco-labels
and environmental management systems can help better signalling the environmental performances of products and processes. In
addition, public procurement can play a major role by creating niche markets for environmental technologies and by allowing
feedbacks between experimental users and the emerging technology producers. These demand-oriented measures may shape
market demand conditions and so the diffusion of environmental innovation (cf. Section 2.2).
This whole set of instruments defines an environmental policy mix, the purpose of which is to promote more sustainable
systems of production and consumption. This policy mix determines the incentives to innovate and the directions of research by
defining some environmental priorities and some objectives in terms of emissions, quality of inputs or products,but also by shaping
market conditions and interactions among actors. Properly designed regulation can thus strengthen both technology push and
market pull effects. Nevertheless, environmental innovations cannot be considered to be a simple and systematic response to
regulatory pressure, since many other factors may influence the technological responses of firms. Anyway, given the important
number of implemented instruments, it may be rather tenuous to isolate the effect of each instrument. Moreover, the impact of a
given policy instrument depends on the wayit is formulated and implemented and on the context of application. It is the reason why
the same instrument may exert different impacts oninnovation according to industrial sectors. The notion of technologicalregime is
particularly relevant here. Depending on the nature of knowledge bases and on the conditions of technological opportunity and
appropriability, firms may experience more favourable conditions for environmental innovations. As a consequence, the impact of
environmental policy instruments upon innovation is also dependent on the features of technological regimes.
15
In order to give more support to environmental innovations, particularly to radical innovations, many works emphasize that
environmental policy needs to be coordinated with research and innovation policy.
16
As a matter of fact, the usual market failure
argument justifying innovation support policy is all the more relevant for environmental innovations that they cause what is called a
“doubleexternality”problem. Environmental innovations produce two typesof positive externalities:usual knowledge externalities in
the innovation phase and externalities in the diffusion phase due to the positive effect upon environment [2]. In other words, the
beneficial environmental impact of these innovations makes their diffusion always socially desirable. It creates a second source of
market failure, which may cause a lack of private incentives leading firms to under-invest in environmental R&D and innovation.
Systemic approaches provide an alternative to the concept of the optimizing policy maker that characterizes the standard
market failure.
17
Systemic approaches focus on the locus of intersection of numerous networks generating knowledge and on the
sources of “system failures”which impede the learning and innovative capacities of a system. These system failures mainly come
from a lack of coordination and complementarity between the research activities of agents, a lack of appropriate institutions
allowing a collective creation and diffusion of knowledge and bad adjustments between the evolution of institutions and
technologies. It may create knowledge processing failures and selection failures leading to inadequate selection processes, in the
sense that they eliminate too rapidly or maintain too long certain technological options, practices or firms [56]. So the negative
consequences of these different types of failures are principally technological lock-in phenomena, lack of diversity in the system,
difficulty in creating new technological paradigms and in warranting the transition from the old to the new paradigm. Various
innovation policy instruments, such as R&D subsidies, diffusion of information, public procurement and cooperative research
programme may be used to correct these failures and to provide favourable conditions to knowledge creation and innovation.
2.4. Overview of the framework
Sectoral system approaches provide a framework for an integrated and dynamic analysis of environmental and innovation
policy. By acknowledging that environmental innovations,like innovation in general, result from a dynamic and interactive process
13
The term ‘radical innovation’is used in the Freeman's sense that is “innovations that are usually the result of deliberate R&D activities and require new
knowledge and practices which may call into question the prevailing technology”[49].
14
For a survey on voluntary agreements, see [50]. See also [51] for a discussion on voluntary agreements as an instrument for escaping “techno-institutional
lock-in”.
15
This is linked to the technological determinism argument put forward by Ashford [52]:“There is a kind of technological determinism that influences not only
what can be done, but also what will be done…Technological developments have their own dynamics and constraints that determine the direction of change
even when stimulated by external forces”.
16
See for example [53] and [54].
17
For a detailed discussion see [55].
5V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
of knowledge creation and diffusion, the emphasis is put on the coevolution between the various elements of the system. Within
this perspective, institutions, technology and industry structure develop in a co-evolutionary relationship in which policies are
adjusted to emerging technological paradigms and paradigms are adjusted to policies [57]. Using such an evolutionary framework,
we propose to analyse the sectoral patterns of environmental innovation as the result of the interplay between technological
regimes, market demand conditions and environmental and innovation policy.
18
Our framework of analysis is depicted in Fig. 1.
3. An application to the French automotive industry: the case of low emission vehicles (LEVs)
3.1. Environmental innovations in LEVs: exploitation of the dominant design versus exploration of alternative technologies
Many empirical works on the automotive industry
19
emphasize an increasing trend in environmental concerns and
innovations. In France, INPI data ([59]) show that 40% of the automotive patent applications are linked to environmental
objectives.
20
Over the period [2000–2006] we can observe an increasing trend in the patent applications linked to environmental
performances of vehicles and to propulsion technologies, with a leading role played by the two French car manufacturers Renault
and PSA Peugeot Citroen. These firms are the two top firms in terms of French patent applications, while Valeo (one of the main
first tier suppliers) ranks fourth [59]. This feature illustrates the very concentrated and cumulative innovative process that typically
characterizes a “routinized”technological regime (cf. Section 2.1).
Fig. 1. Framework for the analysis of sectoral patterns of environmental innovation.
18
In order to be exhaustive, a wide range of other policies should be considered, in particular industrial policy, energy policy and competition policy. But in this
article, we focus on the policies directly linked to environmental innovation, that are environmental and innovation policy.
19
For a survey on the European Automotive industry, see [58].
20
In this section, patent data are used as an indicator of environmental innovative activities. In spite of the well known limits linked to patent data [60], patents
provide a good indicator since the propensity to patent of automotive firms is high. For a survey on the use of patent data as an indicator of environmental
innovations, see [61].
6V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
The environmental research and innovative activities carried out by the automotive industry are driven by multiple objectives. The
main objectives are the decrease in primary polluting emissions (carbon monoxide, nitrogen oxide, particle matters, sulphur
dioxide and volatile organic compounds), the reduction in greenhouse gases (especially CO2 emissions) and the decrease in fuel
consumption. Given the difficulty in combining these objectives, some trade-offs are necessary. Within this context, research and
innovative activities of automotive firms follow simultaneously different technological trajectories looking for a technological
compromise to produce cleaner, quieter and more efficient energy vehicles. Innovations in low emission vehicles (LEVs)
correspond to all the innovations aiming at decreasing global and/or local polluting emissions. These innovations may concern
different parts of a vehicle. But the core of the research and innovative activities in LEVs carried out by automotive firms' concerns
vehicle propulsion systems, with a focus on engine technologies.
In that field, two technological paths are usually distinguished: the continuous improvement of conventional engine
technologies (internal combustion engines) and the development of alternative engine technologies (electric battery vehicles and
fuel cell vehicles).
21
The automotive industry is characterized by a strong and persistent dominant design, which is the internal
combustion vehicle (ICEV). During last decades, the global environmental performances of ICEVs have been significantly improved
thanks to innovations in direct injection technologies, combustion concepts and particle filters. A cluster of innovations has been
developed which has enabled to decrease fuel consumption, polluting emissions and noise rate, and to increase energy efficiency of
engines. This is particularly true for diesel engine vehicles, which are far more powerful, refined and quiet, and so offer an attractive
solution for reducing CO2 emissions and fuel consumption.
Concerning alternative engine technologies, the focus is on electric battery vehicles (EBVs) and fuel cell vehicles (FCVs). In terms
of polluting emissions, EBVs are very efficient since they do not emit any emission during use. But their main disadvantage is linked
to the storage capacity of batteries and soto the limited rangeof use of vehicles. For these reasons, EBVs are limited to certain niche
markets and are not considered as an alternative to ICEVs anymore.
22
In a long-term perspective, the most promising option for a
technology breakthrough is FCVs.
23
A FCV is defined as a vehicle driven by an electric engine, which is powered by a fuel cell. When
the fuel is hydrogen, the system does not generate any pollutants. But the fuelling technology, and more precisely the production
and storage of hydrogen, is the major problem of FCVs. Moreover, the incompatibility with existing infrastructure and the relatively
high costs are major obstacles to the development of FCVs.
The case of hybrid vehicles (HVs) is generally considered as a transition technology between ICEVs and FCVs.
24
Given present
limitations in battery and fuel cell technologies, the most viable powertrain alternatives are hybrid configurations that include a
relatively small internal combustion engine and an electric motor. One of the advantages of this technology is that it is competitive
with ICEVs, in terms of range of use and speed, with a total efficiency, which can be twice as high as the efficiency of the
combustion engine. Moreover, they are compatible with both the available fuel infrastructure and the current ICE system. Hybrid
technologies cover a multitude of possible propulsion architectures, thus providing an attractive technological compromise for
LEVs. Given the number of environmental criteria to be considered, there is no one best solution. Each LEV technology presents
some advantages in terms of environmental performances and/or performances of the vehicle (efficiency, price and range of use),
but no technology is better on all the criteria. As an illustration, Fig. A.1 in Appendix A presents the performances of different LEV
technologies in terms of fuel consumption and CO2 emissions. It shows that the hydrogen fuelled FCV and the diesel-hybrid vehicle
exhibit the largest potential compared to conventional ICEVs. But the performances of FCVs fall down as soon as one puts an on-
board reformer. We also see that for fuel consumption the best potential for reduction is obviously FCVs. The best combination
between fuel consumption and carbon dioxide emissions is obtained by diesel-hybrid vehicles.
The high performances of advanced diesel and of hybrid diesel vehicles certainly explain the strong persistence of conventional
engine technologies. The dominant design seems to be far from being played out. This feature is particularly significant when we
consider patent applications of the French automotive industry. The INPI (2006) study shows that the most part of the patented
inventions of the French automotive industry comes to conventional technologies and to the integration of electronics in these
technologies (cf. Fig. A.2 in Appendix A). Among these patents, 40% are dedicated to improvements in air admission, 30% to
combustion and 30% to exhaust pipe. Fig. A.2 illustrates that the dominant design is still continuouslyunder improvements. Thanks
to the innovative activities of car manufacturers, but also of their component suppliers, a steady rate of technological progress
continues to make ICEVs gradually better and cleaner. ICEVs still represent the core of the innovations carried out by the French
automotive industry. This feature strengthens the argument according to which the car industry demonstrates a preference for
incremental technological change. According to most car manufacturers, particularly the French ones, adopting clusters of
incremental innovations enables to meet environmental performance requirements in the most cost effective manner. To gradually
improve the environmental performances of the dominant design is a way of complying with regulation while going on exploiting
the increasing returns linked to the adoption of the ICEV technology. As a matter of fact, there are many sources of increasing
returns in terms of scale economies in production, learning by using, network externalities and technological interrelatedness
linked to the development of fuelling and transport infrastructure. This latter source of increasing returns is an essential source of
technological lock-in since the development of new fuel distribution and retailing infrastructure involve huge costs. The strong
persistence of the dominant design can also be linked to the complexity of the product and of the knowledge base characterizing
the automotive technological regime, thus favouring technological inertia [58].
21
For a detailed discussion of the competing technologies, see [62].
22
For a detailed analysis of the case of EBVs, see for example [63].
23
For a discussion on FCVs, see [62] and [64].
24
For a complete overview on hybrid vehicles, see [65].
7V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
In spite of the persistence of the dominant design, we can also observe that French car manufacturers have diversified their
technological portfolios. Fig. 2 shows the evolution of French patent applications of the automotive industry in electric and hybrid
propulsions, in fuel cells in general and in fuel cells applied to cars.
Over the period [2000–2004], we can observe that the number of patent applications has significantly increased in the three
considered items: by 80% for electric and hybrid propulsions, by 150% for fuel cells and by 350% for fuel cells car applications. In
spite of the slight decreasing trend since 2004, these figures show that French automotive firms have started to develop innovative
activities in alternative propulsion technologies. This feature can be better illustrated by the evolution of the patent portfolios of
the two main French car manufacturers. Fig. 3 presents the evolution of Renault and PSA Peugeot Citroen patent portfolios over the
period [1990–2005]. Each portfolio is divided into five technological items: electric vehicles (EVB), hybrid vehicles (HV), fuel cell
Fig. 2. Evolution of national patent applications in alternative propulsion technologies (source: INPI-OPI, [59]).
Fig. 3. Renault and Peugeot patent portfolios (source: [66]).
8V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
vehicles (FCV), diesel engine (DE) and internal combustion engine vehicles (ICEV). By using the European Patent Office database,
we applied the keywords method to count the cumulated number of patents filed worldwide
25
for each technological item.
Fig. 3 confirms the argument on the persistence of the dominant design since it shows that the patent portfolio of both firms is
mainly composed of patents linked to ICEV and to diesel engines. The argument is even stronger for PSA since it exhibits in 2005 a
share of 70% of its patent portfolio linked to the dominant design with a strong specialisation on diesel technologies. Nevertheless,
the evolution of patent portfolios also illustrates the on-going technological diversification of car manufacturers, which are more or
less involved in the five competing technologies. Such diversification corroborates the results of Frenken and al.[67], which show
that the evolution of LEVs technologies is characterized by an explorative stage in which firms increasingly widen their patent
portfolios. It also illustrates the pattern of knowledge diversification that characterizes the automotive technological regime (cf.
Section 2.1). In the field of LEVs, firms are active in a range of technical fields along similar search trajectories, but with a certain
variety in their ability to exploit technological opportunities strongly related to their innovative strategy and R&D activities. In the
case of French car manufacturers, patent data show a continuous exploitation of the dominant design, with a specialisation on
diesel technologies, combined with an exploration of alternative propulsion technologies increasingly focused on hybrid
technologies.
3.2. The role of demand in LEVs competition: towards a market segmentation?
The development of LEVs corresponds to a typical case of technological competition between an established technology, i.e. a
dominant design, and a set of alternative technologies. As discussed in Section 2.2, increasing returns to adoption may favour a
technological lock-in of an inferior technology, but the characteristics of demand and the heterogeneity of consumers also
influence the outcome of technology competition and may lead to a shared market. The purpose of this section is to present the
main characteristics of the automotive market and of consumers' preferences in order to analyse their influence upon the
competitive dynamics.
The Western European automotive market is mature and mainly a replacement market with a limited growth potential in
comparison to emerging markets. This in turn implies that the rate at which consumers renew their vehicles is a key factor
affecting the performance of the market. Using Table A.1 in Appendix A, which presents the main characteristics of the French
automotive market in terms of registrations, sales and households' fleet, we can emphasize four features:
•In France, the automotive fleet is characterized by an increase in the average age of cars (7.9years in 2006) and in the share of
second-hand cars, more particularly in the share of second-hand cars above 5years old. These features tend to slow down the
fleet renewal and, hereby, also slow down road safety improvements and environmental performances. It also implies that the
diffusion of new LEVs is very slow.
•The French automotive market is characterized by a very high share of diesel cars. In 2006, diesel cars represent 71.4% of car
registrations and the share of diesel cars in the fleet is continuously increasing. When we compare among European countries,
we can see that this specialization on diesel technologies is particularly important in France (the share of diesel cars is 51.6% for
EU15 [58]). To a certain extent, the share of diesel cars reveals the specific technological trajectory of countries. As discussed
before, French car manufacturers do play a leading role in the production of diesel technologies and in related innovations.
•On the automotive market, the brands are the primary channels through which customers recognize manufacturers, thus
creating brand effects. In France, brand effects are particularly high since there is a strongaffiliation of French car buyers towards
French brands. Even if the share of foreign brands is increasing, we can observe that in 2006 67.1% of the fleet comes from PSA
and Renault. This brand attachment certainly tends to reinforce market concentration and so the leading role of French car
manufacturers.
•A last feature is the specialization of the French automotive market, in terms of production and sales, on low and medium-low
models. Low and medium-low models represent 78.3% of car sales in 2007 and 77.1% of the total fleet in 2006. This specialization
strongly determines the innovative strategy of car manufacturers, which have to develop technological compromises for LEVs
also compatible with these targeted market segments. Given the resulting constraints on the price and weight of vehicles, this
market positioning does not favour the development of alternative (and costly) engine technologies.
As discussed in Section 2.2, the structure of consumers' preferences shapes the competitive dynamics of environmental
technology. The characteristics of the French automotive market presented in Table A.1 give us some insights on the preferences of
consumers, in particular on their preferences for diesel and down-market cars. Furthermore, purchase criteria of car buyers can
also be used as a proxy of consumers' preferences. A French survey on the purchase criteria of car buyers has been conducted on a
sample of private buyers [68]. According tothis survey, fuel consumption and price are the first criteria to be considered in new car
purchase. The decision process of car purchase is divided into three components: price, use (security and comfort) and identity
(brand and design). Meyer [68] shows that fuel consumption is the most important criterion in the price component of the decision
process and that it is completely dissociated from pollution criteria. The study also shows that pollution and environmental criteria
have a weak contribution to preferences. In summary the preferences of mainstream car buyers mainly rely on price, fuel
consumption and security.
25
For a complete presentation of the methodology and of the results, see [66].
9V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
Finally, it appears that the demand characteristics of the automotive market reveal to be more favourable to incremental
innovations and to a reinforcement of the dominant design than to a technological breakthrough towards radically new LEVs
technology. The slow fleet renewal, the specialization on down market segments, the strong preference for diesel cars, as well as
the high sensitivity to price and to fuel consumption create sources of technological inertia. The focus of demand on price and fuel
consumption reinforces the persistence of the dominant design, since gasoline and diesel cars remain the cheapest options for
most market segments. This also implies that the relative price of each LEVs technology remains the main determinant of the
future evolution of the market. Fig. 4 presents a detailed assessment of vehicles retail prices for different powertrain technologies
[69]. The estimated incremental vehicle retail prices are expressed in percentage relative to a 2010 gasoline PISI vehicle i.e. an ICE
vehicle using the Port Injection Spark Ignition technology. Fig. 4 shows that hybrid and fuel cell vehicles would respectively cost
about 20% to 100% more than ICE-PISI vehicles. This increase in vehicle retail prices represents an obstacle to the adoption and to
the diffusion of alternative powertrain technologies.
Given these market demand conditions, what kind of outcome can we expect from this technological competition between
LEVs? Based on the theoretical background presented in Section 2.2, can we argue that demand conditions are more favourable to
technology disruption, technology succession or to a partitioning of the market between the competing technologies? We can
consider HVs, EBVs and FCVs as disruptive technologies in the sense that they provide higher environmental performances than
ICEVs,
26
while being inferior along the dimensions that are the most important for mainstream users, in particular price but also
the range of use which is, for example, very limited in the case of EBVs. But the conditions for a technology disruption to occur are
not fulfilled since the absolute price associated to these new LEVs technology is higher than the price of the dominant design.
Moreover, one cannot argue that there is a ‘performance oversupply’(in the sense of [31]), since empirical works on purchase
criteria of car buyers do not show any shift of consumers' preferences on secondary criteria such as pollution and environment. In
terms of technology succession [34], even if we consider that new LEVs technologies offer new or more “environmental service
characteristics”than ICEV technology, a technology succession seems unlikely since the quality differential between the new and
the old technology, and consequently the gain in direct utility of consumers from the new technology, is very low. This argument is
strengthened by the existence of a strong “sailing ship effect”since car manufacturers continuously innovate in the dominant
design in order to improve environmental performances of ICEVs (cf. patent data in Section 3.1). Advanced ICEVs, especially
advanced diesel technologies, remain very competitive not only in terms of price and mainstream criteria, but also in terms of
environmental performances.
Globally the characteristics of the competing technologies and demand conditionsare more favourable to the persistence of the
dominant design, and so to the co-existence of the various technological options. It means that the competitive dynamics is likely
to lead to a technological market segmentation in which technological hybridizing may play an increasing role. The structure of
demand, in terms of heterogeneity of consumers and asymmetry of preferences among market segments (personal cars, light
commercials, urban buses, over-the-road trucks.), seems more favourable to such a market segmentation than to a complete shift
towards radically new LEVs technology. These arguments support the forecasts according to which the automotive market will
evolve by 2030 towards an engine mix characterized by a persistently high share of ICEVs and a continuing growth of the share of
advanced diesel engines [70]. Given the share of diesel cars and the specialization of French car manufacturers on this technology,
this forecast seems all the more relevant for France. This engine mix will certainly also be characterized by an increasing share of
hybrid technologies since they provide a way of improving the environmental performances of conventional ICEVs while
minimizing incompatibility effects with the existing infrastructures, market discontinuities and destroying competence effects.
Fig. 4. Estimated incremental vehicle retail price (expressed in percentage relative to a 2010 gasoline PISI vehicle) (source: [69]).
26
Even if this argument can be brought into question for certain hybrid configurations or for FCVS with an on-board reformer [69].
10 V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
Moreover hybrid technologies are sources of flexibility since they cover a multitude of possible propulsion architectures,
27
which
may be adapted to the range of use of vehicles, in particular to urban use.
The type of use, particularly the proportion of urban use, is an important determinant of the overall performances of vehicles.
Consequently, the technological market segmentation will be greatly influenced by the radius of operation of vehicles. According to
Maxton and Wormald [70], the developmentof LEVs is determined by the segmentation of vehicle applications, which is presented
in Table 1.
According to Table 1, passenger cars and light trucks could become hybrid-based, but still using conventional engines, while
urban taxis and transit buses could be candidate for an application of fuel cell technologies. Long-distance trucks and buses will
continue to use diesel engines, whose ‘well-to-wheel’efficiency is hard to beat for most of their operating patterns. According to
this forecast, over 80% of drive trains will still be ICEVs in 2030, with a continuous increase in advanced dieselvehicles. Within this
perspective radically new engine technologies will remain restricted to market niches mostly pulled by public captive fleets. These
captive fleets can play the role of experimental users willing to experiment new technology and to pay a price premium. Several
authors
28
underline the role of market niches in supporting the development of radically new environmental technologies in
transportation, while raising the difficulties to extend these market niches and to reach a critical size. To avoid these captive fleets
being doomed to isolated experiments remains one of the main challenges that need to be addressed by public policy.
3.3. Environmental policy and innovation in the automotive industry
Fig. A.3 in the Appendix A shows the evolution of local polluting emissions from road transport in France over the period [1990–
2006]. We can observe a significant decrease in the levels of local pollutants, particularly in lead, particle matters and sulphur
dioxide, but also in nitrogen oxide and volatile organic compounds. These local pollutants are subject to European regulation
through emission standards that have been initiated in the beginning of the 1990s and which stringency has been progressively
increased. In terms of stringency, we can notice that European regulation has been introduced later and in a less stringent way than
in US and Japan. The difference in stringency is particularly significant for particle matters and nitrogen oxide emissions of diesel
cars (in terms of emission limit values), reflecting the dominant role of car manufacturers in diesel technologies.
29
These
performance standards give flexibility towards the means of compliance, but also towards time frames, which are negotiated with
car manufacturers. In that sense, there is no technology forcing effect, but a continuous incentive to improve the current
technology and to decrease the levels of local pollutants. As argued in Section 2.3, this type of instruments is conducive to strong
incentives to innovate and to significant results in terms of pollution, but without calling into question the dominant design.
Regarding CO2 emissions, the European Commission's strategy was mainly based on voluntary commitments from the car
industry, which promised to gradually improve the fuel efficiency of new vehicles. The 1998 voluntary agreement between ACEA
(European Automobile Manufacturers Association) and the Commission included a commitment by carmakers to achieve a target
of 140 g/km by 2008. Although significant progress was made, average emissions fell only from 186 g/km in 1995 to 161 g/km in
2004. In France, we can observe that the average level of CO2 emissions of new passenger cars has significantly decreased (149 g/
km in 2006), while the global level of CO2 emissions has increased by 20% over the period [1990–2006] (see Fig. A.4 in Appendix A).
Because of this continuous increase in CO2 emissions from road transport, the Commission has decided that a binding regulation
was necessary. An ‘integrated approach’has been proposed where average emissions are to be brought down to 130 g/km by 2012
through vehicle-technology improvements and a 10 g/km supplementary cuts are to be achieved by complementary measures,
such as the further use of biofuels, fuel-efficient tyres and air conditioning, traffic and road-safety management and changes in
driver behaviour (eco-driving). Such an integrated approach is strongly supported by car manufacturers, which call for a
27
Hybrid technologies provide a multitude of propulsion technologies according to the percentage of electric energy source. There is a range of hybrid types
from a pure ICE driveline with a bit of additional power and regenerative braking capacity via a special starter-alternator (as the one developed by Valeo), to a
pure battery electric with a small ICE.
28
See for example [71].
29
Euro 5 standards (applied in 2009) will aim at optimising diesel engines by reducing from 80% their particles emission and by generalizing the adoptionof
particle filters.
Table 1
Potential applications of new driveline technologies (Source: [70], page 87)
Urban use proportion Preferred driveline technology
Cars and light trucks 50% Hybrid
Taxis 100% FC
Light commercials 80% Hybrid
Local delivery trucks 80% Hybrid
Over-the-road trucks 10–20% ICE
Urban transit buses 100% FC
Long distance buses 10–20% ICE
11V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
comprehensive strategy involving all relevant stakeholders (car manufacturers, oil suppliers, customers, public authorities, etc.)
and focusing on biofuels, which are considered as the most promising option for the medium term [72].
Obviously emission regulations represent only one piece of a broader regulatory framework that car manufacturers have to take
into account in the design and development of new vehicles. Other environmental problems (end-of-life vehicles, noise, safety,
etc.) generated by cars are subject to legislation. Forexample, the directive on end-of-life vehicles (2000/53/EC) puts the emphasis
on the producer responsibility and has implications on the whole life cycle analysis of a car. Automotive firms criticise that a high
recycling ratio collides with other environmental protection measures, such as fuel consumption. Likewise, other performance
characteristics of cars such as safety or air conditioning devices can come into conflict with environmental objectives. This
multiplicity of environmental objectives and the potential conflicts between the various product performance attributes of cars
raise the necessity for car manufacturers to carry out some trade-offs and to seek for technological compromises. This can nurture
some form of resistance on the part of car firms to go away from the dominant design and calls for a more integrated approach.
As argued in Section 2.3, this environmental policy mix gives incentives to innovate, as well as some orientations in terms of
performance objectives, but is not sufficient to shape the innovation process itself and the resulting technological trajectories. In
transportation, innovation policy is mainly based on large-scale research programmes with public-private partnerships. In France, research
and innovation in transportation are conducted through the PREDIT programme (National Programme of Research and Innovation in
Transportation).
30
PREDIT is an interdepartmental device created in the 1990s which aims at coordinating and developing research and
innovation in the field of transport. The PREDIT 3 programme (2002–2007) is characterised by a focus on energy and environmental issues,
especially greenhouse gas emissions, but also on safety issues. Car manufacturers are strongly involved in these projects, but an increasing
part of them is lead by first tier suppliers. Tab les 2 and 3 summarize the PREDIT 3 programme.
Energy and environmental issues correspond to the priority of PREDIT3 (48% of total funding). Data on the research projects
linked to clean vehicles and funded by PREDITcan give us relevant indications on the directions of research. All the projects consist
in cooperative research projects involving car manufacturers, first tier suppliers, industrial firms coming from other sectors (for
example oil sector), public institutions (for example the French Oil Institute), public laboratories and universities. If we look at the
research orientations of PREDIT 3 (cf. Table 3), we observe that more than 50% of the projects are dedicated to the internal
combustion engine with a major focus on efficiency improvement of conventional engines. By contrast, only 23% of the projects
concern electric and hybrid motors with a strong emphasis on the development of batteries. Noise reduction represents 9% of the
projects and few projects are concerned with the development of auxiliaries like air conditioning devices.
This focuson ICEVs results both from the definition bypolicy makers of the priority research areas, and from the projects submitted
by firms since PREDIT works on a “call for proposals”principle. To a certain extent, this illustrates the coevolution between
technological regime and innovation policy: policy makers define the main objectives and research areas on the basis of the current
technology and research directions, while the research proposals, which mainly come from leading automotive firms, are determined
by the characteristics of the technological regime and the prevailing trajectories. Thistends to create a dynamics inwhich technologies,
institutions and industry structure co-evolve within the dominant technological paradigm. Even if the European regulation on cars
polluting emissions becomes globally more stringent, the multiplicity of objectives, the complexity of the regulatory framework and
the flexibility towards the means of compliance tend to reinforce technological lock-in on the dominant design.
In summary, we can argue that environmental and innovation policies have an influence on the automotive technological
regime, in terms of incentives and R&D support, but in the meantime they are strongly influenced by the technological regime,
which conditions innovative activities. As discussed in Section 2.3, this co-evolutionary relationship may create system failures in
terms of exploitation/exploration trade-off, lack of diversity in the system and transition to new technological paradigms. To be
overcome, these system failures should also be addressed by demand-oriented instruments, which may create more favourable
market demand conditions. The implementation of demand-oriented measures is one of the pillars of the integrated approach of
the automotive regulation advocated by the European Commission and by the European Association of Car Manufacturers [73].
Demand-side measures are needed to shape consumers' perception, to foster demand for clean and safe cars and to encourage
30
This research is also supported by the European Research Framework Programme. The 7th Framework Programme launched in 2007 will allocate €4.1 billion
over the €50.5 billion research budget specifically to transport research activities, such as research on ‘greening’transport and on decongesting transport
corridors. A further €2.26 billion will be allocated to energy research, including research on hydrogen fuel cells and renewable fuel production, such as biofuels.
Table 2
Overview of PREDIT 3 initiatives (source: http://www.predit.prd.fr)
Issues Funding
Energy-environment 48%
Safety 21%
Mobility of people 18%
Transport of goods 13%
Total funding (2002–2006) 290 M€
Which 33 M€on account of the Clean Vehicle Plan (VPE) implemented by the National Environmental Agency (ADEME)
12 V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
changes in consumers' behaviours. The development and the diffusion of LEVs do not only require new technological
competencies, but also new demand conditions that are compatible, in terms of preferences and behaviours, with the new
technological trajectories. In France, different measures taken within the ‘Plan Climat’illustrate this trend: the creation of an
energy label for new cars, a system of bonus-malus at the time of purchase, a greater sensibility to eco-driving during the driving
license and fiscal incentives (tax on registration certificates) to buy low CO2 emissions cars.
4. Conclusion
With its focus on the inducement effect of regulation, the existing literature on environmental innovations neglects the
interdependencies between environmental innovations and industrial dynamics of sectors. As a matter of fact, environmental
innovations are not onlya response to regulatorystimuli, but should also meet competitiveness and demand objectives. This article
proposed to enrich the analysis of the determinants and the patterns of environmental innovations bytaking into account not only
the incentive effects of regulation, but also the role of technological regime and demand characteristics. Within this perspective,
environmental innovations result from the interplay between the three building blocks of sectoral systems of innovation, which
are technological regimes, demand conditions regimes and public policy.
The sectoral system approach developed in this article brings three main contributions to the analysis of environmental
innovations. First, the concept of technological regime enriches the analysis of the micro-economic dynamics of environmental
innovations by taking into account the characteristics of the technological environment at the industry level. It provides a better
understanding of the learning and innovative processes that underlie the development of environmental innovations. Second, by
considering demand conditions as one of the building blocks, sectoral system approaches enable to go deeper in the analysis of
demand side determinants. The role of demand structure in the diffusion of environmental innovations and in technological
competition can be fully grasped. More particularly, a thorough analysis of the impact of consumers' preferences, purchase criteria
and price effects upon the outcome of the competitive dynamics can be developed. Finally, environmental and innovative policy
are considered in an integrated way, so that we can study how they influence the technological regime and demand conditions, and
in the meantime how they are conditioned by these two blocks.
To conclude, a sectoral system approach provides a useful tool in various respects: for a better understanding of the industry
and technological determinants of environmental innovations, for the identification of the factors affecting their competitive
dynamics and for a more integrated and dynamic view of environmental and innovation policy. On this latter point, it may help
policy makers to take into consideration the idiosyncrasies of sectors and to identify mismatches and blocks that impede the
learning and innovative capacities of a system.
Future research on sectoral systems of environmental innovation should explore more thoroughly and more formally the
coevolution of the three building blocks. It may help to go further on the analysisof public policy and to develop proposals on how
to affect the transformation of systems. A comprehensive analysis of co-evolutionary patterns should also cope with the evolution
of the boundaries of sectors, which can be significantly modified by environmental challenges. This issue is particularly relevant in
the case of transport since the transition towards sustainable systems certainly implies to reconsider the place of automobile in
transport systems and to modify the patterns of consumption. Such an evolution may entail significant changes in the competitive
dynamics among means of transport, and consequently in the boundaries of systems as well.
Acknowledgment
This paper has benefited from the insightful comments and suggestions of three anonymous referees. We also thank Andreas
Pyka for his helpful comments on the first draft. The support of the DIME European Network of Excellence is gratefully
acknowledged.
Table 3
Orientation of projects relative to clean vehicles funded in the PREDIT 3 (2002–2007) (source: http://www.predit.prd.fr)
Main area Sub-area Number of projects
Internal combustion engine 79
–Efficiency improvement 38
–Post-treatment of pollution 19
–Alternative fuels 12
Auxiliaries, structure and lightening of vehicles e.g. Air conditioning devices 13
Electric and hybrid engines 35
–Energy storage 15
–Energy management 7
–Development of new vehicles 7
–Power electronics 6
Noise reduction 14
Railway materials and techniques 10
Total 151
13V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
Appendix A
Fig. A.1. Fuel consumption and CO2 emissions for different engine technologies (source: [70], page 85).
Fig. A.2. Evolution of national patent applications in conventional engine technologies (source: INPI-OPI, [59]).
Fig. A.3. Evolution of local pollutants for road transport in France (source: [73]).
14 V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
Fig. A.4. Evolution of CO2 emissions (source: [73]).
Table A.1
Characteristics of the French automotive market (CCFA data, [73])
French car registrations 1995 2000 2005 2006 2007
Number of secondhand cars/number of new registered cars 2.1 2.4 2.6 2.7 –
% of secondhand cars above 5 years old 57 60 60 60 –
% of French brands in total registrations 60.8 59.1 56 54.3 –
% of diesel cars in total registrations 46.5 49 69.1 71.4 –
% of diesel cars in the total car fleet 27.6 35.6 47.7 49.8 –
Range distribution of car sales
Low models (%) 43.6 40.1 38.7 –44.7
Medium-low models (%) 28.2 32.6 34.5 –33.6
Medium-high models (%) 17.3 14.2 11.2 –13 .1
Up-market models (%) 9 7.7 6.3 –8.6
Households' car fleet
Average age of cars (in years) 6.6 7.3 7.7 7.9 –
Brand distribution
PSA (%) 33.3 33.3 30.2 29.8 –
Renault (%) 36.2 35.2 36.4 37.3 –
Foreign brands (%) 30.5 31.4 33.2 32.1 –
Range distribution
Low models (%) 43.4 45.1 44.5 45.1 –
Medium-low models (%) 24.3 27.3 32.2 32 –
Medium-high models (%) 22.2 19.9 16.2 15.7 –
Up-market models (%) 7 7 5.7 5.8 –
References
[1] A.B. Jaffe, R.G. Newell, R.N. Stavins, Environmental policy and technological change, Environ. Resour. Econ. 22 (1–2) (2002) 41–69.
[2] K. Rennings, Redefining innovation —eco-innovation research and the contribution from ecological economics, Ecol. Econ. 32 (2000) 319–332.
[3] J. Horbach, Determinants of environmental innovation —new evidence from German panel data sources, Res. Policy 37 (1) (2008) 163–173 .
[4] M. Porter, C. van der Linde, Toward a new conception of the environment–competitiveness relationship, J. Econ. Perspect. 9 (4) (1995) 97–118.
[5] F. Malerba, Sectoral systems of innovation and production, Res. Policy 102 (2002) 845–859.
[6] J. Van den Bergh, J. Gowdy,Evolutionarytheories in environmental and resourceeconomics: approaches and applications,Environ. Resour. Econ.17 ((2000))37–57.
[7] J. Van den Bergh, Evolutionary thinking in environmental economics, J. Evol. Econ. 17 (2007) 521–549.
[8] V. Oltra, M. SaintJean, The dynamicsof environmentalinnovations:three stylisedtrajectoriesof clean technology, Econ.Innov. NewTechnol.14 (3) (2005)189–212.
[9] R.R. Nelson, S.G. Winter, An evolutionary theory of economic change, The Belknap Press of Harvard University Press, 1982.
[10] G. Dosi, Technological paradigms and technological trajectories, Res. Policy 11 (3) (1982) 147–162.
[11] F. Malerba, L. Orsenigo, Technological regimes and sectoral patterns of innovative activities, Ind. Corp. Change 6 (1997) 83–117.
[12] G. Winter, Schumpeterian competition in alternative technological regimes, J. Econ. Behav. Organ. 5 (1984) 287–320.
[13] F. Malerba, L. Orsenigo, Schumpeterian patterns of innovation, Camb. J. Econ. 19 (1) (1995) 47–65.
[14] K. Pavitt, Sectoral patterns of technical change: towards a taxonomy and a theory, Res. Policy 14 (4) (1984) 182–188.
[15] O. Marsili, The anatomy and evolution of industries: technological change and industrial dynamics, Edward Elgar, Cheltenham, 2001.
[16] O. Marsili, Technological regimes and sources of entrepreneurship, Small Bus. Econ. 19 (3) (2002) 217–231.
[17] J. Schmookler, Invention and Economic Growth, Harvard University Press, Cambridge, 1966.
15V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
[18] B.A. Lundvall, Innovation as an interactive process: from user-producer interaction to the national system of innovation, in: G. Dosi, et al., (Eds.), Technical
Change and Economic Theory, Pinter, London,1988.
[19] E. Von Hippel, The sources of innovation, Oxford University Press, New York, 1988.
[20] J.S. Metcalfe, The diffusion of innovation: an interpretative survey, in: G. Dosi, et al., (Eds.), Technical change and economic theory, Pinter, London,1988.
[21] L. Georghiou, J.S. Metcalfe, M. Gibbons, T. Ray, J. Evans, Post-innovation performance: technological development and competition, MacMillan, 1986.
[22] N. Rosenberg, Inside the Black Box: Technology and Economics, Cambridge University Press, New York, 1982.
[23] B. Arthur, Competing technologies: an overview, in: G. Dosi, C. Freeman, R. Nelson, G. Silverberg, L. Soete (Eds.), Technical change and economic theory, Pinter,
London, 1988, pp. 590–607.
[24] B. Arthur, Competing technologies, increasing returns, and lock-in by historical events, Econ. J. 99 (1989) 116–131.
[25] M. Katz, C. Shapiro, Network externalities, competition and compatibility, Am. Econ. Rev. 75 (1985) 424–440.
[26] J. Farrel, G. Saloner, Standardization, compatibility, and innovation, The Rand J. Econ. 16 (1) (1985) 70–83.
[27] W.J. Abernathy, J.M. Utterback, Patterns of industrial innovation, in: M.L. Tushman, W. Moore (Eds.), Readings in the Management of Innovation, Ballinger,
1988, pp. 25–36.
[28] S. Klepper, Entry, exit, growth and innovation over the product life cycle, Am. Econ. Rev. 86 (1996) 562–583.
[29] F. Malerba, R. Nelson, L. Orsenigo, S. Winter, ‘History-friendly’models of industry evolution: the computer industry, Ind. Corp. Change 8 (1999) 3–40.
[30] F. Malerba, R. Nelson, L. Orsenigo, S. Winter, Demand, innovation, and the dynamics of market structure: the role of experimental users and diverse
preferences, J. Evol. Econ. 17 (4) (2007) 371–399.
[31] C.M. Christensen, The innovator's dilemma: when new technologies cause great firms to fail, Harvard Business School Press, Boston, MA, 1997.
[32] R. Adner, When are technologies disruptive: a demand-based view of the emergence of competition, Strateg. Manage. J. 23 (8) (2002) 667–688.
[33] K.J. Lancaster, Consumer Demand. A new approach, University Press, New York, Columbia, 1971.
[34] P. Windrum, C. Birchenhall, Structural change in the presence of network externalities: a co-evolutionary model of technological successions, J. Evol. Econ.15
(2005) 123–148.
[35] K. Frenken, P. Saviotti, M. Trommetter, Variety and niche creation in aircraft, helicopters, motorcycles and microcomputers, Res. Policy 28 (5) (1999) 469–488.
[36] C. Graham, The ascendency of the sailing ship 1850–1885, Econ. Hist. Rev. 9 (1956) 74–88.
[37] K. Green, A. McMeekin, A. Irwin, Technological trajectories and R&D for environmental innovation in UK firms, Futures 26 (1994) 1047–1059.
[38] R. Florida, Lean and green: the move to environmentally conscious manufacturing, Calif. Manage. Rev. 39 (1) (1996) 80–105.
[39] K.M. Rehfeld, K. Rennings, A. Ziegler, Integrated product policy and environmental product innovations: an empirical analysis, Ecol. Econ. 61 (1) (2007)
91–100.
[40] T.N. Gladwin, The meaning of greening: a plea for organizational theory, in: K. Fischer, J. Schot (Eds.), Environmental Strategies for Industry: International
Perspectives on Research Needs and Policy Implications, Island Press, Washington, DC, 1993, pp. 37–61.
[41] M.E. Drumwright, Socially responsible organisational buying: environmental concern as a noneconomic buying criterion, J. Mark. 58 (3) (1994) 1–19.
[42] J. Connolly, P. McDonagh, M. Polonsky, A. Prothero, Green marketing and green consumers: exploring the myths, in: D. Marinova, D. Annandale, J. Phillimore
(Eds.), The international handbook on environmental technology management, Edward Elgar Publishing, 2006, pp. 251–268.
[43] M.A. Janssen, W. Jager, Stimulating diffusion of green products, J. Evol. Econ. 12 (3) (2002) 283–306.
[44] R. Kemp, Environmental Policy and Technical Change: A Comparison of the Technological Impact of Policy Instruments, Edward Elgar Publishing, 1997.
[45] A. Jaffe, S. Peterson, P. Portney, R. Stavins, Environmentalregulation and the competitiveness of U.S. manufacturing: what does the evidence tell us? J. Econ. Lit.
XXXIII (1995) 132–163.
[46] N.A. Ashford, C. Ayres, R.F. Stone, Using regulation to change the market for innovation, vol. 9, Harvard Environmental Law Review, 1985, pp. 419–466.
[47] G. Heaton, Environmental policies and innovation: an initial scoping study,OECD Report, 1997.
[48] R. Kemp, Technology and environmental policy: innovation effects of past policies and suggestions for improvements, Proceedings Innovation and the
Environment, OECD, Paris, 2000.
[49] C. Freeman, The economics of hope, Pinter Publishers, London and New York, 1992.
[50] C. Carraro, D. Siniscalco, International institutions and environmental policy: international environmental agreements, Eur. Econ. Rev. 42 (3–5) (1998)
561–572.
[51] T. Könnola, G.C. Unruh, J. Carrillo-Hermosilla, Prospective voluntary agreements for escaping techno-institutional lock-in, Ecol. Econ. 57 (2) (2006) 239–252.
[52] N.A. Ashford, An innovation based strategy for a sustainable environment, in innovation oriented environmental regulation theoretical approach and
empirical analysis, in: J. Hemmelskamp, K. Rennings, F. Leone (Eds.), ZEW Economic Studies, Springer Verlag, 20 00, pp. 67–107.
[53] K. Rennings, A. Ziegler, Blueprints for an integration of science, technology and environmental policy (BLUEPRINT), European Commission, 2003.
[54] M. Jänicke, J. Blazejczack, D. Edler, J. Hemmelskamp, Environmental policy and innovation : an international comparison of policy frameworks and innovation
effects, in: J. Hemmelskamp,, et al., (Eds.), Innovation-oriented Environmental Regulation, Theoretical Approaches and Empirical Analysis, 20 00, pp. 125–152.
[55] C. Edquist (Ed.), Systems of innovation –Technologies, institutions and organizations, Pinter Publishers, London, 1997.
[56] L. Bach, M. Matt, From economic foundations to S&T policy tools: a comparative analysis of the dominant paradigms, in: P. Llerena, M. Matt (Eds.), Innovation
policy in a knowledge-based economy, Springer-Verlag, 2005, pp. 17–45.
[57] R. Nelson, The coevolution of technology, industrial structure, and supporting institutions, Ind. Corp. Change 3 (1) (1994) 47–63.
[58] European Commission, Chapter 4: the European automotive industry, in European competitiveness report 2004, Commission Staff Working Document, SEC
(2004) 1397, European Communities, 2004.
[59] INPI-OPI: Automobile et environnement, Etude du département des brevets de l'INPI, (2006).
[60] Z. Griliches, Patent statistics as economic indicators: a survey, J. Econ. Lit. 28 (1990) 1661–1707.
[61] V. Oltra, R. Kemp, F. de Vries, Patents as a measure for eco-innovation”, MEI Report, 2007.
[62] M. Hekkert, R. Van den Hoed, Competing technologies and the struggletowards a new dominant design: the emergence of the hybrid vehicle at the expense of
the fuel cell vehicle, Green. Manag. Int. (47) (2006) 29–47.
[63] R. Kemp, Zero emission vehicle mandate in California: misguided policy or example of enlightened leadership? in: C. Sartorius, S. Zundel (Eds.), Time
strategies, Innovation and Environmental Policy, Edward Elgar Publishing, Cheltenham, 2005.
[64] A. Avadikyan, P. Cohendet, J.A. Héraud (Eds.), The economic dynamics of fuel cell technologies, Springer Verlag, 2003.
[65] P. Christidis, H. Hernandez, A. Goergakaki, S.D. Peteves, Hybrids for road transport- Status and prospects of hybrid technology and the regeneration of energy
in road vehicles, IPTS-JRC report, February 2005.
[66] V. Oltra, M. Saint Jean, Variety of technological trajectories in low emission vehicles (LEVs): a patent data analysis, GRES Working Paper, Cahiers du GRES no.
2006-20, (2006).
[67] K. Frenken, M. Hekkert, P. Godfroij, R&D portfolios in environmentally friendly automotive propulsion: variety, competition and policy implications, Technol.
Forecast. Soc. Change 71 (2004) 485–507.
[68] T. Meyer, A. Clavel, Place de la consommation de carburants dans les critères de choix des véhicules (Particuliers), Rapport final PREDIT 2002–2006,
September 2006.
[69] Well-to-Wheels Report: Well-to-wheels analysis of future automobile fuels and powertrains in the European context, Concawe, Eucar, JRC-EC, version 1b,
January, (2004).
[70] G.P. Maxton, J. Wormald, Time for a Model Change, Cambridge University Press, 2004.
[71] R. Kemp, J. Schot, R. Hoogma, Regime shifts to sustainability through processes of niche formation. The approach of strategic niche management, Technol.
Anal. Strateg. Manag. 10 (2) (1998) 175–195.
[72] CARS 21, A competitive automotive regulatory system for the 21st Century, Final Report, European Commission, 2006.
[73] CCFA: L'industrie automobile française: analyse et statistiques, Comité des Constructeurs Français d'automobiles, 2007.
16 V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
Vanessa Oltra is assistant professor in economics at Bordeaux University (France) and affiliated to the research institute GREThA (UMR CNRS 5113). She made a PhD
at the University of Strasbourg (BETA, ULP, France) on evolutionary simulation modelling of industrial dynamics. Since 2000, she is working on evolutionary
analyses of environmental innovations and industrial dynamics. She is coordinating a working package on “Environmental innovation”within the DIME (Dynamics
of Institutions and Markets in Europe) Network of Excellence (http://www.dime-eu.org /wp25).
Maïder Saint-Jean is assistant professor in economics at Bordeaux University (France) and affiliated to the research institute GREThA. She had completed her PhD
dissertation in 2002 which was dedicated to the analysis of clean technology development within vertical interfirm relationships. Such a work has been followed
by a post-doctoral stay at the IPTS in Seville (Spain) where she could participate to a European project on policy pathways for the development of clean
technologies. Since September 2004, she pursues her research activities at GREThA with a particular interest on the sectoral specificity of environmental
innovations and on the dynamic interplay between environmental regulation and innovation.
17V. Oltra, M. Saint Jean / Technological Forecasting & Social Change xxx (2008) xxx–xxx
ARTICLE IN PRESS
Please cite this article as: V. Oltra, M. Saint Jean, Sectoral systems of environmental innovation: An application to the French
automotive industry, Technol. Forecast. Soc. Change (2008), doi:10.1016/j.techfore.2008.03.025
