Developing Innovative Competences: The role of
institutional frameworks
by
Richard Whitley
(Manchester Business School, University of Manchester)
ABSTRACT
The recent development of the biotechnology and computer industries has
highlighted the variety of ways in which firms in different countries and sectors can
develop innovative competences. Four aspects are particularly important: the degree
of involvement in the public science system, involvement in industry networks,
reliance on specialist skills of individuals, and the ability to change competences
radically. National and regional variations in these result from differences in dominant
institutional frameworks.
These frameworks include the nature of the public science system in addition to the
organisation of capital and labour markets and the structure of inter-firm relations.
Particularly important features of these systems include: the organisation of research
training, the flexibility of researchers and organisations in developing novel goals and
approaches, the organisation of scientific careers and the prevalent science and
technology policies of the state. Particular combinations of these institutional features
have become established in different market economies and led to distinctive styles
of innovative competence development being adopted. These in turn help to explain
continuing variations in patterns of technological change between countries.
2
Introduction
The rapid growth of the biotechnology and computer hardware and software
industries in the USA in the last quarter of the twentieth century has highlighted the
importance of new small firms linked to academic research in developing radical
innovations. Many of these companies were established by highly specialised
research scientists and engineers educated at local and regional universities as spin
offs from established firms or de novo from academic employment (see, e.g.
McKelvey, 1996; Saxenian, 1984). They were, and still are in many cases, funded by
specialist venture capital firms which developed portfolios of investments in high
risk/high return businesses experiencing high rates of technical change. Unlike
innovative firms in other industries that relied on the published scientific literature for
useful formal knowledge and/or collaborative research with applied research
institutes, industry consortia laboratories etc., many of these companies had direct
links with current academic research studying generic processes and phenomena.
Moreover, these connections were seen as key to their competitive success in these
dynamic industries.
At more or less the same time, particularly in the 1990s, the medium and large sized
firms in the coordinated market economies of Germany, Japan and similar countries
that had provided the success stories of the 1970s and 1980s appeared to be less
competitive in these newer sectors. Indeed, many consider that it is the very qualities
that enabled them to be successful in the more traditional machinery and assembly
industries, which inhibit them from developing radical new innovations effectively, as
patent statistics seem to indicate (see, e.g. Guerrieri and Tylecote, 1997; Soskice,
1997; 1999). In particular, their reliance on long term commitments from employees
and business partners to develop distinctive organisational capabilities for integrating
varied kinds of knowledge and skills limit their ability to develop discontinuous
changes in technologies and markets.
This contrast highlights the different ways in which firms can develop innovative
competences, especially in terms of their recruitment and retention policies. It also
emphasises the interconnections between types of organisational competences and
competitive success in sectors with contrasting “technological regimes”, or
characteristic patterns of technical change. Such regimes have been seen as
generating distinctive “sectoral innovation systems” (e.g. Breschi and Malerba, 1997),
in which different kinds of firms prosper to varying extents. Leaving aside the implicit
technological determinism and functionalist reasoning in some of this literature, it
emphasises the variety of ways in which firms can, and do, develop innovative
competitive competences, which are more or less effective in different industries at
different times.
Furthermore, these different types of firms and competences develop in contrasting
institutional contexts and countries (see, e.g. Casper et al., 1999; Soskice, 1997;
Whitley, 1999; Whitley and Kristensen, 1996). Important differences here concern the
organisation and control of labour markets, including the institutions governing skill
development and certification, and the organisation of market relationships, as well
as financial and legal systems. Such variations have been seen as conferring
competitive advantages on firms in sectors with particular technological regimes,
3
while denying them in others, resulting in contrasting innovation patterns across
institutional regimes (see, e.g. Hall and Soskice, 2001).
The logic underlying these relationships between institutions, firms and innovation
patterns has been explored in a number of recent papers contrasting the
technological and sectoral specialisation of Germany, the USA and some other
countries in the last few decades of the twentieth century (see, e.g. Casper, 2000;
2001; Culpepper, 2001; Soskice, 1997; 1999). Stylised in terms of the contrast of
coordinated (CME) and liberal market economies (LME), these contributions have
highlighted how differences in welfare systems, employment law and conventions,
the organisation of industry and trade associations, training systems, financial
markets and legal systems generate different incentives for individuals and firms to
pursue distinctive investment and innovation strategies. These institutional variations
not only lead to different rates of innovation and technical change across industries,
but also encourage different kinds of firms to concentrate on different kinds of
innovations within new industries, such as platform technologies and therapeutic
drugs in biotechnology (Casper, 2000).
Simplifying greatly, CMEs encourage cooperative, long-term investments in firm- and
industry-specific skills that develop organisational competences in coordinating
knowledge and skills across internal and external organisational boundaries to
develop continuous but incremental innovations. LMEs, in contrast, reward more
short-term and adversarial behaviour by both individuals and firms that generates
more generic skills and considerable labour mobility between firms. Such economies
facilitate the rapid use of new knowledge and skills to seize radically new
opportunities. The dominant institutions in CMEs are seen as "solving" the
organisational problems involved in pursuing high quality incremental innovation
strategies, while those in LMEs enable firms to focus more on developing radical
innovations in newly emerging technologies (Soskice, 1997).
As a result, countries that have developed institutional frameworks similar to the
idealised CME, for example Germany, Sweden and Switzerland, have been highly
effective at developing incremental product and process innovations in established
technologies in the chemical and machinery industries. In contrast, societies with
dominant institutions more similar to LMEs, such as the UK and the USA, have been
more effective in developing discontinuous innovations in newer technologies such
as biotechnology and microprocessors, as well as in developing competitive business
service firms that are highly dependent on the specialist skills of particular
individuals.
These models of how institutional frameworks enable firms to deal with the
organisational problems associated with distinctive innovation strategies emphasise
three sets of relationships. These concern the level of commitment and cooperation
between: a) top managers and core employees, b) firms and their business partners,
and c) top managers and the owners/controllers of capital. Different innovation
strategies are here seen as requiring particular institutions to manage these
relationships in distinctive ways to resolve the specific coordination problems they
involve. In other words, contrasting institutional frameworks generate different
incentives for firms and individuals that result in varied patterns of behaviour. These
4
then generate different patterns of technological development and sectoral
specialisation.
This approach to the comparative analysis of institutions and innovations identifies
some of the critical linkages between the organisation of financial and labour
markets, firms' priorities and strategies, and patterns of technical change in two
idealised types of market economy. However, there are important institutional
variations within these two broad "varieties of capitalism," such as the very different
systems of skill formation in Germany and Japan, as well as significant differences in
leading firms' research and innovation strategies across similarly coordinated market
economies. Also, of course, a number of different systems of economic organisation
have developed in late twentieth century capitalism, such as the pattern developed in
France and elsewhere (Boyer, 1997; Hancke, 2001).
Additionally, the variety of firm types and innovation strategies within each kind of
market economy is sometimes greater than this dichotomy would suggest. The USA,
for instance, has developed both large integrated firms pursuing largely autarchic
innovation strategies and smaller specialist research based firms introducing radical
innovations in close cooperation with the public science system. Also, the
coordinated market economy of Sweden developed a considerable number of
specialised software producers in the late 1990s (Casper and Glimstedt, 2000),
which resemble the start-ups of Silicon Valley more than established large firms.
These variations suggest that the simple CME/LME contrast needs further
development and differentiation to encompass the variety of ways in which
institutional arrangements impinge upon firms' innovation strategies and result in
major contrasts in patterns of technological development.
A further important feature of technological development in the last few decades
often twentieth century has been the increasing importance of academic research
skills and knowledge in the development of new industries (see, e.g. Mansfield, 1995;
Narin et al., 1997). This means that variations in the dominant institutions governing
the development and use of knowledge production skills have significant
consequences for the rate and type of technical changes in different market
economies. Additionally, differences in the rate of movement of scientists and
engineers between the public research system and private firms, and between firms,
affect the flow of knowledge and skills throughout the economy. These variations are
especially important in industries where product and process changes are closely
dependent upon the integration of knowledge from many different fields, as they
seem to be in many of the newer sectors. Since national research systems differ
considerably in how they are organised and controlled, such differences constitute an
important part of the institutional environments that explain contrasts in prevailing
patterns of innovation and technological development across countries.
In this paper I explore these interconnections through an analysis of how firms
develop distinctive innovative competences and strategies in different ways. These
differences reflect the pressures and possibilities of different institutional
environments and so help to explain how patterns of technological development and
sectoral specialisation vary between societies. By connecting variations in the
organisation of the public science system with key features of business organisation,
labour and capital markets, this paper presents a framework for analysing how
5
institutions affect innovative firm type and behaviour in different countries. The public
science system here refers to the institutionalised system for generating research
results that are published in order for scientists to gain reputations for contributions to
collective intellectual goals, as distinct from research that is undertaken for
proprietary gain and kept private (Dasgupta and David, 1994; Whitley, 2000).
Initially, I shall discuss four major differences in how innovating firms develop
distinctive competences. These are connected to variations in the kinds of
innovations developed by different types of firms, particularly their cumulativeness
and customisation. The following section suggests how key features of the
institutional frameworks governing labour and capital markets, as well as public
science systems, encourage firms to develop innovative competences in contrasting
ways, and so generate different patterns of technical change in different societies.
The Development of Innovative Competences
In comparing innovative firms across market economies, key differences between,
say, new technology based firms, large integrated mass production companies and
members of business networks, concern the ways in which they develop distinctive
competences. These can be summarised in terms of four basic alternatives that
distinguish the innovation strategies of leading German, Japanese and US firms in
the post second world war period. First, whether to develop innovative competences
internally, keeping knowledge production and skill development in house, or to
develop them in cooperation with external agencies and business partners. Second,
if external partners are involved in developing innovative capabilities, are these to be
the organisations and personnel working in the public science system, or those in the
same industry and using similar technologies, or both? Third, how much do
innovative firms invest in the long-term development of collective organisational
capabilities, as distinct from relying upon more individual specialist skills? Fourth,
how able and willing are firms to change their innovative skills and competences
significantly in the short to medium term?
Considering first the decision about whether to develop knowledge production and
technical problem solving competences internally, as opposed to involving other
organisations, the former ensures that ownership is secured and spillover risks are
minimised. In theory, it also facilitates the integration of varied skills and
competences through a unified authority structure based on ownership. It does,
though, limit access to, and integration of, new knowledge and skills that do not
easily fit into the firm specific technological framework, as well as restrict learning
from suppliers and customers. Such isolation can be particularly disadvantageous in
sectors where the rate of technical change is high and dependent on a wide variety
of knowledges from different fields produced with different research skills. In newly
emerging industries especially, reliance on external sources of knowledge is often
considerable.
For firms that do decide to develop innovative competences in cooperation with
external partners, there are important differences between those in the public science
system and those in the same industry. These principally derive from the quite
different institutional frameworks governing priorities and rewards in public and
private research organisations that can generate considerable cognitive distance
6
between their skills and knowledge. In particular, the competitive pursuit of
reputations from specialist colleagues around the world encourages researchers in
the public science system to focus on generic intellectual problems dealing with
general phenomena because these have more significance for more
colleague/competitors. This means that the results, techniques and intellectual
approaches involved in current research in the public sciences are often quite remote
from current industrial concerns and practices. As a result, firms wishing to access
and use them have to make considerable investments in what Cohen and Levinthal
(1990) term "absorptive capacity", typically by hiring trained researchers and
conducting more generic research than is required for current technological problem
solving activities.
Firms tend to become closely involved in technical communities in the public science
system in situations of high technical uncertainty and rapid technological change. As
well as gaining access to new developments through research contracts and
consultancy, they fund fellowships, establish joint research institutes with university
departments and sometimes engage in cooperative research with other companies
and universities, in addition to regularly hiring highly trained researchers, both at the
PhD stage and in mid-career. As Kenny (1986) has emphasised, these long-term
investments have been particularly evident in the biotechnology industry.
The postwar public science systems of many countries do, however, produce a wide
range of scientific and technological knowledge that varies considerably in its degree
of abstraction and concern with generic processes, as distinct from more specific
ones that deal with particular technologies and materials. Firms can be involved in
the latter without being closely connected to the former, as are many German firms
that have close ties to applied research and technology transfer institutes (Herrigel,
1993; Soskice, 1997). However, in last two or so decades of the twentieth century,
innovating firms have become more directly associated with theoretically focused
research in academic laboratories, and depend more on the skills and knowledge
produced by them, especially in the US biotechnology industry (Gambardella, 1995;
Henderson et al., 1999; McKelvey, 1996). In general, the more companies become
involved in such research networks, the faster they learn about new results and
techniques, and the more able they are to draw upon knowledge from varied fields
and disciplines in developing new products. They are additionally in a better position
to assess the significance of new knowledge and skills than are firms that are more
remote from current research activities.
We can, then, distinguish three forms of involvement in the public science system.
First, there are firms that have only a minimal, rather passive, involvement with
current research, relying essentially on scanning the published journal and patent
literature for obtaining relevant scientific and technological knowledge. Second, other
companies may be more actively engaged with researchers in the public research
system, but these are mostly working on particular technologies and materials that
are most relevant to firms in specific industries. A third group of firms is more directly
involved in current research on generic phenomena and processes. They are
concerned to access relatively general and abstract knowledge, especially the skills
for producing it, to develop key competitive competences. These three varieties of
involvement in the public science system can be termed passive, industry and
technology specific, and generic. Generally, the more firms are involved in such
7
research networks, the more they should be able to search effectively for, and use
appropriately, new knowledge and skills from a wide variety of fields and disciplines.
Innovating firms also, of course, gain knowledge about new technologies, markets
and process improvements from trade associations, industry groupings, suppliers
and customers. Firms vary greatly in the extent to which they share knowledge with
business partners through membership of industry networks. In countries like
Germany and Japan, for example, business groups, industry associations and similar
networks often engage in joint standards setting, knowledge diffusion and technical
development, as well as establishing and policing norms of appropriate firm
behaviour (Culpepper, 2001; Morris-Suzuki, 1994; Odagiri and Goto, 1996; Soskice,
1997; 1999; Tate, 2001). Their members are accordingly able to access new
knowledge and information more quickly than those not involved in such groupings,
which can be especially advantageous to small and medium sized companies
(Herrigel, 1993).
This kind of knowledge sharing with suppliers, competitors and customers can be
termed the degree of involvement in industrial networks. When this involvement is
considerable, it encourages firms to invest in the development of customer-specific
knowledge and skills because they are less likely to lose such connections than in
more adversarial relationships. Through such networks and authority sharing, firms
are also likely to develop strong competences in integrating information from a
variety of industry sources, and in developing innovations that are more customised
than generic. Equally, though, the mutual commitments developed in industry
networks will typically limit the degree of technological change undertaken by firms
since radical, transformative innovations threaten current organisational
competences (Christensen, 1996; Tushman and Anderson, 1986). High levels of
industry network involvement, then, tends to be associated with considerable
technological cumulativeness of innovations.
Cumulativeness here refers to the degree of technical continuity involved in
developing innovations (see, e.g. Ehrnberg and Jacobsson, 1997). Two aspects can
be distinguished (Casper and Glimstedt, 2000). First, the extent to which the skills
required to develop and commercialise new products are stable and predictable
varies considerably between fields. Where such instability is high, in "discrete"
technological trajectories (Breschi and Malerba, 1997), firms do not know which skills
and other resources will be required to develop research programmes until the
results of earlier phases are available. Second, the risks of project failure are much
greater when technical uncertainty is high, both for employers and researchers.
Generally, then, where technological development is more discrete than cumulative,
it is less susceptible to planning, may require rapid and radical changes in skills, and
may threaten existing organisational competences.
Involvement in one of these two sets of external networks that provide new
knowledge and skills can vary independently of that in the other. Firms can develop
close connections with public and private research organisations, hire PhDs and
participate actively in scientific and technological conferences and technical
exchanges, for example, without forming powerful industry associations or long term
ties with customers and suppliers. Many firms in the US computer hardware and
software, and biotechnology, industries seem to combine such considerable technical
8
involvement in the research community with largely adversarial and autarchic interfirm relationships (Chesbrough, 1999).
Similarly, high levels of industry embeddedness and inter-firm dependence, as in
many Japanese industries in the postwar period, can be combined with largely
autonomous knowledge development in relative isolation from the public science
system. Most large Japanese firms, for instance, have restricted their academic
contacts to hiring engineers and scientists with MScs rather than PhDs and informal
research support for individual professors on a relatively small scale (Coleman, 1999;
Kneller, 1999; Westney, 1993; Yoshihara and Tamai, 1999). On the other hand,
many German firms in the chemical and engineering industries seem to combine
strongly coordinated technical exchanges within industry sectors with close
connections to applied research organisations such as the Fraunhofer Gesellschaft
laboratories and technical schools (Herrigel, 1993; Lehrer, 2000).
The third important aspect of competence development considered here concerns
the extent to which firms rely more on the specialist skills of individuals hired as
needed rather than developing distinctly organisational capabilities based on a
relatively stable group of core employees. While all companies develop distinctive
organisational capabilities as emergent properties of each organisation that
transcend the abilities and activities of individual employees (Dosi et al., 2000;
Nelson and Winter, 1982), the degree to which these depend on individuals'
specialist expertise that is not firm-specific varies considerably.
For instance, the role of new firms founded by highly trained and experienced
engineers and scientists in the development of the US biotechnology and computer
industries has shown how relatively small and quickly formed organisations of
specialist researchers and designers can play a major role in developing significant
innovations. Under particular conditions, that is, the ability to create firms that
integrate high level skills around specific goals can generate competitive advantages
in industries undergoing high rates of technical change. Such firms depend greatly
upon the skills and knowledge of project leaders and their teams of specialist staff to
develop innovations, as distinct from developing distinctive collective competences
that are more organisational. Similar sorts of project-based firms also play an
important role in the construction industry in many countries, as well as being a
significant organisational form in the Danish machinery sector (Kristensen, 1992;
1996). The difference here is that these radically innovative organisations are
dominated by highly trained researchers who integrate a variety of different
knowledges and skills to develop highly novel products for a range of customers.
Among the important conditions that encourage such reliance on individually owned
and controlled specialist skills are the ease of appropriating profits from innovations,
for example through patenting in the pharmaceutical industry (Gambardella, 1995),
and the existence of open standards facilitating the development of modular
innovations (Langlois and Robertson, 1995; Langlois and Mowery, 1996). High levels
of appropriability and modularity of innovations facilitate the specialisation of firms in
product design and development without having to invest in complementary assets in
marketing and distribution (Teece, 1986). Coordination of innovative activities in such
circumstances can thus be carried out by project groups of specialist experts rather
than needing extensive organisational routines and procedures. They organise their
9
activities around teams of highly qualified specialised engineers and scientists
focused on short to medium term innovation goals. Competitive advantages and
competences here derive from flexible and speedy responses to new knowledge and
skills, and the ability to integrate new kinds of information and expertise to generate
disruptive products and processes, usually in newly emerging industries.
Conversely, where appropriability is difficult and/or technological change is systemic
rather than modular, innovating firms tend to coordinate product development with
production, marketing and other complementary activities to protect their assets and
integrate components of technological systems. Such coordination involves the
construction of formal organisations with collective capabilities. Knowledge
production and problem solving skills are here more organisational than individually
owned and developed.
Firms in many assembly and machinery industries, for instance, typically coordinate
knowledge from many different fields, both within and outside the organisation, to
develop and commercialise innovations, and establish organisational routines to do
so. Distinctive firm specific skills develop as a result that are not tied to particular
individuals' skills and contributions. These skills become entrenched in distinctive
technological paradigms that guide development trajectories and how engineers
tackle problems in, say, car design and aircraft development. While all work
organisations, then, generate particular kinds of competences by coordinating and
controlling work in a systematic manner, the extent to which these capabilities
depend on the generic skills of specific individuals varies considerably. This aspect of
innovating firms' competences can be characterised as the degree to which they
depend on the high level specialist skills of individuals.
The fourth aspect of competence development to be considered here concerns the
speed with which, and degree to which, firms change their innovative capabilities and
competitive competences. Many companies, for instance, diversify into new
technologies and markets through developing new skills and abilities incrementally,
building on existing ones. In contrast, others are able to change their core
competences more radically by hiring new staff with quite different skills, as in the
case of some US pharmaceutical firms developing biological research skills (Zucker
and Darby, 1997), or by acquiring companies with expertise in new technologies
such as biotechnology and software start-ups.
This characteristic reflects the general willingness of firms to enter into long term
commitments to staff and business partners, as well as their specialisation in
particular technologies and industrial sectors. The more they develop distinctive
competences through investments in employee training and customer specific
knowledge, the more difficult they will find it to develop quite new capabilities, and to
be successful in novel technologies and sectors. Technological changes made by
such firms will, then, tend to be incremental and customer-focused rather than
generic and transformational.
In principle, firms can select whether to focus on developing technologically
cumulative innovations for particular kinds of customers or on creating radically novel
and more generic innovations for a wide range of new customers. The former
strategy is typically pursued through continuous improvement of skills and knowledge
10
about particular technologies and industries, with both employers and employees
making long term commitments to each other, suppliers and customers. The latter
implies greater willingness to change firms' skill and knowledge base quite
discontinuously by hiring and firing staff, and changing business partners, as well as
buying and selling whole businesses.
These four different ways of developing innovative competences are interconnected.
In particular, varying involvement in research and industry networks is associated
with differing degrees of organisational development and rates of competence
change. For example, high levels of involvement in public science research networks
usually requires considerable investment in "absorptive capacity" which implies
recruitment of highly trained specialist researchers, and the encouragement of work
on topics that are similar to those being studied by the scientific community. As Hicks
(1995) and others have pointed out, a major reason for firms to encourage staff to
undertake fundamental research and publish it is to gain credibility with academic
researchers and so facilitate informal access to their work and skills.
If, on the other hand, firms focus on developing highly organisation-specific skills that
enable them to coordinate product development, manufacturing and distribution
effectively, most of their engineers and scientists are unlikely to do the sort of work
that will be published in leading scientific journals, and indeed may become
incompetent to do it. Their ability to absorb and use effectively new knowledge and
techniques will be limited. A very strong emphasis on organisational competences at
the expense of developing more generic specialist research skills, as in many large
Japanese companies (Westney, 1993), may well prevent firms from gaining close
access to current research and skills in the public science system.
Similarly, firms that concentrate on developing technologically cumulative innovations
are unlikely to invest in close involvement in public research networks. Building on
current competences and expertise to introduce new products and processes
incrementally, they have a deep organisational understanding of the dominant
technology in the industry that is remote from the intellectual approaches and
concerns of the public science system. In contrast, when technical uncertainty is high
and research results are unpredictable, existing skills and technological trajectories
may become redundant as new knowledge is produced and so firms are more willing
to develop close connections with public science researchers. In emerging
technologies and industries, especially, the value of particular skills and projects is
difficult to be sure about as new results appear, so that companies innovating in such
sectors tend to invest more in keeping up with academic knowledge production.
Investment in industry networks involves sharing knowledge with business partners
and competitors. This is unlikely to develop without firms being fairly sure that others
are equally committed to improving current technologies, so that opportunistic use of
such information and resources can be meaningfully sanctioned by loss of reputation.
Where companies restrict such commitments, and can change their competences
quite radically through hiring and firing, other firms will not be willing to share
important resources with them. High levels of industry network involvement, then, are
associated with the relatively cumulative development of innovative competences
and not with the introduction of competence-destroying innovations.
11
Conversely, more isolated firms without strong ties to suppliers and customers will be
more able to change their competences discontinuously and adopt radically new
research and technology development skills. Innovations developed by such firms
are less likely to be incremental and will be more generic than customer-specific. This
also means that firms relying heavily on the specialist skills of individuals that can
readily be changed through external labour markets to develop generic innovations
are unlikely to become deeply embedded in such industry networks.
Finally, firms that wish to be able to change their innovative competences rapidly to
seize new opportunities are unlikely to invest heavily in creating distinctive
organisational capabilities, especially if they require long term commitments to
employees. While some innovating companies that do construct integrated
organisations are able to change direction by restructuring their labour force and/or
trading subsidiary units, as in the case of US pharmaceutical firms, it is difficult for
them to change radically in the short to medium term. More decentralised
organisations structured around project teams that can be changed at short notice
are relatively flexible and able to adapt quickly to new knowledge and skills.
These interconnections suggest that certain combinations of these characteristics are
more probable than others, so that firms develop innovative competences in a limited
number of ways. For example, those that focus on producing radical, disruptive
innovations in rapidly emerging technologies and industries, such as computer
hardware and software and biotechnology, are more likely to become involved in
public science research networks than in industry networks. They will also tend to
change key skills and knowledge bases at short notice to deal with high levels of
technical uncertainty instead of developing distinctly organisational capabilities over
the medium to long term. In contrast, firms specialising in the rapid development of
new products and processes within current technological trajectories through highly
coordinated design, development, manufacturing, marketing and distribution activities
are likely to invest in the development of strong organisational competences and
firm-specific skills, but not in close connections to the public science system.
Innovative Competence Development in Different Institutional Frameworks
Firms are encouraged to adopt some of these alternatives, and discouraged from
other ones, by particular combinations of dominant institutions in different market
economies. In addition to those governing labour and capital markets, and the
organisation of employers, unions and similar collective associations, that have been
highlighted in recent comparative analyses of market economies (see, e.g. Hall and
Soskice, 2001; Hollingsworth and Boyer, 1997; Whitley, 1999), the nature of public
science systems also affects the ways that firms develop innovative competences.
Four features of these systems are particularly important for innovative firms. First,
the amount of investment in research training and the way that it is organised.
Second, the flexibility and pluralism of the public sciences in developing new
intellectual goals, approaches and techniques. Third, the way that scientific careers
are organised and the institutionalisation of professional researcher roles. Finally, the
prevalent institutions and policies governing the direction of scientific and
technological research and technology transfer. Combined with the more general
features of dominant institutions, these characteristics of publics science systems
guide firms’ choices and so affect patterns of innovation and technological
12
development in different countries and regions in ways that are summarised in table
1 and will now be further discussed.
TABLE 1 ABOUT HERE
Considering first the factors that affect the degree of firm involvement in the public
science systems of different countries, many of these stem from variations in how
research is organised and controlled. For example, differences in the integration of
research training with current projects, and the amount of such training in knowledge
production, influence firms' ability to access and use new results and techniques
produced in the public science system. By integrating research training with the
production of new knowledge, academic systems produce qualified researchers who
have the tacit knowledge and skills required to undertake independent research and
develop new lines of understanding. This means that firms hiring them can access
new results quicker and use them for technological purposes faster than those that
recruit graduates accustomed to less uncertain work. As Feller (1999: 83) suggests:
"Students also are a means by which new scientific findings and technologically
relevant knowledge are transferred from the campus to the firm. Indeed, as new
technologically relevant research findings become more embedded in the tacit knowhow of students regarding laboratory procedures and software, their importance as
technology transfer agents is likely to increase".
In contrast, higher education systems that separate research training from work in
leading institutes, and/or focus more on training MSc graduates than PhDs, produce
engineers and scientists who may be quite effective in dealing with relatively
precisely formulated problems within current technological trajectories but are not so
capable of researching novel problems and technologies that involve greater
uncertainty. In Japan, for instance, despite the expansion of graduate schools at
many state universities since the 1960s, most of their students have left with MScs
rather than PhDs (Coleman, 1999; Ogura and Kotake, 1999). This is partly because
firms rarely preferred to recruit PhDs, regarding them as being too specialised and
remote from commercial concerns (Westney, 1993). In fact, not only did PhDs not
receive more pay than MScs in most firms, but often they were paid less because
they had less seniority with the company (Sienko, 1997; Yamamoto, 1999). Together
with the limited expansion of universities and other public sector research
organisations in the 1970s and 1980s, this discouraging labour market for
researchers in Japan has considerably restricted the output of knowledge producers,
and so the availability of novel kinds of research skills for many Japanese firms.
The second important aspect of public science systems that affects the level of firm
involvement concerns the ease and frequency with which research scientists and
engineers are able to develop new intellectual goals, fields and approaches, such as
software engineering and molecular biology. Where research objectives and
strategies are varied and changeable, as distinct from being tightly integrated around
established disciplinary goals, frameworks and expertise, it should be easier to
extend and apply new ideas and techniques for technological purposes, and to
develop new areas of research with new skills. The boundaries between theorydriven scientific research and more instrumental knowledge production are more
fluid, permeable and overlapping in such public science systems than in those where
intellectual, skill and organisational boundaries are strongly structured around
13
separate disciplines. As a result, firms find it easier to become involved in research
networks that combine generic research into general phenomena and processes with
more instrumental goals than they would in more stable, discipline-bound research
systems.
This intellectual and organisational flexibility of national research systems is affected
by the nature of the employment system in universities and allied organisations.
Where individual heads of departments and research institutes exercise considerable
control over resources and careers, the rate of change and variety of intellectual
approaches and skills is likely to be less than in employment systems where there is
greater pluralism of power and authority within administrative units. The US, and to
some extent UK, pattern of locating a number of research groups within relatively
large university departments, for example, permits greater intellectual pluralism of
projects than the German and Japanese pattern of individual research groups and
institutes controlled by a single or couple of professors who combine scientific
leadership with administrative responsibility. When this separation of intellectual
production units from organisational ones is combined with extensive reliance on
external funding of research projects, the power of institute heads to direct research
programmes is greatly reduced and competition between groups within departments
encouraged. Diversity of research goals and approaches is therefore greater in such
employment systems than where control over research programmes is more
centralised within departments.
A further feature of public science systems that affects the ways firms develop
innovative competences concerns the organisation of careers and the extent to which
professional researcher roles and identities are institutionalised. Where engineers
and scientists develop careers more within employment organisations than
professional specialities, and cosmopolitan role models are weakly established, firms
will be able to develop firm-specific innovative competences relatively easily, but will
find it more difficult to access current work in the public science system. Conversely,
where mobility is expected in the course of a research career, and researchers are
encouraged to pursue distinctive research strategies separately from department and
institute heads, intellectual pluralism and the significance of specialist individual
identities based on expertise will be increased. Together with a professional labour
market that enables firms to acquire new specialist skills and techniques relatively
quickly, this should facilitate firms' access to public research networks.
The fourth aspect of public science systems that affects firm involvement in public
technical communities, and many other aspects of innovation development, concerns
the predominant way in which states and quasi-public institutions set priorities and
implement their policies through them. Although the contrast between "diffusion" and
"mission" oriented state science and technology policies and practices is too simple
to describe the variety of institutional arrangements and goals that states have
established in the late twentieth century to manage public research, it does highlight
important differences between them (Doremus et al., 1998; Ergas, 1987). In
particular, while diffusion oriented policies are concerned to improve technologies
continuously throughout entire sectors, often through joint research activities,
mission-oriented ones focus on mobilising public and private resources to achieve
major public policy goals without much regard for current industrial practices and
capabilities.
14
A key feature of the diffusion oriented policy style is the strong collaboration between
firms, business associations, state agencies and both public and private research
organisations in developing and diffusing new technological knowledge. Typically, the
state provides basic funding for a range of facilities, such as the laboratories of the
Fraunhofer Gesellschaft in Germany, and encourages firms, both individually and
collectively, to organise and fund research projects in them (Abramson et al., 1997).
By underwriting much of the costs associated with technological research and
involving industry associations in its management, states here encourage firm
involvement in the public science system.
Such involvement, however, is usually limited to work on technologies and materials
that are connected to current problems and trajectories rather than with more generic
research that could lead to quite different technologies. Because the primary goal
here is to enhance and improve current industrial competences, diffusion oriented
policies are unlikely to encourage close links with researchers engaged on more
remote topics intended to produce general explanations of phenomena, especially in
academic systems that are strongly structured around discrete disciplines.
Instead it is the combination of strong mission-oriented science and technology
policies, a flexible and pluralistic higher education system and a large programme of
integrated training and research in fields favoured by dominant objectives that seems
most likely to encourage high levels of firm involvement in public science systems.
Substantial state support of research in priority areas is particularly important
because it both underwrites much of the cost of pursuing risky and open-ended
research projects and funds an expansion of training programmes in new skills that
are often quite different from those currently used by firms.
These features seem to have been highly developed in the biomedical sciences in
the late twentieth century USA. Here, the fluidity and pluralism of employment units
and funding arrangements were reinforced by the provision of considerable research
resources by the state and other organisations, typically allocated through
decentralised, peer-reviewed competitive processes. Additionally, the NIH
laboratories provided alternative sources of employment and elite hierarchies to the
leading research universities, and so limited disciplinary elite control over resources
and careers.
As a result, constraints on intellectual novelty were relatively low in US biomedical
fields, but the high rate of competition for scientific reputations – enhanced by the
high number of qualified researchers produced by the graduate schools (Feller,
1999) – ensured considerable coordination of research results and a willingness to
take intellectual risks. Where the knowledge produced by this kind of research
system was directly relevant to R&D activities in firms, as in biotechnology, this
combination of novelty and competition in the public sciences generated a
continuous stream of potentially useful research results that innovative companies
had to keep up with, and hence needed to be involved with research networks. The
strong institutionalisation of the professional researcher role model and frequent job
changes in a competitive scientific labour market, both within the public science
system and between universities and corporate employers, also encouraged the flow
of knowledge and skills between organisations in the USA.
15
This high level of flexibility and pluralism of organisational goals, boundaries and
labour markets enabled researchers to develop novel ideas and approaches more
rapidly than in relatively rigid systems, and to pursue more technological projects
while remaining in academia. It additionally reduced the risks of undertaking
entrepreneurial ventures by allowing well known researchers to return to public
science afterwards and produced a relatively open environment for firms to develop a
variety of cooperative relationships with individual researchers and organisations.
Turning now to consider the second major aspect of innovative competence
development, the degree of firm involvement in industry networks, this is affected by
the nature of scientists' careers and state policies as well as by general institutional
arrangements that encourage, or discourage, cooperation between companies. The
more researchers pursue professional careers with little organisational loyalty and
continuity, the less likely firms are to share knowledge and risks with each other in
industry networks since competitors could easily acquire key staff and thereby
destroy competitive advantages. Greater employer-employee commitment, in
contrast, is associated with the development of more organisation-focused
competences that limit the ability of other firms to appropriate key technological
expertise through industrial networks. Academic systems that reward the pursuit of
specialist reputations based on individual contributions by offering high incentives to
change employers and institutionalise the poaching of stars are not, then, likely to
encourage firms to become involved in such networks.
Diffusion oriented state science and technology policies, on the other hand, often
involve groups of firms in establishing standards, research consortia and diffusing
technological best practice throughout an industry. They therefore encourage
collaboration in developing and applying new technological knowledge to upgrade
their collective capabilities. These policies are implemented more effectively when
strong industry and trade associations have become established that limit
opportunistic behaviour. In coordinated market economies, these groupings organise
negotiations with unions and agree wage increases, restrict poaching of skilled staff,
establish technical standards and generally facilitate cooperation in particular kinds of
"industrial orders" (Herrigel, 1994; 1996). They embed firms in strong industry
networks and obligations that encourage "voice" rather than "exit" modes of
behaviour, and considerable risk and information sharing between companies
(Nooteboom, 2000).
Societies with skill formation systems that are jointly managed by employers, unions
and state agencies also encourage technology sharing and knowledge development
through institutionalising collaboration and establishing common standards for
practical skills. Organised around current industries and technologies, such training
systems reinforce sectoral boundaries and identities as well as skill upgrading within
existing technological trajectories. By the same token, though, they limit the rapid
development of radically new skills and the adoption of quite different technologies
that transcend current industrial boundaries.
Conversely, market economies where: a) inter-firm relations are largely adversarial,
b) commitments between economic agents are limited in scope and duration, and c)
the education and training system is not collaborative in the sense of being jointly
16
organised by employers, unions and state agencies, encourage the academically
successful to invest in generic, portable skills that are valuable on external labour
markets. They also lead to considerable mobility between employers. In such
societies, the less academically successful find it difficult to obtain recognised
training in valued skills, and are usually dependent upon individual initiative and
funding to gain such expertise. As a result, skills are more individually owned and
traded in the more liberal market economies.
In broad terms, then, firms in economies with strong trade, industry and employers'
associations will be deeply embedded in industry networks that facilitate collective
standards setting, joint research activities and cooperation on a range of issues.
When they also establish common wages structures across the industry and are
combined with institutional constraints on firms' ability to hire and fire, such
associations limit labour mobility - especially poaching - and encourage investment
in firm-specific competences through integrating the bulk of employees into full
organisational membership. Skills in such societies are as much organisational as
individual. They are also difficult to change radically in the short to medium term,
although continual incremental improvement will be the norm as both employers and
employees seek growth opportunities through innovations in technologically and
market - related activities.
Involvement in industry networks is also affected by the nature of the financial
system. Often contrasted in terms of its capital market or credit based characteristics
(Zysman, 1983), more recent analyses of how the institutions governing capital flows
to innovating firms affect their behaviour emphasise the insider-outsider dichotomy
(see, e.g. Guerrieri and Tylecote, 1997; Tylecote and Conesa, 1999). Essentially, this
refers to the extent of lock-in effects between investors and entrepreneurs/managers,
and the consequent close coordination of particular capital providers and capital
users. Insider based systems encourage relatively long-term connections between
banks, families and other groups of owners and firms, while more remote outsiderbased ones facilitate the rapid reallocation of capital between firms, sectors and
technologies, as well as limiting the risks attached to any one investment.
As a result, the former type of financial system favours innovating firms that build
long-term organisational competences with business partners and employees to
develop new products and technologies within existing technological trajectories.
Since majority owners are locked-in to the development of particular firms in this kind
of financial system, they have to develop detailed knowledge of each firm they
control and their industry in order to evaluate risks and opportunities adequately to
deal with their greater exposure. They are therefore able to judge innovation
strategies and competences within established industries in a more informed way
than are investors in outsider-dominated financial systems. This means that they can
evaluate and support incremental and long-term technological developments
relatively effectively.
As Tylecote and Conesa (1999) suggest, such insider-dominated financial systems
should, then, be more competitive in industries where innovations are jointly
developed by employers, employees, suppliers and customers and appropriability
risks are reduced by long-term alliances between key actors. Conversely, outsiderdominated ones find it easier to develop radically novel generic innovations that are
17
competence destroying because they facilitate rapid restructuring of assets and
skills, often through venture capital firms that provide high-risk capital for start-ups in
industries where the returns to successful innovations are high and can be
appropriated by the innovating firm and its shareholders.
These institutional arrangements also affect firms' willingness to invest in developing
distinctive organisational competences around core, long term employees. Where
researchers are trained by working on advanced projects with leading scientists and
engineers, are more loyal to their specialism than to particular employers, and expect
to move to gain career advancement, firms are more likely to rely on project teams of
specialists who can be readily hired and fired than make long term commitments.
These conditions are associated with flexible academic systems and strong missionoriented policies that are implemented in a decentralised competitive manner to
provide the basis for highly individual research careers. Equally, where employers
are reluctant to offer stable employment opportunities, engineers and scientists will
prefer to develop their specialist skills in ways that enable them to move between
organisations, and hence are unlikely to invest greatly in creating firm-specific
competences. Weak industry and employer associations, limited cooperation in
training systems and outsider-dominated financial systems reinforce these
tendencies by discouraging collaboration between firms and stability in industry
membership and ownership.
Conversely, where strong industry associations limit individual wage bargaining and
restrict poaching, mobility between employers tends to be lower than in more "liberal"
market economies, and skilled workers have stronger incentives to invest in
enhancing their firm-specific skills (Culpepper, 2001). Employers here have more
encouragement to integrate workers' skills and knowledge into product and process
improvements since they are unable to change them easily and, in effect, much of
the labour force in countries like Germany and Japan is a fixed cost in the medium
term. Labour mobility of technical staff between employers does not appear to be
nearly as great in these countries as in the UK and USA, not least because the risks
of changing organisation are higher and the rewards less obvious (Casper, 2000;
Streeck, 1997). Likewise, employers are often constrained in many of the more
coordinated market economies from rapidly changing the nature of scientific and
engineering skills through hiring and firing by legal rules, works councils pressure,
strong unions and collective bargaining conventions (Soskice, 1997; 1999). This
means that new technologies and capabilities are built more on existing ones, and
are competences are more organisational than individual.
The nature of public science systems and the organisation of capital and labour
markets similarly affect the ability and willingness of firms to transform their
innovative capacities. The more that research training is integrated with the
production of new knowledge, and is supported on a large scale, for example, the
more firms are able to acquire novel research skills and be able to change their
knowledge producing and using capabilities relatively quickly. Depending on the
extent to which academically constituted identities are preferred by engineers and
scientists to organisational ones, such integration may additionally encourage the
development of strong specialist skills that inhibit the establishment of more
organisational capabilities.
18
On the whole, then, integrating research and training through doctoral programmes
that are well funded facilitates the development of innovation strategies based on
rapid access to new knowledge and skills. When coupled with strong professional
researcher identities and fluid labour markets in technical skills, and a flexible,
pluralistic public science system, such training systems facilitate the development of
new technological knowledge and skills that firms can acquire rapidly. Discrete
technological change is more likely to be developed in these kinds of society
because skills and competences can be altered at relatively short notice, and hence
uncertainty managed more easily.
In contrast, strong and effective diffusion-oriented institutions and policies
implemented through powerful industry associations can limit the rate of change of
technological competences and skills by focusing on the continual improvement of
current capabilities within existing technological paradigms. Acquiring radically novel
skills and knowledge to develop new products and processes in newly emerging
technologies will be difficult for most firms in states pursuing such policies because
they are locked into cooperative relationships with suppliers, customers and core
employees based upon the incremental upgrading of current ones. These tendencies
are reinforced by collaborative training systems and insider-dominated financial
systems.
These features of the institutional environment of innovative firms are interconnected
so that distinct kinds of market economies are associated with significant variations in
how innovative competences are developed. For example, countries with outsider
dominated financial systems rarely have strong industry and trade associations. This
reflects the connection between such capital market dominated economies and arms'
length adversarial relations between economic actors, as well as the typically
regulatory state in these kinds of market economies not actively encouraging their
development. However, not all insider-dominated financial systems are associated
with strong employers' associations, as the examples of France and South Korea
indicate. Here, the strong, not to say dirigiste, state inhibits the establishment of
intermediary associations between it and individual companies.
Similar linkages occur between the institutions of liberal market economies in general
and the dominance of professional researcher role models and labour markets. Weak
and/or fragmented industry and employer associations in these kinds of economies
limit both employer-employee and business partner commitments. Together with
strong capital markets and dominating markets for corporate control, the lack of
strong associations in such economies discourages firms from sharing risks and
knowledge with suppliers and customers, and from developing long term
commitments with technical employees. As a consequence, there is little incentive for
engineers and scientists to invest in firm-specific skills, but considerable
encouragement for them to improve specialist skills that are generic across
organisations. Since such societies are often characterised by strong professional
identities and conceptions of high level expertise based on generic knowledge that
are credentialled and controlled by professional associations independent of the
state, the professional researcher model usually dominates that of the organisational
researcher in liberal market economies.
19
Again, though, the reverse relationship does not always hold. The contrast of
Germany and Japan illustrates the variable linkages between strong market
organisation, insider-dominated financial systems and professional identities and role
models. While both exemplify coordinated market economies, they differ significantly
in the strength of their public skill formation systems, and hence in the
institutionalisation of expertise based occupational identities. Although technical
societies in Germany, and similarly organised European societies, are more
integrated into state structures and do not function as labour market controllers to the
same extent as their Anglo-Saxon counterparts, most Germans have a stronger
sense of occupational status based on formally certified expertise than Japanese
employees (see, e.g. Crouch et al., 1999). Researchers in market economies that
have both high levels of coordination and effective public training systems that
generate prestigious, standardised technical skills combine, then, professional and
organisational role models. Where skill formation is overwhelmingly controlled by
employers, the organisational model dominates.
Finally, diffusion-oriented state science and technology policies and institutions are
often associated with strong industrial associations as the state involves them in
diffusing technological knowledge, developing research programmes and
establishing standards. Mission-oriented policies and agencies, on the other hand,
can encourage flexibility in public science systems when combined with
decentralisation of resource allocation decisions through a peer review system, but
need not always do so, as the case of France indicates.
Conclusions
These interconnections suggest that, while public science systems can and do vary
considerably in some respects between market economies, their general
organisational pattern and consequent effects on the development of innovative
competences reflect broader institutional frameworks and priorities. In Japan, for
instance, the long standing concern with economic and technological catching up
with Western Europe and the USA led to the development of a predominantly
diffusion-oriented science and technology system (Ergas, 1987; Morris-Suzuki, 1994;
Odagiri and Goto, 1996), and an educational system that focused on developing
knowledge acquisition skills as much as knowledge production.
It is then the combination of general institutional arrangements governing capital and
labour markets with particular features of public science systems that encourage
firms to develop innovative competences in different ways, and so follow distinctive
innovation strategies. As the Japanese example illustrates, the combination of weak
professional labour markets with a higher education system that focuses more on
producing engineers and scientists who are able to acquire relevant knowledge from
the published literature than on training them to do advanced research limits firms'
involvement in public science networks. Firms here use the public research system
more to find information to solve specific problems than to develop more generic
knowledge that could be used for a range of new products and technologies. Social
identities and loyalties are not so tied to specialist scientific expertise as they are
where the role model of the academic researcher is well established and prestigious,
and so graduates will be more amenable to developing careers and skills within
organisations in these kinds of societies.
20
Together with strong diffusion oriented science and technology policies, these
features encourage firms to focus on developing new products and processes with
general, firm-specific skills that facilitate organisational integration, rather than relying
on more specialist research skills that could coordinate public knowledge production
with corporate purposes. Typically sharing risks and knowledge with suppliers and
customers, these kinds of firms are embedded in industry networks that encourage
alliances and partnerships. Strong industry associations and insider-dominated
financial systems support such long-term commitments and facilitate the speedy
development of new products with a flexible, stable workforce.
Conversely, where coordinated market economies and diffusion-oriented science and
technology policies are combined with stronger public training systems and expertise
based occupational identities, firms are more likely to become involved in public
research networks, albeit more technologically focused ones than in the previous
case. They also will tend to be more supportive of specialist researchers developing
generic skills and pursuing more fundamental research priorities. Consequently, what
might be termed "absorptive hierarchy" types of firms that integrate knowledge and
skills from the applied sciences and industry networks to improve products and
processes are likely to develop in these kinds of market economy.
In the case of arms' length types of market economy that combine weak forms of
market organisation, outsider-dominated financial systems, and professional
researcher role models with relatively flexible public science systems, and
predominantly mission-oriented state policies and practices, firms are likely to adopt
the following forms of competence development. First, they will become quite highly
involved in public research networks in fields where the state has provided
considerable support and/or risk sharing for knowledge production. Second, by
generating large numbers of highly skilled engineers and scientists who seek to
update and improve their expertise, such societies encourage firms to rely on the
specialised expertise of individuals who can be acquired through fluid labour markets
to change their competences rapidly. Such radical shifts in organisational capabilities
are also assisted through the active market in corporate control in these kinds of
economies. Finally, these institutional features discourage extensive and long-term
involvement in industry networks and hence limit investment in customer specific
knowledge and innovations.
They therefore enable project-based firms developing radically discontinuous
innovations with generic, codified knowledge from a variety of fields and novel skills
to dominate industries where appropriability and modularity are high. This
combination of institutional features also, though, encourages more autarchic
innovation strategies. Focused on developing new products and processes in house,
but with limited investment in firm specific skills, "isolated hierarchy" types of firms
(Whitley, 1999) recruit staff with generic skills that can be integrated through authority
hierarchies to achieve development goals. Both labour market institutions and the
educational system encourage individuals to identify more with their specialist
expertise than with particular employers in such economies, so that firms can readily
acquire new skills and knowledge through the labour market. This does however
make coordination across skill areas and functions more difficult and can slow down
the development of new products.
21
In sum, firms have a number of choices in developing innovative competences and
selecting innovation strategies that are guided by dominant institutions. These
institutions include those governing the development of skills and labour markets,
capital markets and inter-firm relationships as well as the organisation and conduct of
research in the public sciences. As a result, societies with distinctive institutional
frameworks encourage the development of particular kinds of innovative capabilities,
and so manifest contrasting types of technological development and sectoral
specialisation, as the examples of late twentieth century Germany, Japan and the
USA illustrate.
22
References
Abramson, H Norman, Jose Encarmacao, Proctor R Reid and Ulrich Schmoch (eds)
(1997) Technology Transfer Systems in the United States and Germany,
Washington, DC: National Academy Press.
Boyer, Robert (1997) "French Statism at the Crossroads," pp 71 -101 in Colin Crouch
and Wolfgang Streeck (eds) Political Economy of Modern Capitalism, London: Sage.
Branscomb, Lewis, Fumio Kodama and Richard Florida (eds) (1999) Industrializing
Knowledge: University-industry linkages in Japan and the United States, Cambridge,
Mass: MIT Press.
Breschi, Stefano and Franco Malerba (1997) “Sectoral Innovation Systems:
Technological Regimes, Schumpeterian Dynamics and Spatial Boundaries,” in
Charles Edquist (ed) Systems of Innovation, London: Pinter, pp 130-156.
Casper, Steven (2000) “Institutional Adaptiveness, Technology policy and the
Diffusion of New Business Models: The Case of German Biotechnology,”
Organization Studies. 21. 887-914.
--------------------- (2001) "The Legal Framework for Corporate Governance: contract
law and company strategies in Germany and the United State," in Peter Hall and
David Soskice (eds) Varieties of Capitalism.
--------------------- and Henrik Glimstedt (2000) "Market Regimes and Competency
Orchestration within the European Internet Economy," unpublished paper presented
to an ESRI workshop held at Sesimbra, Portugal, September.
---------------------, Mark Lehrer and David Soskice (1999) "Can High-technology
Industries Prosper in Germany: Institutional frameworks and the evolution of the
German software and biotechnology industries," Industry and Innovation, 6. 6-23.
Chesbrough, Henry (1999) "The Organizational impact of Technological Change: A
comparative theory of institutional factors," Industrial and Corporate Change, 8. 447485.
Christensen, Clayton M (1997) The Innovator’s Dilemma: When new technologies
cause great firms to fail, Boston: Harvard Business School Press.
Cohen, Wesley M and Daniel A Levinthal (1990) “Absorptive Capacity: A new
perspective on learning and innovation,” Administrative Science Quarterly, 35. 128152.
Coleman, Samuel (1999) Japanese Science: View from the inside, London:
Routledge.
23
Crouch, Colin, David Finegold and Mari Sako (1999) Are Skills the Answer? The
Political Economy of Skill Creation in Advanced Industrial Countries, Oxford: Oxford
University Press.
Culpepper, Pepper D (2001) "Employers' Associations, Public Policy and the Politics
of Decentralised Cooperation," in Peter Hall and David Soskice (eds) Varieties of
Capitalism.
Dasgupta, Partha and Paul David (1994) "Towards a New Economics of Science,"
Research Policy, 23. 487-521.
Doremus, Paul N., W.W. Keller, L. W. Pauly and S. Reich (1998) The Myth of the
Global Corporation, Princeton, N. J.: Princeton University Press.
Dosi, Giovanni, Richard Nelson and Sidney Winter (eds) (2000) The Nature and
Dynamics of Organisational Capabilities, Oxford: Oxford University Press.
Edquist, Charles (ed) (1997) Systems of Innovation: Technologies, Institutions and
Organizations, London: Pinter.
Ehrnberg, Ellinor and Staffan Jacobsson (1997) “Technological Discontinuities and
Incumbents’ Performance: An analytical framework,” in C. Edquist (ed) Systems of
Innovation.
Ergas, Henry (1987) “Does Technology Policy Matter?” in Bruce R. Guile and Harvey
Brooks (eds) Technology and Global Industry: Companies and nations in the world
economy, Washington, D.C: National Academy Press.
Feller, Irwin (1999) "The American University System as a Performer of Basic and
Applied Research," pp 65-101 in Lewis Branscomb, Fumio Kodama and Richard
Florida (eds) Industrializing Knowledge, Cambridge, Mass: MIT Press.
Gambardella, Alfonso (1995) Science and Innovation: The US pharmaceutical
industry during the 1980s, Cambridge: Cambridge University Press.
Guerrieri, Paolo and Andrew Tylecote (1997) “Interindustry Differences in Technical
Changes and National Patterns of Technological Accumulation,” in C. Edquist (ed)
Systems of Innovation pp 107-129.
Hall, Peter and David Soskice (eds) (2001) Varieties of Capitalism: the institutional
foundations of comparative advantage, Oxford: Oxford University Press.
Hancke, Bob (2001) "Revisiting the French Model: Coordination and restructuring in
French industry," in Peter Hall and David Soskice (eds) Varieties of Capitalism.
Henderson, Rebecca, Luigi Orsenigo and Gary Pisano (1999) "The Pharmaceutical
Industry and the Revolution in Molecular Biology: Interactions among scientific,
institutional, and organisational change," pp 267-311 in David Mowery and Richard
Nelson (eds) Sources of Industrial Leadership, Cambridge: Cambridge University
Press.
24
Herrigel, Gary (1993) "Large Firms, Small Firms, and the Governance of Flexible
Specialization: the case of Baden Württemberg and socialized risk," pp 15-35 in
Bruce Kogut (ed) Country Competitiveness: Technology and the organizing of work,"
Oxford: Oxford University Press.
Herrigel Gary (1994) Industry as a Form of Order “ in R. Hollingsworth, P. Schmitter
and W. Streeck, (eds). (1994) Governing Capitalist Economies, Oxford University
Press.
Herrigel, Gary (1996) Industrial Constructions: The Sources of German Industrial
Power, Cambridge University Press.
Hicks, Diana (1995) "Published Papers, Tacit Competencies and Corporate
Management of the Public/Private Character of Knowledge," Industrial and Corporate
Change, 4. 401-424.
Hollingsworth, Rogers and Robert Boyer (eds) (1997) Contemporary Capitalism: The
embeddedness of institutions, Cambridge University Press.
Kenny, Martin (1986) Biotechnology: The university-industry complex, New Haven:
Yale University Press.
Kneller, Robert (1999) "University-Industry Cooperation in Biomedical R&D in Japan
and the United States: Implications for biomedical industries," pp 410-438 in Lewis
Branscomb et al (eds) Industrializing Knowledge.
Kristensen, Peer Hull (1992) 'Strategies against Structure: Institutions and economic
organisation in Denmark', in Richard Whitley (ed) European Business Systems,
London: Sage Publications.
Kristensen, Peer Hull (1996) On the constitutions of economic actors in Denmark:
interacting skill containers and project coordinators’, in Richard Whitley and Peer Hull
Kristensen, (eds.) The Changing European Firm, London: Routledge, 118-158.
Langlois, Richard and David Mowery (1996) "The Federal Government Role in the
Development of the U.S. Software Industry", pp 53-85 in David Mowery (ed) The
International Computer Software Industry: A comparative study of industry evolution
and structure, Oxford: Oxford University Press.
Langlois, R. N. and Robertson, P. L. (1995) Firms, Markets and Economic Change,
London: Routledge.
Lehrer, Mark (2000) "Has Germany Finally Fixed its High-tech Problem?" California
Management Review, 42. 89-107.
Mansfield, Edwin (1995) “Academic research underlying industrial innovations:
Sources, characteristics, and financing,” The Review of Economics and Statistics, 77.
55-65.
25
McKelvey, Maureen (1996) Evolutionary Innovations: the business of biotechnology,
Oxford: Oxford University Press.
Morris-Suzuki, Tessa (1994) The Technological Transformation of Japan: From the
seventeenth to the twenty-first century, Cambridge: Cambridge University Press.
Narin, Francis, Kimberly Hamilton and Dominic Olivastro (1997) “The Increasing Link
between US Technology and Public Science,” Research Policy, 26. 317-330.
Nelson, Richard and Sidney Winter (1982) An Evolutionary Theory of Economic
Change, Cambridge, Mass: Harvard University Press.
Nooteboom, Bart (2000) "Institutions and Forms of Cooperation," Organization
Studies, 21. 915-940.
Odagiri, Hiroyuki and Akira Goto (1996) Technology and Industrial Development in
Japan: Building capabilities by learning, innovation and public policy, Oxford: Oxford
University Press.
Ogura, Seiritsu and Hiroto Kotake (1999) "Public Policies for Japanese Universities
and the Job Market for Engineers," pp 547-576 in Lewis Branscomb et al (eds)
Industrializing Knowledge.
Saxenian, Annalee, Regional Advantage: Culture and Competition in Silicon Valley
and Route 128, Cambridge, Mass: Harvard University Press.
Sienko, Tanya (1997) A Comparison of Japanese and U.S. Graduate Programs in
Science and Engineering, Tokyo: National Institute of Science and Technology
Policy, Discussion Paper no. 3.
Soskice, David, (1997) “German Technology Policy, Innovation and National
Institutional Frameworks,” Industry and Innovation, 4. 75-96.
Soskice, David (1999) "Divergent Production Regimes: Coordinated and
uncoordinated market economies in the 1980s and 1990s," pp 101-134 in Herbert
Kitschelt, Peter Lange, Gary Marks and John Stephens (eds) Continuity and Change
in Contemporary Capitalism, Cambridge: Cambridge University Press.
Streeck, Wolfgang (1997) "German Capitalism: Does it exist? Can it survive?" pp 3354 in Colin crouch and Wolfgang Streeck (eds) Political Economy of Modern
Capitalism, London: Sage.
Tate, Jay (2001) "National Varieties of Standard-setting," in Peter Hall and David
Soskice (eds) Varieties of Capitalism.
Teece, David (1986) "Profiting from Technological innovation: Implications for
integration, collaboration, licensing and public policy," Research Policy, 15. 285-305.
Tylecote, Andrew and Emmanuelle Conesa (1999) "Corporate Governance,
Innovations Systems and Industrial Performance," Industry and Innovation, 6. 25-50.
26
Tushman, Michael L and Philip Anderson (1986) “Technological Discontinuities and
Organizational Environments,” Administrative Science Quarterly, 31. 439-465.
Westney, Eleanor (1993) "Country Patterns in R&D Organisation: The United States
and Japan," pp 36-53 in Bruce Kogut (ed) Country Competitiveness, Oxford: Oxford
University Press.
Whitley, Richard (1999) Divergent Capitalisms: The social structuring and change of
business systems, Oxford: Oxford University Press.
Whitley, Richard (2000) The Intellectual and Social Organization of the Sciences,
Oxford: Oxford University Press. Second edition.
Whitley, Richard and Peer Hull Kristensen (eds) (1996) The Changing European
Firm: Limits to Convergence, London: Routledge.
Yamamoto, Shinichi (1999) "The Growing Sophistication of research at a Time of
Broadened Participation in Higher Education," pp 531-546 in Lewis Branscomb et al
(eds) Industrialising Knowledge.
Yoshihara, Mariko and Katsuya Tamai (1999) "Lack of Incentive and Persisting
Constraints: Factors hindering technology transfer at Japanese universities," pp 348364 in Lewis Branscomb et al (eds) Industrializing Knowledge.
Zucker, Lynne and Michael Darby (1997) "Present at the Biotechnological
Revolution: Transformation of technological identity for a large incumbent
pharmaceutical firm," Research Policy, 26. 429-446.
Zysman, John (1983) Governments, Markets and Growth, Ithaca: Cornell University
Press.
27
TABLE 1
Connections between Institutional Frameworks and the Development of Innovative Competences
Developing Innovative Competences through:
Features of
institutional
Frameworks
Passive
involvement in
the public
science system
Generic
involvement in
the public
science system
Involvement in
industry
networks
Reliance on
specialist skills
Rate of change
of innovative
competences
Negative
Technologically
specific
involvement in
the public
science system
Positive
Large training
system
integrated with
research
Flexible,
pluralistic public
science system
Strong
professional
researcher role
model
Diffusionoriented policies
Strong industry
and employers
associations
Collaborative
training system
Insiderdominated
financial system
Positive
Varies
Positive
Positive
Negative
Positive
Positive
Varies
Varies
Positive
Negative
Weakly positive
Positive
Negative
Positive
Positive
Positive
Positive
Negative
Positive
Negative
Negative
Varies
Weakly positive
Varies
Strongly positive
Negative
Negative
Weakly negative
Positive
Varies
Positive
Negative
Negative
Varies
Weakly positive
Weakly negative
Positive
Negative
Negative
29