Innovation, Learning Organizations and Industrial Relations

Abstract

Innovation may be seen as a process of knowledge creation and the speed and direction of knowledge creation reflects the organizational set-up of the firm as well as its investments in R&D and training. Establishing ‘a learning organization’ where horizontal interaction and communication inside and across the borders of the firm is a major factor promoting knowledge creation in the context of a learning economy. An important issue is to what extent direct and indirect participation of employees in shaping the new form of organization is critical for its realization. On the basis of a unique data set covering 2000 Danish private firms it is demonstrated that firms combining several of the organizational traits of the learning organization are much more prone to introduce new products than the others. It is also demonstrated that such firms have involved employees in different forms of direct and indirect participation much more frequently than the rest. As more sectors become exposed to the need to engage in incremental product and service innovation the economic potential of diffusing good practices in terms of organization and participation is growing and needs to be reflected in firm strategies and public policies aiming at promoting innovation and knowledge creation.

Full text

DANISH RESEARCH UNIT FOR INDUSTRIAL DYNAMICS
DRUID Working Paper No 03-07
Innovation, Learning Organizations
and Industrial Relations
by
Peter Nielsen and Bengt-Åke Lundvall

Innovation, Learning Organizations and Industrial Relations
Peter Nielsen
Center of Labor Market Studies
Aalborg University
And
Bengt-Åke Lundvall
Department of Business Studies
Aalborg University
bal@business.auc.dk
Abstract
Innovation may be seen as a process of knowledge creation and the speed and direction of
knowledge creation reflects the organizational set-up of the firm as well as its investments in
R&D and training. Establishing ‘a learning organization’ where horizontal interaction and
communication inside and across the borders of the firm is a major factor promoting knowledge
creation in the context of a learning economy. An important issue is to what extent direct and
indirect participation of employees in shaping the new form of organization is critical for its
realization.
On the basis of a unique data set covering 2000 Danish private firms it is demonstrated that firms
combining several of the organizational traits of the learning organization are much more prone
to introduce new products than the others. It is also demonstrated that such firms have involved
employees in different forms of direct and indirect participation much more frequently than the
rest. As more sectors become exposed to the need to engage in incremental product and service
innovation the economic potential of diffusing good practices in terms of organization and
participation is growing and needs to be reflected in firm strategies and public policies aiming at
promoting innovation and knowledge creation.
Key words: Innovation, knowledge creation, learning economy
JEL Codes: L22, O31, O32
.
ISBN 87-7873-139-9

1. The learning economy as context
It is important to reflect on why the impact of new organization forms and innovation capacity
upon economic performance has increased through the 90s. As we get a better understanding of
the mechanisms at work, it becomes easier to sort out what trends will dominate the future. Here
we assume that the changes in the economic environment can best be captured by the concept ‘the
learning economy’ (Lundvall and Johnson 1994).
In various contexts we have introduced an interpretation of what actually takes place in the
economy under the term ‘the learning economy’ (Lundvall and Johnson 1994; Lundvall and
Nielsen 1999). The intention is that the term should mark a distinction from the more generally
used term ‘the knowledge-based economy’. The learning economy concept signals that the most
important change is not the more intensive use of knowledge in the economy but rather that
knowledge becomes obsolete more rapidly than before; therefore it is imperative that firms
engage in organizational learning and that workers constantly attain new competencies. This can
be illustrated by the reference in a recent report from the Danish Ministry for Education to a
German source that claims that half of the skills a computer engineer has obtained during his
training will have become obsolete one year after the exam has been passed, while the ‘halving
period’ for all educated wage earners is estimated to be eight years (Ministry of Education 1997,
p. 56).
A learning economy is thus one in which the ability to attain new competencies is crucial to the
economic success of individuals and to the performance of firms, regions and countries. The
background for the crucial importance of learning is that the combination of globalization,
information technology and deregulation of formerly protected markets leads to more intense
competition and to more rapid transformation and change. Both individuals and companies are
increasingly confronted with problems that can be solved only through new competencies. The
rapid rate of change is reinforced by the fact that the intensified competition leads to a selection of
organizations and individuals that are capable of rapid learning, thus further accelerating the rate
of change.
The transition to a learning economy confronts individuals and companies with new demands. We
see the growing emphasis on new organization forms promoting functional flexibility and
networking as responses to the challenge of the learning economy. In a rapidly changing
environment it is not efficient to operate in a hierarchical organization with many vertical layers.
It takes too long to respond if the information obtained at the lower levels should be transmitted
to the top and back down to the bottom of the pyramid. In many instances relational contracting
and networking enhance functional flexibility.
One important result from what follows is that the new organization forms which tend to support
competence building through learning by doing and interacting tend to speed up product or
service innovation. In the next section we will focus on how knowledge creation relates to
innovation and to the establishment of learning organizations.
2. Innovation and knowledge creation
Most authors using the concept of knowledge creation and knowledge production refer to
technological knowledge and to technical innovation as the output of the process (Nonaka and
Takeuchi 1995). In the new growth theory, the output of the R&D sector is viewed either as a
blueprint for a new production process that is more efficient than the previous one or as a

production of new semi-manufactured goods that cannot not easily be copied by competitors.
A striking characteristic of knowledge production resulting in innovation is the fact that
knowledge, in terms of skills and competencies, is the most important input. In this sense, it
recalls a ‘corn economy’, in which corn and labour produce corn. But it differs from such an
economy in one important respect. While the corn used to produce corn disappears in the process,
skills and competencies improve with use. Important characteristics of knowledge reflect that its
elements are not scarce in the traditional sense: the more skills and competencies are used, the
more they develop. This points to knowledge production as a process of joint production, in
which innovation is one kind of output and the learning and skill enhancement that takes place in
the process is another.
Innovation as the outcome of knowledge production
There are two reasons for focusing on innovation when it comes to understand knowledge
production. One is that innovation represents – by definition – something new and therefore adds
to existing knowledge. The second is that innovation is a process where the innovating unit
operates under uncertainty and therefore regularly is confronted with unforeseen problems.
There are important sector differences in knowledge production. Such differences are reflected in
the character, the mode and the outcome of the innovation process. The taxonomy developed by
Keith Pavitt (1984) represents an important effort to capture these differences systematically, so
far as the manufacturing sectors are concerned. Pavitt defined four categories of firms and sectors.
- supply-dominated sectors (e.g. clothing, furniture)
- scale-intensive sectors (e.g. food, cement)
- specialized suppliers (e.g. engineering, software, instruments)
- science-based producers (e.g. pharmaceuticals, biotechnology, electronics)
For a long time, knowledge production/innovation processes were considered largely as the
province of the fourth category, and still there is a bias in this direction. Often this was combined
with a linear view which assumed that new scientific results are the first step in the process,
technological invention the second step, and the introduction of innovations as new processes or
products the third. There is now a rich body of empirical and historical work which shows that
this is the exception rather than the rule (Rothwell 1977; von Hippel 1988; Lundvall 1988). The
recent models of innovation emphasize that knowledge production/innovation is an interactive
process in which firms interact with customers, suppliers and knowledge institutions. Empirical
analysis shows that firms seldom innovate alone (Lund Vinding 2002).
This is also the background for developing a systemic approach to knowledge production.
Innovations systems are constituted by actors involved in innovation and their interrelationships.
The actors are firms, technological institutes, universities, training systems and venture capital.
Together they constitute the context for knowledge production and innovation. The specific
constellations differ across sectors, regions and nations. They are typically specialized in terms of
their knowledge base, and the specific mode of innovation will reflect institutional differences.
This is the background for the growing literature on innovation systems (Freeman 1987; Lundvall
1992; Nelson 1993; Edquist 1997) and technological systems (Carlsson and Jacobsson 1997).

Innovation systems may be defined as regional or national, as well as sector- or technology-
specific. The common idea is that the specificities of knowledge production reflect unique
combinations of technological specialization and institutional structure.
Competence as the outcome of knowledge production
The change from a linear to an interactive view of innovation and knowledge production has also
been a way to connect to each other innovation and the further development of competence. As
now understood, the innovation process may be described as a process of interactive learning in
which those involved increase their competence while engaging in the innovation process.
In economics, there have been various approaches to competence-building and learning. One
important contribution is Arrow’s analysis of ‘learning by doing’ (1962), in which he
demonstrated that the efficiency of a production unit engaged in producing complex systems
(aeroplane frames) grew with the number of units already produced and argued that this reflected
experience-based learning. Later, Rosenberg (1982) introduced ‘learning by using’ to explain why
efficiency in using complex systems increased over time (the users were airline companies
introducing new models). The concept of ‘learning by interacting’ points to how interaction
between producers and users in innovation enhances the competence of both (Lundvall 1988). A
more recent analysis of learning by doing focuses on how confronting new problems in the
production process triggers searching and learning, which imply interaction between several
parties as they seek solutions (von Hippel and Tyre 1995).
In most of the contributions mentioned above, learning is regarded as the unintended outcome of
processes with a different aim than learning and increasing competence. Learning is seen as a
side-effect of processes of production, use, marketing, or innovation. An interesting new
development, which tends to see learning as a more instrumental process, is the growing attention
given to ‘learning organizations’ (Senge 1990). The basic idea is that the way an organization is
structured and the routines followed will have a major effect on the rate of learning that takes
place. The appropriate institutional structures may improve knowledge production in terms of
competence building based on daily activities (Andreasen 1995).
The move towards learning organizations is reflected in changes both in the firm’s internal
organization and in inter-firm relationships. Within firms, the accelerating rate of change makes
multi-level hierarchies and strict borders between functions inefficient. It makes decentralization
of responsibility to lower-level employees and formation of multi-functional teams a necessity.
This is reflected in the increasing demand for workers who are at the same time skilful, flexible,
co-operative and willing to shoulder responsibility. It is also reflected in relationships with
suppliers, customers and competitors becoming more selective and more intense. ‘Know-who’
becomes increasingly important in an economy that combines a complex knowledge base and a
highly developed, rapidly changing specialization.
Apart from these organizational changes, there is a growing emphasis on making employees and
teams of employees more aware of the fact that they are engaged in learning. It has been
suggested that second-loop learning and deutero learning, i.e. a process in which agents reflect on
what has been learnt and on how to design the learning process, is much more efficient than
simply relying on the impact of experience (Argyris and Schoen 1978).

A blurring of the distinction between production of knowledge as a separate (off-line) activity and
as a by-product of regular routine activities (on-line)
It is useful to separate two different perspectives on the process of knowledge production which
are not mutually exclusive in real life but which can be found, in more or less pure form, in the
literature on innovation and growth. On the one hand, one might look for a separate sector in
charge of producing new knowledge or handling and distributing information. Such a sector
might, for instance, involve universities, technical institutes and government S&T policies, as
well as R&D functions in firms. Here, the production of knowledge is assumed to take place as a
deliberate activity, outside the realm of production.
On the other hand, one might regard the creation and diffusion of knowledge as rooted in and
emanating from routine activities in economic life, such as learning by doing, using and
interacting. Here, the production of knowledge is assumed to take place as a by-product of in
daily activities, through learning by doing or learning by using.
This distinction, between deliberate and non deliberate forms of knowledge production, can also
be referred to as respectively ‘off-line’ and ‘on-line’ learning activities. Above we referred to the
growing focus on establishing learning organizations. Another related new trend is the emergence
of a form of learning qualified as ‘experimental’ using ‘the factory as a laboratory’. Both these
forms relate to learning, taking place ‘on line’ (that is to say, during the process of producing the
good or providing the service) but they reflect deliberate efforts to enhance competence building.
The possibility of moving to this type of learning in more and more activities represents one
important transition in the historical evolution of the learning economy. In effect, as long as an
activity remains located either in a separate knowledge sector or gives rise only to non-reflective
learning there remains a distinct cleavage between those who deliberately produce knowledge and
those who are expected to use and exploit it. When more activities move to higher forms of
learning where the individual can program experiments and reflect on results, the production of
knowledge tends to become a more collectively distributed responsibility. The distance between
the laboratory and the shop floor becomes shorter.
3. Research questions to be illuminated
This paper tests and finds support for the following set of hypotheses:
- the probability of successful product innovation increases when the firm has organized
itself in such a way that it promotes learning.
- organizational forms promoting learning have to be seen as multi-dimensional - they
typically combine several of a number of internal relationships and activities and external
relationships.
- internal relationships and activities include integrative organizational forms, quality
management, human development efforts and compensation systems.
- external relationships include relationships to suppliers, customers and knowledge
institutions.

- firms that go far in introducing organizational practices to promote learning tend to
combine indirect representation of workers with direct representation.
It also tests but finds no support for the following two hypotheses:
- there is a time lag from an organizational set up promoting learning to the outcome in
terms of frequent product innovations.
- the bigger the share of the labor force within the firm that is involved in a specific
learning promoting organizational form the stronger the innovative activities.
In the first part of the empirical analysis we explore factors that lie behind product innovation in
2000 Danish firms belonging to manufacturing, construction and service industries. Most of the
innovations are incremental rather than radical - most of them signal something new for the firm
but not something new for the national economy or the world economy. We do not see that as a
weakness of the analysis, however. Especially in a small open economy of the Danish type,
competitiveness and economic growth will depend on continuous incremental upgrading of
products. ‘Domestic’ radical innovations will not be frequent and there is a high probability that
they will be taken over by foreign big firms before they leave strong marks on the growth pattern
of the whole economy.
We will demonstrate that there are important sector differences in the frequency of innovation and
that big firms innovate more than small ones. Most importantly we will show that bundling a
number of different characteristics having to do with a learning organization has a major effect on
the innovation capability. The functionally flexible firms that engage in networking with
customers and suppliers will ceteris paribus be more active in terms of product innovation than
the firms that operate in a more traditional way.
In the second part of the empirical analysis we explore the direct and indirect participation of
workers. We find that there is a connection between the strength of the influence and at what
stage of decision making employees are engaged in the processes of organizational change. We
also find that while the tendency is towards more direct and less indirect participation the most
advanced firms in terms of establishing learning organizations are the ones that most frequently
use indirect forms in the organization development process.
On the basis of our theoretical considerations and empirical results it is fair to conclude that
incremental innovation and learning are two sides of the same coin. It is true that ‘learning
organizations’ are more apt when it comes to mobilize and utilize different sources of knowledge
when it comes to develop new products. But it is also true that the very presence of incremental
innovation increases the need for an organizational framework that can cope with new problems
as they pop up during the innovation process. In terms of innovation policy it might therefore be a
good idea to combine initiatives promoting organizational change with initiatives that give better
opportunities and incentives to engage in product innovation.
4. Data
The empirical analysis is based on a survey addressed to all Danish firms in the private sector –
not including agriculture - with 25 or more employees, supplemented with a stratified
proportional sample of firms with 20-25 employees. 6991 questionnaires were sent to the firms

selected. This survey were planned to collect information from management as well as from
employee representatives by means of two separate questionnaires, implemented through two
phases of collection in each of the firms selected.
The overall design is comparable to the workplace employee- or industrial relation surveys, which
have been carried out in Britain, Canada, Australia and France (Tomlinson 1999; Coriat 2001).
This means that one aim of the survey has been to establish a set of panel data that connects a
survey from 1996 (DISKO), covering the period 1993-95, to this 2001 survey covering the period
1998-00. Information links between the two surveys are established by comprehensive register
data from the Danish Integrated Database of Labor Market Research (IDA) and the register of
Business Data at Denmark Statistics.
I order to construct the panel the core of this survey sample had to be 1363 firms from the 1996
DISKO survey, which had ‘survived’ as units in the IDA-database. From those 1363 firms,
longitudinal data were already available from 1990 to 1997, covering information on firm
performance and employee education, -development and -turnover. The data collection addressed
to this sample resulted in 637 useable responses. The rate of response among the core group of
firms is 47%, which is similar to the rate of response on the first (DISKO) survey, which was 48%
(Lund 1998; Lund and Gjerding 1996).
In total, 2007 usable responses from management and 473 responses from employee
representatives have been collected and integrated in a cross section data set from this last survey.
This makes the overall response rate of the survey 29%, which is not very satisfying. However, a
closer response analysis broken down on industries and size show acceptable variations on
response rates here, and non-respondent information on some of the potential dependant variables
together with comparison to other surveys, do not indicate unacceptable bias.
To this survey data is linked 1998 and 1999 statistics from the Danish Integrated Database of
Labor Market Research and the register of Business Data at Denmark Statistics in such a way, that
integration between the two main dataset has been possible, and the longitudinal data set from
1990 to 2001 is established. The elements of the ‘Innovation, Organization and Competence
Panel’ are shown in Appendix A, in order to give a general overview of the data set.
5. The frequency of product innovation in the surveyed firms
The dependent variable in most of the analysis below is constructed on the basis of management
response to a question about product or service innovation in the period 1998-2000. It may be
argued that product innovation is important both for the economy as a whole and for the single
firm. It is a necessary precondition for economic growth in a closed economy – just sticking to the
old set of products would lead to saturation of markets as productivity increases (Pasinetti 1981)
and it is certainly important in relation to international competitiveness. For the single firm one
way to cope with increasing competition is to introduce products with new and superior use value
characteristics.
Obtaining a meaningful quantitative measure of innovation and innovative behavior on the basis
of information collected in firms belonging to industries with very different conditions, is not
unproblematic. The phenomenon that firms refer to may vary in relation to conditions and
configurations. The fact is that we are confronted with qualitative change rather than change
easily captured in quantitative terms when we ask the firms whether they, in the period of 1998 -

2000, have introduced new products or services on the market.
Table 1: Product or service innovation 1998-2000 by firm size, industry, group ownership
and production (percent horizontal)
P/S innovati
Not P/S innov
Don’t know
(N)
All Firms
45,4
52,4
2,2
1974
Less than 50
36,5
60,9
2,6
1022
50 - 99 Empl
47,3
52,0
0,7
433
100 and more
62,6
34,7
2,7
487
Manufacturing
58,1
40,8
1,1
723
Construction
21,7
75,7
2,6
309
Trade
41,4
55,6
3,1
549
Other services
31,1
67,1
1,8
164
Business serv.
58,7
37,6
3,8
213
Danish group
48,6
49,6
1,7
693
Foreign group
59,7
38,4
1,8
385
Single firm
36,6
60,7
2,7
882
Table 1 shows that almost a half of the firms introduce at least one new product or new service
over a two year period. It also shows that the bigger the firm the greater the probability that it
introduces a new product. There are substantial sectoral differences as well: Firms belonging to
manufacturing and business services are much more active in terms of product innovations than
the average while construction firms are far below the average. Firms owned by foreign groups
have the highest innovation score, firms owned by Danish groups medium score and single - stand
alone - firms the lowest score in terms of product innovation.
As indicated in an earlier section most of the innovations are not ‘new on the market’. In order to
uncover the various degrees of radical innovation in relation to the market context, we have asked
firms whether their innovations already existed on the national market or existed on the world
market. The result appears from the table 2.

Table 2: P/S innovations and their novelty on the Danish and the world market in percent
(N=814)
Known on worl
d
market
New to world marke
t
Don’t know
Known on Danish
market
74,1
0,9
1,8
New to Danish
market
13,1
6,1
1,7
Don’t know
0,6
0,0
1,6
Table 2 shows that three-fourth of the innovations introduced within the period 1998-2000, were
already known at the national as well as well as on the international markets. Firms representing
this group of innovators may be termed ‘local’, as the product or service innovations are only new
to the firms, which produced them, not new on the market. The next step is the innovations new
on the national market, although already existing in world markets. 13% of the firms have
introduced at least one innovation in this category. Finally, we find the small group of firms (6%),
which have introduced at least one innovation new, both on the national and the world market.
6. The frequency of organizational change in the surveyed firms
The theoretical and practical importance of introducing new forms of management and building
new work organizations is not at all new. This importance is for instance at the core of the
theoretical discussion of developments in techno-economic regimes as understood by Freeman
and Perez (1988). In the management literature, concepts such as quality control, diversification
and flexibility tend to form new global ‘institutional standards’, value systems confronting firms
as clues to competitive advantages in the new international division of labor. They are also
promoted by consultancy firms operating on a world wide basis (Røvik 1992; Røvik 1998). These
are some of the reasons why the empirical research in new organization practices gathered
momentum worldwide in the nineties.
Table 3 gives a picture of the frequency of organizational change in different part of the Danish
economy. It is based on the assessment of representatives of management and reflect a response to
a question regarding the occurrence or not of important organizational change in the period 1998-
2000.

Table 3: Organizational changes 1998-2000 by firm size, industry, group ownership and
production (percent horizontal)
Org. changes No org. chag Don’t know (N)
All firms 52,0 46,9 1,2 1992
Less than 50 emp 40,1 59,3 0,6 1037
50 - 99 employee 56,6 40,9 2,5 435
100 and more 73,2 25,6 1,2 488
Manufacturing 58,9 39,6 1,5 722
Construction 33,4 65,6 1,0 314
Trade 51,1 48,0 0,9 558
Other services 46,7 51,7 1,7 182
Business services 62,8 36,7 0,5 215
Danish group 57,5 41,5 1,0 694
Foreign group 71,1 28,4 0,5 388
Single firm 39,7 58,8 1,6 895
Standard product 56,1 42,9 1,0 725
Customized product
50,9 47,8 1,3 1188
Table 3 shows that slightly more than half of the firms responding to the survey have introduced
significant change in their organization 1998-2000. The frequency in this period is the same as it
was in the period 1993 -95, where the same question was directed to 1900 firms from a similar
(and partially overlapping) population.
There are differences between different categories of firms. The larger the firm the higher is its
propensity to engage in organizational change. Almost three-fourth among the 100 + firms have
undertaken organizational changes. But even among the small firms with less than 50 employees,
two-fifth have engaged in changes.
Differences between the various industries are present without being particularly pronounced, and
they are in harmony with the findings in the mid-nineties. Business service and Manufacturing
score high. Trade is at the average level while Other Service and Construction lie below the
average. Foreign owned companies have a high proportion of organizational change and stand-
alone firms have the lowest proportion. The difference between firms engaged in standard product
and those producing individualized products is small.
As the next step we shall study the various reasons why organization changes were initiated. All
the firms that had reported that they had introduced a major change were asked about what was

their aim with introducing it. Table 4 summarizes the different response pattern.
Table 4: Aim of organizational changes
High exten
t
Some exten
t
Small
extent
Not at all
Don’t
know
(N)
Increase effectiveness of
daily work
71,7
25,1
1,7
1,5
0,1
1016
Improve quality and
customer service
53,6
37,0
6,2
3,0
0,3
1012
Strengthen cooperation and
coordination across the
organisation
51,7
37,0
7,9
3,2
0,3
1003
Adapt to turbulent
environments
47,5
39,8
9,7
2,6
0,4
1002
Strengthen and renew
knowledge
29,7
46,4
17,5
5,4
1,1
1003
Develop new products or
services
28,4
40,3
20,7
9,5
1,1
1000
More than two out of three firms refer to the promotion of work efficiency as an important aim
with changing the organization. It is hardly surprising that such a general aim as ‘efficiency’
scores high. To enhance quality, strengthen internal and external cooperation and increase
adaptability are quoted as highly important by half the firms. Strengthening and renewing the
knowledge of the firm and its know-how together with continuously product/service innovation
are seen as important aims by less than a third of the firms. According to the response pattern
most firms are obviously aiming at efficiency and functional flexibility when changing their
organization. But it is also interesting to note that only one out of four firms says that enhancing
the knowledge base of the firm is of minor or no importance.
One interesting issue in this context is whether and to what extent the empirical dimensions in the
table are correlated and represent common latent factors with multiple dimensions. Dimensions
which all are important for the firms as aims of their organizational changes. This question can be
investigated in a factor analysis, that uncovers latent factors behind empirical dimensions. Results
from the factor analysis are presented in the table 5.

Table 5: Factor analysis of the aims behind organizational change.
Factor 1: Innovatio
n
and knowledge
Factor 2
:
Effectiveness
Strengthen and renew knowledge
0,76
-0,31
Develop new products or services
0,73
-0,41
Adapt to turbulent environments
0,64
-0,13
Improve quality and customer service
0,63
0,27
Strengthen cooperation and coordinatio
n
across the organisation
0,52
0,25
Increase effectiveness of daily work 0,41
0,78
Table 5 shows that the factor analysis uncovers two latent factors of importance as aims of
organizational changes. The first factor covers specific aims, such as “the ability continuously to
strengthen and renew knowledge and know-how”, “continuously to develop new products or
services”, “adapt to turbulent environments”, “quality and customer service” and “cooperation
and coordination across the organization”. The second factor covers only the dimension of
“effectiveness of the daily work”.
This analysis of how the aims behind organizational changes are interrelated first of all gives
evidence of one multiple aim, covering dimensions which represent strengthening of knowledge,
communication and innovation together with quality and adaptability in relation to business
environments. The other factor covers the single aim of enhancing the effectiveness of daily work.
What is of special interest here is that the firms giving strong emphasis to elements involved in
creating a functionally flexible organization ( adaptability to change, internal co-operation and
responsiveness to customers) seem to combine such aims with a strong emphasis on innovation
and knowledge creation. This might be taken as a first indication that aiming at a functionally
flexible organization is also to aim at establishing a learning and an innovating organizations.
This is a perspective that will by analyzed in more detail in the next section.
7. Bundling organizational characteristics and analyzing the impact on
innovation
The globalization of competition and changes in consumer demand through the eighties and
nineties have promoted development of universal values of organizational solutions, such as
delegation of responsibility, decentralized internal and external communication, quality
management and competence building. Researchers and experts have developed and/or confirmed
these values to what has been called “institutional standards” (Røvik, A. 1992 1998), confronting
the firms as solutions of their problems or challenges in the globalized economy.
In the survey, we have measured the incidence of an array of organizational dimensions, which all
directly or indirectly refer to contemporary theories dealing with innovation and functional

flexibility in organizations: Cross occupational work groups, integration of functions, softening
demarcations, delegation of responsibility and self directed teams are empirical indicators,
referring to Moss Kanter’s theory of integrative organization and Burn’s & Stalker’s organic
organizations. Quality circles and proposal collection systems are indicators of Quality
management (TQM) and Knowledge Management (Nonaka & Takeuchi 1995). Tailored
educational system and Educational planning indicate Human Resources Development and
cooperation with eksternal actors refer to Innovation systems (Lundvall 1992). In the figure below
the dimensions are classified in relation to the theoretical aspects they are indicators of.
Figure 1: Theoretical perspectives and organizational characteristics and practices
Theoretical perspective Organizational characteristics and practises
The organic and integrative organization
– focus on internal functional flexibility
Cross occupational working groups
Integration of functions
Softened demarcations
Delegation of responsibility
Self directed teams
Quality management – focus on engaging
employees
Quality circles/groups
Systems for collection of employee proposals
Human development – focus on
competence building
Education activities tailored to the firm
Long term educational planning
Compensation system – focus on
incentives
Wages based on qualifications and functions
Wages based on results
External communication – focus on
external functional flexibility
Closer cooperation with customers
Closer cooperation with subcontractors
Closer cooperation with universities & technological institutes
In order to examine whether the dimensions cover underlying or latent but more general variables,
possibly in accordance with the above-mentioned theories, a factor analysis has been performed
on the use of the organizational dimensions in the work processes of the firms, as the next
analytical step. Besides the theoretical interests it is known from empirical research, that clusters
of dimensions further productivity as well as innovation behaviour in firms (Dyer and Reeves
1995, Huselid 1995, Huselid et.al. 1996, Wood 1999, Osterman 2000, Lund Vinding 2001,
Laursen 2001). Exploring clusters of dimensions should lead us forward in the analysis of the
“anatomy” of the knowledge organizations. The results of the factor analysis are shown in the

table below.
Table 6: Factor analysis: dimensions of organization, communication, human development,
cooperation and compensation
Faktor 1
:
Integra-tive
organization
Faktor 2
:
Human
developmen
t
Faktor 3
:
Eksternal
cooperation
Faktor 4
:
Com-
pensation
systems
Cross occupational working groups
0,66
0,14 0,07 0,12
Quality circles/groups
0,62
0,11 0,02 0,01
Self directed teams
0,61
0,04 -0,02 0,15
Integration of functions
0,59
-0,01 0,08 0,03
Delegation of responsibility
0,55
0,08 0,02 0,38
Systems for collection of employee proposals
0,48
0,11 0,18 0,12
Softened demarcations
0,40
0,10 0,24 -0,06
Education activities tailored to the firm 0,10
0,87
-0,00 0,07
Long term educational planning 0,15
0,84
0,11 0,10
Closer cooperation with subcontractors 0,07 0,04
0,80
0,08
Closer cooperation with customers 0,10 0,04
0,80
0,04
Closer cooperation with universities &
technological institutes
0,26 0,27 0,27 -0,06
Wages based on results 0,06 0,08 0,03
0,78
Wages based on qualifications and functions
0,17 0,04 0,06
0,74
Explained variation 2,38 1,60 1,47 1,37
The 14 different empirical dimensions examined in the analysis load on four factors, after a
varimax rotation. The first factor, called “Integrative and quality work organization”, includes
seven dimensions: Cross occupational working groups, Quality circles/groups, Self managed
groups, Integration of functions, Delegation of responsibility, System for collection of proposals
from employees and Softened or disappeared demarcation between occupational groups. This
factor covers all the central dimensions of integrative organization and quality management. The
next factor includes the human development dimensions: Tailored educational activities and Long
term educational planning. “External cooperation” is the common property of factor 3 which
include the two dimensions: Closer cooperation with subcontractors, Closer cooperation with
customers, but not Closer cooperation with knowledge centers . Finally factor 4 is called
“Compensation systems” and includes “Wages based on qualifications or functions” and “Wages
based on results”.

The factors uncovered in this analysis give the impression of a “theoretically consistent”
application pattern of the organizational dimensions in the firms. It is interesting to note that the
first factor that explains most of the variation brings together internal organizational
characteristics but neither education efforts, systems of wage nor external collaboration. This
implies that building the characteristics of a learning organization may be seen as a specific effort
that may or may not be supplemented by elements going into the other three factors.
Now, it is interesting to examine to what degree the four factors as latent variables affect the
appearance of product or service (P/S) innovation in the surveyed firms. This analytical step is
performed in a logistic regression model with occurrence of P/S innovation as dependent variable
and with organizational factors as explanatory variables. The results are shown in table 7.
Table 7: Logistic regression of factors related to organization and quality (Factor 1), human
development (Factor 2), external cooperation (Factor 3) and new compensation systems
(Factor 4) on P/S innovation. (odd ratio, 95% confidence interval, estimates and P-values)
Dimension Effect Lower Higher Estimate
Chi-sq P-valu
Factor 1 1,91 1,73 2,11 0,65 163,53 <.0001
Factor 2 1,38 1,25 1,51 0,32 43,20 <.0001
Factor 3 1,45 1,31 1,60 0,37 52,06 <.0001
Factor 4 1,31 1,19 1,44 0,27 31,34 <.0001
Table 7 demonstrates that each of the four factors has a significant effect on the innovation
performance of the firm. The first factor “Integrative organization” shows the largest effect, with
almost double the chances of P/S innovation in the firms. The other factors increase chances of
P/S innovation more moderately. This model may be seen as a verification of theoretical aspects
in relation to the innovative behavior of the firm. It indicates that integrative organizations are
more innovative than those that are not integrative. But it also shows that other factors having to
do with respectively external co-operation, education and system of pay tend to work in the same
direction.
We can go further in the analysis and ask whether the organizational characteristics and practices
complement each other and thus increase the chances of P/S innovation cumulatively. Such
complementarity of elements might reflect that there are ‘bundles’ of organizational techniques
that support each other and that it is only when the firm has got all the elements working together
that it will harvest the full benefits in terms of innovative behavior. Another way of understanding
complementarity would be that the more aspects and dimensions the firm implemented in the
organization, the more the internal consciousness in the organization about the importance of
learning and knowledge development among actors in the firm. This would be in accordance with
the empirical findings on “clustering” of aims where there was a strong interconnection between
aiming at learning and at building an integrative organization (see the comments to table 5).
Building on such arguments, an additive index has been constructed applying all the fourteen
organizational characteristics. The empirical distribution of observations (firms) in the additive
index of organization, quality control, human development, compensation and external

communication is shown in the table below:
Table 8: Distribution of firms in terms of the number of organizational practices in terms
of integratio, quality, human development, compensation and external communication (N =
2007).
Index Frequency Percent Cumulative perc.
0 32 1,59 1,59
1 64 3,19 4,78
2 105 5,23 10,01
3 135 6,73 16,74
4 210 10,46 27,20
5 202 10,06 37,27
6 224 11,16 48,43
7 250 12,46 60,89
8 213 10,61 71,50
9 210 10,46 81,96
10 165 8,22 90,18
11 90 4,48 94,67
12 63 3,14 97,81
13 30 1,49 99,30
14 14 0,70 100,00
We have classified the firms in three groups, according to how many dimensions they have
adapted in their organizations - in other words - how many organization, quality, human
development and external cooperation facets are built into the firm’s organization. We have thus
divided the firms into three groups:
- Low level learning organization – includes firms that have introduced zero to four of the
dimensions
- Medium level learning organizations- includes firms that have introduced five to eight
dimensions
- High level learning organizations- includes firms that have introduced nine to fourteen
dimensions.
This quantitative bundling aspect may be assumed to reflect the degree of organizational

sophistication. Applying many dimensions signals consciousness knowledge development. In
other words it signals a culture of change and learning in the firms. In table 9 results of this
construction is shown. Table 9 shows how frequent high level learning organizations are in
different categories of size, industry, ownership and production.
Table 9: Learning organization development by firm size, industry, group ownership and
production (percent horizontal)
Variables: High (9-14) Medium (5-8) Low (0-4) (N)
All firms 28,5 44,3 27,2 2007
Less than 50 employees 18,1 45,9 36,0 1048
50 - 99 employees 35,0 42,3 22,7 437
100 and more employees
45,1 43,3 11,6 490
Manufacturing 36,3 42,9 20,8 725
Construction 14,5 42,8 42,8 318
Trade 24,5 48,3 27,2 563
Other services 19,6 45,1 35,3 184
Business service 41,2 40,3 18,5 213
Danish group 30,1 44,7 25,3 701
Foreign group 40,7 43,8 15,5 388
Single firm 22,3 44,5 33,2 903
Standard product 29,2 45,1 25,7 725
Customized product 29,8 44,9 25,3 1192
By grouping all the firms according to the index of learning organization development we get
27% in the low category, 44% in the medium and 28% in the high category. Table 9 shows us that
this distribution is size dependent. Among firms with less than 50 employees, only one out of five
firms have developed a learning organization at the high level while the same is true for every
second of the bigger firms. With growing firm size, the share of highly developed firms increases.
Table 9 shows as well that the diffusion of high level learning organisations varies between
industries. More than 40% of the firms in Business service are in the category of highly developed
learning organizations, while the same is true for 36% of the firms in Manufacturing. The rest of
the industries lie below the average. It is, however, interesting to notice, that Trade together with
Other Services tend to have higher shares of their in the medium category. This might be seen as
illustrating that either the incentives or the opportunities to develop complete learning
organizations are much less in some industries than it is in others.
Another interesting result coming out from table 9 is that firms owned by foreign groups have
such a high share in the category of most developed. Firms owned by Danish groups are closer to

the general average and single - stand alone - firms are below average in this category. We will
see similar patterns in several of the following analyses. The presence of foreign owned firms
seems to constitute ‘a progressive element’ in the Danish economy while the often cherished
family owned stand alone firms seem to be lagging behind both in terms of technological and
organizational sophistication.
If the production is scale intensive and standardized rather than customer designed does not affect
the distribution between development categories.
What is the importance of this bundling of organizational dimensions for the knowledge
production and learning in the firms, indicated by product and service (P/S) innovations? In table
10 the different categories, representing increasing levels of learning organization will be tested in
a logistic model with P/S innovation as dependant variable, and with control for firm size,
industry ect.
Table 10: Logistic regression of learning organization level categories, size, industry,
ownership and production on P/S innovation (odd ratios, 95% confidence interval,
estimates and P-values)
Variables: Effect Lower Higher Estimate
Chi-sq P-value
High level 5,18 3,90 6,90 0,82 127,30
<.0001
Medium level 2,20 1,71 2,83 0,39 37,11 <.0001
Manufacturing 2,35 1,62 3,40 0,54 38,69 <.0001
Construction 0,69 0,45 1,08 -0,68 28,35 <.0001
Business services
2,27 1,46 3,54 0,51 15,40 <.0001
100 and more 1,61 1,26 2,07 0,30 14,23 0.0002
Danish group 0,76 0,58 1,00 -0,14 3,93 0.0475
Single firm 0,58 0,44 0,76 -0,28 15,85 <.0001
We find a five times higher chance of P/S innovation in the high level category, and even in the
medium category the chance is twice as high as in the low category. Among the other factors
included in the model, Manufacturing and Business services remain significant with 2.3 higher
chance of P/S innovation and Construction is negatively significant with a chance of 0,7. The
effects of large size (100+) is positive but moderate. Danish group ownership and single firms
have a chance below the average. In sum, the model has shown important and significant effects
of the development of what we call learning organization on P/S innovation. This may be taken as
more evidence of validity in the theoretical considerations concerning the construct of “learning
organization”.
For many of the firms, adoption of the operational dimensions or “institutional standards” of the
learning organization is a phenomenon of the late nineties. When we measure the implementation
of the organization dimensions and the P/S innovative behaviour chronological, we confront the

problem that the effect of implemented organization dimensions on behaviour may take some
time. A time lag of 2-3 years, sounds reasonable when focusing on learning practices and
organizational behaviour like P/S innovation. In order to take into account this aspect of
organizational behaviour, a new index and categorizing of the firms has been produced, with the
restriction of three years time lag on the organizational dimensions. The table below shows how
far the firms had developed their organization 1997 - 3 years back in time.
Table 11: Learning organization development implemented 1997 and before (N = 2007)
High (9-14 dimensions) Medium (5-8 dimensions) Low (0-4 dimensions)
9,7 41,3 49,0
When we compare the distribution in table 11 to the distribution in table 9, it is obvious that the
firms surveyed on average have gone through a quite rapid organizational upgrading and that
many of them have moved to a higher level over the last three years. While the category of high
level learning organization was as low as 10% in 1997 it had grown to almost 30% in the year
2000. Since the medium category remained almost unchanged it implies that at least as much as
40% of all the firms moved up one category over the three year period. Perhaps some caution
should be applied here since response regarding the history of the firm may be especially
uncertain.
It is also of interest to examine the effect of this revised model of the learning organization on
innovation. The idea is to see if we can find a time lag in the impact of the organizational set up
and its impact on product/service innovation and the results of the analysis is gathered in table 12.
Table 12: Logistic regression of learning organization development categories implemented
1997 and before, size, industry, ownership and production on P/S innovation (odd ratios,
95% confidence intervals, estimates and P-values)
Variables: Effect Lower Higher Estimate
Chi-sq P-value
High level 3,42 2,42 4,87 0,62 46,53 <.0001
Medium level 1,95 1,59 2,39 0,33 41,41 <.0001
Manufacturing 2,50 1,73 3,60 0,58 46,77 <.0001
Construction 0,65 0,42 1,00 -0.76 37,24 <.0001
Business servic 2,46 1,59 3,79 0,57 20,19 <.0001
100 and more 1,86 1,46 2,37 0,35 20,38 <.0001
Danish group 0,73 0,55 0,95 -0,16 5,40 0.0201
Single firm 0,52 0,40 0,68 -0,32 22,67 <.0001

In this lagged model almost the same factors come out as significant, as in the model without a
time lag. The effects of learning organization level have shrunk here, but the differences between
the two models are not statistically significant. The Industry effects have the same structure but
are weaker, though not significantly weaker. The same is true for size and ownership, where no
significant difference is found between effects in the two models.
All in all, the result of the lagged model did not confirm an assumption that there is a time lag
from organizational development levels to innovation performance. This might be seen as
indicating that the causal links do not go one way from learning organizations to innovation.
Instead there may also be a tendency that firms when they get involved in product innovation
realize the need for organizational development.
Another question that can be illuminated by the data we have access to is the effect of the
proportion employees covered by the implemented organizational dimensions or building blocks
of the learning organization inside the firms. The firms may introduce the individual dimension in
the organization and implement it, only covering a minor proportion of the employees. In order to
examine the importance and effects of a certain amount of employees covered by each dimension,
we have constructed the index and the categories with the restriction that at least 50% of the
employees should be covered by the individual organizational dimensions in the model.
Table 13: Learning organization development where each dimension is covered by 50% or
more of the employees (N = 2007)
High (9-14 dimensions) Medium (5-8 dimensions) Low (0-4 dimensions)
9,1 38,5 52,5
Table 13 shows that this restriction almost has the same effect on the distribution of the
knowledge organization, as what we observed for the time lagged distribution. Of the firms that
have introduced nine or more characteristics only one out of three have done so in such a way that
50% or more employees are covered on each organizational dimension. Of all firms only one out
of ten have reached this very high level of learning organization.
Next we test the effects of the learning organization development where each dimension in the
bundles are covered by 50% or more employees while controling for structural factors in the
model.

Table 14: Logistic regression of Learning organization development categories where each
dimension is covered by 50% or more of the employees, size, industry, ownership and
production on P/S innovation (odd ratios,95% confidence intervals, estimates and P-values)
Variables: Effect Lower Higher Estimate
Chi-sq P-value
High level 3,92 2,69 5,71 0,68 50,33 <.0001
Medium level 2,02 1,65 2,47 0,35 45,81 <.0001
Manufacturing 2,39 1,66 3,45 0,54 40,36 <.0001
Construction 0,69 0,45 1,06 -0.71 31,73 <.0001
Business services
2,29 1,48 3,54 0,50 15,22 <.0001
100 or more 1,93 1,52 2,47 0,38 23,64 <.0001
Danish group 0,74 0,57 0,98 -0,71 4,58 0.0324
Single firm 0,57 0,44 0,74 -0,28 16,98 <.0001
Table 14 shows that the effects of this model fit somewhere in between the original model and the
time lagged model, and the results are not significantly different to the other models. Thus again,
we do get a strong effect of high level learning organizations, and a more moderate effect of the
medium category, when controlling for the structural factors. And the same structural factors stay
significant in the model.
Why does taking account of a more wide use by employees of the bundle of practices not give
stronger effects on innovation performance. When we look in vain for increasing effects in this
model, we may be confronted with the organizational flexibility-stability dilemma (Zaltman et al
1973; Gjerding 1996). The hyper-flexible firm might not be the ideal environment for innovation
and different stages in the innovation process may require different organizational features. While
flexibility may facilitate the initiation of innovation it might on the other hand inhibit the
implementation of innovation. Implementation, on the other hand, demand stability (King and
Anderson 1995). This may be the reason why the model without employee restrictions get slightly
stronger effects on the categories of learning organization development. Here traditional
organization practices may exist in parallel with the new dimensions.
8. Indirect and direct participation
The leading argument in this analysis has been, that the changing and turbulent business
environment in the nineties has promoted development of certain types of organizational solutions
as institutional standards confronting the firms in industrialized countries (Røvik 1992; Røvik
1998). As we have seen, many Danish firms have reacted and adopted several of the institutional
standards in their internal organization and external relations. Focusing on the implemented
standards, with their reference to theoretical considerations of innovation, learning and knowledge
production in organizations, it has been shown, that firms incorporating many of these standards
as dimensions in sophisticated work organization, has a much higher chance of being product or
service innovative on the market. So far the relation between innovative behaviour and

developments of the learning organization is tested and confirmed. An important point, however,
is that the shape of what we call learning organization do vary individually to such an extent, that
it may be difficult to find two firms which have built exactly similar organizations, even though
they have used the same dimensions. Learning organizations are individually shaped and suited to
the firm’s strategic situation on the one hand, and the competences of the employees on the other.
We have stated that innovation theoretically has to do with embedded organizational competence.
Embedded competence relies on the ability of the employees continuously to learn and transform
knowledge into a collective resource, as well as the capability of management to bring new
knowledge and ideas into the practices of the organization. It is in this theoretical light we must
understand the importance of employee involvement and participation in organizational change,
and not least in building the learning organization, as a structural and cultural frame of
knowledge production in the firm . Through the involvement and participation principles applied
in relation to organizational change, necessary commitment among the employees is established.
This commitment is indeed a necessary precondition of continuous competence building, learning
and knowledge production in the firm. From this follows the importance of analyzing the applied
participation instruments and principles implemented, when changing and developing the
organization of the firm.
Table 15: Instruments applied in the cooperation between management and employees in
firms that have been through a process of organizational change (percent)
Yes No Do no
t
know
Not
relevant
(N)
Employee representatives participate in
management meetings
17.6
65.9
0.9
15.6
1254
One or more project group with participants from
management and employees
47.3
39.6
1.4
11.8
1251
The permanent cooperation committee 29.7
43.7
1.5
25.1
1240
Employee representative on firm’s board 13.7
58.9
1.3
26.1
1232
Common meetings with the employees concerned
83.3
11.8
1.1
3.9
1248
Common meetings with all employees 65.0
28.2
1.5
5.2
1243
Direct ad.hoc. consultation with the individua
l
employee
89.4
7.1
1.3
2.3
1262
In the literature on industrial relations cooperation dimensions have been classified in two major
categories (Hyman and Mason 1995):
- Indirect participation (referring to participation through local union representatives and
institutions).
- Direct participation (involvement mainly through communicative and cooperative
relations between management and employees).

It has been argued that over the last decade indirect participation forms have been in decline. To
marginalize trade unions have been seen as an element in ‘modern management’ aiming at
flexibility. In this context it is interesting to examine the participation dimensions applied by
firms in connection with their engagement in organizational change.
Table 15 indeed shows that the direct participation forms are the ones most often used in
situations of organizational change. Almost nine-tenths of the firms use direct contact with the
individual employees. More than 80% of the firms use meetings with the affected employees, and
two thirds use common meetings with all employees in the firm. At the border line between direct
and indirect participation we find ad hoc-project groups with management and employee
representatives. 47% of the firms use this form. Among the classical indirect participation forms
cooperation committees score highest (30%), followed by employee representative joining
management meetings (18%) and finally employee representatives in the company board (14%).
Figure 2: Instruments applied in the cooperation between management and employees
classified in relation to theoretical perspectives.
Direct involvement
Management initiative
From direct individual to indirect collective
involvement
Direct ad.hoc. consultation with individual employee
(89 %)
Common meetings with the employees concerned
(83 %)
Common meetings with all employees (65 %)
One or more project group with participants from
management and employees (47 %)
Indirect participation
Employee initiative
From indirect collective to direct individual
participation
The permanent cooperation committee (30 %)
Employee representatives participate in management
meetings (18 %)
Employee representative participates in firm’s board
(14 %)
In the figure 2 the cooperation dimensions have been classified in relation to the theoretical
perspectives and ordered after frequency in use.The ranking of the cooperation dimensions
perspectives and according to frequency of use. The data seem to confirm, that the direct
participation forms are the ones most commonly used by firms in contemporary organizational
change, and on this basis it is tempting to conclude that the indirect forms are of marginal
importance. It is, however, important to take into consideration, that while the direct forms may
be appropriate as a sort of common foundation the need for the indirect forms may become
realized when organizational changes are becoming more ambitious and complex. The argument

is that the two forms of participation are not substitutes but that they go together especially in
connection with the most advanced efforts to establish learning organizations.
In order to examine this proposition we shall commence by exploring the underlying pattern of
correlations between the various instruments through a factor analysis. Is it possible to uncover
informative latent participation factors in the empirical material?
Table 16: Factor analysis of seven instruments of involvement and participation
Factor 1: Indirec
t
Participation
Factor 2: Direc
t
Involvement
The permanent cooperation committee
0,85
0,01
Employee representative on firm’s board
0,82
0,05
Employee representatives participate in managemen
t
meetings
0,73
0,11
One or more project group with participants from
management and employees
0,58
0,34
Common meetings with the employees concerned 0,18
0,77
Common meetings with all employees 0,14
0,70
Direct ad hoc-consultation with individual employe
e
-0,03
0,66
Table 16 shows that the analysis uncovers two latent participation variables or factors among
firms having experienced with organizational change. The first factor relates to indirect
participation and includes the dimension of project group with participants from management and
employees. The correlated dimensions represent the indirect “corporatist” cooperation. The
second factor correlates the dimensions of direct involvement by individual or collective oriented
measures. The factor analysis indicates that local representatives and indirect participation still is
present and active as a factor in the change processes, building the new organization forms.
We can also analyze how important the different forms of participation are in firms at different
levels of learning organizations. In order to examine this question, a logistic model is applied with
the highly developed category of learning organization as dependent variable, and the individual
cooperation instruments as independent variables.

Table 17: Logistic regression of cooperation instruments, size, industry, ownership and
production on high level learning organizations (odd ratios, 95% confidence intervals,
estimates and P-values)
Variables: Effect Lower Higher Estimate Chi-sq P-value
Employee
representatives
participate in manage
-
ment meetings
2,29 1,50 3,50 0,41 14,74 0.0001
One or more projec
t
group with participant
s
from management and
employees
3,28 2,37 4,53 0,59 51,34 <.0001
Common meetings
with all employees
1,80 1,26 2,56 0,29 10,55 0.0012
100 and more
employees
2,68 1,83 3,94 0,48 17,01 <.0001
Table 17 shows that three of the seven cooperation instruments have statistically significant
effects in relation to chances of high level learning organizations. Project group with joined
participation of management and employee representatives have the strongest effect. Employees
joining management meetings have the second strongest effects, and common meetings with all
employees the third strongest effect. The model indicates that while direct participation may be
the most common form the indirect forms are of crucial importance for firms establishing learning
organizations at a very high level.
But one thing is individual probabilities and effects of the cooperation measures. The factor
analysis showed that the use of the cooperation instruments are correlated and often used
simultaneously in situations of change. How they are “bundled” depends on situational factors in
the individual firms. But we may expect that both direct involvement and indirect participation
instruments are used in various combinations. One simple way to solve this “bundle” problem
empirically is to apply a similar methodology as performed in the analysis of development
degrees in learning organizations. Thus an additive index of the participation instrument used in
the change processes has been constructed. The index only counts explicit “yes” or “no” in the
management responses on use of the seven cooperation dimensions. The result is shown in table
below.

Table 18: Distribution of firms according to the number of participation instruments used
in connection with organizational change (N = 746).
Index Frequency Percent Cumulative perc.
0 1 0,13 0,13
1 45 6,03 6,17
2 93 12,47 18,63
3 218 29,22 47,86
4 205 27,48 75,34
5 117 15,68 91,02
6 58 7,77 98,79
7 9 1,21 100,00
Only few firms use no or all of the cooperation dimensions or instruments, when changing the
organization. More than fifty percent of the firms use three or four instruments, which represent
the midpoint of the distribution. Thus the distribution can be divided into three categories with 3-
4 instruments in the “medium” category, 0-2 in the “few” instruments category, and 5-7 in the
“many” instruments category. Results of this classification is shown in the table below.
Table 19: Multiple use of participation instruments in connection with organizational
change (N = 746)
Many (5-7 instruments) Medium (3-4 instruments) Few (0-2 instruments)
24,7 56,7 18,6
The distribution results in less than one fifth of the firms using 0-2 instruments. This seems to
indicate that the firms do take employee participation seriously and combines the instruments in a
way that suits the situation. In this perspective it is also interesting that 25% use from 5 to seven
instruments. Table 20 relates the frequency of participation instruments used to to the probability
of developing a high level learning organization.

Table 20: Logistic regression of multiple participation instrument categories, size, industry,
ownership and production on high level learning organizations (odd ratios, 95% confidence
intervals, estimates and P-values)
Variables: Effect Lower Higher Estimate Chi-sq P-value
Many instruments 4,99 3,53 7,07 0,80 82,28 <.0001
Medium instruments 1,88 1,47 2,42 0,32 29,91 <.0001
Manufacturing 1,66 1,09 2,53 0,20 4,71 0.0299
Construction 0,84 0,50 1,39 -0,48 10,90 0.0010
Business services 2,51 1,55 4,06 0,62 22,04 <.0001
100 and more
employees
2,53 1,96 3,27 3,39 24,44 <.0001
Foreign group 1,49 1,15 1,93 0,20 9,12 0.0025
The more participation instruments applied in the change processes of the firm, the higher the
chance for the firm being characterized by a high level learning organization. This is the major
result of the logistic regression analysis of bundling the participation instruments. The model is
controlled for structural factors such as size, industry etc. In relation to the high level learning
organization, applying 5-7 participation instruments produces chances more than twice as high as
applying 3-4 participation instruments. This pattern indicates the importance of cooperation in
many forms and by many channels, when work organizations are changed and constructed in high
level learning organizations. In other words, there is evidence of firms that go far toward
establishing learning organizations need a complex and ambitious “cooperation regime” , where
instruments are direct and indirect participation are combined and integrated.
The use of specific instruments may be concentrated in a certain phase of the change process, or
may be spread more evenly throughout the whole process from the idea phase to the
implementation. We shall examine this important issue in table 21 in relation to the three
“bundle” categories.
Table 21: Phase of employee involvement in organization development by multiple
participation instrument categories (percent)
Idea phase Decision phase Implementation
phase
(N)
Few instruments 18,8 18,1 63,0 138
Medium instrument
s
21,4 28,6 50,0 420
Many instrument 33,7 26,1 40,2 184
chisq p = 0,0002 Gamma = -0,23

Table 21 shows that there is a relationship between the number of instruments used and the timing
of the major involvement of the employees. In all the three categories it is a minority among the
firms that involve employees mainly in the idea phase. But while the minority is less than 20%
among the firms using few instruments the corresponding share is more than 30%. Firms using
medium number of instruments have a higher proportion starting in the decision phase, and thus a
smaller proportion starting in the implementation phase. To go a bit further into the problem of
what participation instruments are used in which phase, the table below has been constructed.
Table 22: Phase of employee involvement in organization development by participation
instrument (percent horizontal)
Idea phase
Decision
phase
Implementa
-
tion phase
(N)
1. Employee representatives participate in managemen
t
meetings
38,5
29,8
31,7
218
2. One or more project group with participants from
management & employees
31,2
27,6
41,2
590
3. The permanent cooperation committee
24,8
28,3
46,9
367
4. Employee representative participates in firm’s board
19,5
28,4
52,1
169
5. Common meetings with the employees concerned
25,5
26,4
48,1
1034
6. Common meetings with all employees
25,8
26,0
48,2
799
7. Direct ad hoc-consultation with the individual employee
24,2
26,3
49,6
1120
Table 22 shows a mixed pattern regarding timing of the different types of instruments. Two of the
indirect participation instruments: Participation in management meetings and project groups with
management and employees are used frequently used in the idea phase. But this is not true for the
most formal instrument for indirect participation where there is a representative on the firm’s
board. Actually this form is used less frequently in the idea phase than the direct forms of
participation. The three direct involvement instruments have almost the same structure of
utilization in the phases of the change process. One fourth of the firms start direct employee
involvement in the idea phase, one fourth in the decision phase and the remaining half of the
firms use this kind instruments especially in the implementation phase.

Table 23: Employer assessment of employee influence on change by phase of employee
involvement in organization developments (percent)
Great
influence
Some
influence
Small
influence
No influence
Do not know
(N)
Idea phase 33,7 62,9 3,5 0,0 0,0 318
Decision
phase
14,0 73,9 11,8 0,3 0,0 322
Implementat
i
on phase
6,5 54,5 33,3 4,3 1,4 633
All firms 15,2 61,5 20,4 2,2 0,7 1273
chisq p = <.0001 Gamma = 0,65
It should be taken into account that table 23 is based on assessments made by management of how
much influence the employees have in connection with organizational change. These might
deviate from what employees themselves experience. The table relates the degree of influence to
the timing of involvement of employees. From the table it is evident, that there is a strong relation
between employee influence and phase of involvement in the change process. The earlier the
involvement the greater the influence. When combined with the earlier results it indicates that
influence is strongest in firms that use many instruments and especially in firms where employees
take part in management meetings and participate together with management in project groups.
9
.
Conclusions
In the management literature there is a presumption that certain organizational characteristics
promote learning and competence building. Here we have shown that when the organizational
characteristics having to do with respectively integrative organization, quality management,
human resource development, compensation systems and external network positioning bundles it
has a strong impact on knowledge creation in terms of product innovation. The advanced learning
organizations that combine several of these characteristics tend to introduce product innovations
more frequently than the rest. And the effect is strong also when we take into account differences
among firms in terms of size, sector and ownership.
It cannot be shown that there is a simple causality from the advanced learning organization to
innovation, however. Rather the relationship goes both ways. Firms operating in market segments
where continuous incremental product innovation is a prerequisite for survival and firms pursuing
strategies of continuous product innovation will realize that they need an advanced learning
organization. They will need it in order to organize the different sources of knowledge required
for the innovation and they will need it in order to cope with the unforeseen problems they
encounter as part of the innovation process.
Anyhow the results show that innovation policy needs to combine instruments that promote
product innovation through enhancing technical opportunities and incentives, including advanced
demand, with instruments stimulating the further development and diffusion of the elements
constituting the advanced learning organization.

In the literature on industrial relations there have been different types of arguments favoring the
participation of employees in decision making either directly or indirectly through workers’
representatives and trade unions. One type of arguments refer to economic democracy and
empowerment of employees as positive values that should be promoted. The other type of
argument refers to the presumed efficiency effects from participation. We have found it of interest
to illuminate these arguments in the context of the formation of advanced learning organizations.
The analysis shows that while direct participation is more frequent than indirect participation in
connection with organizational change in the firms the more cooperation instruments applied in
the firm, the higher the chance of highly developed learning organization. Especially the firms
that have moved far toward the model of a learning organization do involve trade unions and shop
stewards in organizational change. Use of the instruments are spread throughout the change
process from the idea phase to the implementation phase and again the active participation of
trade unions and shop stewards at the local level seems to be correlated with involvement at the
early stages of organizational change.
Not all organizational theorists are happy with the uncritical enthusiasm among management
consultants who promote the idea of so-called learning organizations. Our data seem to give
strong support to the hypothesis that establishing several of the characteristics that in the literature
is assumed to constitute such organizations are strongly correlated with product innovation also
when size, sector and ownership forms are taken into account. Promoting the diffusion of good
practices in this respect may therefore be seen as a major task for innovation policy and for
management in sectors where product innovation is important for the competitiveness of the firm.
In the Anglo-Saxon literature there has been a tendency to establish a contradiction by the
management driven efforts to establish new ‘human resource development strategies’ and an
active role for trade unions and shop stewards. The analysis pursued here gives a more optimistic
picture in this respect. In the firms that move very far toward becoming full-blown learning
organizations it seems as if management has found it useful/necessary to engage local trade union
representatives in the process of change.
One elementary caveat is that the data set covers firms operating in Denmark and that the patterns
observed might be (and probably would be) quite different were we to bring in data on firms from
other countries. The Danish innovation system has several peculiar characteristics that might
explain why workers and trade unions are more motivated to support change and managers more
willing to let them do so (Lundvall 2002). We are actually working on such an endeavor where
we together with colleagues from the UK and France will use labor market surveys comparatively
in order to check how far some of the patterns analyzed here differ in these two countries.

Literature
Andreasen, L.E. et al (eds) (1995),
Europe's Next Step: Organizational Innovation, Competition and Employment,
Frank Cass.
Argyris, C. and Schoen, D. (1978),
Organizational learning: A theory of action perspective
, Reading, Addison-
Wesley.
Arrow, K.J. (1962), ‘The Economic Implications of Learning by Doing’,
Review of Economic Studies
, Vol. XXIX,
No. 80.
Carlsson, B. and S. Jacobsson (1997), ‘Diversity creation and technological systems: A technology policy
perspective’, in Edquist, C. (ed.),
Systems of innovation: Technologies, institutions and organizations
, London,
Pinter Publishers.
Coriat, B. (2001), ‘Organizational innovation in European firms: A critical overview of the survey evidence’, in
Archibugi, D. and B.-Å. Lundvall (eds.),
The Globalizing Learning Economy
, Guildford and King’s Lynn, Oxford
University Press.
Dyer, L. and Reeves, T. (1995), ‘Human resources strategies and firm performance: what do we know and where do
we need to go?’,
International Journal of Human Resource Management
, no 6, London, Routledge,.
EIRMA (1993),
Speeding up innovation
, conference papers for the EIRMA Helsinki conference, May.
Freeman, C. (1987),
Technology policy and economic performance: Lessons from Japan
, London, Pinter Publishers.
Freeman, C. and Perez, C. (1988), ‘Structural crises of adjustment: Business cycles and investment behaviour’, in
Dosi et al (eds.),
Technology and economic theory
, London, Pinter Publishers.
Gjerding, A.N. (1996),
Technical innovation and organizational change
, Aalborg University.
Huselid, M. (1995), The impact of human ressource management on turnover, productivity and corporate financial
performance,
Academy of Management Journal
vol. 38 no. 3.
Huselid, M et.al. (1996), Technical and strategic human ressource management effectiveness as derterminants of
firm performance,
Academy of Management Journal
.
King, N. and Anderson, N. (1995),
Innovation and change in organizations
, Routledge, London.
Kolb, D.A. (1984),
Experiential learning
, Englewood Cliffs, Prentice Hall.
Laursen, K. (2001), ‘The importance of sectoral differences in the application of (complementary) HRM-practices
for innovation performance’,
Danish Ministry of Trade and Industry
, Copenhagen.
Laursen, K. and N. Foss, 2000, ‘New HRM Practices, Complementarities, and the Impact on Innovation
Performance’, IVS Working Paper no. 05 (Copenhagen Business School, Copenhagen).
Lund, R. (1998), ‘Organizational and innovative flexibility mechanisms and their impact upon organizational
effectiveness’,
DRUID Working Paper
98 (23), Department of Business Studies, Aalborg, Aalborg University.
Lund, R. and Gjerding, A.N. (1996), ‘The flexible company, innovation, work organization and human resource
management’,
DRUID Working Paper
96-17, Department of Business Studies, Aalborg University, Aalborg.
Lund Vinding, A. (2001), ‘Absorptive capacity and Innovative performance: A human capital approach’, paper
presented at the workshop Innovation, technological change and growth in knowledge based and service intense
economies, at the Royal Institute of Technology, Stockholm February 1-2.
Lund Vinding, A. (2002), Absorptive capasity and innovative performance: A human capital approach, Ph.-D.-
dissertation, Department of Business Studies, Aalborg University, Aalborg.
Lundvall, B.-Å. (1985),
Product Innovation and User-Producer Interaction
, Aalborg, Aalborg University Press.
Lundvall, B.-Å. (1988), ‘Innovation as an interactive process: From user-producer interaction to the national system
of innovation’, in Dosi et al (eds.),
Technology and economic theory
, London, Pinter Publishers.
Lundvall, B.-Å. (ed.) (1992), National Innovation Systems: Towards a Theory of Innovation and Interactive
Learning, London, Pinter Publishers.

Lundvall, B.-Å. (2002),
Innovation, growth and social cohesion, the Danish model
, Elgar Publishers, London.
Lundvall, B.-Å. and Johnson, B. (1994), ‘The learning economy’,
Journal of Industry Studies
, Vol. 1, No. 2,
December 1994, pp. 23-42.
Lundvall, B.-Å. and P. Nielsen (1999), ‘Competition and transformation in the learning economy – illustrated by the
Danish case’,
Revue d’Economie Industrielle
, No.88, 1999, pp.67-90.
Ministry of Education (1997),
National kompetenceudvikling
, Copenhagen: Ministry of Education.
Moss Kanter, R.
The Change Masters
, Unwin, London.
Nelson, R.R. (ed.) (1993), National Systems of Innovations: A Comparative Analysis, Oxford, Oxford University
Press.
Nielsen, P. (1990),
Personale og produktion - menneskelige ressourcer i udvikling af nye fleksible
produktionssystemer
, Aalborg Universitet
Nonaka, I. and H. Takeuchi (1995),
The Knowledge Creating Company
, Oxford, Oxford University Press.
Osterman, P. (2000), Work reorganization in an era of restructuring: trends in diffusion and effects on employee
welfare,
Industrial and Labor Relations Review
, Vol. 53, No. 2
Pavitt, K. (1984), “Sectoral patterns of technical change: Towards a taxonomy”,
Research Policy
, Vol. 13, pp. 343-
73.
Rosenberg, N. (1982),
Inside the black box: Technology and economics
, Cambridge, Cambridge University Press.
Rothwell, R. (1977), “The characteristics of successful innovators and technically progressive firms”,
R&D
Management
, No 3, Vol. 7, pp. 191-206.
Røvik, K.A. (1998),
Moderne organisasjoner - trender i organisationstenkningen ved tusenårsskiftet
,
Fakbokforlaget, Bergen.
Røvik, K.A. (1992), Institusjonaliserte standarder og multistandardorganisationer,
Norsk Statsvitenskapelig
Tidsskrift
, Oslo.
Senge, P. (1990),
The fifth discipline: The art and practice of learning
, New York, Doubleday.
Tomlinson, M. (1999), ‘Learning economies and embodied knowledge flows’,
CRIC Discussion Paper
No. 26,
Manchester, UK, University of Manchester.
von Hippel E. and M. Tyre (1995), ‘How learning by doing is done: Problem identification and novel process
equipment’,
Research Policy
, vol. 24, no. 5.
Zaltman, G., Duncan, R. & Holbek, J (1993),
Innovations and Organizations
, Wiley & sons, New York.
Zuboff, S. (1985), ‘Technologies that Inform: Implications for human ressource management in the computerized
industrial workplace, in Walton, R.E. & Lawrence, P.R. (eds.),
HRM - trends and challenges
, Harvard, Boston.
Wood, S. (1999), Getting the Measure of the transformed high-performance organization,
British Journal of
Industrial Relations
, Blackwell, London.

Appendix A
Danish Work Organization, Innovation and Competence Development Panel
DISKO - Survey: 1993-95
1900 firms (3993)
* Organizational changes
* Job designs
* Qualification demands
* Education and training
* Product and ICT inovation
DISKO - Survey + Register data: 1990-97
1544 firms/workplaces (134.000 - 145.000 emp.)
* Value added - productivity
* Assets
* Turnover
* Job flows
* Worker flows
* Wages
DISKO2/IOC - Survey: 1998-00
637 of 1363 surviving DISKO firms (Panel)
2007 firms (Cross Section)
Questionaries to management & employee representative
* DISKO - measures
* Personnel planning
* Processes of change
* Workplace IR - Participation
* Consequences of change
DISKO2/IOC + Register data: 1990-00
Panel design: 637 firms
Time series design: 1900 + 2007 firms
* ICT variables (e-trade ect.)
* Value added - productivity
* Assets
* Turnover
* Job flows
* Worker flows
* Wages