Developing Firms’ Technological
Capabilities
Erik Arnold
Ben Thuriaux
June 1997
Supporting Companies’ Technological Capabilities
Erik Arnold
Ben Thuriaux
Technopolis Ltd
June 1997
1
Introduction
Many countries have evolved extensive infrastructures to support
innovation and the development of technological capability, especially
among small and medium-sized enterprises (SMEs). These infrastructures
are often complex, and hard for the companies to understand because they
have been built up piecemeal to tackle deficiencies in individual
technological capabilities. The purpose of our paper is to impose some
structure through a needs analysis: What categories of technological
capability should firms develop? Based on this, we can then set out some
principles for better design of the innovation infrastructures which
support industrial development.
Our approach has been to review relevant published literature in the light
of our internal experience of evaluating and operating technology support
programmes and infrastructures. We have focused on the ‘innovation’
literature, which tends to take economic theories of the firm as its point of
departure. This is characterised by a shift from neoclassical to
evolutionary economic approaches in recent years and a growing concern
with ‘national innovation systems’. While technological capability is
treated as important, what the literature has to say about it turns out
generally to be limited.
In the remainder of this introduction, we describe the scope of our enquiry
more closely and describe what we mean by ‘technology’. In section 2, we
then discuss why technological capability is important and consider the
type of shortcomings in capability that exist - partly by enumerating some
of the ways in which firms behave ‘imperfectly’ in comparison with
established economic theory, and partly with reference to they type of
obstacles to innovation that they typically experience. Next, in section 3,
we set out a structured view of technological capabilities and how they
relate to other capabilities needed by the firm. Finally, section 4 draws out
implications for innovation policy and innovation support systems.
1
Over the past twenty or thirty years, it has been increasingly recognised
that innovation involves a great deal more than creating and selling ‘black
boxes’. It is a process that happens in an economic and social context, and
is no longer seen as an ‘automatic’ result of changes in science and
technology. Successive generations of ‘innovation models’ have
characterised innovation as increasingly complex and bound up with
socioeconomic factors such as market linkage and match with the
available infrastructure1 . The technology transfer literature makes it clear
that technology is more than a ‘black box’ but depends on its social and
infrastructural context. If technology is transferred into contexts for which
it is not fitted, it will fail to work.
Within the firm, there is a strong interdependence between technology,
innovation and other activities. Technology strategy is a vital part of
company strategy overall. Innovation is a wider process which does not
always depend on ‘technology’ in the conventional sense of hardware and
software. Following Schumpeter, economists tend to describe innovation
as “a new combination of the factors of production”. This can involve
using results of scientific or technological research, but it can also involve
much more mundane things such as laying out the machines on the
factory floor in a better order, changing the design of the product packaging
or copying ideas from a producer in a distant market in order to create a
local advantage. A key observation, however, is that innovation is a
fundamentally economic process, in which technology may play a greater
or a lesser role.
Our focus in this paper is on the technological capabilities needed for
innovation. These include, therefore, not only the skills associated with
acquiring and manipulating new processes and products but also those
associated with production more generally.
‘Innovation’, like ‘Marketing’ can be defined so broadly as to encompass
almost every activity of the firm. This does not seem to be helpful. For
our purposes, we must first define ‘technology’ in a way which is relevant
to industrial and economic development. The most useful definition is
probably the one given by the Oxford English Dictionary (OED): namely,
“the scientific study of the practical or industrial arts”. There is a great
deal of content packed into this short definition
•
“Arts” here means the production of artifacts - not only ‘works of
art’ but products more generally.2 “Arts” involves an holistic view,
where design is part of production
1
Rothwell, R., 'Successful Industrial Innovation: Critical Factors for the 1990s', R&D
Management,:£ , p 221-239, 1992
In the UK, this old definition of “arts” is frozen in the name of the Royal Society of
Arts - which is more concerned with defining and measuring the skills needed for
economic production than with painting or sculpture
2
2
•
•
•
•
•
•
Technology involves knowledge about doing practical things,
especially producing things. (In a modern context, this must
include both goods and services.)
The definition does not distinguish between engineering and
managerial aspects of the industrial arts. The modern discipline of
management is, historically, an off-shoot of engineering.3 Many
would argue that this fragmentation was a mistake. Some modern
management methods risk being destructive because they are
alienated from production realities.4 The revolution in production
engineering and management over the past twenty years which has
introduced ‘Japanese’ methods and ‘soft technologies’ such as
continuous improvement and lean production can be interpreted as
a move towards reintegrating the engineering and managerial
aspects of technology
Technological knowledge is built up using scientific methods. It is
described in books, tested by experiment and can make deliberate
(and sometimes rapid) progress by doing work in places which are
located away from the production process.5 This distinguishes it
from craft knowledge, which is defined and communicated through
skills and which tends to evolve together with production
Because it is a scientific study, technology involves codification:
writing down knowledge in a systematic way. (In fact, the OED also
lists an older definition of technology as “a discourse or treatise on
an art or arts”.) Uncodified knowledge remains important in
production. But technology involves a battle to try to codify
knowledge which today is tacit, in order to study and improve it
Codification means that a crucial feature of technological
production, compared with craft production, is that its principles
can more easily be communicated or transferred. Craft skills move
‘vertically’ from master to apprentice. Technological knowledge
spreads ‘horizontally’ between technologists
Codification also implies that people who work with technology
must be educated. The idea of a skilled but illiterate craftsman is
familiar from both history and anthropology. But the idea of an
illiterate technologist is self-contradictory
This broad definition of ‘technology’ as something soft as well as hard, and
as including important aspects of management and organisation, guides
the scope of the company capabilities and the policies we have considered.
3
4
5
see David Noble, America by Design: Science Technology and the Rise of Corporate
C a p i t a l i s m, New York: Alfred Knopf, 1977; republished Oxford University Press,
1979
RH Hayes and WJ Abernathy, “Managing Our Way to Economic Decline,” Harvard
Business Review, July-August 1980, pp159-175
Adam Smith grasped the importance of this fact very clearly when he described
how technical progress was enabled by “philosophers or men of speculation whose
trade it is not to do anything but to observe everything; and who, upon that account,
are often capable of combining together the powers of the most distant and
dissimilar objects” The Wealth of Nations, 1776; reprinted, Harmondsworth:
Penguin, 1974, p115
3
2
Improving SME Innovation Performance
The policy desire to improve SME innovation performance arises both
from the general need to make good use of innovation to foster economic
growth and from the particular obstacles to innovation faced by SMEs.
2.1
The Importance of Innovation
Over the last decade we have seen increasing acceptance among policy
makers of the idea that diffusion produces most of the economic benefits
of new technology. It is “not the creation of technological leadership in
itself that affords a nation its competitive advantage, but the rate and level
of diffusion of the technology into economic use”6 . In a series of US case
studies, Mansfield et al7 have shown adopters of even minor technological
improvements attaining a median return on investment of 55%,
compared with 22% for the companies which innovated these same
improvements. The more significant the innovation, the larger the gap
between returns to innovation and diffusion.
Brainard, Leedman and Lumbers summarised the evidence from the
relevant diffusion literature in the form of three significant conclusions
•
•
•
Technology development and diffusion are clearly of considerable
potential economic importance, with diffusion offering particularly
large benefits
Technology diffusion involves far more than the simple
introduction of new machinery into the firm. Additional measures,
such as internal reorganisation of both production and
management processes and upgrading of skills, may be essential to
capturing economic value from investment in new technology
Whereas it may not be necessary to produce technology to reap its
benefits, diffusion is essential to maximise potential national
economic returns. However, realising the benefits of diffusion may
depend critically on broader social and institutional changes, which
may, in fact, represent the most important obstacles of all8
In the light of these conclusions, especially those about the difficulties and
imperfections involved in the diffusion process (which tend to have
particular force for small and medium-sized enterprises: SMEs), it seems
almost surprising that in most countries the greater part of government
funding still goes to the invention and innovation process rather than to
diffusion. In fact, the clear distinctions which tend to be drawn in parts of
6
7
8
Rothwell, R. and Zegfeld, W. (1985), Reindustrialisation and Technology, Essex:
Longman
Mansfield, E. et al, ‘Social and Private Rates of Return from Industrial Innovation,’
Quarterly Journal of Economics, Vol 91, No 2, May 1977
Brainard, R., Leedman, C. and Lumbers, J. (1988), Science and Technology Policy
Outlook, Paris: OECD
4
the innovation literature between invention and innovation are not so
easy to make in industrial practice. Quite where adoption and adaptation
of technology shade into incremental improvement and innovation is
often hard to say, since “any act of adoption involves certain
transformations and is an act of incremental innovation in itself.”9
US data suggest that the small-firm sector has strong growth potential especially through the exploitation of technology. Exhibit 1, which shows
changes in employment over time, may well overstate the case, as small
firms often take on part-time, low-paid workers in service sectors.
Nonetheless, the apparent link with technology use is startling. It suggests
that the rewards are high to countries or regions that can ensure their
SMEs are good at exploiting technology.
Exhibit 1 Employment Growth 1982-7 in US Companies
Technology Use
Lowest
Smallest
Enterprise
Size
Largest
Highest
1
2
3
4
1
55%
68%
79%
218%
2
4%
15%
29%
54%
3
-13%
4%
9%
36%
4
-27%
-8%
-3%
10%
5
-14%
-26%
-19%
-2%
Source: US Department of Commerce
2.2
Economic Theory and Technological Capability
The way in which economic theory treats the firm is necessarily idealised.
Nonetheless, it provides insight into the ways that rational and efficient
firms would operate.
Recent years have seen an important change in the way economists deal
with innovation, and therefore also in the theory of the firm. This has all
the hallmarks of a Kuhnian scientific revolution.10 Thus, while the
neoclassical theory of the firm has actually been refuted11 it is more
9
10
11
Technology and the Economy: The Key Relationships, Paris: OECD, 1992, p48
T S Kuhn, The Structure of Scientific Revolutions, 2nd edition, University of
Chicago Press, 1970
See, for example, Erik Arnold, Competition and Technological Change in the
Television Industry: An Empirical Evaluation of Theories of the Firm, London:
Macmillan, 1985
5
important that a new generation of economists is simply ignoring it.
Instead, they are taking a mixed neo-Schumpeterian and evolutionary
approach. 12 Metcalfe and de Liso say that
... there are four themes which taken together we believe are
having a major impact on the way that scholars approach the
study of innovation. These themes are bounded rationality,
organisational capability, cognitive technological structures, ie
paradigms, and technological systems. Together they lead us to an
empistemological view of the firm as a creative experimental
organisation embedded in a wider network of knowledge-generating
relationships. It is what it believes it is. It does what it does. It
can only innovate by changing what it believes and knows.13
This resource-based approach to capabilities also highlights important
characteristics of the innovating firm. From this perspective, business
enterprise can be understood in terms of learning, path dependencies,
technological trajectories and opportunities, complementary assets,
technology and transaction costs. Performance is a function of intangible
assets, skills and the ability to develop new capabilities over time.14
The upshot of this revolution in the theory of the firm is that companies
are now seen as containing people, with specific histories, habits, skills and
behaviours. They have capabilities, and they are capable of learning.
Exhibit 2 (overleaf) briefly contrasts this modern view with the
neoclassical one, considering: information; rationality; technological
choices; technological progress; and firm behaviour.
It is well understood that, in practice, information flows in a very patchy
way within the economy. Often, it makes sense to keep information (for
example, about prices) secret. In other cases, such as patents, information
is available, but it can only be used at a price. In a society where
Information Technology is widespread, problems increasingly focus on
intelligently filtering out the majority of irrelevant information.
Information cannot be exploited - or filtered - without having the ability to
understand it and act on it. Nor can it be used without devoting effort,
and thereby incurring opportunity costs.
Firms do not always act as rationally as the neoclassical model would
suggest. This may be because they lack the skills or the time to analyse
incoming information effectively, or because that information is
inadequate to their needs. (For example, they may know the prices in the
local market, but be unaware that a new competitor in another country has
developed a cheaper production technique, and can therefore attack their
12
13
14
See, for example, Bengt-Åke Lundvall, National Systems of Innovation: Towards a
Theory of Innovation and Interactive Learning, London: Pinter, 1992
J Stan Metcalfe and N de Liso, “Innovation, Capabilities and Knowledge: The
Epistemic Connection,” University of Manchester (mimeo) 1995
Andrea Prencipe, Technological Competencies in the jet Engine Industry: The Case
of Rolls-Royce plc, MSc Thesis, Brighton: Science Policy Research Unit, 1995, p8
6
market unless they, too, change their production technology.) They may
not know how best to understand their own business and technology
situation. They may not act to maximise profits, acting as ‘lifestyle’
companies, ‘satisficing’ or devoting organisational slack to the interests of
managers rather than shareholders. Or they may take a ‘strategic’ view,
choosing to invest in things which will pay off only in the longer term.
Exhibit 2 Neoclassical and Contemporary Views of the Innovating Firm
Neoclassical View
Contemporary View
Information
Information about prices and technologies
are available to the firm at no cost, and
there are no restrictions on how the firm
uses this information
Rationality
The firm understands the information it
receives and acts rationally to maximise
its profits, based on that information
Technological choices
To produce any given good, there is an
effectively infinite set of technologies
(defined as the proportions in which
factors of production may be combined)
available. The firm can move smoothly
between technologies, incurring
investment costs but not other costs (such
as opportunity or learning costs)
Technological progress
Technological progress is ‘exogenous’.
That is, it is not in the hands of the firm
but just ‘happens,’ changing the
‘production-possibility frontier’ and
allowing the firm to produce more
efficiently. However, because there is
perfect information, the entire set of new
possibilities is immediately available to
the industry as a whole so it is not
possible to monopolise the benefits
Behaviour
The firm is a kind of economic robot,
reacting to circumstances using a small set
of fixed rules
7
Information may be costly to acquire,
inaccurate and hard to understand.
Exploiting it costs money and may require
complementary assets, including internal
technological capabilities
The firm’s rationality is ‘bounded’.
Decisions are based on past experience.
Actions may be strategic, sacrificing short
run profits in order to reach long run goals
Technological choice is limited by the
state of knowledge - both outside and
inside the firm. Change is costly and
risky
Technological progress happens both
‘exogenously’ and ‘endogenously,’ within
the firm. Taking advantage of it requires
that the firm have technological
capability - including the ‘absorptive
capacity’ needed to evaluate and exploit
ideas from the outside
The firm searches for opportunities,
learns and changes its own competences
and behaviour over time
There are, indeed, many technologies available to firms. However, some
will be proprietary. Adopting any of them will involve opportunity costs.
Changing between them may involve overcoming institutional rigidities
such as organisation, skills, market focus and partnerships (for example
use of distributors). It is easier to move between some technologies and
others. Technological progress is, of course, not external but - to a very
large extent - internal to firms and to the networks of relationships within
which companies operate.
Real firms are not at all homogeneous. They vary dramatically, both in
size and in what they are capable of doing. In contrast, all neoclassical
producers act in the same, predictable way. (Historically, this is because the
theory was developed by considering the production of staple agricultural
goods - the ‘corn’ of the textbooks - and their sale on organised
exchanges.15)
The neoclassical model of firm is thus a pretty poor description of the real
thing. While economists disagree about the wisdom of building a
superstructure of economic theory on a model which is easily refuted, the
neoclassical assumptions do at least make it possible to make calculations
about the behaviour of firms and markets. This is much more difficult
with the less predictive contemporary models of the firm.
Paradoxically, while neoclassical economics treats ‘perfect competition’
among small producers in homogeneous markets as the ideal state, real
companies generally have to be large before they are capable of behaviour
which resembles that of the neoclassical firm. Crucial gaps between the
theoretical model and reality are to do with the capabilities which theory
supposes that firms possess. Policymakers talk about ‘market failures’ and
the resulting need for intervention to support SMEs and other firms.
These ‘failures’ are often caused by the inability of firms to live up to the
neoclassical ideal, so the failures are actually in the theory, not in the
market. But they are failures nonetheless. Correcting them as far as
necessary is an important objective of innovation policy.
2.3
The View from the SMEs
It is central to the problem of improving SME performance that these
firms often find it difficult to identify - let alone to express - their needs.
The smallness of SMEs - their defining characteristic - is an important
driver of capability. Typically, these companies have only a weak ability to
interface with the infrastructure. While there are many actors emitting
‘signals’ about technology and about support opportunities, SMEs rarely
have good ‘receptors’ for this information. They tend to have relatively
shaky economics and therefore to focus on short-term problems rather
15
J M Clark, Competition as a Dynamic Process, New York: Brookings, 1961, p166
8
than long-term improvement opportunities. Managements tend to be
small and multi-functional. Often, entrepreneurs run companies singlehanded or take a disproportionate proportion of the key decisions, in
addition to functioning as the general interface to the outside world.
Creating a larger, ‘professional’ management is desirable, but until a
certain size is reached it is difficult to create much division of labour and
therefore to develop specialised interfaces.16 As a result, managements
operate in a vicious cycle of overwork, which leads to inability to consider
and exploit externally-derived improvement opportunities, which in turn
leads to overwork.
The lack of specialised capabilities affects many aspects of firm
performance. Key resources and skills may be completely absent. Notably,
many small firms have no engineers and therefore no ‘intelligent
interface’ to technological changes and opportunities. Other key skills and
resources may be absent. Often, for example, new technology-based firms
have few marketing or business development capabilities.
It tends to be difficult for SMEs to buy their way out of their capability
shortages, except at the most routine level of testing- and maintenancerelated services, because they are immature buyers of consulting and other
advice. Their inability to identify many types of problems or
improvement-opportunities makes it hard for them both to recognise
need and to define a clear consulting brief.
SMEs tend, additionally, to undervalue advice and to be shocked at its
price. In some cases there is a real justification for setting a low value on
advice. Unlike a large company, which can ‘leverage’ strategic advice or
cost-reductions across a large volume of sales, a small firm needs to see
very large percentage-improvements indeed before it makes sense to
invest in a consultant’s fee. SMEs may also be hampered by resource
constraints and lack of experience in trying to implement advice.
Particularly for owner-managed firms or other SMEs run by a strong
leader, there is additionally a risk that advice be seen to undermine
management’s authority. We suspect that this factor, together with an
exaggerated sense of the uniqueness and value of information about the
firm, helps explain the apparent worry among SMEs that using
consultants will lead to breaches of confidentiality and damage the
business. Such worries are far less prevalent among the larger companies
which use the international consulting firms.
16
See Erkko Autio’s systematisation of growth stages in “Supporting the
consolidation and internationalisation of SMEs - The systemic perspective,” in
Osmo Kuusi (ed), Innovation Systems and Competitiveness, Helsinki: Taloustieto
Oy, 1996
9
There are important cases of small, specialised firms which operate
internationally or globally, but in many cases SMEs work within a small
economic ‘space’. This generally means working within a small
geography, but it can also mean a tight linkage to a particular industrial
sector or to the supply-chain of one or a small number of large companies.
Lacking the resources to search widely for support and advice, SMEs tend
in the first instance to work in interpersonal networks defined at these
geographic and sectoral levels. It is therefore natural for their links with
the support system to be most easily defined within their own economic
‘space’: via chambers of commerce, trade associations, research associations
and similar well-established institutions. The corollary is that new types
of support actors have to work very hard if they are to break into SME
managers’ interpersonal networks. In many cases, policymakers have
decided to ‘go with the flow’ and exploit existing interpersonal networks.
Thus, in France, regional development initiatives such as science parks are
strongly centred on chambers of commerce. In the Netherlands, the ‘SBI’
programme which promoted the benefits of IT to SMEs across a wide range
of sectors through demonstration projects and common-interest
development projects did so by using the strong trade associations which
already existed in Holland.
SMEs’ owner-managers are typically very risk-averse. There is good
reason for this aversion. Business failure is likely to affect them in a very
direct way: they are likely to lose both house and pension if the company
goes bankrupt. Many SMEs’ economic positions are genuinely precarious,
especially in relatively undifferentiated industries. Owner-managers also
tend - rightly - to regard company money as their own and to spend with
great reluctance as a result. All these factors encourage SMEs to avoid
uncertainties and novelty - including the use of the support infrastructure
or even innovation more generally.
The Community Innovation Survey (CIS) included a question about the
obstacles that firms see to innovation. While the CIS includes all sizes of
firm, the larger numbers of SMEs in the population means that CIS
responses primarily indicate the views of these small companies. Exhibits
3A and 3B show how Irish and Norwegian companies respectively rated
various obstacles to innovation, and is relatively typical of the type of
response obtained by the CIS and, indeed, other similar surveys.
The most important worry is money. It is easy to jump to the conclusion
that, therefore, more finance should be available. Yet there are few
symptoms of overall capital shortage in the economy, especially at today’s
low interest rates. The real issues are
10
•
•
•
•
SMEs have a high death rate. Lending them money is objectively
risky
Their small scale and often underdeveloped business capabilities
make it hard for them to do the kind of research normally
associated with large-company strategy development. Sophisticated
research and business case development is often necessary in order
to put a reasoned case to a financier
Technological risks and uncertainties may be hard to quantify and
explain
They tend to have insufficient assets to back major loans or
injections of equity
Exhibit 3A Factors Hampering Innovation (Ireland)
Percentage of all CIS respondents saying factor is
very significant or crucial
Lack of appropriate sources of finance
33%
Innovation costs too high
30%
Pay-off period too long
29%
Excessive perceived risk
26%
Innovation potential too small
25%
Legislation, norms, taxation
20%
Lack of skilled personnel
20%
Lack of information on markets
19%
Lack of information on technologies
Lack of opportunities for
co-operation
Innovation costs hard to control
16%
Innovation too easy to copy
16%
18%
17%
Lack of technologicalopportunities
Availability of external
technical services
Lack of customer responsiveness
14%
14%
13%
Resistance to change in the enterprise
12%
Uncertainty in timing of innovation
10%
No need due to earlier innovations
10%
0%
10%
20%
30%
Source: Anne Fitzgerald, Marcus Breathnach, Technological Innovation in Irish
Manufacturing Industry, Dublin: Forfás, 1994
CIS respondents themselves regard innovation as expensive and risky.
Many of the other obstacles identified relate to technology capabilities with the most important being the company’s internal capabilities
(“innovation potential”). Several obstacles (such lack of information on
markets and technologies) relate to inadequacies in companies’ interfaces
with the outside world, perhaps more than a genuine absence in the
environment. We explore these capability questions in the following
section.
11
Exhibit 3B Factors Hampering Innovation (Norway)
Percentage of all CIS respondents saying
factor is very significant or crucial
Innovation costs too high
Innovation potential too small
Excessive perceived risk
Lack of appropriate sources of finance
Lack of skilled personnel
Pay-off period too long
Innovation costs hard to control
Lack of information on markets
Lack of technological opportunities
Lack of customer responsiveness
Uncertainty in timing of innovation
Lack of information on technologies
Lack of appropriate organisation
Lack of opportunities for co-operation
Innovation too easy to copy
Resistance to change in the enterprise
No need due to earlier innovations
Availability of external technical services
39%
37%
36%
29%
26%
24%
17%
16%
15%
14%
12%
11%
11%
9%
9%
8%
7%
6%
0%
10%
20%
30%
40%
Source: Svein Olav Nås, Tore Sandven og Keith Smith: Innovasjon og ny teknologi i norsk
industi: En oversikt, STEP Rapport 4/94, Oslo: STEP Group
Exhibit 4 Evolutionary Model of the Firm
Inputs
• Information
• Resources
Search
Intelligence
•
•
•
Assets
Physical assets
Capabilities
Memory
12
Outputs
• Products
• Information
3
Competences and Capabilities
Modern, evolutionary economics sees the firm as a searching, learning
mechanism. It survives and improves by continually reinventing itself.
The firm consists of two elements (Exhibit 4)
•
•
A pool of assets, including both physical assets and intangible ones
such as capabilities
Intelligence, which learns from the environment and modifies the
resources
Each of these elements can be broken down much further. An important
attribute of the firm’s ‘memory’ is that it comprises a mixture of
knowledge (tacit as well as codified) and of the configuration of assets:
namely, organisation, characteristics of the capital stock, relationships, and
so on.
The idea of ‘core competencies’17 is an attempt to understand how
companies marshal their capabilities. Teece, Pisano and Shuen18 define
core competence as “a set of differentiated skills, complementary assets and
routines that provide the basis for a firm’s competitive capacities and
sustainable advantage in a particular business”. They go on to explain that
core competences are unpredictable because they are competitively
determined. From our perspective, they are possible to identify ex post,
but are not so general that they form a useful basis for making policy.
Core competencies are, however, built up using the more generic
capabilities which concern us here (Exhibit 5). Core competencies define
how firms meet - and create - particular market circumstances. They are
key parts of the corporate ‘memory’ and are defined in relation to the
needs of the market segments in which the firm operates. Since market
segmentation is to a significant extent the imposition of structure by the
firm onto the demand side, the matching and modification of
competences with market segments is one of the central aspects of firm
intelligence or entrepreneurship. Individual products and services are
then ‘instances’ of this matching process.
17
18
GK Pralahad and G Hamel, “The core competence of the corporation,” Harvard
Business Review, 68 (3), pp79-91
David Teece, Gary Pisano and A Shuen, Firm capabilities, resources and the concept
of strategy, CCC Working Paper No 90-8, University of California, 1992
13
Exhibit 5 An Extended View of Core Competencies
Search
Intelligence
Products
Trade
Products
Core
Competencies
Marketing
Managing Capabilities
Market
Segmentation
Capabilities
Total Market
OECD has listed19 major elements of producers’ overall capabilities
(including, but not only, technological capability) as
•
•
•
•
The knowledge and skills required for the process of production,
where shop-floor experience and ‘learning-by-doing’ plays an
important role
The knowledge and skills required for investment, ie the
establishment of new production facilities and the expansion
and/or modernisation of existing ones
The vast area of adaptive engineering and organisational
adaptations required for the continuous and incremental upgrading
of product design and performance features and of process
technology
And, finally, the knowledge required for the creation of new
technology, ie major changes in the design and core features of
products and production processes
Our particular concern here is the role of the capabilities needed for
innovation in all this. While we found many observations about aspects
of technological capability in the literature we surveyed and a smaller
number of discussions about how these develop, there appear to be almost
no attempts to provide an overall view. We need this overview both to
ensure that our description of technological capability is reasonably
comprehensive and in order to understand companies’ technological
capabilities as systems, where individual components are related to each
other.
19
Op Cit, OECD, 1992, p262
14
Two exceptions are Howells20 and Dodgson and Bessant.21 Exhibit 6
shows how they respectively characterise technological capability. Both
recognise that the ability to use and develop technology is deeply
embedded in the ‘soft’ factors which surround the hardware, consistent
with the broad definition of ‘technology’ which we use in this paper.
Exhibit 6 Approaches to Technology Capability
Technology Base
Resources
Tangible Assets
• New products
• Plant
• Equipment
Intangible
Assets
Competence
Formal
• Patents
• Licences
• R&D
• Other IPR
• Training
Informal
• Tacit
Howells, 1994
Innovative
Capabilities
Dodgson and Bessant, 1996
Howells’ description is a static one. His concern is to show the
interdependence of tangible and intangible assets in underpinning firms’
competitiveness. He therefore makes the distinction between these two
kinds of assets central to his model and treats tacit knowledge as a
particularly special category of intangible assets.
Bessant and Dodgson’s approach is dynamic. They define their terms as
•
•
•
20
21
Resources All the assets in the firm which enable firms to operate,
including tangible and intangible assets, skills, knowledge,
organisation, links with other firms
Innovative Capabilities Features of firms and their management
which enable them to define and develop competences to create
competitive advantage
Competences That focused combination of resources which enables
firms to differentiate themselves from their competitors
Jeremy Howells, ‘Tacit knowledge and technology transfer,’ in Gustavo Fahrenkrog
and Patries Boekholt (eds) Public Policies to Support Technology Transfer, EIMS
Report No 8, Luxembourg: European Commission, 1994, p3
Mark Dodgson and John Bessant, Effective Innovation Policy: A New Approach,
London: Thomson, 1996, p12
15
These three elements interact through learning, which is a purposive
search for competitive advantage.
If our analysis of technological capability is to be consistent with a neoSchumpeterian or evolutionary view of the firm, it needs to involve this
combination of resources and intelligence
Capability is much more than ... individual assets. If it were not,
the firm would be no more than a bundle of bilateral contracts
between owner and employee, and rent could not exceed the
differences between current and next-best use value. One thing this
tells us is that the organisation must possess a memory, or a
tradition of practice, so that losses in personnel can be matched
with new employees who can be trained in the firm’s routines.22
The need for a memory drives progressive companies to accumulate tacit
knowledge, to identify its components and to try to codify it as intellectual
capital. Few are as explicit about the process as Skandia Insurance which
some years ago appointed a Director of Intellectual Capital whose business
is to codify Skandia’s businesses processes. Once codified, they can be
analysed, optimised and reproduced - for example when diversifying into
new markets. Nonetheless, the same thing happens in many companies
through, for example, engineering and reengineering, computer systems
development and the articulation of processes and company standards.
Creating intellectual capital - technology in the older sense of “a discourse
or treatise on an art” - in this way is intended to improve the firm’s
effectiveness. But it also tends to lock the company into specific products,
markets and technological trajectories, promoting ‘path-dependency’. The
problems can in principle be reduced if the corporate memory can forget as
well as learn.23
In Exhibit 7, we set out a simple way to think about technological
capabilities which captures both this need for a corporate memory and the
need to connect it with the market. It is based on our literature survey and
shows three kinds of capabilities: internal; external; and strategic. These
are interlinked and interdependent, because they are involved in a
dynamic learning process.
22
23
Metcalfe and de Liso, Ibid
Johanson in Lundvall, 1992
16
Exhibit 7 Key Elements of Technological Capability
•
•
Strategic Capabilities
Search for market opportunities
Understand and manage the fit between the firm’s capabilities
and market needs
Internal Capabilities
Manage tangible technology base
• Products
• R&D facilities
• Appropriate plant and equipment
Develop and manage appropriate
intangible resources
• Codified intellectual capital
• Qualification and skills profile
adapted to the needs of the firm
• Tacit knowledge
Create needed organisation
• Technology management
capabilities
• Change-management capabilities
• Coordination among internal
‘owners’ of capabilities
External Capabilities (Networking)
Access external knowledge
• Science
• Technology, techniques
• Artifacts, practices
• Know-how, tacit knowledge
• Information resources
Manage producer/user relations
Access partners with needed
complementary assets
• Complementary knowledge
• Complementary production
• Complementary supply-chain role
The strategic level provides the intelligence or control mechanism which
allows the firm to manage its capabilities and exploit them via the market.
This meta-level involves the entrepreneur in deliberately stepping outside
the accustomed circular flow of daily economic life, trying to understand
what knowledge makes the business succeed and using this knowledge
about knowledge to increase performance.
In modern industrial practice, the strategic function does not have a
monopoly of learning, but ensures that it takes place at all levels of the
firm, for example through Continuous Improvement groups. Intelligence
is thus distributed through the firm, rather than belonging solely to an
‘heroic’ Schumpeterian entrepreneur.
The second category has to do with the internal capabilities of the firm: its
management’s ability to
•
•
•
Identify and invest in the right physical infrastructure to meet the
competitive requirements of the firm
Analyse its situation, identify and put in place the needed skills
Organise appropriately, and have the vision to understand when
organisation needs to change
17
Inevitably, the right levels of attainment here are competitively
determined - even though it may be convenient in the support system to
think about absolute levels of performance (such as conformity with the
ISO 9000 quality standard).
The third group of capabilities is external - or, more precisely, concerned
with managing the relationship between the firm and the outside
resources which it needs. These are largely the issues addressed by the
current discussions of ‘networking’ in the innovation literature. If
contemporary writers are correct that networking is central to the
innovation process, then the ability to network must itself be a crucial
capability. This means, then, making use of external knowledge, using
partners to access complementary assets and managing the producer/user
relations which have consistently been identified in the innovation
literature as key to innovative success.
Both internal capabilities and external linkages - generally subsumed
under the catch-all term ‘networks’ - are clearly important
Cross-sector studies have found that external sources contribute
around one third of all knowledge used in innovation, with more
being obtained from other companies than [public sector research]
institutions.... Of the two-thirds that are obtained internally, half
is knowledge which is personally held24
Tacit knowledge is therefore an extremely significant part of technological
capability, though its role does vary greatly between sectors. It is a key part
of the distributed intelligence of the firm.
Provided technological capabilities are strong, it is possible for real firms to
move towards the ideal or norm of firm behaviour anticipated in the
neoclassical model (cp Exhibit 2)
•
•
•
•
Networking and the capability within the firm to search for and
understand external signals puts it in a position to make betterinformed decisions
Strategic and learning capabilities in the firm keep decisions rational
and relevant to changing circumstances
Strong capabilities provide firms with the greatest possible freedom
in choice of technology, within the constraints of natural laws
They enable innovation as well as imitation or incremental
development of others’ technologies
Innovation policies which encourage capability development and
exploitation will therefore tend to promote efficient markets as well as
industrial competitiveness.
24
Roy Rothwelland Walter Zegveld, Reindustrialisation and Technology, Harlow:
Longman, 1987: 14
18
4
Hierarchies of Technological Capability
What does technological capability allow firms to do? Trying to answer
this question makes it clear that there are hierarchies or levels of ability.
At its simplest, this hierarchical thinking is embedded in Exhibit 8, which
suggests that technologies can be understood at different levels.25
Exhibit 8 Degrees of Mastery of Technologies
Technology acquired in turnkey form.
No ability to innovate or firefight
Minimum capability to make
applications-based adaptations
Ability to make incremental
improvements to the technology
itself, as well as to its applications
Able to develop significantly new
variants or innovations
For many technologies, a ‘black box’ understanding is perfectly adequate.
Few companies, for example, would benefit from a deep understanding of
the computer or telephone equipment they use every day. On the other
hand, the business’ core technologies generally need to be quite well
understood. Whether they are acquired through internal development or
by transfer, the aim will generally be to move as far towards an ‘unboxed’
understanding as is needed to generate and sustain competitive advantage.
Bell26 has set out a more detailed scheme, oriented to the inward transfer
of technology to developing countries and the creation of local
technological capability (Exhibit 9). He spells out in a level of detail that
appears unparalleled in the literature what it should be possible to do at
different levels of technological development. His levels correspond
closely to those we sketched in the previous Exhibit, but takes on board the
wider questions of the type of engineering skill needed to make
investments, build machinery, develop products and link the firm with
outside sources of technology.
25
26
We owe some of the ideas here directly to Cuneyt Öge
Martin Bell et al, Aiming for 2020: A Demand-Driven Perspective on Industrial
Technology Policy in Malaysia, Brighton: Science Policy Research Unit, 1995
19
20
Inherently, behind these hierarchies of abilities must lie a hierarchy of
capabilities. Exhibit 10 shows a simple hypothesis about a useful way to
segment companies according to their level of research and engineering
capability. It is far from perfect, and at present we have no empirical basis
for mapping it across to the hierarchies of abilities just described. (This
may, in fact, only be possible at the sector or firm level.)
Exhibit 10 Simple Hierarchy of Company Types
Company Types
• Research department or
Research
equivalent
Performers • Able to take long run view of
technological capabilities
Technological
Competents
• Multiple engineers
• Some budgetary discretion
• Able to participate in technology
networks
• One engineer
• Able to adopt/adapt packaged
solutions
• May need implementation help
MinimumCapability
Companies’
• No meaningful technological
capability
• No perceived need for this
• May be no actual need
Low-Technology
SMEs
Our segmentation suggests that there are four reasonably distinct levels in
the development of firms’ engineering and research capabilities. At the
bottom level, there is no meaningful capability and there will tend to be a
presumption that none is needed. At the next level up, the ‘minimum
capability’ level, the firm acquires at least one person able to speak the
language of technology, to monitor and understand the significance of
technological changes happening outside the firm. These bottom two
levels of firm rarely have much contact with universities. They do not
share a common language or interest with them. The professors are likely
to be interested in things which are longer-term than they can consider.
In OECD countries, many larger firms belong to the third level of
‘technological competents’, where there is enough capability to do fairly
serious development work and where there tends to be a specialised
innovation or development function. The highest level firms - ‘research
performers’ - are of two types. Some correspond to the ideal of the very
large company with capabilities in research as well as development and
the strength and vision to work for the long term as well as the immediate
future. Others are new, technology-based firms such as university or other
research spinoffs, many of which exist primarily to do research and will be
21
absorbed by larger companies if their work is successful. These highestlevel firms’ research departments communicate easily with universities.
Third-level firms often have difficulties in doing so.
These hierarchical ways of describing technological capabilities provide
useful clues about the need to segment and to build hierarchy into policies
aimed at developing company capabilities more generally. What we lack
is some equivalent hierarchy of the ‘soft’ capabilities needed to move from
technological change to innovation. A static view of these capabilities is
built into Exhibit 7, but an hierarchical or dynamic view would be more
useful for policy formulation.
5
Elements of Technological Capability
In this section, we elaborate further on individual aspects of technological
capability, raising issues for support infrastructures.
Exhibit 7 does not differentiate between different sizes of firm or stages of
development. Rather, it represents an ideal of capabilities that are needed
for innovative and competitive success. The extent to which companies
actually possess these capabilities tends to increase with firm size.27 This
seems mainly to be because increasing size brings increasing division of
labour, allowing the firm to develop and devote the specialists skills
required for the various dimensions of technological capability. Small,
technology-based firms are the major systematic exception: they tend to
have strong information and technology networking skills, but are not
always as good at business skills. Many of the tasks of the support system
in increasing technological capabilities involve helping SMEs act as if they
were bigger than they actually are. In the best case, this becomes a kind of
self-fulfilling prophecy, where SMEs grow to become large firms.
5.1
Strategic Capabilities
In addition to providing a key part of the ‘search intelligence’ needed to
develop and manage technological capabilities, the strategic functions are
the major interface with the firm’s business capabilities. In particular, this
is where the understanding of customer needs and desires, technological
opportunities and the company’s own capabilities need to be matched
together. This defines the core competencies of the firm. Product strategy
needs to be hammered out here, based on inputs from the three areas
mentioned.
27
For example, Mordchelles-Regnier et al have found that the critical size of firm
needed to support an R&D department is about 1000 employees in low-technology
businesses and 100 employees in high-technology areas. Mordchelles-Regnier et al,
“Le rôle des sociétés de recherche sous contrat vis-à-vis des PME,” Annales des
Mines, July/August 1987
22
We have not extended our review to cover the huge literature on general
business management. The salient point from an innovation perspective
is that the ability to wrap technological issues into business strategy - and
vice versa - is an important capability. This requires an awareness of
business and technology issues that is often not found in the same person:
a problem, especially in small owner-managed companies. Increasingly,
this fact is being reflected in education, awareness and support
programmes aimed at capability development.
5.2
Internal Capabilities
In our model, internal capabilities have three main elements
•
•
•
Managing the tangible technology base
Developing and managing intangible resources
Creating the organisation needed to make effective use of these
assets in pursuit of the company’s business mission
The tangible basis of companies’ internal technological capabilities is their
products and the design and production facilities employed, which need to
be adequate to meet competitive needs. Choice, maintenance and renewal
of these is based on design and engineering skills. In the smallest firms,
these often belong to the people who are also the owners and managers.
The pressure of their combined roles can mean that the technology used is
out-dated or otherwise sub-optimal. As STEP observe in the Norwegian
context28
1
The proportion of innovating firms in a size class rises with firm
size. Among the firms with less than 10 employees, only 16%
engaged in innovation activity, as opposed to 72% for firms with
more than 100 employees. This suggests that the scope for
increasing activity in SMEs may be large
2
When small firms innovate, they must spend much higher
proportions of their total sales on innovation-related activities
than large firms; this suggests that if an innovation fails, the result
is much more serious for small firms than for large firms.
3
Within the small-firm class, innovation activity is distributed
very unevenly. Less than 10% of the small firms account for the
majority of new product sales. Again, this suggests considerable
scope for extending innovation and research performance
Very small firms are typically dominated by a single manager - often the
owner. The capabilities of this single individual therefore have a dramatic
impact on firm performance. Small enterprises which use new
production technologies typically have an owner/director who is well
28
Keith Smith, Policy Options for the Norwegian SME Sector, Draft Report to NFR,
Oslo: STEP Group, February 1997
23
educated or a graduate, or has supervisory staff with higher education.
These people use more sophisticated management practices, including
‘technology watch’ to monitor relevant events in technology.29
Generally, it seems that small firms under, say, 60 employees conduct
research in a very sporadic way, unlike in big companies where a regular
budget can be allocated. However, the SMEs which commercialise
inventions tend to produce more radical leaps than their larger
counterparts. Small innovating firms spend a greater proportion of their
innovation investments than large firms on non-R&D activities.30 ,31
Further, “... there is a clear tendency for the share of firms who acquire
outside technology to rise across size classes. Given that it is usually
believed that SMEs have a greater need for external technology inputs
than large firms, this suggests a role for policy.”32
Tacit knowledge is, almost by definition, hard to manage. Codifying this
knowledge, in order systematically to improve it, is an important objective
of technologists.
Codification may be understood as a process of generalising what is
specific and translating messages into a common and shared
language. It involves the establishment of technical standards and
of basing technical development on general scientific principles. A
special aspect relates to the design of the innovation process itself
where information technology makes it possible to pursue
development work on computers through virtual experiments rather
than through real tests in real laboratories.... This new step in the
codification of knowledge is important because it moves the border
between tacit and codified knowledge. But it does not necessarily
reduce the relative importance of skills, competencies and other
elements of tacit knowledge, however. The more easy and less
expensive access to information makes skills and competencies
relating to the selection and use of information even more crucial
than before. In general, skills related to handling codified
knowledge become more important in the labour markets.... The
most fundamental aspect of learning is perhaps the transformation
of tacit into codified knowledge and the movement back into
practice, where new kinds of knowledge are developed.... At any
point of time a certain amount of knowledge is in the pipeline being
in the process of codification. While some engineers and scientists
are involved in producing innovations and inventions, a much bigger
proportion is engaged in standardisation and in codifying and
generalising knowledge.33
29
30
31
32
33
OECD, Small and Medium Enterprises: Technology and Competitiveness, Paris:
OECD, 1993
Tore Sandven, Typologies of Innovation in Small and Medium-Sized Enterprises in
N o r w a y, STEP Notat No 4, 1996
Tore Sandven, Innovation Outputs in the Norwegian Economy: How Innovative are
Small Firms? STEP Notat No 5, 1996
Ibid
Bengt-Åke Lundvall, “The Learning Economy - Challenges to Economic Theory and
Policy”, Paper presented at EAEPE Conference in Copenhagen, 27-29 October 1994
24
Codification results in the production of operating procedures, manuals
(including quality manuals), standards, norms and even innovations in
machinery. But continuing dependence on tacit knowledge for
innovation arises from
•
•
•
•
The tendency for advances in knowledge and techniques to be
associated with new tacit knowledge
Adherence to previously successful practice
Lack of scientific or technological expertise within certain firms or
sectors
Systems complexity, which makes it hard to provide a
comprehensive codification34
Because tacit knowledge is person-embodied, it follows that managing it is
linked to both engineering and personnel policy. It can be transferred
internally through co-work or externally via interpersonal networking.
Broader workforce skills are clearly a key aspect of firm capability.
Industrial training systems have long been forced to tackle the task of
defining the technological capabilities required by different categories of
individuals. The UK’s recent and rather belated attempt to relaunch the
idea of technological training through a system of National Vocational
Qualifications (NVQs) is typical in trying to define “units of competence”
which can be taught and tested. While undoubtedly important, a
weakness of this ‘bottom-up’ approach to defining and implementing
competences is its lack of connection to company strategy and the specific
needs of the firm. A dynamic mechanism needs to be in place for
reviewing needs and updating the required definitions of competence.35
This is a natural function of the personnel and training department in a
large company, but is more difficult to accomplish in smaller firms, which
may need help from the support system.
The various aspects of technological capability need to be coordinated in
order to pursue the company’s interests, so good internal communications
between research, development, production and marketing are also
necessary for successful innovation.36 In management terms this means
there has to be a good match between business and technology strategies, as
well as means to operationalise this match.
Bessant and Rush identify one such organisational ability as the capability
of management to organise and implement technology transfer. This,
they say, has seven aspects
34
35
36
Jacqueline Senker and Wendy Faulkner, “Networks, Tacit Knowledge and
Innovation,” Brighton, Science Policy Research Unit, 1993
Graeme Currie and Roger Derby, “Competence-based management development:
rhetoric and reality’”, Journal of European Industrial Training, May 1995, 19(5),
pp11-14
Roy Rothwell et al, “SAPPHO Updated - Project SAPPHO, Phase II,” Research
Policy, 1974, 3, pp258-291
25
1
2
3
4
5
6
7
Recognition of requirements for technology through a systematic
and regular audit of its current competences and a comparison of
those which it needs to develop or acquire in order to become or
remain competitive. Essentially, firms should have technology
strategies and be able to plan their growth and development
Exploration of the range of technological options available (there
may be several competing solutions to the problem such as different
machines, different technologies, different suppliers, etc) and
search widely for these so as to get a good fit with their needs
Comparison between all the options available which can be
achieved through some form of benchmarking
Selection of the most appropriate option based upon this
comparison
Acquisition of the technology (either through direct purchase or
via some form of licence, collaboration, alliance, etc) This is likely
to involve extensive negotiation around price, specification,
transfer of knowledge, property rights, etc
Implementation of the technology within the firm. This may
involve extensive project planning and management activities and
require configuration of both technology and organisation to get a
good and workable fit
Operation of the technology and learning about how best to use it.
Over time, this may involve extensive learning and development;
competence is very much the product of this last stage of
accumulation and incremental development, and much of what is
represented by technological competence is highly specific and
often tacit in form37
However, it is difficult to change organisations. They tend to have “sticky
resource endowments”38 - the negative aspects of the ‘memory’ we
referred to above - so firms lack the ability to build new capabilities
quickly. Since some aspects of capability are tacit and non-tradeable, it is
often not possible simply to buy one’s way around the difficulty.
New technology - especially Information Technology - can have
organisational implications. Exhibit 11 schematically shows three levels of
IT adoption, with increasing productive effectiveness but also with
increasingly radical implications for the organisation of work and of the
entire adopting firm.
37
38
John Bessant and Howard Rush, “Building Bridges for Innovation: The Role of
Consultants in Technology Transfer,” Research Policy, 1995 (24), pp97-114
David Teece, et al, Op Cit, 1992
26
Exhibit 11 A Three-Part Model of the IT Adoption Process
Substitution
Understand
and document
the process to
be IT-assisted
Substitute IT
for existing
technology
Reap modest
effectiveness
and efficiency
gains
Enhancement
Redesign,
based on needs
& technology
opportunities
Change the
process with
the help of IT
Obtain larger
effectiveness
and efficiency
gains
Change the
organisation
and potentially
its strategy
Maximise
effectiveness
and efficiency
increases
Transformation
Redefine the
role of the
process
Source: Nagy Hanna, Ken Guy and Erik Arnold, The Diffusion of Information Technology,
World Bank Discussion Papers No 281, Washington: World Bank, 1995
Clearly, companies without the ability to reengineer themselves - whether
based on internal or external expertise - cannot reap the economic benefits
of the ‘transformation’ stage in IT adoption. This reengineering in turn
depends on the ability to do objective and critical analysis within the firm.
It is much harder to achieve in the presence of owner-managers than if it
is done alone by professional technologists.
5.3
External Capabilities (Networking)
External or networking technological capabilities involve
•
•
•
Accessing external knowledge
Managing the producer/user relationship which is central to
successful innovation
Accessing other partners who have useful complementary assets
and capabilities
While it is tempting to dismiss as mere fashion the current tendency to
describe almost any external relationship of a firm as ‘networking’, the role
of such external relations in successful innovation39 and in learning is
extremely important.
39
Rothwell et al, 1974, Op Cit
27
Learning is a social process. It is seldom done individually, without
the support of, or isolated from, interpersonal interactions....
Innovation may accordingly be viewed as a collective activity; an
outcome of communication and interaction between people.40
Networking is not independent of other capabilities. A firm’s ability to
forge effective external linkages depends so no small extent on its in-house
skills, typically in the form of qualified scientists and engineers.41
“Personal relationships of trust and confidence (and sometimes fear and
obligation) are important both at the formal and informal level.”42 They
reduce transaction costs because the parties involved are known to each
other. They make additional transactions between the partners more
likely by reducing the amount of search effort needed before making a
transaction and by providing channels through which ideas and
opportunities are fed to the firm.
Freeman distinguishes ten types of network that are important in
innovation
1
2
3
4
5
6
7
8
9
10
Joint ventures and research corporations
Joint R&D agreements
Technology exchange agreements
Direct investment (minority holdings) motivated by technology
Licensing and second-sourcing agreements
Sub-contracting, production sharing and supplier networks
Research Associations
Government-sponsored joint research programmes
Computerised data banks and value-added networks for technical
and scientific interchange
Other networks, including informal networks43
Other work at the Science Policy Research Unit by Rothwell confirms that
innovators tend to be well networked. An analysis of SMEs represented in
SPRU’s innovations database showed that external R&D relationships of
various kinds were surprisingly important (Exhibit 12). More generally,
the employment of qualified scientists and engineers (QSEs) is strongly
associated with the ability to make use of external scientific and
technological resources - both in the form of people and information.44
Studies in the plastics processing industry suggest that lack of QSEs and
technicians limits companies’ ability to use the technical information
available from suppliers and reduces the quality of products.45
40
41
42
43
44
45
Björn Johnson, “Institutional Learning’” in Bengt-Åke Lundvall Op Cit, p 34
Rothwell, Op Cit, 1991
Christopher Freeman, “Networks of innovators: A synthesis of research issues,”
Research Policy, 1991, 20, pp499-514
Christopher Freeman, Ibid, p502
Jacqueline Senker, “Small and medium sized firms’ access to the science base,”
Brighton: Science Policy Research Unit, 1993
Vivian Walsch et al, Technical Change and Skilled Manpower Needs in the
Plastics Processing Industry, Brighton: Science Policy Research Unit, 1980
28
Exhibit 12 External Linkages of SMEs in SPRU Innovation Database
Linkages
Percent
Contract R&D
Collaborative R&D
agency for other firms
Collaborative marketing arrangements
Used sub-contract manufacturing
Produced under licence
Sponsored student at universities
Took ‘sandwich’ students for training
38%
26%
25%
37%
68%
16%
31%
39%
Source: Roy Rothwell, “External networking and innovation in small and mediumsized manufacturing firms in Europe’” Technovation, 11 (2), 1991, pp93-112
The ability of the scientist to reach into the wider research community for
solutions to problems is one of the bases of the relationship between
science and industry, as Gibbons and Johnson noted in their path-breaking
study of the sources of knowledge for innovation.46 While scientists tend
to rely more on published sources and engineers on oral
communications47 both categories of people are vital for accessing
technological knowledge
Sometimes [technical knowledge] is summarised and stored in a
data bank as a patent or a documented know-how. But usually such
information, useful for an unsolved problem, exists randomly in the
society as tacit knowledge. In this sense it depends on the context of
a special problem. To approach such information, ‘know-who’ is
crucial.48
Companies increasingly take care that they locate in places where there is
an adequate scientific49 and technical infrastructure.
Some companies do, in fact, make considerable use of external
information about technology. Generically, the activities involved in
accessing external knowledge, integrating it with internal research and
absorbing it into the company look rather like those shown in Exhibit 13.
46
47
48
49
Michael Gibbons and Ron Johnson, “The roles of science in technological
innovation,” Research Policy, 1974, 3, pp220-242
J P Lester , “The utilisation of policy analysis by state agency officials,”
Knowledge: Creation, Diffusion, Utilisation, 1993, 14(3), pp267-290
Ken-Ichi Imai in Rothwell, 1987
Ben Martin, Ammon Salter et al, The Relationship Between Publicly Funded Basic
Research and Economic Performance, Report to HM Treasury, Brighton: SPRU, 1996
29
Exhibit 13 A Systems View of the Role of Research
Research Process
Identify
Acquire
Exploit
Communicate
Policy
Decisions
Track
Create
Explore
Delivery Process
Understand
Internalise
or Reject
Standardise
Build Tools
&
Capabilities
Communicate
Improvements
or
Solutions
Design
and
Manufacture
Specify
Problems
Resources
Firms identify opportunities by tracking changes in knowledge and
practice or (probably less frequently) by doing research which creates new
opportunities. This takes effort, because information flows are imperfect.
It also takes skills: both formal skills, represented by qualifications; and
informal skills or experience. They must then test the opportunities, build
any tools needed to understand and internalise them and communicate
the results internally to those responsible for design and production.
As the Exhibit implies, doing all this involves many processes and,
usually, quite a number of people. Few SMEs can afford to do these
activities in-house or in a formal way. Many innovation support systems
therefore include measures which address this weakness of SMEs.
Kleinknecht has confirmed that for Dutch SMEs that the problems of
keeping abreast of technological development by monitoring future
applications, finding technological information and developing employee
skills are all significant at small firm sizes but tail off in importance once
firms grow above 200 employees.50
More generally, we can say that SMEs often lack the resources and desire to
monitor and assimilate external technology in the way shown in Exhibit
10. In practice, policy systems often try to compensate for this lack.
The ‘gap’ between companies and potential external sources of technology
can be wide. Exhibit 14 illustrates the general experience of this gap.
However, the actual width of the gap varies according to the capabilities of
the companies involved, and to their level of engagement with the
50
Kleinknecht, “Firm size and innovation,” Small Business Economics, 1 (2), pp 215222
30
technology supplier. Notably, studies in France, Italy and the Netherlands
have noted that the presence of one or more research centres in a region is
not an important factor in the effective dissemination of information
unless the centre was started or is controlled by the users.51
Exhibit 14 Generic Gaps Between Users and Suppliers of Technology
Users
Gap
Suppliers
• Universities
• Research
Institutes
Company
System
• Technology Centres
• Consultants
• Professional Associations
• Chambers of Commerce
• Competitors
• Suppliers
• Customers
• Trade Press/Exhibitions
The price of access to science - as opposed to part- or ready-packaged
technologies - is high. Hicks stresses that companies and academic
researchers alike “in fast-moving areas of science, to be aware of current
work, must be in contact with leading researchers in the field because by
the time research results are published they are no longer current.”52 Not
only academics but also companies performing research publish in
refereed scientific journals. Hicks argues that companies publish both to
buy an ‘entry ticket’ into relevant scientific “invisible colleges”53 and to
signal the presence of unpublishable, possibly tacit, capabilities which can
make them interesting partners.
Considerable resources are needed to participate in science as a source of
exploitable knowledge. Senior members of company research departments
often take care to secure a part-time professorship in order to be ‘insiders’
in relevant scientific communities. Callon54 has argued that the costs of
using science are so high that it can no longer be regarded as a public good.
A company wanting to use science today needs a well-equipped laboratory
and highly-qualified personnel - in our terminology, they must already be
51
52
53
54
OECD, 1992, Op Cit, p39
Diana Hicks, “Published papers, tacit competencies and corporate management of
the public/private character of knowledge.” Industrial and Corporate Change,
1995, 4(2) p413
Derek de Solla Price, Big Science, Little Science, 1963
Michel Callon, “Is science a public good?” Science, Technology and Human Values,
19 (4), Autumn 1994, pp395-424
31
a ‘research performer’. Another study of 39 research-intensive SMEs
confirmed that
They have significant involvement in in-house research and also
make extensive use of external research.... Internal research is
largely aimed at getting processes and products to market; external
research is for discovering concepts for future generations of products
and processes.55
It is not just university research that is most readily accessed by the most
capable. While Research Associations have traditionally been set up to
serve the needs of smaller firms through shared research facilities, in fact
it is large firms which are best able to make use of them.56 A key
recommendation emerging from work on academic-industry links is that
the numbers of Qualified Scientists and Engineers (QSEs) employed in
industry must be sufficiently high to permit good networking. QSEs build
informal relations with the research sector, bartering goods and services
for access to expertise.57
Links between SMEs and external science can, therefore, be extremely good
in research-intensive areas. The (often new) technology-based firms
which can make this link to research have very different resources and
capabilities than the great bulk of companies of a comparable size. For
more traditional companies, the idea of using scientific inputs remains
remote.58
The importance of close producer-user relations in successful innovation
has long been understood.59 Their role does seem to vary between
sectors 60 but is especially important in new and unstable technologies. A
key element is physical proximity, because inter-personal relations are
most important when things are uncertain.61 Correspondingly, at least in
such fast-moving businesses, an important aspect of capability must be
physical location. This is one of the reasons why such new industries
grow in small geographic clusters rather than at random. Successful
globalisation of markets and research linkages by technology-intensive
55
56
57
58
59
60
61
R Oakey et al, New Firms in the Biotechnology Industry, London: Pinter, 1990
Christopher Freeman, Op Cit 1991
Jacqueline Senker, “Overcoming barriers to technology transfer in small and
medium-sized firms,” Paper to COST Workshop, Milan, February 1-2, 1996
Jacqueline Senker, “Small and medium sized firms’ access to the science base,”
Brighton: Science Policy Research Unit, 1993
cp Roy Rothwell et al, “SAPPHO Updated - Project SAPPHO, Phase II,” Research
Policy, 1974, 3, pp258-291; Rothwell, 1991; Shimshoni
Pavitt, K., and Patel, P., ‘The innovative performance of the world’s largest firms:
some new evidence’, Economics of Innovation and New Technologies, Vol 2: 2, p. 91 102, 1992
Björn Johnson in Bengt-Åke Lundvall, Op Cit
32
SMEs in particular appear often to be grounded in high levels of local
networking with respect to research collaboration and intra-industry
links.62
Håkansson’s survey of networking among Swedish SMEs shows that
producer-user relations are crucial in innovation by these firms. A
significant fraction of the resources allocated to technical development
involves cooperation with at least one external actor. These network
relationships were durable - with an average age of almost 10 years. The
majority (almost 80%) of the cooperation activities involved were
conducted informally, without involving any sort of legal agreement.63
In relation to the broader use of networking as a way to access
complementary assets, Rothwell argues that SMEs treat networks as “more
or less complex systems for the exchange of information horizontally and
vertically.” The networks may be loose, with free circulation of partial
information, or they may be tighter and more organised. In this case,
more information circulates, but there are more ‘strings’ attached. Because
information networks function on the basis of trust, older networks tend
to do better than newer ones.64 More generally, SMEs network in order to
exploit personal contacts, to undertake ‘technology watch’ and to simplify
their otherwise rather complex environment by reducing the size of the
initial ‘search space’ when needing information.65
However, there are also some interesting variations in SME abilities to
network. One pattern is that they seem to network increasingly as they
grow up to some 150-200 employees. Thereafter, the degree of networking
tails off, presumably in response to their developing internal capabilities.
Thus, Ratti and Baggi found that the number of technology agreements
entered into by enterprises grew with firm size up to the 100-199 range,
then tailed off.66 In our (unpublished) study for the British Department of
Trade and Industry of the way SMEs use the state-funded infrastructure
which supports industry, we found a similar effect where the frequency of
use of the support system peaked in the same size range. The most
effective users of the Irish state Technology Transfer and Partnership
programme were also in the range 100-200 employees.67
62
63
64
65
66
67
D Keeble, C Lawton, H Smith, B Moore, F Wilkinson, Internationalisation
processes, networking and local embeddedness in technology-intensive small firms,
ESRC Centre for Business Research, Cambridge Universirty, Working Paper 53, 1997
H Håkansson, Corporate Technological Behaviour: Cooperation and Networks,
London: RKP, 1989
Roy Rothwell, “External networking and innovation in small and medium-sized
manufacturing firms in Europe,” Brighton: Science Policy Research Unit (mimeo),
1990
OECD, Op Cit
R Ratti and M Baggi, “Analyse stratégique et spatiale des accords de coopération
entre les entreprises du secteur industriel,” Revue d’économie régionale et urbaine’
No 3/4, pp 464-478
Erik Arnold et al, The Technology Transfer and Partnership Programme: An
Evaluation, Dublin: Forfás (Forthcoming)
33
Not only knowledge but also other external complementary assets are
helpful to the development of the firm. In fast-moving products,
outsourcing can produce competitive advantage by enabling swift changes
of products and production processes. This implies that ‘networking’ with
sub-contractors and components suppliers can in certain cases be an
important capability. (In contrast, more stable businesses lend themselves
more to vertically integrated structures.) “Most firms designate a group of
‘privileged’ sub-contractors with whom to build these close relationships
[of collaboration, extending into design]. This group normally includes the
20 percent of suppliers which provide 75-80 percent of the value of its
components.”68 Industrial procurement orthodoxy has been shifting from
adversarial to ‘partnership’ models over the past twenty years or so,
especially in industries where co-development is desirable, as for example
between a car manufacturer and an automotive components company.
The other side of this coin is that SMEs are easily dominated by a single
customer, and can readily become ‘locked’ into a particular configuration,
serving that single customer.69 This may lead to ‘arrested development’
and over-dependence, so that policy measures to encourage diversification
of the customer base become desirable in order to promote growth.
6
Technological Capability and SME Support
Traditionally, arguments for intervention through a technology support
structure relate to three locations of ‘market’ failure: scientific research;
technology diffusion; and infrastructure (Exhibit 15). Aspects of support to
develop SME technology capabilities fall under each of these headings.
This leaves a crucial question of how the interface should work between
SMEs and the support system. This section ends with some suggestions.
Ken Arrow probably best captured the traditional argument for
government intervention in research in his famous 1962 article.70 He
identified three major sources of market failure which made it useful for
government to fund research
68
69
70
AnnaLee Saxenian, “The origins and dynamics of production networks in Silicon
Valley,” Research Policy, 20, 1991, pp 423-437
Erkko Autio, “Supporting the consolidation and internationalisation of SMEs - The
systemic perspective,” in Osmo Kuusi (ed), Innovation Systems and
Competitiveness, Helsinki: Taloustieto Oy, 1996
Ken Arrow (1962), “Economic Welfare and the Allocation of Resources for
Invention,” in Nathan Rosenberg (Ed.) (1971), The Economics of Technological
Change, Harmondsworth: Penguin
34
•
•
•
Indivisibility, because of the existence of minimum efficient scale
Inappropriability of the profit stream from research, leading to a
divergence between public and private returns on investment
Uncertainty, namely divergences in the riskiness of research
respectively for private and public actors
Exhibit 15 Types of Failure Addressed by Technology Policy
Diffusion
Bridging
R&D
• Indivisibility
• Indivisibility
• Information Failure
• Inappropriability
• Capability Failure
• Uncertainty
Infrastructure Failure
With respect to technology diffusion,71 there are also three major
categories of failure
•
As with research, problems of indivisibility arise
•
In addition, diffusion can be impeded by capability failure, namely
-
•
Information failure. Timely and accurate information is a
prerequisite for rational behaviour. Information failure
therefore comprises
-
71
A shortfall in the technical skills needed to adopt a new technique,
when adoption would otherwise be economically rational
Organisational inadequacies, which prevent the rational
exploitation of new techniques
Deficiencies in business skills and understanding on the part of the
firm or the infrastructure on which it depends (eg banks), which
prevent the firm from taking rational decisions
Inadequate availability of information about new technological
opportunities for entrepreneurs to know a choice-of-technique
decision is possible
Presentation of the needed information in a way which is not useful
or credible to the potential adopter
Erik Arnold and Ken Guy, “Diffusion Policies for IT: The Way Forward,” (mimeo)
Paper prepared for OECD/ICCP Expert Group on the Economic Implications of
Information Technologies, Paris, November 7-8, 1991
35
The aspect of rationality is extremely important. Particularly among SMEs,
there are substantial barriers to the rational behaviour assumed in
mainstream economics and in a great deal of international policy
discussion. Unless firms can operate rationally, markets must operate
imperfectly.
The other potential area of failure is in the infrastructure of education,
metrology, information and so on, including the parts of the infrastructure
which exist to provide bridges between companies and the R&D system.
We can perform a first, crude check on our analysis of technology
capabilities by seeing if they have already been identified by policymakers
and addressed by support programmes. Exhibit 16 maps policies we have
identified72 in practice onto the technology capability categories. Based
simply and crudely on the numbers of policies shown, most policy
attention goes to fostering firms’ external links. Comparatively little
attention is paid to education and development aimed directly at the
entrepreneur, establishing a bedrock of people in firms who have formal
qualifications in science and engineering, or internal and organisational
questions such as identification, development and maintenance of
intellectual capital. Based on what the literature has to say about the
importance of these aspects of technological capability, it may be useful to
devote more policy attention to them.
Ideally, we would have liked to find results in the literature which could
help refine an hierarchical or dynamic approach to capabilities. Based on
the current state of knowledge, we have had to settle for a more static
approach, which essentially provides a checklist of desirable capabilities.
This leaves the question of how much of each capability is needed by
different types of firm to be resolved largely case by case. This may not be
all bad. Much experience suggests that when the state tries to overdetermine economic situations, things tend to go wrong.
The complexity of technology capability, and the limits to our
understanding of it and its role in the overall innovation system, suggests
that interventions may be needed at multiple points in the system. There
is not a single, simple lever that policymakers can pull in order to
improve capabilities and performance at a stroke.
72
Our chief sources are work we conducted for the World Bank, surveying IT diffusion
policies in eight countries, which is reported in Hanna, Guy and Arnold, 1995 (Op
Cit); supplemented by more recent work - Erik Arnold, Technology Support for SMEs
in Northern Sweden, Stockholm: NUTEK, 1996
36
Exhibit 16 A Policy Repertoire for Improving Technological Capabilities
•
•
•
•
Strategic Capabilities
Business capability development, especially marketing
Business and technology audits; mentoring
Awareness programmes, including visits and comparisons
Feasibility assessments
Internal Capabilities
External Capabilities (Networking)
Manage tangible technology base
• Product development assistance
• R&D tax breaks
• State-subsidised R&D programmes
• Manufacturing consultancy
Access external knowledge
• Innovation ‘cheques’ or credits
• Science parks
• Technology centres
• Research institutes and
associations
• Technology development networks
• Technology transfer programmes
and brokerage
- Research-industry
- Company-to-company
• University liaison officers
• Faculty industrial placements
• Subsidy to university/industry
R&D collaborations
• Technology information services
• Metrology programmes and
services
Develop and manage appropriate
intangible resources
• Quality programmes
• Placements of qualified personnel,
eg engineering graduates
• Loans of research personnel
• Training needs analysis and
training programmes
Create needed organisation
• Technology management courses
Manage producer/user relations
• Procurement programmes
- State procurement
- Company supplier development
Access partners with needed
complementary assets
• Partner-search programmes
• Inter-company network
programmes
The heterogeneity of firms, too, militates against such a simple policy
approach. Policymakers must continue to struggle to segment firms into
groups with generically similar needs. However, the design of policy
delivery systems needs also to take account of the uniqueness of firms as
individuals. Some state support infrastructures are (in our view, rightly)
taking an approach of diagnosing needs at the level of individual
companies as part of their approach to developing companies and
capabilities.
37
The need for such a diagnostic approach is built into the ‘learning paradox’
associated with capabilities. Those with limited capabilities also have
limited ability to identify their own problems and opportunities. If it is to
help, the state needs to be proactive with those who cannot yet decide what
to do for the best. Despite the attractions of ‘hands-off’ policies which do
not involve the state in making firm-level decisions (’picking winners’), it
is therefore difficult to imagine many effective ‘hands-off’ measures to
improve capabilities (especially among weaker firms). This means that
progress in policy depends not only on finding the right economic levers
but on closer engagement with firms and technological practice.
Generally, quite a number of actors and programmes share the task of
developing technology capabilities. However, the support system needs to
operate as an effective whole. Exhibit 17 illustrates how such programmes
can relate together into a developmental staircase, even if the services
need not map onto Exhibit 17 in a one-to-one manner.
It is important that individual programmes operate in conjunction with
the other parts of the support system to
•
•
•
•
Obtain cost synergies, especially in needs diagnosis
Enable cross-referrals to and from other parts of the system
Enable an holistic approach by the system to company development
Avoid fragmentation and build the scale necessary to provide highquality, specialist capabilities
Exhibit 17 A ‘Staircase’ for Developing Company Capabilities
Research
Services
Technological
development
Sector-specific capability
development
Basic capability
development
Proactive
mentoring
38
Capability
Development
A complete Capability Development system would tend to have services
capable of moving firms some considerable distance up the capability
staircase. At the top of the staircase, research-performing firms will be well
integrated into global sources of science and technology. There is not a
necessity for local or regional actors to meet their needs - though it may
nonetheless be helpful, for example in influencing company decisions
about where to locate R&D.
A good innovation support infrastructure would have the following
services available:
1
Proactive mentoring One of the most difficult problems to be
overcome in helping SMEs develop capabilities is caused by the
‘learning paradox’. That is, that until you have learnt something
you cannot properly specify what you need to learn. Someone in
the infrastructure is needed who has a brief to guide firms especially those with limited technological capabilities - in
identifying their needs and finding ways to satisfy them. In Ireland,
this role is played by Forbairt regional staff, who visit companies on
their ‘patch’ periodically to identify needs and suggest ways in which
the state infrastructure would help. The former Engineering
Industry Training Board’s regional training advisors used to play
this kind of role in the UK. In agriculture the world over it tends to
be performed by staff in agricultural extension services.
2
Basic, general-purpose capability development services To raise
SMEs’ competences, not only in technology but also in the basics of
business, there need to be sources of practical help and training close
at hand. Issues such as Quality, simple manufacturing strategy and
use of IT are generic, yet these are areas where many SMEs need
help. Some of these services are useful to firms at the ‘Low-tech
SME’ stage in development; all have relevance to the ‘Minimum
Capability’ stage, and provide an important basis for moving firms
up the capabilities staircase
3
Sector- or technology-specific capability development services may
not be more sophisticated than those considered under 2 above.
However, for reasons of scale, they are certainly more difficult to
deliver across the generality of the economy unless target firms are
present in ‘clusters’ - especially where the sectors they address are
relatively narrow
4
Technological development services such as contract R&D can be
bought by almost any firm with money to spend. However,
adequately specifying and making good use of them requires a fair
degree of internal capability. By the time a firms climbs to this point
on the staircase, the questions are no longer about creating a level of
internal technological capability but about making best use of it
39
5
R&D services include collective R&D, research information and
services to link companies with capabilities in universities and
research institutes. These presuppose quite high levels of
technological capability on the part of users. Most SMEs are not in a
position even to have a conversation with university or research
institute researchers, so users of this type of service are quite special
Ireland is an interesting case where the bulk of the support system is
managed by a single state organisation. This provides a unique
opportunity to manage the integration of programmes into the overall
system (Exhibit 18). There are important caveats in such a system,
however
•
•
•
The quality, capabilities and commitment of personnel in the
different parts of the system need to be high in order to generate
useful help
The requirements of staff in the ‘diagnostic interface’ or 'proactive
mentoring’ role are particularly high
There need to be multiple entry-points for companies into the
system
Exhibit 18 Forbairt (Ireland) Interface to SMEs
Forbairt Service Portfolio
Start up and expansion
Management capability
Companies
Client
Executive
Marketing/networking
Technology
Operations
• Diagnostics and
development trajectories
• ‘Owns’ the client firm
Stand alone grants/equity
Based on studying programme delivery in a number of countries and on
international practices in support systems, we are skeptical of the idea that
a single person can adequately mentor companies across the full range of
business and technology support needs. Systematically, technologyfocused mentors under-deliver on business support while businessfocused mentors underestimate the importance of technology.
40
Two examples where attempts have been made to build single customer
interface to support systems underscore this need for a twin approach. A
negative example is the Swedish regional development fund. This
historically delivered a range of support services focused on finance and
improving business capabilities. Efforts over the past decade have failed
adequately to embed a technology support function. Technology audits
and referral services have foundered on the organisation’s inability to
internalise the technology function. Swedish policy is now moving
towards a ‘first-stop shop’ cross-referral system, where key support
deliverers are trained and networked. The idea of a single customer
interface has been discussed, to some degree tried, and appears impractical.
A potentially more positive example is the UK experience of Business
Links. These (originally called ‘one-stop-shops’) have been set up to
interface between small firms and the whole support system - both
business and technology. They have some limited resources for diagnosis,
but their main mission is to refer companies to the right points within the
larger support system.
They use a ‘twin interface’ model (see Exhibit 19). All Business Links have
Personal Business Advisors and separate Innovation and Technology
Counsellors. These people are responsible for monitoring and proactively
mentoring local small companies as well as for responding to externallygenerated requests for information and help. They may work with clients
separately or together, depending on client needs. Some Business Links
additionally have Design and/or Export Counsellors, providing more
direct links into these specific forms of expertise.
Exhibit 19 UK Business Links Structure
SuperNet
• 75 members
• Mostly RTOs,
HEIs
Business Links
Service
delivery
‘footprint’
• Personal Business Advisors
• Innovation and Technology
Counsellors
• Design Counsellors
• Export Counsellors
NearNet
• Service
Providers
• Mostly local
Business
Links Net
• Emerging
• Methods,
practices
41
We expect that successful links and referral systems between SMEs and
support systems will be built using this type of ‘twin interface’ model,
because of the skill requirements at those interfaces. Systematic
diagnostics - simple technology audits and business audits - are coming
into use to codify the process of bringing promising SMEs into support
systems and set them on a path which allows them to develop their
capabilities.
42