The Unintended Consequences of Innovation Studies 1
Benoît Godin
INRS (Quebec), Canada
Innovation has a long history. In fact, men have innovated from the very beginnings of
humanity. Similarly, the word “innovation” is very old, dating back to antiquity. One
finds early thoughts on innovation among Greek philosophers and historians, then among
the Romans (Godin and Lucier, 2012). The word came into widespread use in both
religion and politics after the Reformation. Yet at the time, innovation had a meaning
totally different from that of today. In the second half of the twentieth century, innovation
came to be studied in the context of economic progress and was theorized about. This
gave rise to what is now called “innovation studies”: the economic, management and
policy study of innovation (Fagerberg and Verspagen, 2009).
This paper takes a critical look at innovation studies and its representation of innovation.
While innovation has been understood for centuries as “introducing change in the
established order”, particularly the political order, it came to mean commercialized
invention (technological innovation) in a matter of decades. This has had important
consequences on the place innovation holds in society. This paper highlights three
characteristics of innovation as a concept over the centuries, then discusses some
assumptions of innovation studies and their consequences on the study of innovation
generally and on policy.
Intellectual History of Innovation
For 2,500 years, innovation was an eminently contested idea. It was pejorative and
negative (Godin, 2012). An innovator was a deviant: one given to popery (in England), a
1
Paper prepared for a communication presented at “Policy Implications due to Unintended Consequences
of Innovation”, Special Track at EU-SPRI, Madrid, 10-12 April 2013.
2
revolutionary, a socialist. Gradually over the nineteenth century and twentieth century,
the study of innovation gave rise to a totally different representation. Innovation became
an object of praise and fashion. The representation gave rise to almost exclusively
positive assumptions about innovation. From a vice innovation became a virtue. To most
people, including researchers, innovation is always good. “Researchers have implicitly
assumed”, claimed sociologist Everett M. Rogers in the first edition of his classic book,
“that to adopt innovations is desirable behaviour [rational] and to reject innovations is
less desirable [irrational]” (Rogers, 1962: 142).
Yet, for centuries innovation was essentially a political concept, regulated by Kings and
laws. In 1548, Edward VI in England imposed punishments and imprisonment on
“innovators” (Godin, 2010). In contrast, in the twentieth century innovation started being
looked at from a policy-oriented framework (Godin, 2012). Innovation is now considered
instrumental to national progress. Government should play a role and stimulate
innovation, many claim. Such a representation made innovation into a panacea for
solving every socioeconomic problem.
For most of its history, the concept of innovation has had a fuzzy meaning. Innovation
has meant different things to different people. From the eighteenth century onward, it
served accusatory purposes: denying an enemy the right to change things. Then in the
twentieth century, the concept acquired a dominant and spontaneous meaning. Innovation
became “restricted” to technological innovation – a representation that many actually try
to broaden with still-fragmented conceptual schemes (e.g.: social innovation).
Innovation Studies
What role does innovation studies play in these conceptual developments? Three
assumptions deserve mention, for they have been influential both on our current
representation of innovation and on policy.
3
Normativeness. Innovation studies’ representation of innovation is essentially normative
– and un-reflexive. It evolves around a key ideology of modernity, namely economic
issues and the “positive” contribution of innovation to economic growth. As a
consequence, there is little if any questioning of what innovation really is, but a
normative perspective is offered from the start. 2
Performativity. In contrast to scholarly disciplines like history and sociology, innovation
studies is a policy-oriented field. It studies innovation as a phenomenon but also works to
persuade policy-makers (and others) of the desirability and inevitability of innovation,
said to be the latest or more recent stage of development (e.g.: the knowledge-based
economy). Together with national governments and international organizations like the
OECD and the European Union, social researchers develop narratives, conceptual
frameworks and “models” aimed at supporting innovation as THE solution to
socioeconomic problems.
Technology/Market-Centeredness. Innovation studies’ representation of innovation is
essentially technological – and market-oriented: innovation is the commercialization of
technological inventions. Invention is not innovation unless it is commercialized.
Imitation (adoption of existing technological innovation) is not innovation because it is
not original (first introduction to the market). The study of types of innovation other than
technological (cultural, social and political) is relegated to the residual, if discussed at all,
and “societal” studies are excluded from the field as it is defined (constructed) in recent
bibliometric studies.
2
Very few researchers stop to examine what innovation means, starting rather with their preferred
definition. In the 1960-70s, this led people from many different horizons to criticize the concept and to
suggest abandoning it. “Every change becomes an innovation simply because it has not been done before ...
Such a definition sterilizes the term innovation” (Becker and Whisler, 1967: 463); innovation “has come to
mean all things to all men” (Ames, 1961: 371); “we shall do better without the word innovation” (Machlup,
1962: 179); “the use of the term innovation is counterproductive” (Roberts and Romine, 1974: 4); “One of
the vogue words these days is innovation. For some people it is even more – it is a value word that implies
something good and positive. As with most popular words, it is misused and has different meanings for
different people” (Holt, 1971: 235).
4
Alternative Visions
What would a different representation of innovation look like, and what policy would
ensue? Let’s conduct a thought experiment, namely imagining alternative views of
innovation and discussing the policy issues that logically arise from a different
perspective on innovation:
- What policies would we have if innovation studies had been critical and reflexive
instead of normative?
- What policies would we have if innovation studies had started with the study of
social problems rather than presenting innovation as the a priori solution?
- What policies would we have if innovation studies had included in their analyses
a broader range of innovations than only the technological?
A Different Regime of Incentives and Appropriability
Currently, the theory of innovation makes a contrast between innovation and imitation.
Imitation is not innovation but rather an undesirable way of deriving profit from an
innovation developed by another firm. Such is the essence of David Teece’s much cited
article: “Innovating firms often fail to obtain significant economic returns from an
innovation while customers, imitators and other industry participants benefit” (Teece,
1986: 285). As a consequence, theories of innovation are concerned with ways of
preventing imitation or “keeping imitators/followers at bay” (Teece, 1986: 290), that is,
how can firms get the full benefit of their innovation: “how the “innovator is to avoid
handing over the lion’s share of the profits to imitators” (Teece, 1986: 292). Teece
discusses the “strategies the firm must follow to maximize its share of industry profits
relative to imitators and other competitors” (Teece, 1986: 300-301).
Such a theory of innovation neglects a few important things. First, imitation is part of the
competition game. It gives other firms a further incentive to innovate. Furthermore,
economic growth occurs because of diffusion (imitation) as much as innovation. As
5
Edwin Mansfield put it long ago: “an innovation will not have its full economic impact
until the imitation process is well under way” (Mansfield, 1961: 762). Second, there may
be legitimate reasons for (some) firms not to innovate. A firm may not (always) need to
innovate; imitation may suffice. In this sense, what is missing in the literature is studies
of imitators and non-innovators. Studies narrowly focus on innovators and best practices.
The focus on an a priori solution (innovation) to growth rather than on the diverse issues
firms face explains this “bias”.
One has to turn to a different literature for the study of imitation (diffusion). From
Gabriel Tarde in the late nineteenth century to Everett Rogers, imitation has been the
sociologists’ definition of innovation. The innovator is someone who adopts a new idea,
thing or behaviour – regardless of the fact that someone else uses it already. The
innovator may simply imitate, but at the same time he innovates with regard to his own
past behaviour. Unfortunately, such a definition is perhaps too subjective for the
economist.
A New Type of National System of Innovation
Despite its focus on “cultural” factors, the very popular idea of a National System of
Innovation (NSI) remains quite restricted. It evolves around or centers on the suppliers of
innovations, namely the firms, around which the other sectors (government and
university) evolve or to which they (should) contribute. Technological innovation is the
ultimate and uncontested aim of the system. Take for example the OECD’s recent
strategy of innovation on the issue of development – the strategy claims its debt to the
idea of NSI (OECD, 2010a). To the OECD, “the last half-century has seen different
approaches to development which have achieved varying degrees of success” (OECD,
2010a: 14). In their place, innovation should now be considered a strategy for
development: “most current social, economic and environmental challenges require
creative solutions based on innovation and technological advance” (OECD, 2010a: 30;
32). But is this really the case and how exactly? The document, as with most of the
literature on innovation, starts with innovation as a panacea, not with problems of
6
development or the extent (and limitations) to which innovation is or is not a (THE)
solution.
What would the idea of a NSI be if the study of problems or demand (social needs) rather
than supply (innovation) was the focus of interest? Paradoxically, innovation may appear
not to be the universal solution. The majority of developing countries are, first of all, and
for better or worse, consumers of knowledge and technology produced elsewhere
(imitation). There is therefore a need to emphasize these countries’ efforts to absorb
(imitate) what comes from outside as much as their own inventive and innovative efforts.
To this end, one must shift his attention from an exclusive focus on firms as innovators.
Usually, there is little concern for “people” as innovators (doing things differently) in
theories and policies except, again, as introducers of new inventions to the market or as
buyers of new inventions. Certainly, the consideration of people as innovators in the
larger sense gets some hearing in the OECD document, like the discussion of the
informal sector. However, the issue is discussed entirely in terms of the market. As if
every solution to health, poverty and education needs a firm, a technology, a market.
How do people change their behaviour in response to new knowledge (like AIDS)? How
do organizations (schools, hospitals, etc.) contribute to people adopting new behaviours?
In the past thirty years or so, demand was transformed to mean economic demand, and
was stripped of any relationship to social need (Godin and Lane, 2013). Yet, supply
(innovation) would play a different role than it currently plays in theories and policies if
theories placed the emphasis on needs and the beneficiaries of innovation. Rather than the
dichotomy of either universities or firms being the drivers of innovation systems around
which other participants play the role of “context”, the emphasis would be on 1.
consumers, citizens and their community associations, 2. public managers and programs,
3. government departments, public organizations and policies (see Appendix 1).
7
More Refined Policies
Over the last sixty years, governments and international organizations have promoted the
idea of innovation policy as horizontal policy: a policy encompassing broad, integrative
and common issues, including those common to different government departments.
Recent innovation policy has now embraced still more horizontal issues: those relating to
globalization. Yet, innovation policies have failed with regard to the sectoral department
or various government departments. Innovation policies are essentially technological
policies, because the dominant ideology of innovation is technological. But not every
department needs a technology policy. It may need to innovate in other dimensions of its
activity, or may need to stimulate other types of innovation than the technological.
Innovation policy makes no sense. Currently, theories on which innovation policies are
based are too macro and not detailed enough to address the needs of specific sectors. This
reminds me of theories of “social change” before 1970: too general, too theoretical, with
few applications. 3
Today, such theories have more or less disappeared from the
literature. 4
Yet in its place, social innovation – the new buzzword, although an old
concept – has appeared. Theories of social innovation seem to correct many of the above
limitations. First, they concern a broader range of innovation than just technological
innovation – although this is not a novelty: many of these types of innovation have been
studied under other names for decades (e.g.: organizational innovation). Second, social
needs define what social innovation is: “new ideas that work in meeting social needs”
(Mulgan, 2007: 8). Other participants and government departments than those linked to
the economy are thus invited to join the fashion of innovation.
3
As Wilbert Moore put it long ago: “a ‘pure’ theory of social change, independent of substantive
identification of the patterns undergoing transformations [is] uninteresting” (Moore, 1960). Similarly, to
Raymond Boudon: “the sociology of social change ... suffer[s] from a misplaced conception of generality”.
To Boudon, the general and formal models, frameworks and systems of concepts of theories of social
change can be applied to no specific social process (Boudon, 1983).
4
For better or worse, many theories of social change have instead embraced the concept of innovation. For
example, modernization theories have adopted innovation as a central concept (Hill, 1974; 1975;
Himmelstrand, 1982; Zapf, 1991).
8
Yet most of the theories on social innovation are far from serious to date (Godin, 2013).
They follow (adapt) the dominant innovation paradigm (models, etc.) to their own cause,
with very few conceptual “innovations” or novelties. They remain normative, as
normative as theories of technological innovation. The theories include every change –
social change! – but have nothing to say about particular change. Social innovation is a
slogan like innovation tout court is. It allows a new class of researchers to catch the
attention of policy-makers.
Conclusion
Innovation studies, its assumptions and its representation of innovation have had
unintended (are they really unintended?) consequences on policy. Innovation policy is
entirely concerned with technological innovation, with firms and the market and with
innovation as originality (first commercialization) 5
rather than imitation. Furthermore, it
focuses on innovation as a universal, uncontested and a priori solution rather than
studying the specific problems (needs) of society. Social problems are taken for granted.
At best, they are studied as an afterthought to the study of innovation, and the problems
studied are those that arise due to innovation: social problems as a result from a lag
between innovation and society (like unemployment and skills) and the need for society
to adjust to innovation.
Given the “naturalness” of these assumptions (they have become a given, part of the
credo of innovation studies) one may identify and study them only if one takes a
comparative (historical) and critical view of the concept of innovation, as I have done
here. This is certainly a fruitful lesson to be learned from the intellectual history of
innovation, a field that is vastly understudied today.
5
On the (paradoxical) absence of statistics on “commercialized invention”, see Appendix 2.
9
Cited Works
Ames, A. (1961), Research, Invention, Development and Innovation, American
Economic Review, 51 (3): 370-81.
Becker, S. W., and T. L. Whisler (1967), The Innovative Organization: A Selective View
of Current Theory and Research, Journal of Business, 40 (4): 462-69.
Boudon, R. (1983), Individual Action and Social Change: A No-Theory of Social
Change, British Journal of Sociology, 34 (1): 1-18.
Fagerberg, J., and B. Verspagen (2009), Innovation Studies: The Emerging Structure of a
New Scientific Field, Research Policy, 38: 218-33.
Godin, B. (2010), “Meddle Not With Them That Are Given to Change”: Innovation as
Evil, Working Paper no. 6, Project on the Intellectual History of Innovation,
Montreal: INRS
Godin, B. (2012), Innovation Studies: The Invention of a Specialty, Minerva, 50 (4): 397-
421.
Godin, B. (2012), καινοτομία: An Old Word for a New World, or, The De-Contestation of
a Political and Contested Concept, in Karl-Erik Sveiby, Pemilla Gripenberg and
Beata Segercrantz (eds.), Challenging the Innovation Paradigm, London:
Routledge: 37-60.
Godin, B. (2013), Social Innovation: Utopias of Innovation from c.1830 to the Present, in
A.-M. Rieu, The Social Turn of Research and Innovation Policies, Paris: Éditions
des archives contemporaines, forthcoming.
Godin, B., and J. P. Lane (2013), Pushes and Pulls: The Hi(S)tory of the demand Pull
Model of Innovation, Science, Technology and Human Values, forthcoming.
Godin, B., and P. Lucier (2012), Innovation and Conceptual Innovation in Ancient
Greece, Project on the Intellectual History of Innovation, INRS: Montreal.
Hill, E. B. (1974), Innovative Processes in Social Change, Current Sociology, 22 (1-3):
123-26.
Hill, E. B. (1975), Innovation Processes in Social Change (Modernization II). An
Introduction, International Review of Community Development, v-xi.
Himmelstrand, U. (1982), Innovative Processes in Social Change: Theory, Method and
Social Practice, in T. B. Bottomore et al. (eds.), Sociology: The State of the Art,
Beverly Hills: Sage: 37-66.
Holt, K. (1971), Social Innovation in Organizations, International Studies of
Management & Organization, 1 (3): 235-52.
Machlup, F. (1962), The Production and Distribution of Knowledge in the United States,
Princeton: Princeton University Press.
Mansfield, E. (1961), Technical Change and the Rate of Imitation, Econometrica, 29 (4):
741-66.
Moore, W.E. (1960), A Reconsideration of theories of Social Change, American
Sociological Review, 25 (6): 810-18.
Mulgan, G. (2007), Social Innovation: What It Is, Why It Matters and How It Can Be
Accelerated, SKOLL Centre for Social Entrepreneurship, Oxford, Said School of
Business.
OECD (2005), Guidelines for Collecting and Interpreting Innovation Data. Paris: OECD.
10
OECD (2010a), Innovation and the Development Agenda, E. Kraemer-Mbula and W.
Wamae (eds.), Paris: OECD
OECD (2010b). Measuring Innovation: A New Perspectivity. Paris: OECD.
OECD (2010c). The OECD Innovation Strategy: Getting a Head Start on Tomorrow.
Paris: OECD.
Roberts, R. E., and C. A. Romine (1974), Investment in Innovation, Midwest Research
Institute, Kansas City, Report prepared for the National Science Foundation.
Rogers, E. M. (1962), The Diffusion of Innovation, New York: Free Press.
Teece, D. J. (1986), Profiting from Technological Innovation: Implications for
Integration, Collaboration, Licensing and Public Policy, Research Policy, 15: 285-
305.
Zapf, W. (1991), The Role of Innovations in Modernization Theory, International Review
of Sociology, 1 (3): 83-94.
11
Appendix 1.
What Would a Policy of Innovation Look Like
If One Starts With a Demand View?
- It would address and focus on specific and precise problems (social needs) or
areas of development, not innovation in general and broad terms (“percentage of
enterprise that introduced innovation”).
- It would concern end-users, not just producers.
- It would cover individuals, groups and organizations as well as firms.
- It would measure diverse kinds of innovation: ideas, behaviours and things (and
compare the new to the old) and their sources.
- It would ask what use, if any, is made of the innovation and by whom?
- It would identify the mechanisms through which innovations diffuse and their
presence or absence in a country: How does knowledge about X gets into country
Y? What lags occur? Why?
- What are the effects, including the negative ones? To what extent is the
innovation adapted to a country’s needs?
- Evaluate the role of government as innovator in matters of policy (not just as a
hindrance): what infrastructures, policies and programs exist in country Y to
support innovation?
12
Appendix 2.
On Innovation as Commercialized Invention
Technological innovation has for a long time been defined as the commercialization of
discoveries and inventions in the form of new or improved products or services. To the
OECD Oslo manual, “an innovation is the implementation of a new or significantly
improved product (good or service), or process, a new marketing method, or a new
organizational method (in business practices, workplace organization or external
relations)” (OECD, 2005: 46). In the case of products (technological innovation)
implementation means “introducing to the market” or commercialization (OECD 2005:
47).
Given this official definition of technological innovation, one would logically expect to
find some indicators of commercialization in the statistical series addressing innovation.
What one finds instead is everything but commercialization. Take the OECD as example
again. In the recently published Measuring Innovation: A New Perspective, there is one
and only one indicator on commercialization (or rather one indicator called
commercialization): patents. Yet, a patent does not at all represent commercialization. It
is a national certification of ownership over a claimed invention. The OECD states that
patents are included for the “likelihood of [a patent] being commercialized” (OECD
2010b: 104-5). But probability is not certainty. The OECD ignores its own basic
definition of technological innovation as the actual commercialization of technological
invention. It offers no direct indicators on commercialization in its statistical scoreboard.
Since the 1990s, the OECD has professed an intention to broaden the scope of this
representation of innovation by articulating a systemic approach (NSI). “It is necessary”,
claims the OECD, “to analyse the role of public policies, economic incentives,
organisations, market structure, foreign trade and investment, and other institutional
factors, along with the complementarities and synergies (OECD 2010c: 29).” Right.
However, in practice the organization is only adding indicators of activities peripheral to
13
commercialization. The systemic approach shifts (unconsciously?) the study (and
measurement) of technological innovation away from market commercialization. To be
sure, commercialization is not everything. But it is surprising that, given the decades-old
definition of technological innovation as commercialization, there exists not one direct
indicator of commercialization in the OECD statistical toolbox, and that little effort is
devoted to improving the situation. 6
Despite all the limitations of the early surveys of technological innovation, like the
OECD’s Gaps in Technology in the late 1960s (counting the origin and diffusion of
technological innovations from a selected list), the statisticians of the time were truly
measuring technological innovation as production and commercialization. The
measurement of innovation today should perhaps look back to this example for
inspiration and in order to further develop the statistical series.
6
Some possible indicators on commercialization are mentioned in the OECD Oslo manual as factors
(objectives and effects of innovation activities), but not as indicators of commercialization. Above all, these
factors rely on subjective reporting from managers, including yes/no answers or ordinal multiple choice
scales (OECD 2005: 107-8).