«ITEC Working Paper Series 07-26 December 2007 CBR (University of Cambridge) – ITEC (Doshisha University) Joint Papers Innovation Policy as Cargo ...»
Innovation Policy as Cargo Cult: Myth and Reality
in Knowledge-led Productivity Growth
ITEC Working Paper Series
CBR (University of Cambridge) – ITEC (Doshisha University) Joint Papers
Innovation Policy as Cargo Cult:
Myth and Reality in Knowledge-led Productivity Growth
Institute for Technology, Enterprise and Competitiveness, Doshisha University
Working Paper 07-26
University of Cambridge
Centre for Business Research Judge Business School Building Trumpington Street Cambridge CB2 1AG Email: email@example.com and Visiting Fellow Institute for Technology, Enterprise and Competitiveness, Doshisha University Kyoto, Japan
This paper questions the current emphases in innovation policy on a particular interpretation of US performance which emphasises R&D intensive high technology producing sectors, spin-offs from the science base and private sector venture capital.
Whilst recognizing the important role they may play it is argued that it has been greatly exaggerated to the neglect of other key factors. One is the importance of the diffusion and use of ICT as a general purpose technology beyond the ICT and other R&D intensive high-tech producing sectors. A second is the dominant role which performance transformation in existing firms plays in driving industry level productivity compared with the direct role of new entrants. A third is the diversified role played by universities in knowledge exchange which extends beyond a narrow focus on spin offs and licensing to encompass the creation of human capital and a wide range of formal and informal business interactions. Finally there is the major role that public R&D procurement policy has played in the US in the effective provision of public rather than private sector venture capital. The paper provides a broad overview of evidence on each of these factors and considers some broad implications for innovation policy which might be drawn on the basis of that review. In particular it concludes by arguing that the crafting of innovation policy in the context of any specific national innovation system requires a careful consideration of the structural features of that context and the particular opportunities and challenges facing policy practitioners in it. An imperfect interpretation of the experience of one country’s system is unlikely to be an appropriate guide to innovation system failure or success elsewhere.
JEL codes: O31, O33, O38 Keywords: Innovation Policy, University-Industry Links, Productivity Growth
This paper has benefited from a number of sources of support. The author would like to thank Hugh Whittaker and Yoshifumi Nakata of ITEC, Doshisha University, Kyoto, Japan, for providing an institutional base to work in this area in 2005 and 2007 as a Visiting Professor and then Overseas Visiting Fellow at ITEC. The author is also grateful to Richard Lester, Andy Cosh and Michael Scott Morton for many stimulating discussions in this area, to Anna Bullock for help in data preparation, to the Cambridge MIT Institute for financial support for the survey research into US and UK innovation on which this paper draws, and to the EPSRC for financial support under grant ITEC Working Paper 07-26 EP/EO23614/1 IKC in Advanced Manufacturing Technologies for Photonics and Electronics – Exploiting Molecular and Macromolecular Materials, a ‘public space’ experiment in fostering commercialisation activities. I am also grateful for the congenial working environment provided by Mark Dodgson at the University of Queensland Business School and for the opportunity to present some of the ideas contained in this working paper at a series of innovation summit events he organised in Australia in the autumn of 2006 when the author was a visiting professor at the University of Queensland Business School. A version of this paper will appear in Bessant, J. and Venables, T. (eds) (2008) Creating Wealth from Knowledge. Meeting the innovation challenge. Edward Elgar: Cheltenham.
In the immediate post-World War II years a series of millenarian movements known as cargo cults 2 swept through Melanesia. They emerged in the aftermath of intensive American contact in the course of the Second World War. These contacts led to a substantial increase in the material goods available to Melanesian islanders, but the end of the War meant that such material goods became less available as military withdrawal occurred. In these circumstances cargo cults emerged in which prophets would promise the return of cargoes of material goods by their ancestors (often expected to take the form of the Americans) with cargo typically shipped in the airplanes which had been such a common feature of the war experience. The means by which the return of the cargo was to be encouraged varied between different cults in different islands, but frequently involved the ritual preparation and construction of a variety of structures such as airfields, storage facilities, landing strips and associated paraphernalia. Cult members were encouraged to abandon previous cultural practices and often mimicked the behavioural characteristics of Americans (Worsley (1957), Jarvie (1964)). The emergence of these cults did not lead to the return of material cargo.
There is in my view a danger today that the evolution of innovation policy structures based on copying perceived cultural characteristics and structures of the US innovation system will also fail to deliver the goods. In the case of innovation policy the cargo is improved economic welfare through improved productivity growth based on enhanced innovation performance. The key “ritual” structures are increased R&D expenditures; an emphasis upon the commercialisation of science through university based spinouts and licensing routes in high technology producing sectors; the promotion of entrepreneurship and new business entry and a supposed US entrepreneurial culture based on the subsidisation of risk taking in venture capital investment and of the development of the SME sector more generally.
These perceived key elements feature centrally in policy debates. For example in March 2000 the EU adopted the “Lisbon” strategy to make, within the next decade, the EU the most dynamic and competitive knowledge-based economy in the world. The strategy was explicitly positioned as a response to the observed superior performance of ITEC Working Paper 07-26 1 the US economy which had in the previous decade substantially outperformed the European economies. It also explicitly accepted the view that this superior US economic performance was based on the emergence of high technology sectors such as ICT and biotechnology as key totems of the new knowledge-based economy of the US (European Commission (2004)). Despite the subsequent bursting of the dot.com bubble and an increased awareness of the emerging threat to Europe from India and China rather than the US these key elements of the innovation and technology strategy connected with Lisbon continue to be emphasized. Thus, in 2004, it was asserted that “There is overwhelming evidence of the vital importance of boosting R&D as a prerequisite for Europe to become more competitive. To fail to act on that evidence would be a fundamental strategic error …”
Similarly, it was asserted that entrepreneurship is required to take advantage of technological developments “Increasingly, new firms and SMEs are the major sources of growth and new jobs.
Entrepreneurship is thus a vocation of fundamental importance, but Europe is not ‘entrepreneur-minded’ enough.”
Both of these arguments were followed by calls for greater tax subsidization of high technology investment, R&D expenditures and enhanced policies aimed at boosting entrepreneurship and new entry and reducing risk aversion and the “stigma of failure” (European Commission (2004)).
In relation to enhancing the role of universities the policy emphasis on spin-offs and licensing “US style” is often noted;
“In recent years, spurred by the experience of the US in particular, policy makers, enterprises, investors and academics throughout the industrialized world have paid increasing attention to the role of universities as drivers of innovation. Many universities have established formal offices and processes for identifying promising discoveries made within their walls and turning them into revenue streams through licensing or spin-outs. 3”
In this paper I wish to question these emphases on R&D intensive high technology spin-offs from the science base and entrepreneurial science. In doing so it is not my intention to argue that R&D or new entry or the growth of venture capital or university spin-offs do not matter4. My contention is rather that they have been greatly exaggerated to the neglect of other key factors when one considers the innovation system as a whole.
One of these factors is the importance of the diffusion and use of ICT as a general purpose technology beyond the ICT and other R&D intensive high-tech producing sectors. This has enabled “unexpected” user sectors with negligible conventional R&D spend such as retailing to dominate movements in US aggregate productivity growth. A second factor is the dominant role which performance transformation in existing firms plays in driving industry level productivity compared with the direct role of new entrants. A third is the diversified role played by universities in knowledge exchange which extends beyond a narrow focus on spin offs and licensing to encompass the creation of human capital and a wide range of formal and informal business interactions.
A further factor related to this is the predominant role of customer supplier interactions in open innovation systems (Chesbrough (2003)) rather than direct university business interactions. Finally there is the major role that public procurement policy has played in the US in the effective provision of public rather than private sector venture capital and the high value placed by US firms on public sector sources of knowledge for innovation.
The paper attempts in the space available to provide a broad overview of evidence on each of these factors and to consider some broad implications for innovation policy which might be drawn on the basis of that review. In particular it concludes by arguing that the crafting of innovation policy in the context of any specific national innovation system requires a careful consideration of the structural features of that context and the particular opportunities and challenges facing policy practitioners in it. An imperfect interpretation of the experience of one country’s system is unlikely to be an appropriate guide to innovation system failure or success elsewhere.
ITEC Working Paper 07-26 3 Interpreting US Economic performance Since so much policy is linked to references to US economic performance it is useful to begin with a brief overview of it in the recent past. Table 1 shows that the most dramatic feature of United States’ performance since World War II is that its recent improvement is heavily concentrated at the end of the last century and at the beginning of this century, when it returned to its long run trend performance after two decades of relatively low growth performance. The dramatic improvements in productivity growth after 1995 are now, however, due to the direct performance of R&D intensive high technology industries.
This can be seen if we decompose the aggregate performance into its components.
An industry’s contribution to the aggregate depends on its own change in productivity growth, and on its size, because the economy is a weighted average of the different sectors.5 Table 1. US PRODUCTIVITY GROWTH 1947-2003
Source McKinsey Global Institute, (2001), Farrell, Bailey and Remes (2005) Decomposing productivity growth in the first period from 1995 to 2000 reveals that six of fifty-nine sectors accounted for the whole of the acceleration in productivity growth. The top three key sectors in the United States economy on this basis were wholesaling; retailing; and security and commodity broking. Their joint contribution was twice as great as the next three, which were electronic and electric equipment (semiconductors), industrial machinery and equipment (computers) and telecoms (McKinsey Global Institute (2001)).
None of the top three are technology intensive sectors in any conventional sense. In the second period, the most recent years for which decomposition data are available, seven sectors accounted for 85% of all the productivity growth. These were retailing;
finance and insurance; computer and electronic products; wholesaling; administrative ITEC Working Paper 07-26 and support services; real estate; and miscellaneous professional and scientific services.
None of these, with the exception of computers and electronics, are in any sense conventionally R&D intensive (Farrell, Bailey and Remes (2005)). It’s a Wal-Mart not a Microsoft led turnaround. The traditionally identified R&D intensive sectors have not carried most weight.