Probably most innovation studies fall into other categories such how they work, why they work or even do they work which aren't really related to their architecture, but there is still a wealth of knowledge of the structure of innnovation systems.
Problematically, as Machlup noted 50 years ago the word 'structure' is often just a jargon word that obscures meanings as much they enlighten. It appears to me that we can create a taxa of at least three classes of 'structure'. The first world be shift/share analysis which look at the changing composition of industry or innovation patterns. A second class would be 'specialisation' which we would typcally see in analysis of national industry or export patterns. The third and final group of studies would be those that are focussed upon the relationships between actors in the innovation system.
Shift Share.
I am not aware of any particular pieces of work that have linked shift share analysis, which is typically associated with the changes in the sectoral breakdown of GDP over time, and the innovativeness of economies.
Specialisation
Often specialisation has been used as a term related to relative geographic concentrations.
Great work on this topic has been done on innovation and specialisation by people like: Keld Laursen.
(2000), Trade Specialisation, Technology and Economic Growth: Theory and Evidence from Advanced Countries, Cheltenham, UK and Lyme, US: Edward Elgar, ISBN 1 84064 385 4.
That book deals with export specialisations, many other have considered technological specialisation.
Recent work has begun to create a clearer distinction between specialisations and concentrations of activitiy. See for example: Cutrini(2010) 'Specialization and Concentration from a Twofold Geographical Perspective: Evidence from Europe', Regional Studies, 44: 3, 315 — 336.
Relationships
However, most of the innovation studies interested in structure have focussed most on relationships. Different kinds of relationships have been described in the literature on Innovation.
Some examples include:
Redrawn from. Geroski, P. (1994) Market Structure, Corporate Performance and Innovative Activity. New York: Claredon Press – Oxford
.
In this diagram we can see a complex system of innovation interdependencies. A couple of primary ‘sectors’ – mechanical engineering and chemicals are key developers of innovations that flow onto a number of other sectors. Some sectors such as electronics and instruments are intermediate players, both benefiting from innovations generated elsewhere but also providing innovations to others. Finally, there are a range of largely ‘recipient’ complex integrating sectors such as auto, aerospace, paper, food and construction etc which are typically end points for a wide range of technological inputs.
Another presentation.
Redrawn from:
Silvani, A., F. Cancemi, C. DeBresson and X. Hu (1996) ‘Innovative self-sufficency or regional interdependence: Lombardy and the Rest of Italy’ in DeBresson, C. (ed) (1996) Economic Interdependence and Innovative Activity. Cheltenham, Edward Elgar.
Other surveys of the producers and users of innovations have revealed similar patterns. Silvani et al. (1996), for example, have shown quite a similar pattern exists for the innovation interdependencies within just a single Italian region (Lombardy). Figure 2 reveals not only the innovation flows but also key flows of goods (combining an innovation producer-user matrix with standard input-output matrix data. Again, auto, paper, and textiles are recipient sectors, while machinery and chemicals are important innovation generating sectors. In this region the wood and rubber sector has important innovation interdependencies with the chemicals and machinery sectors.
Finally, the analysis of Cesaratto et al. (1996), based on similar data to Silvani et al., is quite interesting. These authors categorise industries into a quite a different format than has typically been developed, although principally reinforcing the emergent understanding of the technological division of labour and the interdependencies. Their taxonomy groups industries into design based capital goods, investment based intermediate goods innovators, complex innovators, marketing oriented innovators, cost oriented consumer goods innovators, construction, services and consumption.
In my own work I kept the interactions down to a single industry but looked at the relationships across borders.
So in the diagram above I show that I-O interactions within Europe for the transport industries favour Germany (the arrows are backwards flows of finances) goods flows in opposite direction.
Thus innovation studies have contributed significantly to our understanding of the relationships within the economy.
