Research and technology outsourcing and innovation systems: An exploratory analysis

It is also revealing trying to define a system around the formal and informal innovation outsourcing market, which includes both service and manufacturing firms and organizations. Delimiting such a system obviously cuts across many technological systems and national systems of innovation that have already been highlighted in earlier studies. The key elements in such a system would be both CRTOs and their "manufacturing" (but also service) customers (forming in turn a wider "organizational population" of different actors and agencies; Reddy and Roa 1990; see also Phillips 1960). The boundary of the system could be defined in straightforward "market" terms, but as has been indicated above, there are strong informal and non-traded elements in the CRT "market" which involve high levels of reciprocity and trust and which parallels much of what Dore (1986) would term as "relational" networks. In terms of this wider system, many of the links described would not be based on purely monetary transactions and formal contracts, but involve informal know-how trading and patterns of exchange (see, for example, Von Hippel 1987; Kreiner and Schultz 1993). A clear nesting of the wider system is also apparent hierarchically with sub-systems evident in relation to specific technological systems where the research and technology outsourcing market may operate, such as energy or environmental "services", but also narrowing down to specific national or regional systems of research and technology outsourcing. This has outlined the basic framework of a system, but it has crucially left out the institutional, as well as more general cultural, social and economic, elements in such an innovation system. Without these elements the framework that has been described is more akin to a simplified innovation network (Hakansson 1987) or linkage structure (Andersen and Lundvall 1997). Institutions are central to the notion of any kind of system of innovation (Edquist 1997). They provide the dynamic context (the routines and "guide posts"; Lundvall 1992) through which systems evolve and change.

Furthermore, although the innovation systems literature has so frequently stressed the dynamic qualities of such systems and the approach, it then usually goes on to treat such systems in a static fashion. The wider innovation outsourcing system, or one of its sub-systems, in period nl is going to be different in period n2. Key elements or actors will have disappeared between the two periods and have been replaced, leading to new and different linkage patterns and changes in the overall shape and configuration of the system (with some of these changes being measurable and therefore allowing systems to be compared).

Indeed much of systems theory development has been driven by its ability to describe and map different contexts in a dynamic fashion. What might be considered to be the appropriate "system" in one time period or for one set of analyses is unlikely to be the same for a different period. Appropriate systems modelling should also allow the ability to move between different hierarchies or "nests" so that particular phenomena or processes can be the focus of analysis or conversely the analysis can be widened out to allow a broader view. In transport systems modelling the focus can be on the whole transport system covering all transport modes (car, rail, air and sea) on a national perspective, a single transport mode such as car traffic within for example a single urban system, or indeed the car itself can be treated as a system (or indeed the electrical "system" of the car can be viewed as a system). Much is lost from systems analysis by sticking rigidly to viewing and conceptualizing one particular system and innovation systems, it is argued here, are no different from other systems. One of the chief merits of systems theory is its hierarchical nature and ability to flexibly move between these hierarchies over time, when conditions allow or when analysis requires this. Any view taken of what a system is, therefore, depends on what phenomenon (or what attribute of a phenomenon) we wish to study. This determines what the "element" (or basic unit) within the system will be.