Climate Change 2001:
Working Group III: Mitigation
Other reports in this collection

5.3.1 Technological Innovation

Many governments and firms have focused their strategies for GHG mitigation on encouraging technological innovation – various processes of research, experimentation, learning, and technology development. Innovation may lead to improvements in technology performance, reductions in GHG emissions per unit of service provided, or reductions in cost for low-GHG technology, all of which can contribute to GHG mitigation. Innovation can help to raise the technological, socio-political, economic, and market potentials for adoption of low-GHG technology, and for GHG mitigation. Identifying the barriers to, and opportunities for, technological innovation depends on understanding the innovation process. Since the IPCC SAR, there has been a rapid growth of interest in the theory of innovation, and in the development and application of models to evaluate climate mitigation policies that take account of endogenous technological change (Azar, 1996; Goulder and Mathai, 2000). The Innovation Process

Until the 1980s, policy analysts generally viewed innovation as a linear process from R&D through to demonstration and deployment. Policies were focused on “science push” and “demand pull” for new technologies (OECD, 1992). Over the last twenty years there has been a growing recognition of the interconnectedness of the many processes involved in technological change, and the possibility of finding new insights or knowledge anywhere from the research lab to the customer service department.

Technological change can take many different forms including: (1) incremental improvements in existing technology; (2) radical innovation to introduce completely new technology; (3) changes in a system of linked technologies, and (4) changes in the “techno-economic paradigm” involving widespread re-organization of production and consumption patterns (Freeman and Perez, 1988). These four types of innovation have different dynamics. Thus, the first type is likely to occur continually through the accumulation of experience, selection of successful techniques and adaptation to a changing economic, legislative and socio-cultural context. The second and third types of technological change involve more positive creativity, being linked to new information in the form of a discovery, idea, or invention; or to a creative application of an existing invention. The fourth type, again, involves creativity but, because it involves a radical change in culture and markets, may also depend on these being “ripe” for change – on a general perception of a major challenge requiring a radical response.

Technology diffusion, the spread of existing technology through the population of potential users, can be distinguished from innovation – the first commercial application of a new technology. At a local level, however, there may be little difference between the two. Wallace (1995) notes the importance of an active and creative absorption process in the uptake of the new technology.

Technological change is a complex process. It occurs through a variety of interdependent mechanisms (Nelson et al., 1967; Rosenberg, 1982; Dosi, 1988; OECD, 1992; Rosenberg, 1994; Lane and Maxfield, 1995), which can include:

Because of the complexity of the technological innovation process, there are many different ways of looking at it. A variety of theories or models may be helpful, depending partly on specific circumstances.

From the perspective of neoclassical economics, innovation can be seen as the result of a process of investment in “knowledge capital”, in the form of R&D to develop both formal and tacit knowledge (Griliches, 1979). The former includes the scientific literature and patents; the latter includes the skills and experience developed by those involved in developing new technology and can also be viewed as “human capital”. Increasing capital, again, tends to feed into higher levels of economic output and improved efficiency. Sometimes this may contribute to GHG mitigation, but more often the improvement is in labour productivity, leading to increases in GHG emissions. In so-called “new growth” theory economic models (e.g., Grossman and Helpman, 1991, 1993), new knowledge may be assumed to result directly from R&D spending which, in turn, can be modelled as a result of the expected returns from the investment. In this framework, firms and research institutes are treated as rational investors in R&D. The size of their investment will depend on the opportunity cost of capital and the expected return from R&D. While new growth theory has generated useful insights into the sources of national differences in competitiveness at an aggregate or sectoral level, it is less useful for describing technology innovation for GHG mitigation.

In addition to R&D investment, knowledge capital can also be accumulated through the process of “learning by doing” (Arthur, 1994; Grubb, 2000). Empirical studies show that the cost of a generic technology such as solar photovoltaic cells tends to fall with the level of existing investment in that technology, including spending on R&D (Christiansson, 1995; Messner, 1996; Nakicenovic, 1996).

An alternative to the neoclassical investment approach to innovation is that pioneered by Nelson and Winter (1982), to view technological change from the perspective of the firm, as a stochastic process of search, imitation, experimentation, and learning (Winter et al., 2000). Recent developments in agent-based modelling adopt this type of “evolutionary” framework, helping to bring out the role of information networks, the importance of existing experience, and also some of the spatial aspects of technology development and diffusion.

Finally, several analysts have adopted models of technology competition and diffusion analogous to those used to represent species competition and diffusion in ecosystems. Regularities have been found, for example, in the market succession of technology in energy supply, transport, and the iron and steel industry (Häfele et al., 1982; Grübler and Nakicenovic, 1991; Nakicenovic, 1996). However, no approach can hope to foresee reliably the form of the next “wave” of technology in any of these sectors.

Other reports in this collection