Three Policies to Improve Productivity Growth in Canada

30

in Canadian provinces, ranging in 2004 from a low of 40 per cent in Alberta to a high of
202 per cent in Quebec. Very generous provincial tax credits to R&D explain these large
subsidies.

To be sure, R&D is important for innovation, particular in certain high tech
sectors such as communications equipment and aeronautics, and governments must
contribute to support R&D activities. But in 2002 only 12,272 firms in Canada actually
reported performing R&D, with 100 firms accounting for 56 per cent of total R&D
(Statistics Canada, 2006:17). Thus less than 1 per cent of firms perform R&D in Canada.
If the term “innovation” were reserved for firms that undertook R&D, very few Canadian
firms would be considered innovative.

However, an innovative firm is more appropriately defined as one which
introduces new production processes and products. By this criterion, 81 per cent of
manufacturing firms in Canada can be considered innovative according to the 1999
Survey of Innovation (Arundel and Mohnen, 2003:58). Thus from the point of view of
fostering productivity growth, R&D is not relevant for the vast majority of Canadian
firms. What is relevant is the adoption of best practice technologies. Yet this path to
productivity improvement receives much less attention than R&D.¹⁹ A distinction
between innovation in the sense of knowledge creation (invention) and innovation in the
sense of the adoption of existing technologies (diffusion) is needed.

Lipsey, Carlaw and Bekar (2005:518) point out that economists have not
appreciated the tradeoff between invention and diffusion as they assume diffusion occurs
instantaneously or costlessly. Innovators introduce new stand-alone technologies which
diffuse through the economy in unchanged form. But invention and diffusion are separate
activities. New technologies must be adapted for particular uses and require supporting
technologies and facilitating structures.

Lipsey, Carlaw and Bekar (2005:520) point to sunk costs, defined as costs that
have already been incurred and which cannot be recovered to any significant degree, as a
barrier to diffusion and argue that this situation may justify public policy intervention.
According to these authors:

“Sunk costs are important for the development of new products and processes;
they are equally important for acquiring codifiable knowledge about new
knowledge, as well as tacit knowledge about how to operate given technologies.
One major policy implication is that governments can effectively disseminate
technological knowledge by operating on a scale that makes the sunk costs
bearable, or even trivial, where they would otherwise be prohibitively high for
small firms. This is the objective of the Canadian Industrial Research Assistance
Program (IRAP), which seeks, among other things, to help firms to identify
existing technologies that are of potential value to them and to assist them in
adapting these technologies to their specific needs.”

¹⁹ For an overview of the issue of the diffusion and adoption of advanced technologies, see CSLS (2005).

<<   27   28   29   30   31   32   33   >>

More intriguing information