Archived — Occasional Papers
Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats on the "Contact Us" page.
No. 9: Science and Technology: Perspectives for Public Policy
by Donald G. McFetridge, Professor, Carleton University, under contract to Industry Canada as part of the Science and Technology Review, July 1995
The conventional justification for a government's role in innovation activities (or in any other economic activity) is the market failure rationale. This approach demonstrates that under specific conditions, market institutions do not allocate resources as efficiently as is theoretically possible. Failures are often identified in markets that are characterized by at least one of the following: externalities, public goods, indivisibilities, imperfect information, incomplete markets and imperfect competition.
This paper thoroughly discusses the theoretical explanations of why markets and governments may allocate insufficient resources to various forms of innovative activity and knowledge creation. This paper also argues that government support of S&T can be effective in correcting market failures. However, intervention may not improve on market outcomes if the programs or support do not follow the micro-economic principles of efficient policy design.
It is generally agreed that knowledge has several properties that are likely to lead to its under-provision by private decision makers. One of these properties is that knowledge can be only imperfectly appropriable. This means that individuals or firms creating knowledge might be unable to prevent others from making use of it, and making a profit from it. Knowledge has two other characteristics that lead to a social interest in its diffusion. First, if it is to have any value, knowledge from one source must often be used in conjunctioin with knowledge from other sources. Second, knowledge is a public good, which means that its use in one purpose does not preclude its use in another setting.
The paper looks at externalities or spillovers, defined as the excess amount a user would have been willing to pay over and above what he had to pay for a good. The sources of spillovers which are reviewed include: imitation, demonstration effects, experience and incubation.
Evidence on spillover benefits from R&D spending and innovation is also examined. This discussion begins by surveying the case study evidence which attempts to estimate the difference between the social and private rates of return to innovative activity. In looking at government sponsored R&D in particular, the paper finds that government sponsored, precompetitive R&D yields good rates of return while the record of government-supported proprietary R&D is much less impressive.
The paper then surveys the evidence that points to a decline in R&D spillover rates in recent years. It also discusses local agglomeration effects and concludes that agglomeration economies exist and that their most likely sources are labour market and infrastructure (especially transportation) specialization. It finds, however, that there is little to indicate that firms located at a distance from their rivals are technologically disadvantaged.
International spillovers are then examined. The existence of international spillovers complicates prescriptions for efficient resource allocation. Policy should favour activities with large domestic spillover benefits. The types of innovations that generate the largest domestic spillover benefits might be those that add value to unique domestic resources.
Various forms of legal mechanisms, particularly patents, are evaluated in their function to protect intellectual property rights and increase the appropriability of knowledge. The evidence suggests that patent protection now is, on balance, stronger than is required. However, there are in Canada some concerns about the terms of access that justify a somewhat stronger patent protection system.
The paper then goes on to discuss the policy instruments aimed at offsetting the inappropriability of knowledge, in particular, tax incentives, subsidies, procurement, and concessionary financing. It finds that the efficacy of policy instruments indicates that tax incentives and concessionary financing may be more effective than direct subsidies, although the empirical evidence is limited.
In concluding the paper, models of innovation are examined in conjunction with public policy. The linear model of innovation characterizes the innovative process as a flow beginning with basic research and proceeding to applied research and then to development and commercialization. Scientific discovery is regarded as exogenous. The public policy response in the face of such a model is a progression from substantial direct government support at the early stages of the model to limited direct support at the later stages. The feedback model, on the other hand, emphasizes the cumulative and interdependent nature of the innovative process. Here, the simple rule that basic research is the sole responsibility of government, no longer applies. The feedback model of innovation appears to be a more realistic representation of the true process.
The implications for government policy in the face of the feedback model are complex. Overarching government plans are not necessary. In many cases, governments may be merely participants in an ongoing evolution. The feedback model sees the business sector as an important potential source of support for university research. The role of government is to accord universities the flexibility to participate in theses arrangements. Also, research conducted in government labs may have commercial applications in this model. This suggests the need for more cooperative research involving contracting-out, contracting-in, alliances or consortia. If the feedback model applies, the internal organization and incentive structures of government labs may have to be altered if they are to interact effectively with other components of the innovation system.
- Date modified: