Report of the Advisory Panel on Efficiencies - Chapter 4 Promoting dynamic efficiency as public policy

This chapter examines the promotion of dynamic efficiency — innovation — as public policy. The first part of the chapter deals with the place of innovation on Canada’s public policy agenda. The second part of the chapter looks at the relationship between competition and innovation, in particular research that relates industry concentration and innovation.

Innovation and Canada’s public policy agenda

The pursuit of greater productive efficiency permeates the current economic agendas of governments across Canada. Indeed, increasing productive efficiency is the key to improving Canadians’ standard of living, since it results in more value being produced from existing economic resources.

Greater efficiency is achieved by various means. Broad policies, such as fiscal and monetary policies, expanding trade, manpower training and tax policies, encourage investment and the expansion of economic activities and thus target both employment creation and productivity. Moreover, given that capital intensity is a major factor that explains Canada’s lower productivity, policies encouraging investment — and in that regard, the spectrum of policies can be very broad — will tend to increase the productivity of the Canadian economy and lead to better use of labour. In fact, any policy that aims to make the economy more competitive bears on productive efficiency.

There are also policies that are specifically directed at innovation, which aim to improve productivity through the better use of existing factors. Throughout the world, governments have developed innovation agendas, driven by the recognition that ideas spur economic growth. The Organisation for Economic Co-operation and Development (OECD) regularly surveys the innovation policies of its member states and publishes a scorecard tracking countries’ innovation performance.²²

In Canada, the federal and provincial governments have placed great value on innovation in recent years. Governments offer tax credits for research and development expenditures, fund commercialization strategies for university-based research and development, encourage greater participation in post-secondary education and lifelong learning, facilitate the development of clusters and encourage more collaboration between the academic and business communities.

In 2002, the federal government put forward its Innovation Strategy (Canada 2002a, b). The strategy calls on the public, private and non-profit sectors to collaborate to achieve the objective of “increasing Canada’s innovation capacity and creating a culture of innovation and learning.”²³ Each of the four pillars of the strategy — knowledge performance, skills, the innovation environment and strengthening communities — has detailed goals and targets, and action plans.

Provincial governments have also implemented strategies to enhance productivity and innovative capacity. For example, in recent years successive Ontario governments have introduced a number of programs designed to enhance human capital, cluster development and greater collaboration between the business and academic communities. These programs include the Biotechnology Cluster Innovation Program, the Medical and Related Sciences Discovery District, the Ontario Commercialization Advisory Committee and the Task Force on Competitiveness, Productivity and Economic Progress.²⁴

Despite these efforts, Canada still trails other countries in research and development investment, which may partly explain Canada’s recent poor productivity growth. In 1981, Canada made investments in research and development amounting to 1.2 percent of gross domestic product (GDP). This compares to 2.3 percent in the United States and the OECD average of 1.9 percent. By 2002, Canada had increased spending to 1.9 percent of GDP, narrowing the gap but still trailing the United States (2.7 percent) and the OECD average (2.3 percent) (OECD 2004b). Data also indicate that Canada trails the world leaders in other innovation-related factors such as per capita research and development expenditures and the proportion of the workforce engaged in research and development (OECD 2004b).

Innovation strategies tend to focus on the structures that support innovation, in particular building human capital and collaboration between the various sectors of society. This suggests that policy makers believe that innovation policy should focus more on support than pressure for innovation. Pressure for innovation will come from customers and rivals, which is the realm of competition policy.

Competition and innovation

Competition between firms is one factor that influences the rate of productivity-enhancing innovation in an economy. Competition creates pressure conducive to innovation, since firms that increase their productivity to surpass that of their competitors will increase market share and profits. This was illustrated, for example, in Figure 7 in Chapter 3, which showed that exporting firms had experienced larger productivity gains than had non-exporters, reflecting the stronger competitive pressures resulting from international trade.

But given that many other factors are involved in innovation, it is not surprising that international evidence about the relationship between market concentration (i.e. the number of competitors), firm size and innovation (typically measured by research and development spending) is mixed.²⁵ Studies by the OECD examining competition and innovation came up with a variety of findings with no clear pattern. Some of these findings are as follows.

• There are both negative and positive relationships between industry concentration and the level of research and development spending.
• There is a positive relationship between concentration and firm size and research and development up to a point and a negative relationship thereafter (an “inverted U” relationship; see below).
• These relationships differ by industry. For example, there appears to be a positive relationship with large firms in concentrated and relatively capital-intensive industries and a positive relationship with smaller firms in relatively skill-intensive industries (Ahn 2002).

The absence of a clear linear relationship between innovation and productivity, on the one hand, and firm size and industry concentration on the other suggests an inverted-U shaped pattern. Thus, within the same industry, both a positive relationship and a negative relationship could hold, depending on how big a firm is or on how much concentration there is. In other words, there is an optimal size at which productivity is at a maximum. Firms falling on either side of that optimal point are less productive and less capacity for innovation. The same pattern can also be applied to industry concentration, implying that both extensive fragmentation and high concentration are probably associated with both lower productivity and with less innovation. This is illustrated in Figure 1 (Aghion et al. 2003).

In area A, innovation and productivity are below the optimal level because the firms are too small and cannot muster the necessary resources. In area B, by contrast, the firms are too large and suffer from bureaucratic inertia. The same can be said about concentration. An industry characterized by low concentration does not have the leaders and large enough firms to stimulate innovation-based competition. On the other hand, competition in an overly concentrated industry (area B) is too weak to stimulate productivity and innovation.

Since firms can be on either side of the inverted U, and the shape of the curve varies between industries, it is impossible to establish a linear relationship between efficiency or innovation and either concentration or firm size. Not surprisingly, a leading authority on innovation and productivity, Dr Sanghoon Ahn, concludes that “there is little empirical support for the view that large firm size or high concentration is strongly associated with a higher level of innovative activity.” He adds, “All in all, empirical evidence does not support the view that market concentration is an independent and significant determinant of innovative behavior and performance” (Ahn 2002, 16).

In Canada, Charlene Lomno at Statistics Canada has done research that suggests no relationship exists between concentration of research and development spending in an industry and total amount of research and development spent by industry (Lomno 2003). Moreover, it is very difficult to establish a pattern between industries. As one author noted, “R&D intensity and market structure are jointly determined by technology, the characteristics of demand, the institutional framework, strategic interaction and chance” [emphasis added] (Symeonidis 1996, 1).

Research also indicates that the number of competitors is not necessarily an accurate indication of the intensity of competition between firms, at least with respect to innovation. A high degree of market concentration does not imply a low level of competitive pressure in all circumstances. For example, an analysis by the Centre for the Study of Living Standards notes that “a high market share held by a small number of firms is not necessarily inconsistent with intense competition. Concentration can improve productivity through achieving economies of scale, and it can also boost productivity if it allows a small number of large firms to compete intensely with each other” (Kellison 2004, v).

Firm size and innovation

In contrast, size matters when it comes to innovation. The belief that small firms are more innovative than large firms is a myth. Statistics Canada’s Survey of Innovation and Advanced Technology suggests a positive relationship between size and innovation. In particular, the smallest manufacturing firms (fewer than 20 employees) report innovation at roughly half the rate of the largest firms (2,000+ employees) (Baldwin and Hanel 2003).

Further, an examination of the results of Statistics Canada’s Survey of Electronic Commerce and Technology shows that organizational and technological improvements are positively related to firm size (measured by the number of employees) in both the goods and services sectors. This “suggests that larger firms [500+ employees] can more readily absorb the costs associated with technological change which include not only the initial lay-out for technological acquisition but also the often associated training, work interruption due to installation as well as potential short-term loss of productivity” (Earl 2004, 18).

This indicates that large Canadian firms have more support for innovation than do small firms. It should also be noted that large manufacturing firms (250+ employees) are more likely to be exporters than are smaller firms (Baldwin and Gu 2003b). As a result, larger firms face greater competitive pressures than do smaller firms, motivating larger firms to be innovative.

Interestingly, research has also shown that “overall, the services sector’s inclination towards technological change mirrored that of the goods sector” (Earl 2004, 10). Given the widely held view that, in general, border effects and the characteristics of many service industries result in less competitive pressure in the services sector than in the goods-producing sector, this research suggests that factors other than competition play a significant role in innovation.²⁶

These findings underscore the fact that innovation occurs within a complex set of interactions, making it difficult to isolate and evaluate the value of its specific elements. As a result, “In many market circumstances there is so much serendipity in research and development that it is impossible to predict the sources of innovation with reasonable certainty” (Gilbert and Sunshine 1995, 588). Moreover, developing and adopting new products and technology does not necessarily imply commercial success. As Baldwin and Sabourin have stated, “Plants may adopt new technologies in order to improve relative productivity and gain market share. But the growth process is on the whole stochastic, depending not only on production success but also on the whims of consumer demand” (Baldwin and Sabourin 2004, 29).

Conclusions

Innovation, or as it is sometimes called in the merger literature, dynamic efficiency, is the result of numerous factors interacting and contributing to a more valuable output out of the combination of existing production factors. Government policies, such as the federal Innovation Strategy, focus mainly on increasing support for innovation through enhancing human capital development. But innovation is a complicated process, and firms require both market pressure and support to be innovative. In certain industries it may be that an increase in competitive pressure, alongside increases in support for innovation, is what is needed to create the environment most conducive to innovation and, as a result, higher productivity.

However, while competitive pressure is important to a successful economy, infrastructure, education, attitudes and many other variables also play an important role. Overall, the mixed results of economic research on competition, firm size and productivity-enhancing investments in innovation, and on the difficulties of successfully implementing new technologies, suggest that deliberately increasing competition (or, alternatively, deliberately encouraging market concentration) would not, by itself, have a predictable and replicable impact on Canada’s innovative capacity. In some cases, increasing concentration in an industry may lead to more innovation and benefit the economy; in other cases, increasing concentration may have a negative effect. This suggests that a one-size-fits-all approach to enhancing dynamic efficiency through competition policy will not work. Indeed, the effect of a more or less restrictive merger policy, or a merger policy with more or less consideration given to efficiencies, will vary both in significance and in the direction of its impact from one case to another.

As a result, it is impossible to have a detailed and comprehensive roadmap to determine how and when competition policy should take innovation (dynamic efficiency) into consideration. There are no universal criteria for determining when innovation would be relevant or how much weight it should be given. Concentration (and the associated intensity of competition) is one factor among many. Size is sometimes positively correlated with innovation, sometimes negatively. The circumstances in each case will be unique. For these reasons, the effects of a merger on dynamic efficiency must be assessed case by case.

An additional complexity is the often long time horizon for realizing dynamic efficiencies. The economic impact of innovation more often than not does not materialize within the two-year time frame that is commonly used to assess the impact of mergers.²⁷ This poses more difficulties for measuring and assessing the impact of a merger on the efficiency of the economy, even in cases when this impact is deemed significant.

Innovation is a critical component of any public policy framework aimed at enhancing the efficiency and productivity of the economy. Governments strive to promote innovation through a wide array of policies. There is nothing to suggest that competition policy should be excluded in that regard. On the other hand, competition policy should not have to shoulder more than it is able.