Complicated relationship between competition and innovation

Noboru Kawahama

22 March 2012



More than 20 years have passed since the collapse of the Soviet Union. Although various views exist about the cause of the failure of the socialist economy, stagnation in innovation is arguably one of the fundamental reasons. It is true that the Soviet Union once achieved a level of basic technology high enough to lead to the Sputnik crisis. But under a regime that did not allow for opportunities to cultivate consumer demand or use profit opportunities as leverage to utilise technology, innovation in general did not develop. Under a free market economy, profit incentives can entice firms to discover consumer demands and to attempt various experiments to provide products that meet such demands at lower prices. The free market economy thus functions as an ‘innovation machine’ (Baumol 2002).

The free market economy performs better through competition. However, as has been well known since the time of Schumpeter, a monopolistic market is more desirable for innovation than a competitive market. A certain amount of market power is required to make investments and take risks in research and development (R&D). Furthermore, profit isolated from competition after successful innovation is necessary for investment incentives.

The opposing view, often associated with Arrow, is that a competitive market better facilitates innovation because innovation incentives for monopolistic firms decline due to the replacement of existing products, and active competition leads to enhanced incentives for innovation in order to exit the competition.

Although there is a diverse body of research, an inverted-U relationship is generally assumed, as shown in the study by Aghion et al (2005). This research finds that such a relationship exists between product market competition, as measured by the price cost margin, and innovation using data from the UK market.

Several patterns are conceivable for the relationship between the intensity of product market competition (whether measured by concentration or price cost margin) and innovation, depending on the situation of the competition in the market in question (asymmetric or not) and the type of R&D assumed (drastic or not). Recent theoretical analyses tell us the effect those incentives have in specific situations. No unilinear answer is considered to exist for this question, as is the case with other questions such as the extent to which patents are conducive to incentives for R&D or what is the optimal scope and term of patents.

Misunderstanding concerning competition policy

The complicated relationship described above is considered to be the cause of a major problem in competition policy. Competition policy often takes the formation, maintenance, and reinforcement of static market power as the criteria for intervention. Doesn't this take into account dynamic competition which is more important to the national economy? Doesn't such a competition policy pursue short-term efficiency at the cost of greater efficiency?

In a sense, these questions arise from misunderstanding competition policy. In an intervention pursuant to competition law, the existence of static market power itself is not regarded as problematic, but what is regarded as problematic is when market power either is or threatens to be formed after an adverse impact on the process of competition such as competition avoidance or exclusion. Tension between competition policy and dynamic competition arises only where the formation of market power via an act harming the competition process is identified as necessary for innovation. For instance, the formation of market power by a price cartel is unlikely to promote innovation. A merger that brings about static market power is unlikely to be helpful to innovation unless efficiency through R&D is clearly indicated. Of course, it is not impossible that a concerted action taken for facilitating R&D has a short-term effect of competition avoidance on one hand and an efficiency-improving effect on the other, but cases in which hardcore restraint of competition is necessary for successful R&D are rare.1

Toward competition policy for innovation

Isn't the important issue to promote dynamic competition through competition policy? Apart from whether or not static competition leads to innovation, there are cases in which a certain type of action suppresses competition in terms of innovation. For instance, among the practices that were condemned in the Microsoft antitrust case in the US, the one against JAVA hindered the development of JAVA-related technology which was likely to have grown as a multi-platform technology. What was regarded as a problem was the prolonged monopolisation of the operating system market caused by the suppression of innovation. In the US, the innovation market has played an important regulatory role since the Antitrust Guidelines for the Licensing of Intellectual Property in 1995. The guidelines assume an evaluative space apart from product and technology markets in order to assess competitive activities at the stage of R&D. In Japan also, laws reducing free competition by lowering R&D incentives, as observed in the Microsoft Non-Assertion of Patents provision or the Qualcomm case, have become regarded as problematic in recent years.

These examples concern laws that directly exclude competitive actions in innovation. For instance, the Horizontal Merger Guidelines of the US, revised in 2010, clearly state that such a merger is problematic and results in the acquired firm's reduced incentive for R&D. There is a question of whether the concept of an innovation market often employed in US merger regulations falls under the "particular field of trade" in Japan. It is possible, however, that if the number of innovations identifiable without using the concept is small, incentives that stagnate innovation may occur depending on the given situation. It is certain that merger regulations in Japan tend to concentrate on short-term effects, and some work is required to examine whether there was any oversight in this regard.

In evaluating laws reducing competition in innovation, we need to judge the extent to which such acts influence the ability and incentives for innovation in a specific context. This may be a complicated job, but isn't it a problem worth addressing with the help of recent theoretical developments in the complicated relationship between innovation and competition?

Editor’s Note: This column has been reproduced with permission from RIETI. Original text in Japanese.


Baumol, William J (2002), The Free-Market Innovation Machine: Analyzing the Growth Miracle of Capitalism, Princeton University Press.

Aghion, Philippe et al (2005), “Competition and Innovation: An Inverted U Relationship”, Quarterly Journal of Ecnomics, 120:701.

1 If there is a large spillover effect, concerted research activities enhance incentives for R&D activities. Normally, the hardcore avoidance of competition is not required, however. If concerted R&D has the effect of avoiding competition in the product market, this poses a difficult problem. This, however, is a problem with non-hardcore cartels in general.



Topics:  Competition policy

Tags:  US, competition, Japan, innovation

Faculty Fellow, RIETI; and Professor, Graduate School of Law, Kyoto University