This is the first of two blog posts that summarise some of the discussion that took place at an IEA seminar on EU competition policy with particular reference to Intel. The next post will deal with the specifics of that case.
Traditional ways of looking at competition policy are problematic in general, but especially in innovative industries. The focus is on the structure of the market as if it is a static entity. There has been more progress in recent years by competition authorities to try to understand the application of different concepts of market definition. But, nevertheless, we have had enquiries into supermarkets and banks in the last decade where the authorities have tried (in the case of the banks) to calculate their super-normal profits using the capital asset price model and (in the case of the supermarkets) taken market definitions to a ludicrous degree of granularity. From this sort of flawed analysis they have then developed policy.
The Austrian view of entrepreneurship sees the market as a dynamic process of entry and exit with entrepreneurs discovering new information, new products and new forms of production in ways that are welfare enhancing. The key issue to an Austrian is whether the legal framework allows the process of competition – both free entry into the market and exit from the market for failed firms (in the wake of the banking crisis, we should not forget to stress the latter).
Competition authorities with their neo-classical models have sometimes adapted to this way of thinking in a limited way. They have sometimes replaced models of market structure with models of contestability and looked at barriers to entry. Some in the competition authorities, such as John Fingleton, have also been keen to focus on barriers to entry imposed by the state. Interestingly, the EU takes a dim view of barriers to entry imposed by individual member states against businesses in other states though not of barriers to entry imposed by the EU itself.
But this does not take the Austrian perspective far enough. As Vince Cable suggested in some disparaging remarks he made about business, it could almost be said that the purpose of businesses is to create monopolies. What Vince Cable did not say, though, is that, except when governments create monopolies, those private monopolies come under continual attack from potential entrants. Competition comes both from actual entry and the threat of entry. The structure of the market at a particular point in time does not necessarily indicate the degree of competition.
What can Austrian economics tell us about competition policy? As ever, the main function of the economist, as Hayek put it, is to help us understand how little we know. One of the chapter headings of Law, Legislation and Liberty read: ‘If the factual requirements of “perfect” competition are absent, it is not possible to make firms act “as if” it existed’. In other words, if we do not have perfect competition there are unexploited opportunities for welfare enhancement that entrepreneurs still have to discover. If they have not been discovered we do not know what they are and therefore we do not know how to correct for the market failure.
This is clearly a problem for the authorities. But, there is another problem. The potential for monopoly profits must be a spur to innovation, research and development. If this were not the case, then why do we grant patents? If we restrict the monopoly profits that arise before a monopoly is contested or before innovation makes the monopoly irrelevant then we will reduce invention and innovation. We can never, know, of course, by how much invention and innovation will be reduced because it is impossible to know what might have been invented in different circumstances. We only know what has been invented: not what has not been invented. I wonder, therefore, if we should have a less restrictive patent regime – especially given the legal uncertainty that most patent regimes lead to – and a less restrictive competition policy regime. To some extent, both are cancelling each other out whilst creating greater legal uncertainty. But, the main point is that competition policymakers are working in the context of two huge unknowns and they are, as Hayek put it, pretending that they know more than they do.
The competition policy authorities might respond by arguing that they need to use some models and that the static models of market structure help them muddle through in this difficult area. The alternative point of view was put by Kirzner who said: 'Now the mere failure of a theoretical picture to replicate with precision all features of the reality it seeks to explain, is not necessarily fatal for the usefulness of that theoretical picture. But mainstream theory filters out of the picture those aspects of reality which are at the core of an adequate explanation for market phenomena.'