Competition law (which travels under the name antitrust in the United States) rests on the proposition that the most efficient allocation of goods and services takes place within the framework of a competitive market, where many buyers and sellers each seek their best deal. The genius of this situation is that the pursuit of individual self-interest will lead to a maximisation of social welfare if all players have to abide by sound institutional rules. One such key constraint requires that firms on the same side of the market not collude with each other to raise prices or reduce output. The central office of competition law is to identify and limit that collusion.

Yet competition law has a dark side, which is darker at present in the EC than it is in the US. Thus it is commonly held that unilateral actions by single firms can also create and extend monopoly power, even without collusion, as when a single firm holds the market dominant position when new entry is limited. But pushing hard against unilateral behaviour will land us in hot water if we do not guard against the risk of excessive liability that dampens innovation. Frequently, firms achieve dominance through innovation that their less astute competitors have overlooked. Imposing extensive liability on these firms could easily reduce the pace of innovation.

I fear that the EC’s hot pursuit of Microsoft for the dominance of its operating system, which I discuss at length here, has just this consequence. Under competition law, the shape of the remedy is often as important as the finding of liability. Requiring interconnections to a dominant server is one thing, forcing a firm to share its intellectual property with another is, however, overintrusive. We must never let competition law become the enemy of new forms of competition that break up established ways of doing business.