The Decision of the Commission of 13 May 2009 in the Intel case: Where is the Foreclosure and Consumer Harm?

Abstract

On 21 September 2009, the European Commission published a provisional non-confidential version of its 13 May 2009 Decision in which it condemned Intel to a record fine of € 1.06 billion on the ground that it had granted conditional rebates and payments to a number of OEMs and a large retailer of consumer electronics purchasing its x86 CPUs, and that it had paid OEMs to delay, cancel or in some other way restrict the commercialization of specific AMD-based products. This paper shows that the Commission Decision contains a number of flaws. They include the facts that the Decision: (i) relies in substance on a per se prohibition of conditional rebates recognized by the formalistic case-law of the Community courts, notwithstanding that the Commission had clearly indicated in various important policy documents, including its Guidance Paper on Article 82 EC, its intention to move away from this approach for an effects-based analysis; (ii) states, contrary to sound policy, that it need not conduct an “as efficient competitor” test, but conducts a misguided one anyway; (iii) insufficiently supports its speculative theory that the OEMs’ purchasing policy was influenced by their understanding of Intel’s alleged intention to reduce or eliminate their rebates should they buy x86 CPUs from AMD; (iv) fails to demonstrate its contention that Intel’s rebates harm competition and consumers; and (v) conducts an excessively restrictive analysis of the efficiencies created by Intel’s rebates. The Intel decision thus stands for the dangerous proposition that any dominant firm is at risk under Article 82 EC if there exists evidence that employees of a customer believe that reducing present purchases from it could have repercussions with regard to the availability and terms of future purchases, even if the belief is ambiguous, equivocal or contrary to written assurances of the firm or its executives, and without any showing of foreclosure. While the foregoing may be considered as an overstatement and that an “agreement” on conditions (not a mere unilateral belief on the part of the customer) is necessary to find a violation, the Commission accords itself so much latitude on how it collects, interprets and weighs evidence that the distinction is illusory. The compatibility of the Commission Decision with EC competition law will now be examined by the Court of First Instance of the European Communities to which Intel lodged an appeal. Because of the wide-ranging implications of this Decision, not only for Intel but for all large corporations having to negotiate price incentives with their customers, it is to be hoped that the Court of First Instance of the EC will review this decision carefully and hold the Commission to the same rigorous standards it has applied in the merger control area. An important question (that will not be addressed by the Court of First Instance, but which is nevertheless relevant from a policy standpoint) is whether antitrust intervention was at all needed in a market characterized by increasing output, decreasing prices and sustained innovation. These characteristics alone should raise serious doubt about claims of anti-competitive foreclosure and consumer harm, especially when they are made by competitors. These characteristics also question the Commission’s wisdom of investing large enforcement resources in what turned to be a long and protracted investigation. As this paper will demonstrate, the market for x86 CPUs was competitive and there is no convincing evidence that Intel’s conduct was anti-competitive and foreclosed AMD and harmed consumers.