Antitrust for Dummies
Caveat: I am not a lawyer but am an economist who specialized in industrial organization (re antitrust) in grad school.
The Clayton and Sherman Acts are the foundation of current US antitrust law.
Prohibitions in US code against anticompetitive behavior generally fall in two categories - a general prohibition against actions with an anticompetitive purpose, and specific prohibitions against certain acts (termed pro se violations).
Generally all of these restrictions apply only to the dominant firm or firms in an industry - ie it is illegal if Intel does it but not if AMD does it (sorry bees). AMD actions might be illegal if they were intended to exclude 3rd firm entries (Transmeta, Via etc from the market).
The first category, anticompetitive actions, is kinda a blurry mess. Basically it forbids dominant firms from certain kinds of strategic behavior. Intel is allowed to compete in order to increase it current sales and profits. It can compete for future sales and profits, but it cannot cut prices now in order to hurt AMD so that AMD won't be able to compete in the future. These sorts of claims are difficult to prove - they involve intent and demonstrating that the action was both anticompetitive in intent and result.
The second category - per se violations, is much cleaner. A firm with market power (term of art - essentially any firm that can effect industry wide pricing or terms of sale - including but not restricted to monopolies, definitely including Intel in the x86 processor market although it will be a matter of fact to be adjudicated) is prohibited from certain types of behavior which might have the anticompetitive effects - regardless of the actual effect or intent of the company. Some per se violations include tying (where the company requires that customers purchase additional products to get the ones they really want), discriminatory pricing (where different customers get different prices based on non-economic factors), side payments for exclusive dealing etc. AMD alleges all of these in their complaint. Intel claims variously that they are not subject to Clayton restrictions, that their actions do not violate the law, that they have not acted as alleged, and that the US court lacks jurisdiction for the alleged violations. We do not yet have specific Intel responses to AMD claims so it is not possible to evaluate Intel's defense.
There are also non-antitrust claims in AMDs action for things like tortuous interference. This is a tort claim that a third party interfered in the legitimate business of AMD and its customers. If Intel paid Supermicro not to sell AMD products, that would support a claim of tortuous interference regardless of Intel's market status.
Finally we have the question of jurisdiction and the recent ruling. The judge accepted Intel's argument that the US court lacked jurisdiction over some of AMDs claims because either the suspect behavior or damages arising from it occurred outside the US. This is a big deal both for the suit and for antitrust law generally. It is not entirely clear whether the judge intends to exclude only claims for damages to business outside the US or damages to US business that resulted from suspect activities that occurred outside the US or some combination. Read strictly that US law covers only US damages from actions that occur in the US effectively guts US antitrust law, when most markets are international in scope. Even if it only effects damages so that if AMD prevails on the merits they need to pursue claims for damages in each country where their products are sold it would be a huge victory for Intel. I suspect it will be overturned on appeal, but the mere fact that the judge put out such silliness tells us he is not good for AMD.
Couple additional thoughts - on the question of declaring Intel a monopoly - many of the restrictions under Clayton are contingent on the firm having market power (not necessarily monopoly). This is a matter of fact which must be proved by a preponderance of evidence by the plaintiff (AMD in this case, the govt in an enforcement action). If a court has already made a finding on the facts in evidence, then the burden shifts to the defendant to reverse the finding. That is why Microsoft was so concerned about "being declared a monopoly" It lowered the burden for all the private plaintiffs looking for a piece of them.
Last word - while I don't know how this case will turn out, on either the facts or the law, the positions of the Intel defenders here reflect a truly embarrassing ignorance of the law and standards of evidence. The judge has so far made no rulings on the merits of the case besides the positive one to let it proceed to discovery and tentatively to trial. The violations alleged by AMD do not involve the actions of thousands of low level sales people, they are about relations with key OEMs and suppliers involving the top handful of people at Intel. The allegations from AMD are specific person x at Intel told person Y at ASUS (or HP or Gateway etc) that they should do z, or Intel offered a first dollar rebate amounting to x% of total purchases contingent on reaching a specific threshold at a specific time. The response is general, Intel didn't break the law, the alleged behaviors are not illegal, nobody threatened anyone. That is not surprising, Intel certainly doesn't want this tried in public and doesn't want to release any specific info until they have to, and should not lead us to believe they have no defense to the charges, but it is not a basis for asserting as some here do that Intel has denied AMDs charges and therefore it is a matter of whom to believe.
If I say that I saw George shoot Tom in the back with a 9mm handgun and George responds that it was me, that is a matter of dispute with no obvious basis for assessing credibility. If he responds that he did not, was not in fact there, was only defending himself anyway and used a 44 instead of a 9mm, we are entitled to make certain judgements.
e |
