The titular question is posed by Louis Kaplow (Harvard) in a recent piece in the Harvard Law Review that I suspect will attract a fair amount of attention. I may have more to say about this later, but for now, here is the abstract:
Competition law is dominated by the market definition / market share paradigm, under which a relevant market is defined and pertinent market shares therein are examined in order to make inferences about market power. This Article advances the immodest claim that the market definition process is incoherent as a matter of basic economic principles and hence should be abandoned entirely. This conclusion rests on four arguments. First, meaningful inferences of market power in redefined markets cannot be made. Second, the paradigm relies on an unarticulated notion of a standard reference market whose necessity and prior omission signal a serious gap. Third and most important, determining what market definition is best is impossible without first formulating a best estimate of market power, rendering further analysis pointless and possibly leading to erroneous outcomes. Finally, the need to define markets engenders a mistaken focus on cross-elasticities of demand for particular substitutes rather than on the market elasticity of demand, which further reduces the quality of resulting market power inferences. Although the inquiry is conceptual, brief remarks on legal doctrine suggest that creating conformity may not be unduly difficult.
Check out the whole thing.