Horizontal Mergers: Law, Policy, and Economics

While there is considerable room for disagreement about the implications of economic learning for merger policy, it is believed that there is a sufficient theoretical and empirical basis for a structural merger policy that accords proper weight to considerations of entry and efficiencies. The Cournot, the Bertrand, the dominant-firm, and other models support the prevention of at least very substantial mergers in most cases, as does the limited empirical evidence. Evidence indicates that, although collusion of various sorts does occur and is a real concern in merger cases, collusion concerns do not justify a merger policy significantly more restrictive than that suggested by noncollusive models. The economist's basic tools applied to the detailed facts of an actual industry may yield valuable insights. However, estimation of the price and output effects of mergers (assuming no efficiencies) is rather crude, and estimation of the magnitude of unique efficiencies and the prediction about entry are even less reliable.