RECENT decisions suggest that the United States Supreme Court is beginning to look upon integration as illegal per se, under the antitrust laws. It may be presumed, in so far as this inference is valid, that the Court believes that integration necessarily reduces competition "unreasonably." No sharp distinction is made by the Court between vertical and horizontal integration. (Joseph J. Spengler, "Vertical Integration and Antitrust Policy", Journal of Political Economy, Vol. 58, No. 4 (Aug., 1950), pp. 347-352.)This was written in 1950 and things have changed since, at least with respect to vertical integration,
Because there are often efficiencies to vertical integration, the Antitrust Division typically requires a showing of market power before it considers whether a vertical arrangement poses serious competitive concerns.At least part of the reason for the change in view is due to the work of Oliver Williamson,
Williamson was skeptical of the conventional wisdom of the time [1960s], which presumed that the purpose and effect of many vertical practices was the enhancement of market power and the erection of entry barriers. Contrary to this view, which was widely adopted by antitrust lawyers and courts in the 1960s, Williamson could see rationales for various vertical practices that were based instead on economic efficiency.The view of horizontal integration has changed less, but the work of Hart and Holmstrom may be a step in the direction of greater change even here. In their model, two firms are in a lateral relationship and yet they can show that there are conditions under which integration is optimal. So they can show that even horizontal integration can be efficiency enhancing.
This result follows on from a similar conclusion found by Williamson, arrived at via a different framework of analysis, who argued that there is a trade-off between reduced market competition and increased efficiencies due to a horizontal merger. Hart and Holmstrom gives us another reason to consider the efficiency side of the argument.