Appraising Merger Efficiencies

Abstract

Mergers of business firms violate the antitrust laws when they threaten to lessen competition, which generally refers to a price increase resulting from a reduction in output. However, a merger that threatens competition may also enable the post-merger firm to reduce its costs or improve its product. Attitudes toward mergers are heavily driven by assumptions about efficiency gains. If mergers of competitors never produced efficiency gains but simply reduced the number of competitors, a strong presumption against them would be warranted. We tolerate most mergers because of a background, highly generalized belief that most or at least many produce cost savings or improvements in products or service. This article considers the current approach of merger enforcement policy to merger-induced efficiencies. Merger analysis today takes efficiencies into account in two ways. First, it makes assumptions about efficiencies in determining where the line for prima facie illegality should be drawn. Second, it recognizes an efficiencies "defense" once prima facie illegality has been established, with the burden of proof on the defendant. The rapidly growing empirical literature on post-merger performance suggests that merger policy today is more likely to permit an anticompetitive merger than to prohibit a harmless one. At the same time, however, the fault appears not to lie with the efficiencies defense. The defense has almost never successfully defended a merger after the government has made out a prima facie case of illegality. In that case the under deterrence problem must lie in the prima facie case itself. Welfare tradeoff models attempt to assess the welfare effects of mergers by comparing consumer harms and producer gains. One problem with the well known welfare tradeoff model developed by Oliver E. Williamson is that the efficiencies it contemplates occur at output levels that are lower than they were prior to the merger. While efficiencies at lower output levels are possible, they properly require additional proof. Of course, efficiencies might be so substantial that post-merger output is higher, and prices lower, than at premerger levels. But in that case there is nothing to trade off -- both producers and consumers would benefit from the merger. Williamson's model also assumed a market that was perfectly competitive prior to the merger but monopolized thereafter. Virtually no challenged mergers today fall into that territory. Most mergers occur in moderately concentrated markets where pre-merger prices are already substantially above marginal cost. In that case consumer welfare losses are much larger and efficiency gains must be spread over a much smaller output. The 2010 Horizontal Merger Guidelines also require that efficiencies be "merger specific" -- that is, that they could not reasonably be brought about except by the merger. Under a general welfare test that trades actual consumer losses against producer gains that approach makes sense, but under the consumer welfare test that the Merger Guidelines apply it is perplexing. First, if the efficiencies are not of sufficient magnitude to offset fully any propensity toward a price increase, then the efficiency defense will be rejected whether or not the claimed efficiencies are merger specific. However, if the efficiencies are in fact of sufficient magnitude to predict that the post-merger price will be no higher than the pre-merger price, then why do we care? Such a merger does not harm consumers, and as a result is not anticompetitive.