May 23, 2011
Comments of Roger Noll on the Proposed AT&T/T-Mobile Merger
Posted by Roger Noll
Superficially, the proposed acquisition appears to run seriously afoul of the merger policy of the antitrust enforcement agencies. The first step in merger analysis is typically to measure concentration—an indicator of the extent of competition. The standard measure of concentration is the Hirschman- Herfindahl Index (HHI), which is the sum of the squares of the market shares of the sellers in the market. According to the Federal Trade Commission and Antitrust Division Merger Guidelines, if the post-acquisition HHI exceeds 2500, an acquisition that causes the HHI to increase by more than 200 is likely to cause a significant reduction in competition if the market also has substantial barriers to entry. At present, nationwide concentration in wireless telecommunications services is roughly equal to the 2500 threshold, and the acquisition would increase the HHI by more than 600. These numbers probably understate the effective concentration in the industry for two reasons. First, only the four major carriers can serve customers who seek mobile access in most of the nation. For connections outside their service territories, smaller carriers often resell Sprint’s service. Second, the available data do not distinguish between mobile voice service, which is more competitive, and high-speed data service, which is more concentrated. Based on the high concentration of the industry, the Antitrust Division will try to determine if the merger would cause an increase in consumer prices. Agency staff will examine the degree of competition between AT&T and T-Mobile. T-Mobile generally charges lower prices than the other major carriers. Whether T-Mobile’s low-price strategy is disciplining the prices and service offerings of the other major carriers is an empirical question. For example, if AT&T serves mostly high-end consumers and competes most intensively with Verizon, while T-Mobile serves low-end consumers and competes most intensively with Metro PCS and other smaller carriers, the merger would not be viewed as having the same anticompetitive effects as it would if AT&T and T-Mobile compete intensively for the same customers. Because wireless carriers provide numerous products, determining how and why customers switch from one provider to another seemingly is difficult; however, due to number portability, most customers who switch providers keep their wireless numbers. As a result, it may be possible to estimate empirically and precisely the competition between AT&T and T-Mobile because the Antitrust Division will have access to data about customer switching behavior between carriers. Another concern of the Antitrust Division will be the effect of acquisition on competition in specific local markets. While several carriers operate in major metropolitan areas, smaller cities generally have fewer carriers. The nationwide concentration of the industry reflects an average between less concentrated major markets and more concentrated small towns and rural areas. In the past, the Antitrust Division has approved some wireless mergers only after requiring divestiture of one party’s assets in local markets where the merging firms have high market shares. For example, in its acquisition of Alltel, Verizon agreed to spin off assets in 85 markets in which the two companies overlapped. This kind of compromise is less likely for the AT&T/T-Mobile deal because, as discussed elsewhere, the principal rationale for the deal is to enable AT&T to obtain more spectrum in areas where it currently has capacity constraints.
May 23, 2011 | Permalink
TrackBack URL for this entry:
Listed below are links to weblogs that reference Comments of Roger Noll on the Proposed AT&T/T-Mobile Merger: