An Examination Of Competition In The Wireless Industry PDF Download

Last year's Annual Report and Analysis of Competitive Market Conditions with Respect to Mobile Wireless broke new ground by not concluding, as had prior reports, that the wireless services market was “effectively competitive.” This year's report did the same. The 14th and 15th reports review a wide variety of evidence, both direct (how firms and customers behave) and indirect (industry concentration measures) in making its competitive assessment. The reports are silent on how to interpret this evidence. In contrast, modern antitrust analysis relies far more on direct evidence. In failing to put more weight on the relevant direct market evidence to reach an informed competitive assessment, the 14th and 15th reports invite erroneous conclusions about the real state of competition in wireless markets. We are concerned that these erroneous conclusions eventually could adversely influence regulatory policy in wireless markets. Before economists came to rely on direct measures of market power, they relied on indirect measures, such as market share in the relevant markets, the Herfindahl-Hirschman Index (HHI), and market definitions. The 14th and 15th reports downplayed direct evidence of competition - namely, aggressive pricing behavior, robust entry, and continued long-term reductions in price, all of which strongly support a conclusion of “effective competition.” Instead, the FCC focuses on inferences of market power based on market shares. For example, the FCC makes much of the combined share of the top four wireless providers generally, and of the top two wireless providers, AT&T and Verizon, in particular. To test the FCC's presumed relationship between market structure and prices in the wireless industry, we analyze the TNS Telecoms database of cellular telephone bills. We find no statistically significant relationship between a household's monthly wireless bill and the HHI of the economic area in which the household resides. Thus, market concentration does not appear to have an impact on what the customer actually pays. This finding, along with the fact that wireless prices have declined over time as industry concentration has increased, undermines the structure-conduct hypothesis that undergirds the FCC's market-share analysis. Finally, we discuss the policy implications of our findings for handset exclusivity and spectrum allocation.