“If we had any more innovation [in wireless] I think our heads would explode.”
– Professor Gerald Faulhaber, comment at wireless conference in Berkeley, April 2010.
Hats off to the FCC for its new approach to evaluating wireless competition. Its latest report on wireless competition explicitly recognizes that wireless services now include such a broad range of industries, activities, and linkages to other sectors that it no longer makes sense to think of wireless as a single, overarching “industry.” Many observers believe—happily or indignantly, depending on who they are—that by failing to apply the phrase “effective competition” to everything wireless the FCC is sending a signal that it sees reasons to be concerned.
Perhaps that is the Commission’s intent. Perhaps not. I’ll leave divining its intentions to the Kremlinologists. Instead, let’s step back and take a look at some of the economics underlying the analysis and the report’s central conclusions.
Until the 1980s economic analysis relied on the so-called “structure-conduct-performance paradigm” (SCPP) in which market structure was taken as given and a concentrated market was assumed to allow firms to behave as monopolists and therefore raise prices and reduce output. Therefore, a small number of firms was, by itself, cause for concern. It sounds reasonable, and policymakers still seem implicitly to embrace the SCPP. But a funny thing happened along the way to testing this seemingly obvious theory. The empirical relationship between market structure and firm performance turned out to be weak.
It remains true that it is easier for a smaller number of firms to collude to raise prices and lower output than it is for a larger number of firms, so estimating market concentration can be a useful starting point for analysis. However, economists realized in the 1980s that analyses of competition had to recognize that there is no straight line between market structure and performance, other factors are involved, and, indeed, firm performance itself plays an important role in determining market structure. That means analyses of competition must focus on firm behavior and actual market outcomes to determine whether an industry is competitive.
Which brings us back to the FCC’s latest wireless competition report.
The report compiles lots of data on both market structure and the various aspects of behavior and performance of firms related to wireless.
Any concerns about the industry must come from certain features of market structure. In particular, the report notes that the weighted average national Herfindahl-Hirschman Index increased to 2848, which indicates a concentrated industry, though the indicator varies widely across geographic areas.
Key indicators of behavior and performance in the industry—and therefore the most relevant features to evaluating its competitiveness—are eye-opening.
- Churn (a measure of how many people switch providers and therefore an indicator of switching costs): increased slightly from 1.9% to 2.1% per month. A recent study suggests that this rate of churn is higher than the average across industries.
- Prices: The annual cellular consumer price index (CPI) decreased by 0.2% from 2007-2008, while the overall CPI increased by 3.8% during the same time period.
- Pricing plans: New pricing plans emerged, based on prepaid and unlimited models, in addition to the more standard post-paid “bucket” of minutes.
- Average Revenue per User (ARPU): ARPU has been flat, in nominal dollars, for years at about $47.
- Profit margins: The report notes, “While the seven largest mobile wireless service providers all had EBITDA margins over 20 percent during the second quarter of 2009, only four – AT&T, MetroPCS, T-Mobile, and Verizon Wireless – had EBITDA margins greater than 30 percent, and the two largest providers had the highest EBITDA margins.” The weighted average EBITDA for 2008 Q4 (the latest data the report provides that allow us to create the weights) is about 35 percent, down slightly from about 36 percent three years earlier.
- Investment inputs: Wireless providers invested somewhere between $20-$25 billion in their networks in 2008 (the report notes that different sources had different estimates), which represents either a small decrease or increase from the previous year and a decrease in terms of investment as a share of revenue.
- Investment outputs: The number of cell sites increased from about 18,000 in 2007 to almost 29,000 in 2008.
- Advertising: In 2008 the wireless industry was the sixth-highest spender on advertising among product categories. It was the second-highest spender of advertising on Spanish-language television.
- Handsets: Between 2006 and 2009 the number of manufacturers selling mobile handsets in the U.S. increased from 8 to 16, and the number of available handset models increased from 124 to 260.
- Device innovation: In addition to the rise of smartphones, the report notes that entirely new wireless devices, like mifi cards that receive a cellular wireless signal and transmit a wifi signal, and machine-to-machine hardware are emerging.
- Entry and exit: We see lots of both, with entry by providers like Leap, Metro PCS, and Clearwire, and exit through mergers.
- Call quality: Problems per hundred calls decreased to its lowest level ever in 2008 and remained there in 2009. Moreover, the gap in quality across providers decreased.
Other data are more ambiguous. The report notes, for example, that some providers have increased early termination fees, but that those increases seem to be associated with higher handset subsidies. The report further notes that the same handsets are available without early termination fees at much higher prices.
In short, we see in wireless an excellent example of why economists have largely abandoned the SCPP approach to evaluating competition in favor of looking at actual outcomes. Thus, even if we accept the premise that the market for wireless providers has become more concentrated, we nevertheless see an incredibly dynamic market that is yielding new devices, new services, and lower prices. Professor Gerry Faulhaber remarked at a conference on wireless in Berkeley last month, “if we had any more innovation I think our heads would explode” (see video at 9:35).
The FCC made a smart decision to gather lots of data about the myriad components of wireless and to focus most of its efforts on examining outcomes. This approach will allow the Commission to make changes to its reports almost in real-time—a necessity given the rate of change in the industry itself.