Last year, when Congress was debating the voluntary incentive auction provisions of the Middle Class Tax Relief and Jobs Creation Act, many argued—including FCC outgoing Chairman Julius Genachowski—that the Commission should have the authority to adopt auction participation rules so that it could prevent an “excessive concentration of licenses” under Section 309(j)(3)(B) of the Communications Act.  While Congress did not include any specific auction participation rules in the Middle Class Tax Relief and Jobs Creation Act, Section 6404 of the new legislation states that “Nothing … affects any authority the Commission has to adopt and enforce rules of general applicability, including rules concerning spectrum aggregation that promote competition.’’  Given these Congressional parameters, the FCC has subsequently issued a Notice of Proposed Rulemaking to modify and tighten its spectrum screen, a policy change that could create de facto spectrum caps and exclude the largest CMRS players from the broadband spectrum auctions.  Similarly, some constituencies have filed comments in the incentive auction docket for the Commission to establish outright bright-line spectrum cap rules, though some broadcasters have encouraged the agency to impose no limits on participation.

The argument for a spectrum cap is familiar and is well summarized in a statement by Greg Rosston, former FCC Deputy Chief Economist and, more recently, Senior Economist for Transitions at the FCC:

… the FCC has tools to make facilities-based competition more likely and more viable.  First and foremost, the FCC should get even more spectrum out into the marketplace.  And it is probably important that the spectrum not continue to go into the hands of the two incumbent landline telephone companies that also have by far the most valuable wireless spectrum.

Put simply, the argument for a spectrum cap holds that by directing spectrum resources away from larger firms and toward smaller firms the latter will become more successful and this will make consumers better off (perhaps by lowering price).   The argument sometimes extends to new entrants, whereby new rivals can be created by keeping spectrum out of the hands of incumbents.

If only it were that simple.  By the Commission’s own admission in its more recent CMRS Reports, innovation and competition in the mobile sector has outgrown the simplistic economic frameworks the agency has favored in the past (and many still encourage it to use), and the case for a spectrum cap is perhaps the most simplistic and naïve argument around.

The argument for spectrum caps rests on many assumptions, but two of the more obvious include the assumption that (1) the equilibrium number of firms serving the wireless industry is determined solely by spectrum holdings; and (2) the assumption that the quality of services does not depend on spectrum holdings.  Obviously, the first assumption is invalid.  Spectrum is but one input into the production of wireless services—giving a firm spectrum does not ensure its financial success (as we have seen, repeatedly).  The construction and operation of a mobile wireless network requires billions in capital expenditures every year.  While the companies spend billions on spectrum in auctions and acquisitions, the data indicates that for each $1 spent on spectrum wireless carriers spend about $5 on network build out.  The sizable investments in infrastructure limit the number of firms that can serve the market (see this paper for an explanation). The construction and operation of a mobile wireless network requires billions in capital expenditures, and these investments limit the number of firms that can serve the market.  As for the second assumption, it is well understood that the provision of advanced services requires greater capacity, which is created by more spectrum (or investment in greater spectral efficiency).  It follows, therefore, that dividing a fixed amount of spectrum across a larger number of firms may reduce the quality of mobile wireless services, thus harming customers even if price were to fall.  The tradeoff between price and quality changes from spectrum policy is detailed in our recent paper A Policy Framework for Spectrum Allocation in Mobile Communications, published in the Federal Communications Law Journal.  You can see George’s summary of the paper at this blog post.

Another problem with the spectrum cap argument is its lack of empirical support.  As demonstrated by the figure from our Spectrum Allocation paper (replicated below), over the past two decades the amount of spectrum has risen (both overall and for each of the major U.S. wireless carriers), yet industry concentration, as measured by the concentration ratio, has not declined.  Thus, heeding the call for more data and less theory, historical evidence does not support the notion that “more spectrum” automatically means “more firms” or “more successful small firms.”  Critically, while concentration has risen over this interval, the price of mobile telephony has contemporaneously fallen.  Therefore, historical evidence does not support the notion that higher concentration leads to higher prices.  As stated plainly by the FCC in the Sixteenth CMRS Report and discussed in George’s recent blog, “High market concentration is not synonymous with a non-competitive market or with market power.”

Furthermore, as we show in our paper Wireless Competition Under Spectrum Exhaust, if wireless firms face a binding spectrum constraint as the FCC argues, then the case for spectrum caps becomes even weaker.  Under spectrum exhaust, increasing the number of competitors will actually increase prices, increase congestion and reduce quality—turning the standard view that more competitors leads to lower prices on its head.  So, even if a spectrum cap were to increase the number of competitors (which is it unlikely to do), in doing so it could make consumers worse off.  George summarizes the analytics of this paper here.

By all accounts, upcoming voluntary incentive auctions are going to be the most complex ever attempted by the Federal Communications Commission.  Before adding the complexity of incumbent exclusion rules and likely reducing the attractiveness of the auction to broadcasters, the Commission should put the argument for spectrum caps to a serious cost-benefit analysis.  I suspect the idea won’t survive honest scrutiny.