Foreclosing Incentives: Spectrum Set-Asides in the Incentive Auction

The months are ticking down to the historic spectrum swap between broadcasters and wireless providers. With time running out to craft the incentive auction rules, a coalition flying the banner “Save Wireless Choice” is pushing for additional spectrum to be set aside for those carriers who have not acquired much in the way of airwaves below 1 GHz. To be clear, we are talking about an additional spectrum reservation; the FCC already plans on setting aside up to 30 megahertz of valuable 600 MHz spectrum only accessible by bidders with less than 45 megahertz of sub-1 GHz spectrum.

It is worth digging into the specifics of the Save Wireless Choice ask. The group is asking that the FCC raise the maximum reserved spectrum available for carriers with limited spectrum below 1 GHz in the incentive auction from 30 megahertz to 40 megahertz. While this may sound simple, the ask is surprisingly bold, though it takes a bit of unpacking to explain why.

The FCC’s incentive auction is an unprecedented attempt to coordinate a two-sided auction, playing match maker between spectrum-hungry mobile carriers and TV broadcasters willing to part with their frequencies. This is an important auction for several reasons.

For one, we are talking about quite a bit of spectrum. Estimates vary, and success depends on a dizzying number of factors—mainly carrier’s willingness to bring the cash and broadcasters willingness to give up the goods. The FCC explored possible scenarios ranging from 42 to 144 megahertz repurposed for mobile broadband, but by most measures we are looking at more than 100 megahertz cleared—a considerable chunk of the 500 megahertz the National Broadband Plan called to be repurposed by 2020.

Second, this is primo, beachfront spectrum. The 600 MHz band has great propagation characteristics, able to punch through clutter and travel farther with less power. And moreover, this is the last auction of this type of low-band spectrum likely for the foreseeable future. Thus the extensive fights over the details of auction design—this one matters.

The amount of spectrum set aside for carriers with less than 45 megahertz of sub-1 GHz spectrum in a particular market has been a key issue for Sprint and T-Mobile. The advocates of Save Wireless Choice argue that this spectrum is an important factor in the competitive dynamics of the wireless marketplace (it is) and if we don’t set aside a sufficient amount of spectrum for the little guys, the market will slip into stagnant duopoly (it won’t). If this all sounds familiar, that’s because it is.

We already had this debate with the initial mobile spectrum holdings report and order a year ago. There was an extensive back and forth on this very issue, the FCC already decided on a compromise. The FCC recognized the importance of 600 MHz spectrum to the competitive landscape, but also recognized that the primary advantage of auctioning spectrum is to discover the firm who values a particular license most. We use this private value as a proxy for the firm best able to put the spectrum to its most socially valuable use. Therefore competitive bidding is key to a successful auction, and all the more important in an incentive auction. After all, if the carriers don’t bring enough money to the auction, less spectrum will be repurposed.   And of course, open, competitive bidding maximizes revenue for the U.S. Treasury.

Thus the FCC’s compromise: after a certain price threshold is reached, the auction bifurcates, with those possessing less sub-1 GHz spectrum (think Sprint, T-Mobile) bidding on one bucket, and the remaining spectrum available for anyone to bid on. But the FCC set a maximum cap on the reserved spectrum at 30 megahertz—three licenses. Carriers are eager for 20 megahertz chunks for LTE, so by setting the reserve at 30 instead of 40 even those bidding on reserved spectrum would be forced to compete, at least with each other. The Commission made this rationale explicit, stating:

A maximum of 30 megahertz of reserved spectrum could permit at least two reserve-eligible bidders to acquire 600 MHz spectrum licenses for deployment of next-generation networks, with one of the bidders potentially acquiring 20 megahertz of reserved spectrum for such deployment. Moreover, a maximum of 30 megahertz of reserved spectrum, an odd number of 10-megahertz blocks, will facilitate competition among bidders seeking to acquire 20 megahertz.

The FCC knows that if 40 megahertz was set aside, Sprint and T-Mobile (and possibly Dish) would easily dominate even smaller bidders and likely split the reserved spectrum in two and call it a day.

This is why I call the recent Save Wireless Choice push bold. The FCC already heard all the arguments and said, “alright, we can ensure no one carrier sweeps the auction, but we are still going to make the reserve participants bid against each other.” To which Sprint and T-Mobile reply, “but, you see, we’d really rather not bid against each other.” This is what the calls for a 40 megahertz reserve amount to—not just a set-aside, but a set-aside these companies don’t have to fight over.

Note, though, the private value measured by willingness to pay at auction is an imperfect signal of the party that will put the spectrum to its most socially valuable use. As the DOJ has argued, in concentrated markets with high margins there is potential that an incumbent will have a higher value spectrum higher for their private purposes than the “use value” that flows to consumers’ benefit. This additional private value can be a result of “foreclosing” potential competition.

This is the question to be asking—to what extent do carriers value upcoming 600 MHz spectrum, not because they need to deploy it and bring valued services to market, but to hamstring competitors. But this question should be answered on a rigorous assessment of the likelihood of foreclosure, the ability of bidders to foreclose, the effect of bidding restrictions on the incentive auction (both in terms of revenues and spectrum cleared), past and future availability of sub-1 GHz spectrum, importance of this spectrum to competition, etc.

After all, we don’t want to reward rent-seeking nor should we punish those the market sees as successfully utilizing this resource.

The evidence generally points towards AT&T and Verizon as the most spectrum constrained operators. These companies are not hoarding spectrum to foreclose competitors, but aggressively deploying. Furthermore, this is an industry in the midst of a price war with margins falling. In fact, T-Mobile appears to be doing quite well by focusing on capacity in urban areas. Issues around rural coverage in this debate are largely red herrings—the need for more spectrum, and, frankly, the money, is still in cities. T-Mobile has deployed its 700 MHz spectrum in urban areas like Houston, Washington, D.C., Dallas, Minneapolis, and, soon, NYC.

By the way, that 700 MHz spectrum T-Mobile is deploying was purchased from Verizon. To be fair, at time of sale the lower A Block had its challenges, but it certainly seems odd for a company supposedly willing to spend billions to keep others out of 600 MHz spectrum to happily sell T-Mobile 700 MHz spectrum.

Instead of presenting rigorous evidence around the foreclosure issue, Save Wireless Choice relies more on slick PR and the same populist rhetoric, hoping to ride the wave that carried us to Title II and nixed Comcast-TWC. For example, this op-ed that frames the issue as “the next battle for an accessible Internet,” essentially arguing that if you like Title II and blocking the Comcast merger, you’re going to love . . . increasing the incentive auction spectrum reserve to 40 megahertz. Good luck drumming up 4 million comments on this one.

Perhaps it makes sense to ensure we have at least three carriers with significant sub-1 GHz spectrum, and a 30 megahertz reserve can certainly do the trick. But a 40 megahertz handout, and especially one based on trumped-up ideology, is unjustified.