5G Needs Better Markets, Not More Buildout Requirements
Behind all those Super Bowl commercials about “5G” has been a years-long push to get the wireless frequencies needed for the next generation of wireless service ready for widespread use. One of the biggest recent steps in that direction is the Federal Communications Commission’s auction of licenses to operate in the lower portion of the 3.7-4.2 GHz band, sometimes called the “C band,” which has ideal characteristics for 5G deployment.
Last week, however, former Alphabet executive chairman Eric Schmidt sought to sound the alarm about the auction. Writing in the Financial Times, he laments that the licenses “imposed no meaningful requirement to build necessary network infrastructure” and that mobile companies who win in the auction will likely face “divestment and downsizing.” This doomsaying, however, is unlikely to come true.
To begin with, the auctioned C-band licenses do come with construction requirements. But beyond that, construction requirements are an ineffective and counterproductive way to ensure productive use of spectrum.
Imagine that instead of spectrum licenses, companies were spending billions to acquire plots of land. By Schmidt’s logic, the fact that someone spends a lot of money for land is evidence that they won’t build on it: They have spent all their money on the land and now have no plans to develop further. The only solution is to have stronger government mandates to force the landowner to build.
But that way of thinking is wrong: If someone spends a lot of money to buy a piece of land, we expect them to develop it quickly so they can get a return on their investment. If they suddenly don’t want to build, then they will probably just sell the land to someone who will develop it. There will be no shortage of offers from interested buyers, especially if it’s quite valuable land. To hold on to the land in the face of these offers would mean a big opportunity cost: if you turn down a billion dollar offer, then your alternative use (or nonuse) must be worth more than that.
Likewise, C-band spectrum, as everyone including Schmidt acknowledges, is extremely valuable. But it is valuable because it’s good for 5G deployment. Leaving it fallow wouldn’t make anyone much money. So even if a mobile carrier doesn’t want to actually deploy in the band, others (probably the same ones that bid up the price in the initial auction) would be happy to take over. If there is a functioning market available, the spectrum won’t go unused.
But that’s a big “if.” If the market isn’t functioning properly, then it is possible that spectrum usage rights will get stuck in unproductive arrangements. And frictions abound in the spectrum economy. A few examples:
- Often spectrum licenses are not so easy to sell. Indeed, just to get to the current state of the C-band required a multi-year FCC proceeding. The same is true for other bands. Buying a spectrum license is a much more cumbersome (and often politicized) process than simply finding a willing seller and agreeing on a price. The time and expense of getting FCC approval are significant transaction costs that, if too high, make the whole transaction unprofitable leaving the rights frozen in unproductive configurations. Thus, the prevalence of these transaction costs keep spectrum rights from flowing to more productive uses, and they could indeed result in it being disused. Reducing these transaction costs, therefore, should be a goal of those who want to prevent underuse of spectrum.
- Fortunately, C-band licenses will be for “flexible use,” but in other bands, the particular use to which a band must be put (e.g. FM radio or space-to-earth satellite) is prescribed by the FCC. This diminishes the supply of spectrum available for any given use. Part of the reason opening up the C-band is such a big deal is that other bands in the neighborhood have incompatible usage rights which gums up the market, like how land zoning distorts and diminishes the market for housing. American leadership in 5G could get a boost just by making spectrum licenses more flexible.
- Supply is also shrunk by the fact that the federal government controls more than half of the most sought-after airwaves despite not having to pay their market value. And even when private users do have rights to a band, federal agencies often become uncooperative, throwing around vague slogans about government interests, often without the engineering facts to back them up.
Fixing these kinds of issues would create a market that is more resilient and less prone to leaving spectrum fallow. A liquid, competitive market would reduce the need and the incentive for licensees to hold on to spectrum for financial or anticompetitive reasons.
But none of these problems would be ameliorated by increased buildout requirements. Instead, those mandates simply encumber licenses further and treat the symptoms of a sickly spectrum marketplace without curing the root causes that keep spectrum rights mired in bureaucracy rather than being unleashed for consumers.
President Biden’s FCC should prioritize creating better-functioning markets and more flexible licenses so that we can use every band productively. The C-band is a step in the right direction, and we should expect it to contribute to American 5G networks, not hold them back.