Riding the Wrong Wave

Why the government’s plans to auction off spectrum to wireless carriers won’t help innovation

Your television, cell phone, and GPS talk to the rest of the world over electromagnetic waves. All of the wave frequencies used for commerce and government communication—roughly from 9,000 cycles per second to 275 billion cycles per second—are together referred to in Washington as “the spectrum.” Two agencies manage it: The Federal Communications Commission handles commercial and state use, and the National Telecommunications and Information Administration runs federal applications. The FCC licenses pieces of the spectrum to private companies, such as radio and television broadcasters, but all of it belongs, legally speaking, to the U.S. There is a finite amount of it, and everybody wants some.

That is causing problems. Last February, Cisco Systems estimated that mobile data will increase 26-fold from 2010’s numbers by 2015. Almost one-third of this will move through Wi-Fi networks, which use unlicensed spectrum and don’t burden wireless carriers such as AT&T or Verizon Wireless. But the carriers have adopted the phrase “spectrum crunch,” designed to make vivid the pain of a hypothetical moment when there are more data than the available spectrum can handle.

The Obama Administration has decided that wireless carriers need more spectrum. Since the Radio Act of 1927, the U.S. has had, in what ultimately became the FCC and the NTIA, a kind of ministry of electromagnetism, allotting licenses for the specific, restricted use of spectrum among industries and government agencies. In 1994, rather than grant all licenses for free, the FCC began auctioning rights to pieces of spectrum, mostly to wireless carriers. Now all the easy pickings in spectrum have been auctioned off, according to Blair Levin, who headed last year’s national broadband plan for the FCC. And so the Administration has adopted Levin’s idea for opening up more spectrum to wireless companies: “incentive auctions.” Television broadcasters will be offered the chance to give up some of their spectrum in return for an as-yet-unknown percentage of the auction proceeds.

It’s a fair solution, but only if you think the FCC need operate within the narrow range of options encouraged by the companies it regulates. The FCC has decided to move spectrum to what it sees as a higher-value use, an ardently wished-for desire of the wireless carriers. What the commission should be trying to do instead is get out of the business of making decisions altogether. The problem is not that spectrum is in the wrong hands, but that there’s no real market for it.

When Larry Summers, then head of the President’s Council of Economic Advisers, delivered a speech in 2010 announcing the incentive-auction plan, he opened with a tribute to Ronald Coase.* In 1959, Coase, then an economist at the University of Virginia, wrote a paper titled “The Federal Communications Commission,” which essentially argued that the FCC didn’t need to exist. Before 1927, electromagnetic spectrum was akin to a piece of green prairie, where homesteading broadcasters created property rights by using it.

Spectrum, according to Coase, was like land, and Congress’s decision to nationalize it had shuttered a growing market. Free licenses were a subsidy, both to broadcasters and government users such as the Pentagon, and there was no incentive to use spectrum efficiently, since there was no opportunity cost to holding too much of it. The FCC, he concluded, should define a property right for spectrum, auction it to private holders, and let them trade among themselves. “It is true that some mechanism has to be employed to decide who, out of the many claimants, should be allowed to use the scarce resource,” he wrote, “but the way this is usually done in the American economic system is to employ the price mechanism.” Coase won the Nobel Prize for work that began with his 1959 paper on spectrum. He is the Adam Smith of spectrum; everyone who has followed him must somehow respond to his arguments.

Spectrum auctions merited only a few lines in Coase’s paper, but they generated the most government attention. They tend to gain adherents when Washington is looking for revenue, and this year is no different. The President, the FCC, and members of Congress inevitably describe them as a win for everyone. An infographic titled “Spec it out!”—commissioned by the FCC from a private market research firm—features men and women in office chairs eagerly raising their hands to volunteer spectrum and a line pointing to the seal of the Treasury Dept. with the single word “profits.”

Levin, now a Communications & Society Fellow with the Aspen Institute, nods to Coase when he acknowledges that the FCC can’t just reassign the television spectrum to wireless carriers, since broadcasters gather a kind of property right over time through use, despite a clear FCC statute that says they don’t. However, Levin—and the 112 economists who have signed a letter in support of the incentive auction plan now awaiting approval by Congress—says government must serve a market-making function. “In the places that have the best wireless technology,” he says, “the government on a periodic basis manages to put together spectrum in a way that we had the first analog phones, then digital phones, and now 3G. It is entirely due to government action that there is sufficient spectrum for a 4G footprint.”

Spectrum, in this view, demands a kind of industrial policy that picks new technologies and clears the way for them. The incentive auctions, if they work, will reward the Administration with a trifecta: It can claim to have made money, encouraged innovation, and—as the wireless carriers promise—improved Internet access in rural areas that are expensive to wire.

What’s not to like? Plenty, if you’re already an innovator providing rural wireless Internet access. The Administration’s approach ignores the fact that innovation, and the broader economic benefits that come with it, can’t always be predicted. Consider Brett Glass, who built and runs a wireless Internet service provider in Laramie, Wyo. He puts up antennas over private homes in exchange for Internet access and connects them using the same Wi-Fi technology that runs home networks. He would like to buy spectrum to increase his speed and range, but he can’t. He took part in an FCC spectrum auction in 2008. “I should not have bothered,” he says. “The price was bid up so high that I’m convinced that no one could make money off of it.” The sliver he wanted, the lower-band B block of the 700 megahertz service in Casper, sold to Verizon Wireless and now lies fallow.

There is no way to tell how often this happens. The FCC has a map of its spectrum licenses, but has no comprehensive record of whether they’re being used. An auction, says Glass, is not a market. It’s just a single transaction, one that rewards the well-capitalized companies that can afford the upfront lump sum of a multiple-year license. In his experience, a secondary market can never work; the incentives to sell are just too low. In effect, the auction system perpetuates the dominance of large carriers and crowds out entrepreneurs. “I don’t have much respect for Ronald Coase’s vision,” says Glass.

Still, Glass has more in common with Ronald Coase than he might think. Coase is 100 now. When reached by phone, he politely demurred on the subject of the FCC’s plans for incentive auctions, saying only, “I’m not up to date with things, you know.” But his original paper shows sympathy for some things that are important to Brett Glass. The goal of spectrum policy, Coase wrote, should not be order, but efficiency. By that standard, Washington should be favoring entrepreneurs like Glass rather than the big wireless companies. The fixed wireless signals Glass is sending from masts atop houses in Laramie wring more bandwidth out of each available megahertz than do the less targeted, more powerful signals of the mobile carriers. Glass complains that “this FCC pretends that mobile wireless is the only wireless that matters.” In his view, “there’s no real spectrum crunch … [but] if you deploy a less spectrally efficient service, you can say ‘Hey, I’m using it all,’ and keep competition out.”

In the pre-1927 spectral Eden that Coase described in his paper, broadcasters created property rights not through purchase, but through use. Glass, too, would like license holders to lose their rights if they don’t exercise them. There are ways of doing this. Britain, for example, has begun to charge spectrum rent, which imposes a yearly fee on some license holders to introduce an opportunity cost to hoarding. This encourages them to keep less spectrum and use it more efficiently. The FCC praised this approach in its broadband plan, as has the Government Accountability Office, and the Obama Administration has requested the authority from Congress to carry it out. But there’s a difference between requesting a policy change and picking a battle. By staking its spectrum strategy on the incentive auctions that will most immediately benefit the wireless carriers, the Administration is thinking too small. Innovation comes not out of decisions, but out of market incentives. The FCC should consider applying more of them.


    * Summers referred to a phrase coined in 1977 by two former FCC commisioners, who said the odds on Coase’s ideas ever coming to pass were “about the same as those on the Easter Bunny in the Preakness.” The U.S. and other countries now conduct spectrum auctions, which means the bunny at least placed.

    Before it's here, it's on the Bloomberg Terminal.