Communications Daily (sorry, no link; it’s subscriber-only) reported today that FCC officials may be interested in auctioning off nationwide blocks of spectrum as part of the next round of spectrum auctions. The story quoted Bryan Tramont, chief of staff for FCC Chairman Michael Powell, saying that the FCC has learned from past auctions that there is “a huge transaction cost on constructing networks” by buying and patching together licenses in various cities: “This is not a local business, by and large.”
Bryan has it exactly right. While spectrum auctions were a huge leap forward from the old era when we just gave spectrum away or handed it out randomly through lotteries, the problem form the beginning with almost all the auctions was the way the wireless properties were aggregated for sale. The FCC, obsessed with the notion of smallness in communications, foolishly decided to carve wireless markets into tiny geographic chunks and auction as many licenses as possible. (The agency also imposed spectrum caps on the overall amount of spectrum one company could hold in a region.) Anyway, you can see the (feeble-minded) logic that was at work here: more licenses = more competition.
As a handful of telecom gurus like Tom Hazlett and Peter Huber predicted at the time, what the industry instead needed was the ability to purchase spectrum properties that covered broad geographic territories to create truly competitive and reliable nationwide wireless systems. But the FCC would have none of it and auctioned off spectrum in small chunks instead of large footprints.
The results were predictable. The industry would spend the next decade re-aggregating wireless licenses in an attempt to built the ubiquitous nationwide networks that public demanded. Again, as Tramont notes, “This is not a local business, by and large.” When we travel from city-to-city or state-to-state, we don’t want to worry about whether or not our cell phone will work once we get to our destination. Today, we demand that it does, and that has only been made possible by the increasing aggregation of the old spectrum licenses into larger national networks. And the mergers and acquisitions of recent years have helped too. Bigger operators have delivered bigger, more reliable networks and services.
The lessons for regulators are simple: Big is not per se bad. Sometime scale counts, and counts a lot. And transaction costs must not be ignored. Nowhere has this been the case more than in the field of spectrum. As the FCC ponders the future of spectrum auctions and the hope of ubiquitous wireless broadband, it would be wise to keep these lessons in mind.
(And before I get some hate mail from the unlicensed spectrum crowd… yes, yes, I know that unlicensed spectrum can play a role here too, but I remain unconvinced that it will be able to provide the goods as quickly or efficiently as private spectrum providers. I’m all for opening up some more spectrum on an unlicensed basis for greater experimentation. But we need to let the big boys build the privately-held networks too and, in the long run, my money is on them to get the job done right.)