Spectrum Auctions

Definition

The usable radio-frequency spectrum is finite: only so many non-interfering channels exist over a given place and time, and their value comes from natural scarcity plus the public's grant of exclusive use, not from anything the licensee produced. That makes spectrum a textbook commons with land-like properties — fixed in supply, publicly owned, valuable by location (frequency band) and coverage. A spectrum auction sells time-limited exclusive-use licences to the highest bidder, so the scarcity rent of the airwaves accrues to the public treasury rather than being handed to incumbents through administrative assignment ("beauty contests" or lotteries).

The idea is due to Ronald Coase, whose "The Federal Communications Commission" (Journal of Law & Economics, 1959) argued that the FCC's practice of assigning frequencies by administrative fiat wasted the resource, and that spectrum use should instead be "determined by the pricing system and awarded to the highest bidder."[1] The proposal was heterodox for decades; the US Congress did not authorise the FCC to assign licences by competitive bidding until 1993, and the first auction ran in 1994.[1][2]

Why It Belongs in the Geoist File

Spectrum is one of the cleaner non-land rents. There is no production story to damage: the frequencies exist whether or not anyone bids for them, so capturing their scarcity value has no supply-side incentive cost — the same argument that makes location rent the clean Geoist case. Auctioning is the market mechanism for discovering that rent and routing it to the public, exactly the motion Georgists apply to land and William Vickrey's auction theory anticipates. Per the rent gradient, spectrum sits near the clean end — with one important design caveat, below.

The Record

Spectrum auctions are now the global standard and have raised very large sums:

• The United States: since the first auction in 1994 the FCC has run more than 100 auctions and raised over $200 billion for the Treasury (the FCC reports a cumulative figure above $233 billion).[2] The record single event was the C-band auction (Auction 107, 2021), which sold 280 MHz of mid-band spectrum for net winning bids of about $81.1 billion — roughly 170% above prior industry projections.[3]
• The design that made it work: the FCC's format is the Simultaneous Multiple Round Auction (SMRA), invented by Paul Milgrom and Robert Wilson (with Preston McAfee). By offering all licences at once and letting bidders raise offers over repeated rounds, it dampens the winner's curse and lets bidders assemble the geographic/frequency packages they need. Milgrom and Wilson received the 2020 Nobel Memorial Prize in Economics partly for this work; the FCC adopted their design almost in full, and Canada, the UK, Germany, India and many others followed.[4]

Honest Limits — the "Sell Once vs Lease" Tension

• One-time capture, not recurring. A standard auction sells a licence for a fixed term for a lump sum — the present value of the rent over that term. This is the spectrum analogue of selling land rather than taxing it: it captures the rent priced in today, but any increase in the frequency's value over the licence term accrues to the holder, not the public — the same limitation the wiki flags for public land leasing. A recurring spectrum-usage fee (a "spectrum rent") would track the rising rent the way a land value tax tracks land, but recurring fees are politically harder and less common than one-off auctions.
• The Coase-vs-"Coasians" debate. Thomas Hazlett and others argue the value comes from creating genuine, flexible property rights in spectrum, not merely from auctioning restricted licences; on this reading auctions capture rent but the FCC's use-restrictions still leave efficiency on the table. Critics on the other side worry that full private ownership of a public resource forecloses future public uses (unlicensed/"commons" bands like Wi-Fi are the counter-model).[1]
• Concentration. Auctions can entrench a few large incumbents able to outbid entrants, a competition concern distinct from the revenue question.

See Also

• Congestion Pricing — the parallel road-space commons case
• William Vickrey — the auction-theory lineage
• Public Land Leasing — the same "capture the rent, but recurring vs one-shot" tension
• Geoism — the umbrella program and rent-domain table
• Resource Rents — the broader natural-resource-rent family

Sources

1. Ronald H. Coase (1959), "The Federal Communications Commission," Journal of Law & Economics 2, 1–40 — used for the origin of the pricing/auction proposal, the 1993 congressional authorisation, and the property-rights-vs-licence debate (C/A-claims; verified via multiple sources this session). JLE
2. Federal Communications Commission, "Auctions Summary" — used for the ">100 auctions since 1994" and the cumulative revenue figure (over $200 billion; FCC reports >$233 billion). FCC
3. Federal Communications Commission, "FCC Announces Winning Bidders in C-band Auction" (Auction 107, 2021) — used for the ~$81.1 billion net-winning-bid record (A-claim). FCC release
4. Royal Swedish Academy of Sciences, "The Prize in Economic Sciences 2020" (Milgrom & Wilson) — used for the SMRA design, its role in the FCC auctions, and the global adoption (A-claims). NobelPrize.org