According to Wikipedia,

A spectrum auction is a process whereby a government uses an auction system to sell the rights to transmit signals over specific bands of the electromagnetic spectrum and to assign scarce spectrum resources.

In 2000, for example, the UK government auctioned off five 3G licenses and raised a total revenue of $£22.5$ billion -- or $2.5$ percent of the UK's Gross Domestic Product at the time. In The Undercover Economist, Tim Harford argues that the auction could not have resulted in higher prices for mobile phone users, because

firms will charge the maximum they possibly can, under any circumstance [whether the licenses are cheap or not]. We also know that their ability to do this is limited by their scarcity power... In the UK there was scarcity built into the amount of radio spectrum available: five licenses were the most the technicians could offer. It does not make any difference to the customer how much the licenses cost, but it does matter to the taxpayers, who would like the government to get a lot of money for this valuable public resource.

Let's say for the sake of simplicity that each license sold for an equal price, $£4.5$ billion (the prices varied, but this doesn't affect my argument). This amount had to come from somewhere. Let's say they borrowed this money. The bills they send to their customer will be affected by three things: infrastructure costs and wages, profits, and the repayment of the loan used to pay for the license. In a free market, the companies would have roughly the same costs and profits. Harford is right that neither of these is influenced by how the licenses were distributed: scarcity is baked into the amount of spectrum available, and only scarcity power determines profit.

But consider the last of these costs. Over time, companies would recoup the licenses' cost by charging their customers. Sure, they could postpone this recoupment, but they will get it someday (and will pass on the interest to the customer). It should "make a difference to the customer how much the licenses cost". If my thinking is right, the windfall from this auction was right out of the taxpayer's purse.

In reality, telecom companies in the UK nearly went bust after the auction. But that needn't have happened. The telecom companies without 3G licenses were the most seriously afflicted. Elsewhere, too, telecom companies have paid exorbitant sums for spectrum licenses and come away just fine.

Tim Harford got his economics degree from Oxford. I, on the other hand, haven't ever read an economics textbook. I'd bet the error in reasoning is on my part rather than Harford's, but I'd like to know where.

  • 3
    $\begingroup$ The question title doesn't seem to match the body of the question. The title is "is .. a tax", but the body is "does the price paid for the resource affect the price charged for it." $\endgroup$
    – James K
    Commented Apr 5, 2021 at 10:32
  • $\begingroup$ "nearly went bust" = didn't go bust, therefore they didn't over-pay for a scarce resource. $\endgroup$ Commented Apr 5, 2021 at 11:40

1 Answer 1


Suppose you write some software that you can then freely sell at practically no cost per unit (the wonders of the internet). You want to make as much profit as possible. Since you have almost no per unit cost, maximizing your profit will amount to maximizing the revenue, the price per unit times the number of units sold.

Note what did not enter the calculation: The cost of developing the software. How much money you might eventually get might influence whether you put in some effort to create the software, but it will not influence your pricing decision once you made the software.

The companies bidding for spectrum are essentially in the same situation. Once they have gotten their spectrum rights, how much they paid will not influence how much they charge.. Of course someone has to pay, but these are companies bidding for monopoly rights. Apart from the cost, the license gives them a huge profit and this profit is what the cost is paid from.

  • $\begingroup$ Thanks! Disregarding collusion, though, won't profits eventually fall to zero? I'm guessing the clincher is that profits can exist for a long enough time -- if a company lowers its prices, other companies will take a while to respond to it because of sticky ads, etc. Is this correct? $\endgroup$ Commented Apr 4, 2021 at 21:33
  • $\begingroup$ There is no free entry here. The spectrum is inherently limited, which is why it is auctioned off. There is nothing that drives profits to zero. $\endgroup$ Commented Apr 4, 2021 at 22:07
  • $\begingroup$ @MichaelGreinecker I'm confused. The cost to develop the software certainly did enter the calculation. The profit maximizing price is heavily dependent on the cost to enter the market. As a trivial example, if the cost to develop the software is zero, the profit maximizing price will be very close to zero because otherwise a competitor will underprice you. $\endgroup$ Commented Apr 5, 2021 at 10:02
  • $\begingroup$ @DavidSchwartz Maybe in a world without enforced monopoly from patents and copyrights. $\endgroup$
    – pipe
    Commented Apr 5, 2021 at 10:56
  • 1
    $\begingroup$ @DavidSchwartz The cost for a competitor to enter the market is the cost of couping the government (or getting laws changed); exclusive use of the spectrum has been sold, so nobody else can enter the market. $\endgroup$
    – wizzwizz4
    Commented Apr 5, 2021 at 12:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.