Testimony of Thomas W. Hazlett before the Senate Commerce Committee, Subcommittee on Communications

The case of Northpoint Technology reveals ongoing infirmities in the U.S. spectrum allocation system. In a more efficient world, innovative wireless companies such as Northpoint would simply buy the spectrum they need, much as any company buys labor, raw materials, and capital inputs. Barriers to entry would be low, and new competitors could quickly test their technologies with consumers in the marketplace.

But, despite license auctions, the U.S. does not have anything approaching competitive bidding for spectrum. Firms attempting to offer novel wireless services must first convince Federal Communications Commission planners that it would be in the public interest for spectrum to be allocated for their project.

It is an understatement to say that this is a costly and time consuming process. In it, the applicant must reveal its business plan, negotiate endless regulatory details with agency staff, negotiate spectrum sharing rules with incumbent users, and do virtually all of the heavy lifting in surmounting regulatory barricades designed not to welcome rivals but to foreclose them.

License auctions ironically lift the walls. Should a competitive technology, however improbably, get past the allocation contest, it must now buy back its business plan from the federal government. In this it will compete with established service providers which have strong incentives to outbid the potential entrant simply to lessen the impact of new competition. In Northpoint’s case, I have estimated that either of the two incumbent satellite TV suppliers would bid several billion dollars above the highest bid Northpoint might plausibly make for MVDDS licenses, owing to the economic gain associated with avoiding price reductions. On an annual basis, cable and DBS subscribers would expect to pay at least $2.75 billion less in subscription fees. It should also be noted that these gains have been lost for several years simply due to the delay imposed on Northpoint’s entry by the FCC rulemaking process.

License auctions are very useful for assigning rights among roughly comparable applicants. The firm willing to pay the most to offer service is, all else equal, the firm that will likely do the best job of providing service to the public.

But where applicants are distinguished, other assignment rules can be more efficient. Indeed, many if not most FCC licenses are still assigned by non-auction procedures because of the disruptive effect auctions would have on telecommunications investment. Specifically, the FCC does not invite competing bids for incumbents’ license renewals. To do so would destroy incentives for licensees to invest in technology, capital, or a reputation for high-quality service. All such investments could be substantially appropriated by others (including the government) were a license to be put up for bids at expiration. Policy makers realize this, and avoid economic losses by simply renewing licenses for law-abiding licensees.

But FCC regulators do not recognize that the same economic misallocation occurs when new entrants are appropriated after investing substantial resources in gaining a spectrum allocation -- the case of Northpoint Technology. The message it sends to innovators is: give up early. If you seek to move the FCC to allocate spectrum to a new technology you will invest years of hard labor, and millions of dollars, for the right (if you succeed!) to bid for a license against other interests who contributed not a kopek. Innovation grinds to a halt while entrepreneurs wait for some one else to shoulder spectrum allocation burdens.

This tragedy of the commons leads to under-investment, and consumers lose valuable new wireless options while paying higher prices for less competitive services. The long-term solution is to reform spectrum allocation, allowing markets to efficiently shift bandwidth from less useful enterprises without government approval. In the near-term, applicants who do invest substantial resources to bring about productive new uses of radio spectrum should not be appropriated but rewarded. I have outlined a simple two-part test in my paper, “Anticompetitive Uses of Competitive Bidding: The FCC’s MVDDS Rulemaking,” which would allow regulators to distinguish when license auctions are appropriate under the current spectrum allocation process. I am happy to make this paper available to any interested party today, or electronically. My address is: twhazlett@yahoo.com