“Ease” and “simplicity” do not spring to mind as the reader slogs through the dense, highly technical NPRM.
Of course, the design phase of the Incentive Auctions is necessarily complex because of the extraordinary complexity of the ultimate goal. That goal includes encouraging as many TV and Class A licensees as possible to surrender their spectrum for repurposing in the most efficient manner possible while what’s left of the TV industry is repacked into less spectrum. And then there’s also the goal of reconfiguring the freed-up spectrum and selling it to wireless providers. The NPRM provides all interested parties an opportunity to attempt to shape the auctions’ final design.
With that in mind, we present the first of a series of posts summarizing various elements of the NPRM. Our series will not address all of the NPRM. Rather, we will attempt to highlight aspects that appear to us to be particularly prominent and worthy of consideration by folks likely to be affected by the process. Comments in response to the NPRM are currently set to be filed by December 21, 2012; reply comments are due by February 19, 2013. We encourage all parties interested in the Incentive Auction program to take a careful look at the NPRM and weigh in with their thoughts.
Auction Design Overview – A chicken-and-egg problem, on steroids, in three dimensions
The Incentive Auction process will include two separate-but-interrelated auctions: a “reverse” auction in which TV and Class A licensees will agree to relinquish some or all of their spectrum rights in return for cash, and a “forward” auction, in which prospective mobile licensees will bid for the right to use portions of the spectrum freed up by the “reverse” auction.
Sounds simple, but wait.
The precise spectrum to be bid on in the “forward” auction won’t be known until the “reverse” auction is completed. And Congress has mandated that the proceeds from the “forward” auction must cover all the payments to successful “reverse” auction bidders, plus a number of other administrative and reimbursement costs. So there’s a threshold interdependence between the two that poses conceptual problems (sort of a regulatory equivalent to M.C. Escher’s Drawing Hands).
But that chicken-and-egg problem is further complicated by the fact that the value of any particular broadcaster’s to-be-relinquished spectrum is likely to be different from any other broadcaster’s. The differences arise from a host of factors, some of them easily calculable (e.g., population covered, perhaps whether any particular relinquishment would create “white” or “gray” areas), some not so much (e.g., extent to which that particular spectrum will facilitate (a) repacking of the TV band and/or (b) repurposing of the spectrum). And, of course, the value of the spectrum available in the “forward” auction will depend on how much spectrum is available and where it can be used.
With respect to determining the value of to-be-relinquished TV spectrum, the FCC is considering a couple of computer programs that might serve to update the likely value of each participant’s to-be-relinquished TV spectrum constantly through the course of the “reverse” auction. Remember, while participating broadcasters will want to keep that value up, it’s in the Commission’s interest to keep it down, so as to minimize the overall pay-out (and, thus, maximize the government’s ultimate take from the “forward” auction).
One of the two computerized approaches under consideration (the “Integer Programming Algorithm Approach”) “would, for a specified amount of spectrum to be cleared, minimize the sum of the reverse auction bids accepted and the relocation costs of stations that are reassigned to new channels.” But that particular approach would not necessarily lead to “optimal” results – although the Commission advises that results would be “within a certain tolerance of optimality” that the Commission, at least, could find acceptable. Oh yeah, and those results might not be easily reproducible and, thus, “less than fully transparent”.
The second computerized approach – the “Sequential Algorithm Approach” – would, as best we can understand it, assess for each auction participant prior to each auction round the feasibility of assigning that participant’s station to some channel in its pre-auction band. As long as the station can be assigned to the pre-auction band, the participant can opt to continue in the “reverse” auction. (Alternatively, of course, it could bail from the auction at any time as well – but heads up: a decision to exit the auction would be irreversible.) The program would determine for each participant the “least-cost” move (“least-cost”, that is, to the Commission). If no re-assignment within the participant’s pre-auction band is possible, then that participant’s compensation would be set at the last price offer it accepted for its last preferred relinquishment option. This approach may be more easily replicated than the Integer Programming Algorithm, but it’s apparently more complicated and less efficient.
As far as operation of the “reverse” auction goes, the Commission is looking at two alternatives: either (a) a single round, one-and-done, put-your-final-bid-forward approach, or (b) a somewhat more conventional multiple round clock format. In the latter, in each round a progressively lower bid amount would be presented by the Commission and each bidder would indicate its willingness to accept that amount. It appears that each FCC-set bid would be unique to each auction participant (based on the particular attributes of that participant’s to-be-relinquished spectrum) and disclosed confidentially only to that participant. The idea is not to have broadcasters bidding against each other, but rather to enable the FCC and each broadcaster to arrive at a mutually agreeable cash value for the relinquishment of that broadcaster’s spectrum.
Another interesting point: the Commission is considering establishing “reserve” prices, i.e., a maximum that it would be willing to pay for the relinquishment of spectrum rights. The maximum would likely vary from broadcaster to broadcaster, depending on the relevant characteristics (e.g., audience served) of the broadcaster’s station.
The “forward” auction would be considerably simpler, but still not without its quirks. The Commission is tentatively planning on using a standard multiple round ascending auction typical of other spectrum auctions. Bidders would be bidding on “generic” categories of licenses (e.g., paired or unpaired) in particular geographic areas, probably in 5 MHz blocks. Specific frequencies would not be involved; those would be assigned once the “forward” and “reverse” auctions have been completed and the repacking process has permitted the identification of specific frequencies available in specific areas. We’ll be posting a separate installment addressing in more detail “forward” auction issues.
For those readers who find the NPRM’s description of the auction process a bit too daunting, you may be better off by starting with the “Incentive Auction Rules Option and Discussion” included as Appendix C to the NPRM. Prepared at the FCC’s request by Auctionomics and Power Auctions, it provides a somewhat more accessible view of what the FCC has in mind. It’s still not quite “FCC Incentive Auctions for Dummies”, but we found it helpful.
Even those who are not interested in participating in the auction may want to take the opportunity to comment. Whether or not they choose to participate, all TV broadcasters will be affected by the Incentive Auction. That’s because, in order to package more desirable spectrum blocks for the “forward” auction, the FCC will likely be forcing most, if not all, remaining full power broadcasters to change channels or implement other modifications. While the Commission is required to take “all reasonable efforts” to protect full-power stations’ existing service areas, and to reimburse relocations costs (to be paid from the forward auction proceeds), the mechanics and implementation of the repacking are, not surprisingly, rather complicated. For low power television licensees, the situation is much worse, as their services will not be entitled to any protection during the repacking process.
Obviously, there remain plenty of questions relative to all auction mechanics. Comments on any and all such questions are invited in the NPRM. Again, comments are currently due to be filed by December 21, 2012 and reply comments by February 19, 2013.
[UPDATE: As we have separately reported, on November 29 the Commission extended the comment and reply comment deadlines to January 25, 2013 and March 12, 2013, respectively.]