STATEMENT OF COMMISSIONER AJIT PAI,
APPROVING IN PART AND CONCURRING IN PARTRe:
Expanding the Economic and Innovation Opportunities of Spectrum through Incentive
Auctions, Docket No. 12-268, Notice of Proposed Rulemaking
In 1939, Winston Churchill famously described Russia as “a riddle, wrapped in a
mystery, inside an enigma.” Seventy-three years later, I believe that this is an apt description of
the challenges we confront in implementing our legal obligation to hold incentive auctions. It
has often been said that this will be the most complicated set of spectrum auctions ever held by
any country. In fact, it has been repeated so often that it has probably become a cliché. But this
Notice of Proposed Rulemaking (NPRM) makes clear the magnitude of the task in front of us.
There are many pieces of the puzzle that have to fit together for this enterprise to end in
success, including a reverse auction, a forward auction and a repacking plan. Doing any one of
these things individually would be a significant undertaking for the Commission. Doing all of
them in conjunction is a daunting proposition indeed.
If we at the Commission make the right decisions, each piece of the puzzle will fit
together seamlessly. We will free up badly needed spectrum for mobile broadband, which will
promote infrastructure investment, economic growth, and job creation. We will preserve a
vibrant and free, over-the-air broadcast service, including by providing needed funding for those
that wish to stay in broadcasting but choose to channel share or move to Very High Frequency
(VHF) spectrum. And we will raise financial resources to help build a nationwide public safety
broadband network, reduce the federal budget deficit, and advance next-generation 911 service.
If, on the other hand, we do not get this right, we could end up with a Rube Goldberg contraption
that will produce a failed auction.
Fortunately, the Commission has assembled an immensely talented team to work on this
proceeding. They have put in long hours to produce this NPRM, and I would like to thank each
of them for their service. In particular, given my past work in the Office of General Counsel, I
would like to recognize Bill Scher of OGC for his leadership on this project. Based on my prior
experience with Bill, I am not surprised that he has conducted himself with such distinction
working on this item.
But yeoman staff effort alone will not be enough to produce rules that will lead to a
successful incentive auction. Rather, it is critical that we receive a wide range of input from all
affected stakeholders. Given the complexity and prominence of this matter, rarely has the
“comment” part of “notice and comment” rulemaking been more important. How will the
proposals that we are making today work in the real world? Are there alternatives that will lead
to a more successful outcome? You need to tell us. And we need to listen.
This morning, I will vote to approve in part and concur in part with this NPRM. As that
suggests, I agree with some proposals contained in this item and not necessarily with others. But
I think we all agree that today’s NPRM serves as an invaluable starting point for an important
public conversation. It frames numerous issues for discussion and commences a rulemaking
process that we are compelled, by law and by the marketplace, to launch.
Unfortunately, this NPRM does not ask all of the questions that need to be asked. While
I appreciate the willingness to include some of my suggestions in the NPRM, I am disappointed
that this item does not include other critical questions that I believe we must pose. I will
therefore ask them here, in the hope that parties will respond to them in the record.
First, the forward auction as proposed stops as soon as it generates proceeds sufficient to
pay successful bidders in the reverse auction, cover the Commission’s administrative costs, and
cover the estimated costs of reimbursements required by the statute.1 This is essentially like
ending a traditional auction as soon as the reserve price is met. In other words, the NPRM
envisions an auction with no net revenues. This means no money for the First Responder
Network Authority (FirstNet) to build out a nationwide, interoperable public safety broadband
network;2 no money for state and local first responders;3 no money for public safety research;4
no money for deficit reduction;5 and no money for next-generation 911 implementation.6
Congress mentioned each of these items in the Spectrum Act, which makes it difficult to square
that legislation with an auction that yields no net revenues. I thus ask stakeholders how we can
amend the auction design and incorporate closing conditions that might maximize the net
revenues raised by the auction. For example, traditional auction theory suggests that an auction
should remain open until no bidders come forward. This is because ending an auction while
there are still willing bidders may allocate spectrum to bidders that value it less and leave
revenues on the table. How can we design the auction so that it does not close prematurely?
Second, the Spectrum Act provides that non-participating television broadcasters and
cable operators must be compensated for costs incurred during the repacking process.7 At the
same time, it limits the Broadcaster Relocation Fund to $1.75 billion.8 The NPRM
acknowledges that this fund may be insufficient to compensate broadcasters and cable operators
for their reasonably incurred costs,9 but it does not seek comment on how this limitation affects
when we close the auction. So I will ask: Does the Spectrum Act require us to include this
limitation as a closing condition? If not, should we require that the estimated costs of repacking
television broadcasters and cable operators be less than that sum as a closing condition? Either
way, what impact, if any, would such a condition have on the efficiency of the broadcast
Third, I hope stakeholders will help us flesh out the appropriate size of guard bands. I
agree with the NPRM that six megahertz guard bands are sufficient to shield against interference
between television broadcasting and the Long Term Evolution (LTE)-based services likely to be
offered using the newly licensed spectrum. But it also may be more than sufficient given the
power and out-of-band emission limits we place on this newly licensed spectrum, not to mention
the availability of improved filters. Because the Spectrum Act directs that “guard bands shall be
1 NPRM at paras. 67–69.
2 Spectrum Act § 6413(b)(1), (3).
3 Spectrum Act § 6413(b)(2).
4 Spectrum Act § 6413(b)(4), (7).
5 Spectrum Act § 6413(b)(5).
6 Spectrum Act § 6413(b)(6).
7 Spectrum Act § 6403(b)(4).
8 47 U.S.C. § 309(j)(8)(G)(iii)(I).
9 NPRM at para. 345.
no larger than is technically reasonable to prevent harmful interference between licensed services
outside the guard bands,”10 I think we must ask the obvious question of whether a smaller guard
band would be “technically reasonable.” For example, would a five megahertz or four
megahertz guard band suffice? And if so, would adopting six megahertz guard bands run afoul
of the Spectrum Act?
Fourth, the NPRM assumes that we need not license and auction the guard bands,11 but I
am not sure this is consistent with the Spectrum Act. At several points, the Spectrum Act
appears to contemplate that all reallocated spectrum will be licensed and auctioned. Our general
incentive auction authority, for example, only lets the Commission “encourage a licensee to
relinquish voluntarily some or all of its licensed spectrum usage rights in order to permit the
assignment of new initial licenses.”12 Our broadcast incentive auction authority is a subset of
that authority and allows us to share proceeds with broadcast licensees “in order to make
spectrum available for assignment through a system of competitive bidding.”13 Similarly, the
Spectrum Act requires a forward auction in which “the Commission assigns licenses for the use
of the spectrum that the Commission reallocates.”14 Do these provisions require the Commission
to license and auction all spectrum reallocated from the television broadcasting service,
including guard band spectrum? What would the value of the guard band spectrum be if licensed
On the other hand, the Spectrum Act also states that we “may permit the use of such
guard bands for unlicensed use.”16 Is that provision best read as allowing the Commission to
authorize unlicensed use of guard bands instead of licensed use (as the NPRM seems to assume)?
Or is there another way to reconcile these provisions, perhaps by licensing and auctioning the
guard bands but allowing unlicensed use therein on a non-interfering basis?17 Could we license
the guard bands and auction it to a band manager for unlicensed use?18 If we must comply with
10 Spectrum Act § 6407(a).
11 To be clear, I am pleased that the NPRM seeks comment on other approaches to guard band spectrum, including
whether we should license and auction it. But my primary concern is fulfilling Congressional intent, and thus I
believe questions addressing our legal authority under the Spectrum Act are necessary.
12 47 U.S.C. § 309(j)(8)(G) (emphasis added).
13 Spectrum Act § 6403(a)(1).
14 Spectrum Act § 6403(c).
15 In Auction 33, six megahertz of non-contiguous guard band spectrum raised more than $500 million at auction.
See FCC, Auction 33, Upper 700 MHz Guard Bands, http://go.usa.gov/YatQ" title="http://go.usa.gov/YatQ">http://go.usa.gov/YatQ. Here, the Commission would likely
have two six megahertz guard bands available for auction, which could also be paired to create a 6 + 6 pairing in
each market. What value would bidders likely place on these guard bands, paired or unpaired?
16 Spectrum Act § 6407(c).
17 Notably, in discussing guard bands and unlicensed use, the Spectrum Act specifically requires that unlicensed
service use a database to ensure non-interference with licensed services. Spectrum Act §§ 6407(d), (e).
18 Cf. Spectrum Policy Task Force Report, ET Docket 02-135, at 63 (Nov. 2002) (suggesting that the FCC make
unlicensed spectrum available through a band manager or frequency coordinator).
a license-and-auction requirement, are there other means to encourage flexible unlicensed use of
the guard bands?19
Fifth, I hope stakeholders will help us explore alternative approaches to channel sharing.
For example, should we require all parties to a channel-sharing arrangement to file pre-auction
applications and participate in the auction? This would allow each party to submit its bids for
sharing, which may reduce the need for pre-auction negotiations on the precise value of sharing
for each party (negotiations that may be difficult under the Commission’s normal prohibited-
communications rule). And since reverse-auction participants could be expected to incorporate
any relocation costs into their bids, it may increase the transparency of the repacking process by
allowing the Commission to better understand the total cost of a particular channel-sharing
arrangement. On the other hand, requiring both parties to participate (with potentially differing
bids) may make a descending clock auction more difficult to implement. What are the costs and
benefits of this option? Are there any other issues that we should consider regarding channel
sharing agreements that may affect who should apply to participate in the reverse auction?
Similarly, I wish the NPRM had raised the question of whether we might encourage
channel sharing by low-power broadcast stations in markets where they are likely to be
displaced. Although such sharing may not be an option if, for example, two low-power
broadcasters operate in different geographic areas, it seems like a worthy question to ask.
Sixth, I would have preferred that the NPRM more directly address the unique situation
that translators face. These low-power television stations convey broadcasts to rural
communities that would otherwise be unserved. What we can do within the constraints of the
Spectrum Act so that this vital communications link for rural America is not broken?
Seventh, I am disappointed that the NPRM does not seek input on a realistic schedule for
completing this proceeding. For example, in order to begin the incentive auction by June 30,
2014, which I believe should be our goal, by when do we need to issue final rules and auction
procedures? If we develop a timetable for completing each of the steps that must be taken prior
to the incentive auction, there is a much greater chance that we will make progress in a timely
These are just some of the questions that I wish had been asked. But I would be remiss
not to mention items I proposed that the NPRM does include, most notably questions about
relocating channel 51 broadcasters,20 bidding on specific frequencies,21 prioritizing paired
spectrum,22 and generally simplifying the incentive auction. I view all these questions as part of
a larger challenge: How do we ensure that all wireless providers—including small businesses
with limited experience in complicated auctions—have a full and fair opportunity to participate?
* * *
19 See, e.g., A Market-Based Approach to Establishing Licensing Rules: Licensed versus Unlicensed Use of
Spectrum, FCC Office of Strategic Planning Working Paper 43 (Feb. 2008).
20 See NPRM at para. 165.
21 See id. at para. 64.
22 See id. at para. 182.
In the final analysis, the questions that we pose and the proposals that we make in today’s
NPRM are important. But they are not as important as the feedback we will receive. If the
incentive auction is going to be a success, we need to listen to the input of stakeholders, and we
can’t be afraid of asking the hard questions.
While there are plenty of challenges in front of us, I remain optimistic that we can
conduct a successful incentive auction. And as we go forward in this proceeding, I believe that
there are four principles that will be critical to our success. First, we must be faithful to the
statute passed by Congress. It is our job to implement this legislation, not to rewrite it to
conform to our policy preferences. Second, we must implement the law in a manner that is fair
to all stakeholders. This is especially important because the incentive auctions will fail unless
both broadcasters and wireless carriers choose to participate. Third, we need to keep our rules as
simple as possible. The incentive auction is inherently complicated; we don’t need to introduce
unnecessary complexities. Rules that are perfect in theory may turn out to be disastrous in the
real world if market participants don’t understand them or don’t like them. And fourth, we need
to complete this proceeding in a reasonable timeframe. We haven’t conducted a major spectrum
auction since 2008, and there is an urgent need to make additional spectrum available for mobile
broadband. Prolonging uncertainty for broadcasters isn’t in their interest either. So it is my hope
that we will be able to conduct a successful incentive auction in mid-2014.
I will end as I began, with a quote from Churchill, one that sums up where we find
ourselves today. “[T]his is not the end. It is not even the beginning of the end. But it is . . . the
end of the beginning.” With this morning’s vote, we complete the first step in the rulemaking
process. This is a significant accomplishment, but the hardest work remains ahead of us. I look
forward to joining with my colleagues and the diligent Commission staff in the time to come in
pursuit of our common goal.
Note: We are currently transitioning our documents into web compatible formats for easier reading. We have done our best to supply this content to you in a presentable form, but there may be some formatting issues while we improve the technology. The original version of the document is available as a PDF, , or as plain text.