M2Z Networks recently filed a study I prepared for them in the AWS-3 service rules proceeding (07-195) before the FCC.
In this study I identified how a coordinated effort between T-Mobile and the Rural Telecommunications Group threatened to wreck the AWS-3 auction by writing rules excluding technology proposed by key potential new entrants, including M2Z Networks, and adopting disastrous combinatorial bidding rules like those which provided a nearly half-billion dollar windfall for Verizon in the
In brief, T-Mobile has proposed a “bandwidth maximization plan,” first mooted in this filing and elaborated here. The T-Mobile plan would split the J Block in half, giving 5 MHz for uplink and joining the other 5 MHz of J Block with 20 MHz of AWS-3 spectrum for downlink. This would force abandonment of the Time Division Duplex (TDD) technology envisioned by the FNPRM in favor of Frequency Division Duplex (FDD) technology favored by T-Mobile.
That might seem innocuous enough at first glance, but it eliminates consideration of a technology which is both more efficient and more robust than T-Mobile’s FDD alternative, and it is never a good idea to throttle new technologies at the bidding of vested incumbents. However, it is more pernicious still in that it aims at excluding the TDD technology on which Sprint, Intel, Arraycom, and M2Z proposed to build a nationwide network, effectively erecting entry barriers to major competitors to T-Mobile.
The irony is that T-Mobile proposes to kill TDD technology in AWS-3 on the pretext of preventing interference between AWS-3 and AWS-1 spectrum (T-Mobile was a major acquirer of AWS-1 spectrum). However, the FCC’s Office of Engineering and Technology conducted extensive testing and found that such interference presented no significant problem. T-Mobile’s justification for the technologically-discriminatory erection of this entry barrier is, thus, a lie.
But it gets worse.
More below…
On December 2, 2008 the Rural Telecommunications Group (RTG) had its general counsel conduct a conference call with Commissioner Adelstein’s staff (1) to endorse the T-Mobile plan, (2) recommend a plan which “would allow a combination of blocks of spectrum to be aggregated in an auction provided certain safeguards are in place for the bidder to opt out of a block if the bidder did not successfully aggregate the needed spectrum”, and (3) recommend auctioning the majority of the AWS-3 spectrum in CMA-sized units.
The RTG later claimed that it had not reversed its earlier opposition to combinatorial in this filing, but it is difficult to give that claim much credence, since any rule system which “would allow a combination of blocks of spectrum to be aggregated in an auction provided certain safeguards are in place for the bidder to opt out of a block if the bidder did not successfully aggregate the needed spectrum” by definition involves combinatorial bidding.
Combinatorial bidding was a catastrophe in Auction 73 (700 MHz Band). The intersection of the combinatorial bidding, activity, dropped bid, eligibility, and minimum acceptable bid rules allowed Verizon, when the 50 states package was broken, to obtain the two most populous and lucrative REAGs in C Block for nearly a half-billion dollars less than other bidders were willing to pay earlier in the auction. The FCC would have to be daft to embrace combinatorial bidding after it let the U.S. Treasury take that kind of a hit, particularly under current financial circumstances.
RTG advocates combinatorial bidding and allocation of the AWS-3 spectrum in CMA-sized units supposedly because such rules would advantage smaller rural providers in overcoming the broadband deficit which rural America suffers. This almost completely misunderstands the problem of rural deployment in the U.S.
There is significant evidence that previous broadband auctions – the Lower 700 MHz Band auction (44, 49, 60), the AWS-1 auction (66), and the Upper 700 MHz Band auction (73) – have not only offered ample spectrum at the CMA and EA levels, but that rural bidders have been remarkably successful in obtaining broadband spectrum. The majority of rural bidders bid on a single license while the number of rural bidders bidding on more than five CMAs in any auction is, on average, vanishingly small. Furthermore, rural bidders are seldom challenged by major incumbent providers except in those areas where predominantly rural CMAs are directly contiguous to major urban areas.
The problem facing rural customers is not availability of quality spectrum or opportunity for rural providers to participate in broadband spectrum auctions. The problem is that the large incumbents with the resources to effectively deploy and provide relatively high quality of service tend to avoid predominantly rural areas because they are notoriously less lucrative as profit centers.
The 2000 census reports that 79.22% of the U.S. population lives in urban areas, while only 20.78% live in rural areas. Examination of winning bidders by whether the CMA/EA/REAG is predominantly urban or rural shows that major incumbents have concentrated 96.24% of their bidding activity on urban areas. This is to be expected, since deployment in equally-sized urban and rural geographic areas will provide, ceteris paribus, almost 2.81 times more potential subscribers in urban areas than in rural for the same deployment cost.
What governs the asymmetry in urban/rural deployment rates and quality of service is economic value. That is not to say that rural providers neither deploy nor attempt to provide quality service, but they are at a huge disadvantage because they are incapable of taking advantage of economies of scale and complementarities available to the major incumbents precisely because they are comparatively undercapitalized and operate in areas which perforce generate revenue streams significantly inferior to those available from the areas in which major incumbents predominantly deploy.
RTG’s proposal to auction a significant portion of AWS-3 licenses at the CMA level simply dooms existing rural providers and similarly sized and capitalized new entrants to replicating their failure to provide high quality broadband service to rural America.
Conversely, offering a single nationwide license at auction for AWS-3 (1) with strict build-out requirements to ensure that rural areas are not systematically slighted as they have been hitherto and (2) with a sufficiently low reserve price to provide incentives for major incumbents to bid (in effect compensating them for the lower expected revenue stream from rural deployment) offers the best opportunity to attract either major incumbents with the resources to deploy in rural areas or new entrants sufficiently well capitalized to do so. The current policy of primarily allocating rural broadband spectrum to providers who are structurally and financially incapable of deploying high quality service in parcels too small to permit economically efficient enjoyment of economies of scale and synergistic complementarities has failed.
It is time for the FCC to try something new.
excellent demonstration of the complex interplay among policy goals of rural broadband, auction design and economic efficiency
under (unlikely) ideal conditions, a “broken package” would indicate the absence of scale-scope economies and declining cost across the entire package – otherwise, at least one bidder capable of serving the full package would value it more than any sum of individual bids
which is what a single license would emulate, given serious problems of flawed auction design and strategic subversion of the combinatorial bidding process