Having suffered from a famine in the valuable midband 5G spectrum range, Verizon and AT&T responded like starving animals and paid record-breaking sums for C-band airwaves in the FCC’s recently concluded Auction 109. But that is not the end of the story – hard on the heels of the CBRS and C-band sales, the FCC is planning to kick off Auction 110 in October, offering another 100 MHz of spectrum in the 3.45 GHz band.
With T-Mobile USA, courtesy of its Sprint acquisition, sitting on large quantities of 2.5 GHz spectrum – which provides an even more ideal 5G balance of coverage and capacity – it is possible that there will be greater restraint in bidding in the new auction. The FCC has set a proposed reserve price of $14.7bn, similar to the reserve price in the C-band process, but way short of the actual amounts bid for those airwaves (almost $81bn).
It seems likely that the big three will be less active in the 3.45 GHz auction than other operators that sat out the C-band sale, or retired from it when prices went too high. But there are plenty of alternative mobile players in the newly diverse 4G/5G market in the USA, including new mobile entrant Dish Network; the large cablecos such as the Comcast/Charter partnership, or Cox; Tier 2 MNOs like US Cellular; and potentially some other players focused on enterprise, wholesale or cloud models, such as AWS.
The level of participation will depend partly on the final rules that are imposed. The allocation of this spectrum has not been as fraught (yet) as that of the C-band, whose terms for relocating the incumbent satellite operators were very hard-won. But it has still provoked plenty of disputes about how it should be allocated and used.
The FCC has published its proposals and is now seeking comment on them. Its draft plan would divide the 100 MHz of spectrum into 10 MHz blocks and allocate them based on Partial Economic Areas (PEAs), which would equate to 4,060 flexible licences nationwide. As with the C-band, there will be two phases in the auction – a bidding stage, followed by an assignment phase in which precise frequencies and locations are allocated to winning bidders.
Acting FCC chair Jessica Rosenworcel said the rules would help to ensure that licences were winnable by a variety of players, by contrast with the C-band result, which did nothing to change the balance of power in the market as the vast majority of licences went to AT&T and Verizon. “Today’s decision commits to the idea that successful auctions have many bidders,” she said, particularly as there will be a 40 MHz cap on potential spectrum gains.
However, while these PEAs are smaller than the licence areas adopted for the C-band (which had about 400 geographies), some newer entrants to 5G wanted even smaller areas, closer to those used for the auction of the CBRS licences.
This used county-sized licences, with over 22,600 available, more than five times the numbers suggested for 3.45 GHz. That enabled smaller players to afford to bid and made it harder for a single operator to gain a dominant position. Importantly, the very localized CBRS licences are well-suited to operators that want to deploy 4G or 5G for specific locations such as for a smart city or an industrial complex, without having to pay for spectrum for a far wider area. This was seen as an important way to stimulate the adoption of 5G for enterprise networks and to support services that were not central to the priorities of a national MNO.
A group of cablecos – led by the NCTA (National Cable and Television Association) and by major cablecos Charter, Comcast and Cox – has joined with Google and some other industry players to lobby Congress to apply the rules adopted for CBRS to the 3.45 GHz band also. The group is concerned that the FCC’s draft rules will limit participation.
“The large partial economic area licenses proposed would not foster the kind of robust participation and innovation that made the CBRS auction a success and could also result in tepid investment and build out in rural areas where Americans need broadband more than ever before,” the group told Congress last month.
Of course, the telcos argue that fragmented spectrum ownership compromises economies of scale and leads to slow roll-out and failed projects. 5G services will reach more people, at higher quality and in shorter order, if most spectrum is held by national players, they say. Dish Network will probably be the operator with the biggest interest in acquiring 3.45 GHz spectrum on a near-national basis, since it ended up affording only one C-band licence; T-Mobile USA, despite its significant 2.5 GHz holdings, may also decide to bid in this sale, to improve its spectrum advantage still further (especially to deliver its promised converged TV offering), and perhaps to keep airwaves out of the hands of potential challengers such as cablecos.
Analysts at New Street Research wrote in a client note in February that TMO and Dish should take full advantage of the new auction, saying that the 3.45 GHz spectrum is “almost as important as the C-band … It’s the same spectrum, available sooner” (because it does not have such a complex refarming challenge as C-band, which is occupied by satellite providers, while 3.45 GHz is used by naval radar).
However, there will still be costs associated with helping the incumbent federal users to clear the band – the NTIA has estimated these costs will be more than $13.4bn. Those won’t be borne by the spectrum winners, unlike some of the C-band clearance costs, but that sum effectively means the auction will have a total reserve price of about $15bn.
Dish has also been petitioning the FCC to change its proposed rules, though in different ways from those demanded by the cablecos. And it is going further, also asking the FCC to change the CBRS rules retrospectively.
“Dish urges the Commission to think holistically about midband spectrum in order to maximize the efficient use of these valuable airwaves,” it wrote.
Dish acquired only one licence in the C-band auction but spent $913m on 5,492 CBRS licences last year, prioritizing breadth over depth to end up with about 20 MHz across the whole country. This complemented its existing holdings in sub-GHz and midband spectrum well, but now it wants a change of rules related to power levels, to make it easier and more cost-effective to build a network in its CBRS airwaves.
It argues that current power limitations would limit CBRS build-out to small cells alone. It wrote in a filing with the FCC: “The current CBRS EIRP [radiated power] limit of 47dBm/10 MHz … is likely to thwart a wide variety of desirable use cases, effectively limiting CBRS users to small cell deployments … Dish’s technical analysis found that no party is likely to be harmed by higher power use cases in the CBRS band.”
The changes Dish wants would not only allow for larger cells and stronger signals, but would enable it to align its CBRS spectrum more easily with any airwaves it acquires in the 3.45 GHz auction – which has similar power limit rules to those of the C-band.
Dish said that keeping the CBRS band “relatively encumbered” with lower power limits leaves it “sandwiched between the 3.45 GHz band and the 3.7 GHz band, both of which would have services rules optimized for large-scale, wide-channel 5G service offerings … This is akin to connecting two cities with a new eight-lane high-speed roadway, but constructing a stretch of single lane road in the middle. The current US approach undermines the overall usefulness of the entire 3 GHz band, and places the United States at odds with our major global competitors, especially China.”
Dish added that TDD synchronization would have to be mandated across the CBRS band and the 3.45 GHz band to prevent interference.
“If the FCC were both to require TDD synchronization and increase CBRS power levels, the US could create 530 MHz of contiguous midband spectrum (3450-3980 MHz) with compatible power levels, which as explained above would position the US to compete at global scale for 5G technology and services,” Dish concluded.