T-Mobile USA is not demonstrating radical thinking in its first stage of 5G deployment. Unlike some operators, which are using 5G to enable new use cases, like Verizon’s fixed wireless; complex architectures like network slicing; or capacity hotspots to complement 4G, TMO is going for a coverage-oriented, national build-out in low frequency spectrum. It was by far the biggest buyer of 600 MHz former broadcast spectrum in last year’s incentive auction, and is already deploying 4G in this, and claiming a roadmap to 5G which will start next year.
Of course, if TMO succeeds in acquiring Sprint, it will have a far richer and more complex range of spectrum and services. Sprint’s 4G and 5G strategies are heavily focused on capacity bands like 2.5 GHz – which it has in plentiful supply – and it has always been more interested in wholesale and enterprise services models than its rival. But for now, TMO needs to prove that it can enhance the commercial success of its consumer-focused, low cost model, complete with unlimited data plans, by building out in the relatively narrow bands and limited capacity of a sub-1 GHz band.
There are two big question marks over TMO’s strategy. One, does it really need another layer of near-nationwide coverage so soon after rolling out 4G? And two, is it even getting the affordability associated with low frequency bands and their excellent propagation, given that 600 MHz is currently an unusual choice?
In many parts of the world, 700 MHz and 800 MHz remain the chief coverage bands for mobile networks. At 2015’s World Radio Conference, the band (470-694 MHz) was identified as IMT spectrum for the Americas and Asia-Pacific but so far, the US is the only country to have conducted auctions, while Canada has one scheduled for 2019. Some Latin American countries, like Mexico, and some Asia-Pacific operators have expressed interest, but they have not yet agreed regional band plans or possible allocation dates.
This means that TMO is somewhat isolated in terms of accessing the economies of scale of a major global device and equipment ecosystem, especially as it was the only one of the big four US MNOs to buy 600 MHz licences (Sprint stayed out of the auction; Verizon participated but did not buy; AT&T has sold its existing 600 MHz licences). It is not a great technical challenge to adapt 700 MHz products for this band, but it does require some effort, which will not be a vendor priority unless there are multiple customers. For LTE, the only handsets supporting TMO’s 600 MHz network are three Android models from Samsung and four from LG (the Samsung Galaxy S8 Active, S9 and S9+; and the LG G7 Thin Q, K30, V30 and V30+). The issue may become even more serious in the early years of 5G, especially if Apple shows no interest in the band.
This may be why TMO is spending more on building out 600 MHz than expected, according to analysts at Oppenheimer – and that may bring back uncomfortable memories of TMO’s 3G deployment, when it was one of the only operators to be rolling out W-CDMA in the AWS band (it has since repurposed that for 4G, accessing a far wider choice of devices, including an iPhone, in the process).
In a recent client note, Oppenheimer’s team wrote: “We spoke with TMUS management, who appeared to be focused on building out its 600 MHz spectrum for increased coverage/capacity, the correct route in our opinion, but this will require more capex and we increase our estimates.” The firm raised its estimates for TMO capex by about $200m for the second quarter of this year, to $1.55bn.
For the full year, TMO has said it will spend between $4.9bn and $5.3bn in capex on its network. It spent about $8bn on its 600 MHz licences.
Despite the challengers of being a trailblazer in a particular band, TMO is deploying at a fast rate. It has built 4G in the band in 900 cities and towns, it says, 120 of them previously not served at all by the operator. Among the major centers to get at least some coverage from 600 MHz LTE are Omaha, Nebraska; Augusta, Maine; and Los Alamos, New Mexico; as well as Puerto Rico.
The operator is using 600 MHz to extend its coverage in new markets, where it does not hold 700 MHz licences, or to improve coverage still further in 700 MHz regions. It is branding its service ‘Extended Range LTE’, claiming that the signal can travel twice as far from the tower, and support four times better in-building penetration, than midband LTE. From this month, it will start to implement carrier aggregation across 600 MHz and its midband 4G spectrum in AWS.
It also says 600 MHz will be the first spectrum it will use for 5G and earlier this year, it said Ericsson and Nokia will build a 5G network spanning 600 MHz, 28 GHz and 39 GHz in 30 cities—including New York, Los Angeles, Dallas and Las Vegas—during 2018.
That ambitious timescale for early 5G roll-out, however limited that may prove to be in the early days, begs yet another question – how TMO expects to get strong return on its 4G investment when that is happening so close to 5G migration? It has not yet detailed its 4G/5G coexistence strategies, but it is to be hoped these include a great deal of flexibility to allow both technologies to live in the same spectrum and infrastructure, and to be accessed interchangeably by different services.