The market for new smartphones is slowing. The global market is approaching saturation, where everyone who might use one has one, and annual sales are dropping. The pace of improvements is slowing, too. The marginal attraction of new apps and more powerful and faster hardware is diminishing.
In 2018, smartphone sales numbers stopped growing, according to two data analysis companies, Strategy Analytics and Counterpoint Research. Strategy Analytics executive director Neil Mawston wrote in his guide to the latest figures that it’s the “first time ever in history the global smartphone market has declined on a full year basis. It is a landmark event”…
This was a five-percent drop over the 1.51 billion sold in 2017, and when you’re talking about billions of phones, a five-percent drop is relatively substantial.
That’s a problem for smartphone manufacturers, but hope is on the horizon. 5G networks need 5G-capable smartphones, and over the next five years that will be the primary driver of upgrades and new phone sales.
A mass market stampede toward 5G phones won’t happen until mass market 5G service is available. That build out will happen slower than mobile carriers have led city councils and county boards of supervisors to believe. And it will be far from comprehensive – the true benefits that will justify a kilobuck smartphone purchase will only be available in urban areas with high revenue potential for carriers.
The big technical question that hasn’t been answered is battery life. 5G service requires more intensive processing, which burns up energy, as do faster bit rates generally. The first units on the market won’t be optimised – can’t be until real consumers start using and abusing them in the wild – so it will be at least another year – 2021 – before manufacturers and carriers really understand power budgets. But 5G smartphones will burn through battery life faster than 4G phones, and that’s a problem yet to be solved.
It’s one thing to promise the moon to customers and city councils, but quite another to mislead Wall Street. Creating outrageous expectations there can land you in jail. Which, presumably, is why two top executives from Verizon and T-Mobile are walking back expectations of a universal 5G wonderland.
According to a story by Sean Hollister in The Verge, it’s about the new frequency bands that mobile companies plan to use for high speed, low latency 5G service. Those bands are way up the spectrum chart, in the millimeter wave range, where data capacity is high but range and penetrating power is low. So to make it work, mobile carriers need to build a lot of small cell sites. Which is expensive and only pencils out where revenue potential is equally high…
“We all need to remind ourselves this is not a coverage spectrum,” Verizon CEO Hans Vestberg told analysts on the company’s Q1 2019 earnings call on Tuesday — just one day after T-Mobile CTO Neville Ray decried Verizon’s 5G rollout as one that would “never reach rural America.”
“Millimeter wave (mmWave) spectrum has great potential in terms of speed and capacity, but it doesn’t travel far from the cell site and doesn’t penetrate materials at all. It will never materially scale beyond small pockets of 5G hotspots in dense urban environments,” Ray wrote.
5G technology can be used on any frequency band, and over time – decades, likely – it’ll replace 4G and older equipment. And there are plans to use it on a few lower frequencies with less data capacity and greater reach in the near term. But without network densification – lots of short range small cell sites – 5G will just be a tech upgrade, and not a quantum leap into a hyper connected world.
It’s a tech upgrade that will bring significant benefits, as did upgrades from 2G to 3G, and 3G to 4G, but it will take a long time for rural and suburban California to notice the difference.
The question of whether mobile broadband will replace wireline service reared its ignorant head again at a California Public Utilities Commission broadband discussion in Sacramento last week. Citing his wife’s preference for a mobile phone, CPUC president Michael Picker questioned the idea that “broadband to the home” is a good way of getting service to under and unserved communities, via the state’s primary broadband infrastructure subsidy program, the California Advanced Services Fund.
The panel’s best response came from Ana Maria Johnson, a program manager with the CPUC’s public advocates office. Echoing findings already made by the CPUC as well as the Federal Communications Commission, she said you need both…
A wireline broadband connection is essential. It’s not a substitute for your mobile broadband. When you have a wireline connection coming to your home you set up a wireless router where you can use your different devices – your laptop, you have your your desktop, you connect all the devices that you need. Students, in doing homework, need that wireline connection. They’re using a Chromebook, or they’re using a tablet, but they’re connecting through their wireless router on the wireline connection, because of the speed and capacity that they need to be able to do that work. Your mobile phone is essential as well, but it’s a complement to your wireline connection. So I don’t think it’s one or the other, but we know that the wireline connection is critical.
The idea that wireless networks can, via the magic of 5G and elsewise, provide all the capacity residential users need is a favorite talking point of mobile carriers, and particularly AT&T, which wants to rip out its rural copper networks. Depending on which stats you look at, in-home wireline data use is one or two orders of magnitude greater than mobile data consumption, and both continue to grow. Mobile carriers are pushing, and investing, as fast as they can just to keep pace with the demands of smartphones and other devices that can’t be reached any other way.
T-Mobile’s closing case for the California Public Utilities Commission review of its merger with Sprint boils down to trust us, it’ll be glorious. Opponents, led by the CPUC’s Public Advocates Office (PAO), say you gotta be kidding. T-Mobile (and Sprint and the California Emerging Technology Fund, but T-Mobile is the lead dog in that pack) filed final arguments on Friday, saying the CPUC should approve the merger. The PAO, the Communications Workers of America (CWA), TURN and the Greenlining Institute urged commissioners to deny it, because consumer prices will rise and rural communities will be left out, among other ills. Links to the “reply briefs” are below.
The PAO argues that any benefits to the public – as opposed to rents extracted by special interest groups – “will occur without the merger, if at all”. Its rebuttal dismisses T-Mobile’s promises as “vague” and “not specific, measurable, verifiable, and enforceable”.
Just so. In Friday’s filing, T-Mobile included a long list of what it calls “commitments”, but what a careful reading shows to be mostly meaningless fluff.
For example, it promises to offer lifeline service “indefinitely”, which sounds like “forever” but could also mean “for an unspecified period of time”. High sounding goals are guarded by weasel words like “strive to” and “good faith efforts”. Any firm promises – for example pricing, wholesale terms and data caps – are limited to no more than three years, which is something like the minimum amount of time required to fully integrate the two companies.
The grand network buildout T-Mobile first promised will be limited to “90% of the cell site locations” identified in the plan it submitted earlier as evidence of its good intentions. As CWA points out, the original pledge was for a 99% build out, but the California Emerging Technology Fund helpfully negotiated it down to 90%.
And 90% of cell site locations is not the same as 90% of California’s land area. Given the far higher density of cell sites in cities and suburbs, it’s not hard to guess where the unfortunate 10% will be. Roughly 95% of California’s population lives on 5% of the land and cell site deployment correlates with population density. So eliminating the least profitable 10% of cell sites from T-Mobile’s aspirational powerpoint presentation could leave the majority of rural Californians in the dark.
By omission, T-Mobile’s brief confirms (if confirmation was needed) that it won’t offer the full benefits of 5G service to rural Californians. It talks a lot about using low and mid-band spectrum in rural communities, but not the high capacity, high frequency millimeter wave bands or high density, low latency deployments that will be used to improve service in urban areas where affluent customers are thickest.
It also tries to rebut a key point made earlier by an economist working for the PAO, Lee Selwyn, who said that T-Mobile’s promises of a rural 5G renaissance were bogus because even its mid-band coverage claims were based on unrealistic assumptions about how many new cell sites would be built. T-Mobile offered a quibbling response about coverage patterns, but didn’t address the core economic question of whether there’s enough revenue potential in, say, Kings County to justify building a significant number of new towers or small cell facilities.
Selwyn also leaned on conventional anti-trust analysis to show that reducing the mobile market from four carriers to three will reduce competition and result in increased prices. T-Mobile hit back at that conclusion by touting the supposedly superior (and certainly more creative) economic model produced by three economists it hired.
Who to believe? If you think the new, merged company would, as T-Mobile plausibly claims, have lower marginal – e.g. operating – costs, then the question becomes whether those savings would be used to lower consumer prices or increase corporate profits. The answer depends in large part on whether T-Mobile, AT&T and Verizon reach a comfortable pricing equilibrium instead of fighting a bloody price slashing war. Profitable equilibriums are even easier to find in a three player oligopoly than in a more heated four player market. As the PAO’s rebuttal puts it, “AT&T and Verizon already have substantially higher prices” than T-Mobile. That’s despite having lower marginal costs. Without Sprint nipping at its heels, T-Mobile can join that club too.
T-Mobile’s rebuttal also goes on at great length about its plan to offer in-home service via its 5G network, but doesn’t explain where the necessary capacity will come from, or why it would want to sell 5G bandwidth at residential prices when there’s more money to be made filling the booming demand for mobile data. The fundamental business case for 5G deployment is based on mobile revenue streams. In-home data consumption is a couple orders of magnitude greater than mobile usage, so it’s hard to see how residential service will be a mainstream offering, rather than a tactic to offload temporarily surplus capacity. Wall Street analysts are not buying residential 5G pitches, not least because what’s known publicly about Verizon’s Sacramento experiment is not encouraging.
Some of T-Mobile’s rebuttal focuses on legal issues, particularly its claim that the CPUC has no business reviewing a merger between two mobile carriers because, among other things, it says that’s the Federal Communications Commission’s job. T-Mobile’s objections will set the stage for court challenges to any adverse decision the CPUC might reach, which could kill any conditions or restrictions the CPUC might impose.
There are a few procedural loose ends to tie up, but the substantive evidence and lawyerly pleadings are in the record. The next step is for the CPUC administrative law judge managing the review, Karl Bemesderfer, to draft a proposed decision for the commission to consider. If he does that in the next couple of weeks, commissioners could vote on it as soon as 27 June 2019. It’s a complicated case and there are other potential bumps in the road, though. T-Mobile and Sprint’s agreement to extend their self imposed deadline to the end of July was a wise move.
“Reply briefs” regarding T-Mobile’s acquisition of Sprint, filed with the CPUC on 10 May 2019:
(Technically, two CPUC reviews are underway. One concerns Sprint’s wireline operations in California, and the other involves mobile services. The two reviews were combined into a single proceeding, but T-Mobile is trying to split them up again. The wireline transfer is relatively uncontroversial, but is squarely within the CPUC’s jurisdiction; the mobile merger is hotly contested, but the CPUC’s authority is less certain. It would benefit T-Mobile if the two issues were handled separately).
Instead, CETF and T-Mobile (and technically Sprint, but it’s T-Mobile that’s running the show) negotiated their deal and then submitted it to Bemesderfer, along with a request from CETF to be allowed to change sides in the case and “enthusiastically and wholeheartedly support” the merger. He said that’s how it’s been done at times in the past, so the agreement can be used to support T-Mobile’s push for CPUC approval of the Sprint merger, but that’s all…
However, all parties should be aware that granting the Motion merely permits CETF and Joint Applicants [T-Mobile and Sprint] to enter their MOU into the record of this proceeding and changes the litigation position of CETF from opposing the Sprint-T-Mobile merger to supporting it. Granting the motion does not pre-judge the question of whether the merger is in the public interest though it adds weight to the argument of Joint Applicants for that conclusion.
Bemesderfer is yet to rule on a flurry of motions filed in the past week. TURN and Greenlining either want the CETF deal excluded from consideration, or be given more time to offer a rebuttal. No decision yet on that request, but Bemesderfer’s latest ruling might be read as an indication of where it’s heading. Also pending are a motion by the PAO to exclude T-Mobile’s not very detailed offer of 1,000 new jobs at a Central Valley call center if the merger is completed, and another request by Sprint and T-Mobile for the CPUC’s immediate approval.
At this point, the only immediate action to expect is another weekend with plenty to read – rebuttal arguments from all sides are due later today. Assuming all goes to plan, I’ll post an update on that on Monday.
The California Public Utilities Commission’s review of the deal could run until then. Its public advocates office (PAO) and two consumer groups – TURN and the Greenlining Institute – are pushing hard to kill it. Last week, the PAO asked the administrative law judge managing the case to exclude a vague new offer of Californian jobs made by T-Mobile in a meeting with a commissioner and in a recent filing.
In its opening brief, and an ex parte meeting disclosure filed the same day, T-Mobile claimed it will “build a new customer experience center in the Central Valley that alone will create approximately 1,000 new jobs in the state”. T-Mobile neglected to mention this incredible benefit in the thousands of pages of evidence and arguments it previously entered into the record, or during hours of cross examination by opponents. The purpose of all that give and take is to flesh out sketchy statements, and get specific and enforceable promises into the record. Without that opportunity, the commission won’t have “a firm understanding of [the offer’s] nature and truthfulness”, the PAO argues.
CETF and T-Mobile are also taking hits on procedural grounds.
When an organisation opposes, say, a proposed merger, but then negotiates a deal and stops objecting to it, there is a specific settlement process that the CPUC lays out for incorporating it into the final decision in the case. That process isn’t always used. As T-Mobile and CETF point out, sometimes the commission will give its blessing to an agreement that settles a particular dispute, without formally calling it a “settlement”. That happened during the CPUC’s review of Frontier Communications’ purchase of Verizon’s wireline systems in 2015 and Charter Communications’ acquisition of Time Warner and Brighthouse cable systems in 2016.
T-Mobile’s takeover of Sprint is different. No one took great exception to the various agreements with Charter or Frontier, but the PAO and the two consumer groups are sharply challenging the substance of T-Mobile’s deal with CETF. On Friday, TURN and Greenlining asked for more time to do that.
I helped put together the City of Gonzales’ settlement with Charter in 2016. We didn’t follow the formal process when we presented it to the CPUC. We weren’t out to get a megabuck payday either. Take it for what it’s worth.
California is either heading for a proletarian broadband paradise or an economic meltdown of Venezuelan proportions. Following months of testimony, document dumps and stupid lawyer tricks, on Friday the companies and their opponents laid out arguments for why the California Public Utilities Commission should or shouldn’t approve the deal.
In two separate filings, T-Mobile (and technically Sprint, but it’s T-Mobile that’s running the show) mostly reiterated the same points and pleadings they’ve been pushing since the beginning: the CPUC is sticking its nose where it doesn’t belong and the merger will benefit everyone – Californian consumers, rural communities, low income and disadvantaged people, job seekers, employees hoping to keep their jobs, and the list goes on.
Over the course of a couple hundred pages, T-Mobile repeated claims made in earlier filings and testimony about how the new, combined company would lower prices, increase speeds, widen coverage and add jobs. Much of it is argumentative, and none of it comes with an enforceable pledge to make good on those promises.
Instead, T-Mobile mostly relies on a hyperbolic salad of marketing superlatives: revolutionary opportunities, incredibly successful, especially striking, fundamentally transform, world-renowned experts, unique pioneering. Much of it involves claimed benefits, such as Internet of things support and low income lifeline services, that are, or should be, course of business for any mobile carrier, merged or not.
The primary opponent to the deal is the CPUC’s public advocates office (PAO). It concludes that the merger is not in the public interest and “will lead to higher prices, reduced capital expenditures in California, stifled innovation, poorer service quality, reduced rural coverage, elimination of low-income plans…and deteriorated consumer privacy”.
The PAO’s sharpest points comes from an economist, Lee Selwyn, it hired to review the merger and, particularly, some of T-Mobile’s microeconomic claims. He makes two important points.
First, contrary to the fragile and convoluted case offered by T-Mobile’s world renowned experts, fundamental anti-trust analysis supports the otherwise obvious conclusion that reducing the mobile telecoms market from four carriers to three means less competition and, ultimately, higher costs and fewer benefits to consumers. His well supported observation that T-Mobile and Sprint mostly compete against each other, rather than against AT&T and Verizon, indicates that the damage done to the market will be that much greater.
Second, even if the especially striking coverage and revolutionary opportunities that T-Mobile promises materialise, the benefits will be limited to densely populated urban areas. Using T-Mobile’s own figures, Selwyn adeptly and quantitatively dismantles visions of a 5G mobile renaissance in rural communities, concluding…
[T-Mobile’s and Sprint’s] claims that the merger will bring coverage to rural areas – and their attempt to buttress such claims with maps that purport to display projected coverage areas at the county level – cannot be squared with the projected capital investments that a merged New T-Mobile anticipates making in each California county through 2024…Rural areas are not served because they are costly to serve, and this fundamental economic reality is not materially changed by the merger. The maps and coverage area projections advanced by T-Mobile in its rebuttal testimony are not credible and should be afforded no weight by the Commission.
Another problem with T-Mobile’s merger case, which gets less attention from opponents than it deserves, is the disconnect between its broadband capacity and speed promises, and the frequencies it plans to use. T-Mobile has less “millimeter wave” spectrum than AT&T and Verizon, which, as it admits in its own filing, will be “largely limited to densely populated urban areas”. Millimeter wave bands – 30 GHz or so and above – are what make many 5G improvements possible.
Everyone involved has two weeks to file rebuttals. In the meantime, some side skirmishes need to be resolved, including a demand by the Communications Workers of America, which opposes the deal, to delete information about contract negotiations from the record, as well as objections to the $35 million payoff to the California Emerging Technology Fund’s (CETF), which led it to switch sides and “enthusiastically and wholeheartedly support” the merger. CETF did not submit arguments for or against it on Friday.
“Opening briefs” regarding T-Mobile’s acquisition of Sprint, filed with the CPUC by 26 April 2019:
One issue in dispute is whether it is a formal settlement, which has to be negotiated and reviewed under CPUC rules, or something else. Which is what T-Mobile and CETF seem to think it is, because they didn’t follow those rules, according to the filings.
But the substance of the deal also came under fire. The objections noted, as I did last week, that T-Mobile’s promises of good behavior and grand public benefits were either recycled (in a somewhat melted form) from earlier statements or were so vague and subject to T-Mobile’s discretion as to be no promise at all.
The single significant new commitment in the agreement was $35 million, to be paid to CETF over five years by T-Mobile. The money is supposed to go towards what the contract calls “digital inclusion policy and programs”, with $22 million earmarked for various non-profits and public agencies, in a manner to be determined by CETF and T-Mobile. CETF keeps the remaining $13 million to spend on its ongoing operations.
The PAO asked the commission to reject the deal. Noting that CETF “receives a disproportional amount of funding” that “exceeds any commission approved operating costs percentage”, the PAO said it…
…has determined that the agreement is not in the public interest or reasonable on its face…
The Agreement requires New T-Mobile to provide $35 million over 5 years to CETF’s “Digital Inclusion Policy and Programs” projects without any basis in the record to evaluate, verify, and monitor these programs to ensure that the amount of $35 million is appropriate. While the Public Advocates Office strongly supports efforts to close the digital divide, as described above, additional hearings are necessary to investigate these proposals. The record does not sufficiently describe what these programs do, the amount of money necessary to properly fund them, who operates them, or any other details about them.
CETF and T-Mobile have ten days to respond. One possible outcome is that the administrative law judge managing the CPUC’s review could order new hearings to delve into the details of the agreement. That has the potential to further delay an inquiry that has been extended by at least a couple of months because of earlier cheap lawyer tricks by T-Mobile.
It looks like 2020 will be the year that genuine 5G smartphones will finally be in the hands of consumers. Two developments this week cleared away significant uncertainty about who will be offering 5G phones, when it will happen and whose technology they’ll use.
The two companies settled a long running legal dispute over intellectual property rights to core 5G technology, including a deal for Apple to buy modem chips, which do the heavy processing work of wrangling radio waves into data streams at one end and reading them at the other.
The second announcement came shortly afterwards. Intel said it’s giving up its quest to build competing modem chips and leaving that market segment to Qualcomm. Not the entire market, though. There are a lot more kinds of chips that go into smartphones, 5G and otherwise, and Intel still plans to make them.
One of the benefits, if you want to call it that, of a monopoly is faster standardisation. Which reduces supply chain uncertainty for manufacturers and simplifies technical challenges for carriers, increasing the odds that predictions of mass market 5G product and service availability by the end of 2019 will come true.
Those early handsets won’t be made by Apple. Major Android phone makers are pushing to have 5G products in the market for this year’s Christmas selling season, but Apple didn’t make the same promise. Now, it can’t. Apple won’t be able to design and tool up to make Qualcomm-based iPhones until 2020, perhaps not until the second half of the year.
But there’s finally a clear roadmap for all major smartphone makers to make the jump soon enough to begin building a meaningful 5G user base in 2020. Mobile carriers will be judged on the basis of how well they deliver on the hype and the deceptions they’ve relied on so far. We’ll finally know what 5G really means.
The odds of T-Mobile getting permission from federal and California regulators to buy Sprint are getting longer. The Wall Street Journal is reporting that the federal justice department is reluctant to approve the deal in its current form. That has a familiar ring to it – it was the same kind of antitrust concerns that led to the justice department and Federal Communications Commission killing Comcast’s bid to take over Time Warner’s cable systems and do market consolidating swaps with Charter in 2015.
T-Mobile seems to be trying to pick up the pieces in California. Its lawyers filed a notice yesterday saying that company representatives will meet with California Public Utilities Commission commissioner Martha Guzman Aceves next week. They didn’t say what they planned to talk about, but it’s not much of a reach to suppose they’ll try to divert attention away from the microeconomic, antitrust harm the deal will do to all Californians, and towards the special benefits that a few have managed to extract for themselves.
The Journal’s story also kicked off a new round of damage control by the companies and speculation on what a deal that would satisfy anti-trust concerns would look like. A story in Investor’s Business Daily speculated that some kind of hybrid wholesale model, where both companies retail service via a consolidated network, might fly.
The CEOs of T-Mobile and Sprint jumped on Twitter to make what amount to non-denials.
John Legere, T-Mobile’s chief, issued a tightly spun response in which he objected to “the premise of this story, as summarised in the first paragraph”. Translation: the facts reported in paragraph two, three, four and more are true. Marcelo Claure, CEO of Sprint, simply said the article “is not accurate”. As in, I wouldn’t have put it quite that way.