We remain short shares of Straight Path Communications. As we argued earlier this month [1], Straight Path is dramatically overvalued relative to the most appropriate comparable transaction: Verizon’s potential future acquisition of XO Holdings’ millimeter-wave (mmWave) spectrum. Reaching this conclusion does not require any particular skepticism about the prospects of mmWave 5G. However, some investors – apparently taking marketing fluff at face value – insist that mmWave 5G matters so much to the future of mobile networks that carriers will eagerly pay a premium for Straight Path’s holdings, even when so much similar spectrum will soon be put up for auction.
This unquestioning attitude toward the industry’s most ambitious claims about mmWave 5G makes little sense. 5G research and standards-writing are still in their earliest phases, and no one actually knows yet whether large-scale mmWave deployments will generate enough value to attract customers and produce acceptable returns on investment. Indeed, just last week, Verizon – the carrier that has made the loudest noises about mmWave 5G, albeit with a strong emphasis on non-mobile applications – confirmed that its efforts still amount to nothing more than “precommercial pilots,” aiming to answer basic questions like “what’s the cost to roll out…how many homes can you cover from any particular node, what speeds can you get.” Far from committing to an aggressive build-out, Verizon’s CFO was noncommittal:
[W]e’ll see how those trials go. It’s the same as any investment…We’re going to look at the revenue opportunity we think we can get based off of the results of the tests that are underway right now and compare that with the cost.
Verizon’s wait-and-see approach helps explain the structure of its spectrum deal with XO: not an outright purchase but a more cautious short-term lease coupled with an option to buy. If even the most aggressive 5G promoter in terms of rhetoric is this tentative in terms of behavior, it doesn’t bode well for Straight Path’s mandatory spectrum sale.
Here, as a corrective to all the uninformed 5G hype circulating in the market, we review recent independent, third-party analyses that call into question whether mmWave 5G will ever flourish commercially. This viewpoint is no short-seller’s fiction; it’s shared by many seasoned telecom veterans. The high costs and unclear benefits of mmWave 5G will translate to slow, patchy deployments at best, tightly circumscribing Straight Path’s value. Talk is cheap, but when the time comes for carriers to open their checkbooks, reality will prevail, and the Straight Path bubble will deflate.
Q: Why is everyone talking about 5G now?
A: Because the equipment vendors want to. Honestly, that’s the whole problem with our industry — they have to be real careful. I know they are hungry for business, but the trouble is they are driving the industry into the ground. We’re barely recouping our investments, and now they’re asking us to plow more money in there for a purpose that is not yet entirely clear. It’s based on speculation about the Internet of Things, or whatever else comes along, and all this hype about speed. Please! What is the biggest growth in mobile networks? Everyone knows it’s video, but can you watch a video quicker? Do you want to fast forward to view it? No you don’t. You don’t need that much bandwidth.
—Douglas Li, former CEO of the billion-dollar Hong Kong–based carrier SmarTone, August 31, 2015
I’m concerned that the spectrum all looks pretty horrible, and I can’t see how the economics of 5G are going to look very attractive.
—Richard Feasey, telecom consultant and 25-year industry veteran, at the second Cambridge Wireless 5G debate [2], November 8, 2016
AT&T’s Vision of Ultrafast Wireless Technology May Be a Mirage
—New York Times headline [3], October 26, 2016
When we released our first report [4] on Straight Path in October 2015, 5G hype had yet to spread very widely. Today, the term “5G” is on everyone’s lips, whether during quarterly earnings calls or industry events like the annual Consumer Electronics Show. Predictably, snazzy demonstrations and utopian pronouncements have edged out critical assessments. Recently, however, one industry participant has clearly set forth the limitations and shortcomings of the prospective technology in a book entitled The 5G Myth: When Vision Decoupled from Reality [5]. The author, a telecom consultant named William Webb who previously served as head of R&D for Ofcom (the UK equivalent of the FCC) and has also been deeply involved in IoT technology in the private sector,[1] [6] makes a series of stark claims that he summarizes as follows:
5G as currently envisaged will not be realised. …
a. Demand for higher speeds is unlikely to continue to grow.
b.Technological improvement will no longer be able to deliver higher speeds without huge increases in cost.
c. Users are unwilling to pay more for their mobile communications limiting the appetite of mobile network operators (MNOs) for investment.
d. Alternative futures are more appealing, such as delivering consistent connectivity which would both be preferred by users and more likely to improve productivity than the 5G vision.
Though Webb’s argument flies in the face of conventional wisdom, his logic is sound. Video is already the main driver of cellular traffic today, but the vast majority of wireless viewing takes place in the home using Wi-Fi; there are only so many hours in the day when the typical user has the ability and willingness to consume video on the go. While high-resolution video can require high data rates, the success of T-Mobile’s Binge On and AT&T’s copycat Stream Saver, both of which convert video streams to lower resolutions in order to reduce traffic, clearly show that few consumers actually care about getting the highest possible resolution. (The human eye often can’t even discern high from very high resolution when looking at screens as small as smartphones.) Beyond the 1 to 5 megabits per second necessary to sustain a basic video stream, then, users see little value in ever higher speeds.
Yet the main appeal of mmWave 5G is supposed to be extremely high speeds enabled by very wide bandwidths. Straight Path, for instance, has labeled its small research facility the “Gigabit Mobility Lab,” an allusion to the fact that 5G will deliver data rates on the order of a gigabit per second. Yet as recent announcements [7] from T-Mobile, AT&T, and Sprint show, the existing 4G LTE standard already supports gigabit speeds, and every major US carrier will provide them in at least some markets in 2017 and beyond. If real-world users are satisfied with a few megabits per second and LTE can reach a thousand, how much additional value will exotic 5G technologies really offer? It’s no wonder T-Mobile’s CTO recently emphasized that “LTE has massive legs. This industry is going to compete around LTE for much of the next decade.”[2] [8]
But if users ascribe little value to speeds beyond those of LTE, large-scale mmWave deployments are never going to make economic sense. While mmWave propagation may vary widely from place to place (and even from time to time within a given place, affected by phenomena like rainfall and plant growth), coverage radii will be on the order of 100 meters – tiny by traditional standards. Either carriers will have to build out enormous quantities of small cells, which in turn will require ubiquitous high-capacity backhaul, or mmWave service will never be available in many places and thus can never be an important part of a carrier’s business or marketing. Why bother incurring huge capital expenditures with no realistic prospect of higher revenues?
In laying out his case that mmWave 5G as currently conceived will fizzle out, Webb anticipates and rebuts the objection that, if the technology was so unlikely to pan out, the industry wouldn’t be working on it:
The manufacturers and operators are powerful companies. … Many believe that if such large companies state that they are going to pursue a particular vision or object then this is certain to occur. History suggests otherwise. MNOs have in the past declared strong support for:
- Video calling
- Picture messaging
- Location-based services
- Femtocells
- Internet/walled-gardens/WAP
- Widgets/own-brand apps stores
- eHealth
- Mobile payment
None of these came to be provided by the MNO. Video calling was delivered over-the-top (OTT) by Skype. Picture messaging became part of Apps with increasingly easy ability to embed pictures in Tweets and emails. Location-based services were delivered by Google using data gathered by the phone and eventually GPS location. Walled garden Internet was rapidly overtaken by devices like the iPhone able to access mainstream sites. Widgets became redundant with the advent of the Apps store. MNOs proved unable to make headway into eHealth and mobile payment was eventually delivered by the banking sector and device manufacturers. Simply because the industry declares it supports a vision does not imply it will happen.
(Other examples would include abortive yet initially much heralded technologies like WiMAX and LTE Broadcast.) But Webb views 5G more cynically than past failures: “In this case…it seems likely that the MNO support is lukewarm. While they indicate a desire for 5G they will await its emergence before making decisions as to whether to invest. Major players can be wrong, but here it is more a case of them sending out stronger signals of support than they believe internally.” In other words, while past disconnects between MNO rhetoric and real-world outcomes arose from blunders, here MNOs privately understand the problems even as they publicly endorse the vision.
Why? Webb points out that, at this early stage, almost everyone in the industry benefits from happy (if unrealistic) talk about 5G. Equipment vendors hope to sell cutting-edge tools. Carriers and regulators hope to portray themselves as facilitators of innovation. Academics and other researchers get funding to investigate fascinating technical problems. In short, “[i]t is in nobody’s interest to rock the boat.” While that may suffice to keep the 5G myth going for a few more years, it doesn’t mean that operators will actually commit real money to the technology (for instance, by purchasing Straight Path’s spectrum at a price of $500 million). With so many question marks looming over the whole enterprise, operators have every incentive to bide their time.
While Webb’s 5G Myth is unusually clear in its debunking of industry hype, others have made similar points. Kim Larsen, a senior vice president in the technology group at Deutsche Telekom (the large German operator that owns a majority of T-Mobile), recently released an essay [9] on “5G economics” that also throws cold water on mmWave 5G (emphasis not added):
Clearly, ubiquitous 5G coverage at those high frequencies (i.e., >3 GHz) would be a very silly endeavor (to put it nicely) and very un-economical.
5G, as long as the main frequency deployed is in the high or very high frequency regime, would remain a niche technology. Irrelevant to a large proportion of customers and use cases.
…In my own opinion, if 5G will be launched primarily in the mm-wave bands around and above 30 GHz, I would not expect to see a very aggressive 5G uptake.
… Would it not be economically more relevant to boost the customer experience across our macro-cellular networks, that actually serves our customers today? As opposed to augment the existing LTE network with ultra hot zones of extreme speeds and possibly also an extreme number of small cells.
If 5G would remain an above 3 GHz technology, it would be largely irrelevant to the mass market and most use cases.
… Maybe more importantly, will the 5G customer experience from the very high speed and very low latency really be noticeable to the customer?
Straight Path bulls, often unaware of this skeptical perspective on 5G, simply cite positive comments from leaders at Verizon and AT&T that seem to signal strong faith in the technology. But much of what these companies are pursuing – still in very small-scale trials – is fixed wireless access (FWA), which, by definition, doesn’t involve mobile devices and thus pertains to a much smaller target market. Fixed wireless can, in principle, enable wireline providers to connect to homes that are too expensive to run fiber to – for instance, in protected historic neighborhoods. But the history of fixed wireless is littered with failures (as well as excuses for why this time was different). Indeed, the 28 and 39GHz bands now authorized in the US for mobile use were originally viewed as domains for fixed wireless access but fell into disuse precisely because the technology could not successfully compete with wireline broadband.
Whether or not Verizon moves forward with the pre-standards 28GHz fixed wireless service it’s now piloting, the company already has enough spectrum locked down via the XO transaction to support its ambitions. For Straight Path, then, that leaves AT&T and T-Mobile (in light of Sprint’s continued financial weakness and spectrum surplus). But many observers believe that AT&T’s newfound optimism about fixed wireless is, at least in part, a ploy aimed at securing approval for its acquisition of Time Warner. By claiming it’s going to create an alternative to wireline broadband and challenge Charter and Comcast, AT&T casts itself as a pro-competitive force, not a potential monopolist.
Even setting aside the question of ulterior motive, AT&T’s fixed-wireless efforts already demonstrate a strong willingness to consider any source of spectrum, including unlicensed bands (used [10] in the company’s “5G business customer trial” in Austin) and the lightly licensed 70/80 GHz bands (used [11] in separate fixed-wireless trials, including one in Minneapolis). Google, for its part, is testing [12] fixed wireless service in Kansas City using the shared 3.5GHz band. With so many other options available to it, a player like AT&T just doesn’t need Straight Path’s spectrum to provide fixed wireless service, even if that is its genuine goal. As for T-Mobile, a recent FierceWireless headline put it well:
T-Mobile reiterates apathy toward fixed 5G services [13]
As much as spectrum bulls like to fantasize about new entrants shaking up the wireless market, it would be economic suicide for anyone to build a de novo network on a mmWave foundation; it would be a nightmare to deploy and, outside of select hot spots, it would offer no coverage. Besides, if any new entrant had serious plans to go toe-to-toe with the incumbents, the best opportunity to achieve scale would be by acquiring low-frequency spectrum in the 600MHz incentive auction – yet as the (still unfolding) results of that auction show, there’s little demand for such spectrum.
For Straight Path, then, AT&T and T-Mobile are the only realistic buyers for the bulk of its spectrum, and their only realistic motives would be to provide mmWave mobile service. As Webb and Larsen’s analyses indicate, however, mmWave mobile networks pose such enormous difficulties and promise such modest benefits that they will likely never take root outside of small niches. For major carriers, small-scale mmWave experimentation may nonetheless make sense as a hedge, but with large mmWave auctions on the horizon and additional bands in the pipeline, there’s no reason to pay up for Straight Path. To be sure, Straight Path is egregiously overvalued even if 5G succeeds commercially, but no one should take that outcome for granted.
Endnotes
[1] [14] For the avoidance of doubt, Kerrisdale has no relationship with Mr. Webb.
[2] [15] T-Mobile call, January 5, 2017. Transcript source: Capital IQ.