The latest salvo from Kerrisdale Capital
License to Fail: A Review of Globalstar’s Experimental Licenses for TLPS
Recently Globalstar, Inc. (NYSEMKT:GSAT) applied to the FCC for a new experimental license to test its Terrestrial Low Power Service (TLPS) concept in California. Though the initial irrational enthusiasm over this relatively unimportant piece of news has already started to die down, the history of Globalstar’s experimental-license activity illustrates the many shortcomings of TLPS and provides further evidence that TLPS will never be commercially viable.
On October 20th, Globalstar, Inc. (NYSEMKT:GSAT) applied to the FCC for an experimental license to use the Channel 14/TLPS frequencies in San Carlos, California, using two Cisco access-point (AP) models, the Aironet 1570 and 3700. The 1570 is specially designed for outdoor usage; the 3700 is high-end and indoor-oriented; and both are capable of 802.11ac, the latest Wi-Fi standard, which is defined only for 5GHz and which TLPS will therefore never be able to offer. An exhibit attached to the application described the “Research and Experiment Objectives Specific to this License Application” as follows:
- “Determine the efficacy of TLPS for outdoor small cell use in diverse semi-urban environments.”
- “Determine the efficacy of TLPS for latency intolerant/high priority applications.”
If granted, this TLPS-related experimental license would not be the first, as the table below shows:
Charlie Munger: Invert And Use “Disconfirming Evidence”
Globalstar, Inc. (NYSEMKT:GSAT)’s partner Jarvinian has had at least one experimental license covering Channel 14 for more than 18 months. Yet not only is Globalstar still seeking to make sense of “the efficacy of TLPS” for different use cases, but Globalstar has never seen fit to release detailed results from any of these prior tests, with the arguable exception of the thin and misleading publicly disclosed “results” (obtained under the initial Cambridge license) that we criticized in a prior report and two ex parte letters to the FCC.
Since Globalstar, Inc. (NYSEMKT:GSAT) has no new technology to keep secret – just a single channel of its own – we find no good reason for withholding test results. But Globalstar may not want to acknowledge certain basic facts about TLPS. For example, in a scenario in which TLPS’s primary user devices are smartphones, its maximum throughput – shared across all devices that are in range of each other – will be at best ~35 Mbps, equivalent to roughly seven concurrent high-definition Netflix streams.* This is the same as any other 20MHz 802.11n channel. The higher effective range of 2.4GHz Wi-Fi, touted by Globalstar and Jarvinian as a great advantage, is in fact a disadvantage when the goal is to offer high capacity and support many users. But if TLPS can’t support large numbers of users or high-value applications, let alone outdo existing public Wi-Fi and cellular coverage, how can it possibly warrant a multi-billion-dollar valuation? As we’ll discuss later in this discussion, Globalstar’s aimless testing efforts provides further evidence that the company itself has no clear vision of how TLPS would generate meaningful value.
Cisco Is Not a Logical Partner and Has Publicly Belittled TLPS
Some of the excitement over the experimental-license application stemmed from the notion that Cisco Systems, Inc. (NASDAQ:CSCO) was somehow involved. However, all that the application itself says is that Globalstar, Inc. (NYSEMKT:GSAT) will use Cisco APs. (By that standard, Kerrisdale has partnered with Cisco to test TLPS: we used the Cisco Aironet 1262 in one of our lab tests.) The attached exhibit includes the phrase “Cisco Test Area” in small letters on Figure 5 but is otherwise silent on the topic.
Cisco does operate a small public Wi-Fi network in San Carlos, a pleasant but modest suburb with a population of 28,000 that serves as a test case for Cisco’s “Smart+Connected Communities” concept. Though Globalstar, Inc. (NYSEMKT:GSAT) grandiosely declares that “[t]he city of San Carlos is used by industry as a test bed for advanced 802.11 network and application testing,” that characterization came as a surprise to our Wi-Fi and technology contacts, including one who works directly on what he characterized as the only unusual Wi-Fi application in the city: a “smart parking” system. This system operates across a few hundred parking spaces. Buried sensors detect when a car is present and communicate this information (not via Wi-Fi) to “gateway” units that relay it to a central management system; in some cases, drivers can view this data in real time to determine where to find an open space. In other cities, these gateways use cellular connections from the likes of AT&T Inc. (NYSE:T) and Verizon Communications Inc. (NYSE:VZ); in San Carlos, because the municipal Wi-Fi network already existed, they use Wi-Fi. (Note that Wi-Fi is not a necessary component of a smart parking system, since other cities utilize LTE).
See full report here.