A note from US investment analyst Oppenheimer has been doing the rounds and a few US publications have picked up the idea that a new service from start-up Starry, would be cheaper for fixed wireless than the 5G pushes of either AT&T or Verizon.
Certainly Starry is charging less, and says that it is offering 200 Mbps for $50 a month all in.
We reported Starry back in the middle of last year, run by Chet Kanojia the one-time CEO of the now defunct Aereo. Kanojia is an expert in antennas, and essentially this is an antenna play. It uses 600 MHz of spectrum in the 37 to 37.6 GHz band, which the FCC has not yet finalized rules for, and then uses a box inside the home to convert the signal to a 802.11ax WiFi signal, which after all companies like AT&T and Verizon are going to have to do anyway, since most broadband in-home is delivered wirelessly.
Financial research firm Oppenheimer said, in a client note, that it calculates that it will cost Starry about $2 billion to cover 70% of the densely populated areas of the US with a fixed wireless network. To do the same with 5G would cost $35 billion to $50 billion, it calculates and be twice that if the operator also needs to deploy fiber.
The problem is that Starry does not have any of its own fiber and will have to backhaul this service somehow, almost certainly using cable or fiber from one of the existing telcos of cablecos in the US.
“Starry’s deployment costs multiples less than 5G fixed wireless. Management stated generously it would cost $10K per mile to lay fiber, whereas Starry’s capex costs are $20 per home passed in dense areas,” wrote the Oppenheimer analysts. In addition, it said Starry’s Beams (base stations) do not require municipal approval as they are only 18 inches square.
The report continues: “Starry’s roadmap is much more economical than 5G fixed wireless … and speeds are comparable without the expense/complexity that accompany the deployment of thousands of 5G small cells.”
We think this analysis is disingenuous in that a fixed wireless service uses line of site in the same spectrum, not small cells and these are only needed to deploy a full mobile 5G experience. No-one has yet seen the size of the antennas that AT&T and Verizon fixed wireless will use and calculated their deployment costs.
Starry has tested the service in Boston and plans to expand them to 16 major markets this year and has just launched in beta in LA and Washington and will add New York, Cleveland, Chicago, Houston, Dallas, Denver, Seattle, Detroit, Atlanta, Indianapolis, San Francisco, Philadelphia, Miami and Minneapolis later this year.
Kanojia is nothing if not ambitious and took the US TV broadcasters to the supreme court after establishing Aereo to deliver free to air TV channels to portable devices, before his effort was shut dow0n.
Oppenheimer says that Starry’s equipment and installation costs total $480, and the firm aims to reduce that to $360 by 2019.
The analysts wrote: “Starry is at nine sites in Boston and will go to 18 by year end. This will blanket Boston, covering 600K homes passed at an incremental capex cost in the $25 million per area. A quick analysis puts the cost for Starry to cover 70% of the dense parts of the US at $2 billion in capex ($20 per home passed x 120m homes x 70%), which is well below the potential cost of running fiber at anywhere from $85 billion (120m homes x 70% x $1,000 per home passed).”
It will take the involvement of a major funder to make this take off for real, but stranger things have happened.