A report from investment bank Exane/BNPP stated that today’s various satellite operators are supplying some 2,000 Gb/s of capacity (as of December 31, 2020, and using Northern Sky Research data). The bank suggests that this will grow ten-fold by the end of this year and 30-fold before the end of this decade.
This growth includes capacity from Jeff Bezos’ Project Kuiper despite – as yet – there are no detailed launch plans published from Kuiper.
Exane/BNPP’s media analyst Sami Kassab said the Telesat Lightspeed LEO fleet contract announced on February 9 will bring an additional 15,000 Gb/s of bandwidth when it goes fully live around the 2023 to 2024 timeframe.
“We estimate that SpaceX’s Starlink will bring over 20,000 Gb/s by the end of 2021. OneWeb will add a further 4,000-5,000 Gb/s by year end while SES mPower will add another 7000-10000Gb/s between 2022/2024 with Telesat adding a further 15,000 Gb/s by end of 2024. HTS satellite capacity will grow by a factor over more than 30x by the end of the decade when Amazon launches,” said the bank.
“The two key questions for the satellite industry are: will the demand be there and who wins the demand? Capacity demand is largely (Euroconsult, NSR) expected to grow but at a slower pace than supply. COVID 19 is currently dampening capacity demand in the commercial aviation and cruise industries; two key growth drivers of the industry. Slow demand growth is likely to put further pressure on capacity prices,” added Exane/BNPP.
“Our SES revenue forecasts assume that HTS capacity pricing falls by an order of magnitude from currently $300-400 per Mb/s per month to $30 in 2025. However, our forecasts also assume that pricing decline is more than offset by volume growth as we forecast a 40 percent capacity utilisation rate for SES mPower in 2025,” noted Kassab.
“Who wins the demand?” asks the bank. “We think that three key variables could play a critical role in who wins the new demand. Bandwidth economics, bandwidth density and vertical integration. Our analysis suggests that SES, Starlink, Telesat have comparable capex per Gbps per year at c$30k in their Non GSO projects. SES has actually not precisely disclosed its constellation capacity and we believe it could even have better bandwidth economics by the nature of its orbital position (MEO vs. LEO). We have Viasat 3 on around $40k. At an estimated $200k, OneWeb looks least well placed. Being able to reduce LEO antenna costs looks like another critical development, especially for residential broadband markets.”
The question of antenna costs is crucial. The bank said, “We also believe that bandwidth density will be an important driver as providing enough throughput over hotspots looks key to win contracts. Inmarsat or Eutelsat cannot play in the Cruise industry because their infrastructure does not currently provide enough bandwidth density. SES O3b can and is winning. Finally, with the recent vertical integration of Intelsat, Viasat, Eutelsat, we believe that access to end customers will also be key. This may require SES to allocate C-band proceeds to the acquisition of a large service provider (Global Eagle?, Speedcast? both are currently in Chapter 11) or to continue building its service business as it has successfully done in Cruise and Fixed Data. SES has attractive bandwidth economics, is making progress on antennas (ie Isotropic deal) and has proven bandwidth density.