Starlink competition - a big merger ahead?

In the wake of COVID-19, OneWeb and Intelsat filed for bankrupcy. It is becoming increasingly clear that satellite communication business is passing through the major crisis mode. And it is not caused by COVID-19. Falling revenues from broadband video, Internet capacity glut caused by aggressive frequency reuse, high debt made any satellite communication growth out of the question. 

And then SpaceX, raising round after round of additional capital, is on the verge of launching Starlink. It has three distinct advantages over any other constellation. First it has significantly lower cost of launch. In a recent interview, Elon Musk quoted marginal cost of Falcon-9 launch of just 15 million USD. That is one tenth of price just a decade ago. 

The second innovation of Starlink is unparalleled mass production of satellites, which is estimated to bring price down to 500000USD per satellite. There are cubesats that cost more. And the third innovation is high density satellite packaging, which maximizes usage of the available fairing volume and LEO mass capability of Falcon 9.

All that promises cost structure of just 750000USD per satellite, or 1.2 billion investment for initial full constellation of 1600 satellites. The big unknowns are R&D, ground segment and user terminal cost structure, but SpaceX seems well capitalized to finish its initial constellation and provide the Internet broadband service. 

So how is competition standing? Amazon (and Blue Origin) are years behind. Leosat dropped from the race. No one else can really compete with SpaceX on the cost of space segment, due to higher cost of launch and no volume/weight optimization of satellite stack to Falcon 9 capabilities. After just 10% of fleet deployment, OneWeb (despite its 48000 satellite constellation request), filed for Chapter 11 bankruptcy and with much higher cost structure will not be able to compete. Intelsat is also in Chapter 11. Telesat...unusually quiet about actual constellation plans. Not a good sign. Eutelsat? They are conservative, playing just in GEO. Inmarsat? Focused on different market segment, covering the oceans. Hughes Network Systems? They bet the wrong horse (OneWeb). 

That leaves Viasat and SES. And (surprise, surprise) Boeing.

Viasat has aggressive GEO based Internet broadband plan based around Viasat-3 constellation. It also announced expansion to MEO and possibly LEO constellation using the same frequency reuse technology. Its Viasat-3 Americas was announced in 2017, as the first terabit satellite with the launch postponed to 2021. Main supplier is Boeing, which is building the satellite bus based on its standard 702 platform. And later Viasat announced additional two Viasat-3 satellites. based on the same technology and supplier. Ah yes, Viasat also offer MEO connectivity, partnering with SES 03b constellation. 

In 2017, SES announced O3b successor - mPOWER constellation, built by Boeing on the same 702 satellite bus platform as for Viasat-3. With 1.4 terabits of capacity per satellite. And who will provide flat phased-array antennas and ground terminals? Yes, Viasat again. 

When Mark Dankberg, Viasat CEO, discussed its MEO and LEO play, they were already using SES 03b for seamless MEO to GEO service integration. And possibly LEO too. In 2016 they announced plans for 24-satellite MEO constellation. Later downsized to twenty satellites. But since then they have been very silent about it.

Now comes 2020. COVID-19 caused major downturn, especially in maritime and airline connectivity markets. SES announced in March that it might separate 03b (SES Networks) into a separate company with IPO in sight. So they are seeking external investment into mPOWER. They announced 24-satellite MEO expansion of mPOWER in 2018 using inclined orbit. And just recently they requested small scale LEO expansion (using 500km orbit, essentially causing trouble for Starlink) with just 36 satellites? That is insufficient for any serious connectivity due to extreme inclination for any ground based gateway. It is obvious that Viasat and SES are partnering on every level and seem to be very careful not to compete with each other. Viasat existing broadband subscriber base is under serious threat from Starlink, especially due to GEO high latency issue. By providing low-cost phased array antennas and terminals connected to the mPOWER (also called 03bN constellation) they could address the latency issue and provide seamless integration between MEO and GEO especially for rural areas. 

At the same time, highly reduced number of MEO satellites compared to Starlink, their total space segment investment would be on the similar level to SpaceX. Plus they could utilize existing and upcoming fleet of high-throughput GEO satellites. And existing gateway locations. At the same time, MEO would enable much wider dislocation of user terminals and gateways/teleports, which can easily span thousands of kilometers. For Starlink without inter-satellite link capabilities, any gateway has to be near the user terminal (less than 500km), significantly reducing Starlink suitability for isolated island, maritime and transcontinental air traffic. Thus Viasat-3 and mPOWER are the only high capacity Starlink competitors that are close to launch/operations, with comparable price per bit and unique short term advantages compared to Starlink. 

What are the short-term advantages of SES/Viasat partnership? Existing customer base of rural subscribers (Viasat) and backhaul (SES Networks) customers. Established sales channels. Great distance between gateways and user terminals. Existing gateway locations with good Internet connectivity. Global reach. Good government relationships outside of USA. Decades of experience operating satellite communication networks. Two way satellite based Internet broadband and frequency reuse? SES had that working twenty years ago. Together, Viasat and SES stand a good chance against Starlink. Decisive point might not be the satellites or price per launch itself as number of MEO satellites reduces any strategic advantage that SpaceX has. The most important aspect will be the cost of the user terminal, which represents major part of the user acquisition cost. But they (Viasat) have to reach similar price point with phased array antennas. Other option is to wait for SpaceX bankruptcy. But many have bet against Elon Musk. And so far they all failed. So the merger of Viasat and SES (Networks) presents low-risk approach to address the largest disruption of satellite based telecommunications market since...1965?