It’s easy to forget that until recently, building a network of tens of thousands of communications satellites in low earth orbit was seen as an insane idea.
When SpaceX proposed building Starlink, the odds were decidedly against success. Most previous attempts had ended in bankruptcy or failure—the complexity and expense of launching so many computers into space on a strict regulatory timeline leaves little margin for error, but delays of all kinds are common in rocket land.
Still, the growing appetite for connectivity, the increasing power of miniature electronics, and the reduced cost of getting to space convinced a few businesses to make this bet. After a falling out, Elon Musk and satellite entrepreneur Greg Wyler started rival efforts. Wyler’s OneWeb operated as a traditional satellite company, hiring rockets on the open market. Musk’s Starlink, on the other hand, could rely on SpaceX’s rockets and technology. A few years later, Jeff Bezos proposed a satellite network called Kuiper—but as a subsidiary of Amazon, not his rocket company, Blue Origin.
Nearly a decade later, Starlink is providing connectivity to millions. OneWeb went bankrupt, but emerged with new investors and is beginning to offer service. And Kuiper is nearly ready to launch—but now some investors are looking askance at the project, most obviously in a new lawsuit against Amazon brought by a pension fund that manages about $425 million investments on behalf of its union members in Ohio.
In 2022, Amazon made the largest-ever purchase of rockets in history to get its satellites into space. The deal was big for space, but also for the e-commerce giant—according to the the lawsuit, it marked the second-largest capital expenditure in its history, behind only the $13 billion acquisition of grocer Whole Foods in 2017.
The pension fund’s complaint is simple: The board members overseeing this expenditure failed in their fiduciary duty to shareholders when they approved the launch deals because they didn’t consider hiring SpaceX to do the job. Instead, they approved contracts with United Launch Alliance, Arianespace, and Bezos’ Blue Origin. And because Blue Origin is building rocket engines for ULA’s new rocket, both of those deals are related-party transactions. Per the fund, Amazon’s board approved the deals in less than 40 minutes without consulting any outside experts. The approach is, in theory, more expensive than launching with SpaceX, in part because 78% of the satellites are set to launch on Blue Origin rockets that have never taken flight.
“The claims in this lawsuit are completely without merit, and we look forward to showing that through the legal process,” Amazon said in a statement.
Hanging over all this is a deadline imposed by the Federal Communications Commission, which supervises satellites in the US: Amazon needs to launch 1,618 of its Kuiper satellites by July 30, 2026. For a comparison, SpaceX launched 1,145 Starlink satellites in the three years after launching its first prototypes. Amazon’s first prototypes are supposed to launch sometime this fall, which gives the company a tight window before its deadline to test, manufacture, and launch all the rest.
Generally, US courts give a fair amount of leeway to businesses to make their own decisions, even with related party transactions—Musk himself won a lawsuit challenging Tesla’s acquisition of Musk’s firm Solar City in 2016. And Amazon will make the argument that it rejected working with SpaceX because Starlink is Kuiper’s biggest competitor, despite SpaceX frequently launching satellites for rival satellite networks, including ViaSat and OneWeb. It’s also not clear if SpaceX, with its own busy launch cadence, has the capacity to fly all of Kuiper’s satellites even if Amazon was inclined to hire its rival for the job.
However the case resolves, it highlights the strange relationship between Amazon and Blue Origin. Amazon executives say that the idea of the network bubbled up organically in brainstorming sessions as a complement to the company’s other businesses, including e-commerce and digital infrastructure like Amazon Web Services. Still, it’s hard to imagine Bezos would greenlight such a project without considering his space company’s potential role.
It also highlights the importance of vertical integration in the space industry. SpaceX has succeeded where others have failed largely because it is a rocket company first. Consider that SpaceX sells satellite launches for around $60 million, but its cost to launch a reusable rocket is thought to be closer to $40 million or less. When SpaceX launched 34 Starlink missions in 2022, the company saved on the order of $680 million compared to what a third party might pay it for those launches. And since the 68 rockets purchased by Amazon for Kuiper are more expensive vehicles, perhaps more than $100 million each, Amazon will pay significantly more just to get its business off the ground.
The Lunar Reconnaissance Orbiter has done its now-traditional duty: finding the crater of a crashed lunar mission. This time around, LRO has spotted the likely crash site of Luna 25, Russia’s recently failed attempt to land a robot softly on the surface of the Moon.
Elon Musk borrowed $1 billion from SpaceX. The unusual loan was made while Musk was sorting out the finances for his purchase of Twitter, and while it was repaid, underscores the financial strains that Musk’s other ventures apply to his space company.
Ariane 6 started its engines. The European rocket completed the first of two test fires ahead of a still-unscheduled debut launch; the delayed launch vehicle will compete with SpaceX’s Falcon 9.
Virgin Galactic scheduled its next flight. Another revenue-generating flight is scheduled for Sept. 8; read this in-depth interview with VG president Mike Moses on the company’s flight cadence and plans for a second generation of vehicles.
Lunar start-ups raised new funding. This week, US firm Intuitive Machines (NASDAQ:LUNR) picked up $20 million through public stock sales to shore up its balance sheet, while Japanese robotics company GITAI raised $15 million from venture investors to support its work on future lunar missons.
American Airlines avoided contrails. Space data is only as valuable as its use cases, and a fascinating one emerged this year: Airlines can significantly reduce their climate impact by dodging regions of the atmosphere where contrails form, using weather satellites and new software developed by Breakthrough Energy and Google.
Last week: SpaceX found a new kind of export control problem.
Last year: New Shepard’s fiery failed launch is also a test for US regulators.
This was issue 194 of our newsletter. Hope your week is out of this world! Please send your shareholder concerns, tips, and informed opinions to email@example.com.