BroadbandBusinessLaunch

Morgan Stanley Note Emphasizes SpaceX’s “Double Flywheel” of Starship, Starlink

A pair of Falcon Heavy boosters landing. Morgan Stanley emphasized the double flywheel nature of SpaceX's launch and Starlink business
A pair of Falcon Heavy boosters landing. Photo: SpaceX

Can anyone catch SpaceX?

That’s the question posed by a Morgan Stanley Research team in a recent, updated note to investors. Spoiler alert: The analysts seem to believe the answer is no. 

Payload read and digested the updated memo, which was first covered Tuesday by CNBC. For starters, the report is front-loaded with choice quotes from investors: 

  1. “More than one client has told us if Elon Musk were to become the first trillionaire…it won’t be because of Tesla.”
  2. “Others have said SpaceX may eventually be the most highly valued company in the world.” 
  3. “As one client put it, ‘talking about space before Starship is like talking about the internet before Google.’” 

Two peas in a pod

SpaceX’s rocket and Starlink businesses reinforce one another, the analysts write. As launches get cheaper and more frequent, it gets easier to deploy a full megaconstellation without breaking the bank. Since a full deployment requires the launch of thousands more satellites, Starlink represents an ultra-reliable “customer” for SpaceX’s rocket business. 

The $100 billion space company has a double flywheel, the analysts write. It works something like this:

  • As more rocket components graduate from expendable to reusable, launch costs could drop more. And launch cadence increases. Those unlocks, coupled with improved payload capacity and propulsion capability, could win SpaceX share in the heavy-launch market. 
  • Starlink gets better with scale, both from the network and end user perspective. Scale (and intersatellite links) help improve the coverage area, which in turn expands the addressable market of users and use cases. And all this, in theory, drives down production costs (TBD).  

Starlink projections

In Morgan Stanley’s bull case, Starlink alone eventually reaches ~$133 billion in enterprise value (base case = ~$81 billion, bear = $0). That’s not the only eye-popping projection: The analysts estimate that Starlink’s buildout will cost ~$240 billion, burn $33 billion from 2020–2030, and become cash-flow positive in 2031. 

  • Under this projection, user terminals would be the most expensive line item, responsible for ~83% of all Starlink buildout costs. 

Risk factors: What could get in the way of SpaceX’s plans for intergalactic domination? The analysts cite four types of risks: regulatory, execution, financial, and competition.

Related Stories
BusinessLEO

Vast Inks Deal with The Exploration Company For Future Cargo Missions

The agreement opens the door for Europe to begin commercial and research-based missions in LEO.

BusinessPolicy

The US Wants to CHIP In On Rocket Lab’s Solar Cell Factory

Rocket Lab is one of three major players in the Western market that builds space-grade solar panels, and the investment will grow production 50% in the next three years.

LaunchMilitary

Scout Space Taps ABL for Tech Demo Launch

Scout needs to get its Owl optical imaging system on orbit to prove it is the right product for information-hungry national security agencies.

Launch

Starship Splashes Down Successfully After Fourth Test Flight

The successful journey from Starbase to the Indian Ocean brought SpaceX many steps closer to its goal of reliable Starship flights with reusable hardware.