At the same cost SpaceX pays for a launch? That seems improbable to me.

depends what you mean. If you think of it as a "replacement" of another launch they could have sold, then yes, it's the same cost. Lost profit opportunity is lost profit is a cost.

But if it means they can increase the # of rockets they build, they might pick up some efficiency gains on the back stretch. I'm pretty sure they're already struggling to keep up with demand, tho.

They're getting to the point where they're demand-limited rather than supply-limited - plowing through the backlog quite fast, and the GTO launch market is in a downturn.

SpaceX keeps rockets other customers "paid" for after they've landed (the customer pays for the flight, not the rocket). They've amassed a fleet they never had to pay a dime to construct.

I believe they have eight block 5 Falcon 9s that landed successfully. More if you count previous generations, but those likely won't be reused.

Does the price of a launch completely cover the cost of construction (excluding R&D)?

It does, for the competition, and there is no reason for SpaceX to go much lower as long as they don't build excessive overcapacity that needs to be filled.

Yes. They actually improved a lot of rocket construction and that's why they were cheaper then everybody else BEFORE re-usability. Now re-usability they just increase the profit.

Until somebody else challenges their price, they can get huge profit from every launch.

Not if the opportunity is to be ahead of your rivals to a new market frontier and guarantee a head start

Even then you're trading one opportunity cost for another.

So then make the competitor pay for it: each time they launch for a competitor, add a margin equating to the total cost of deploying a Starlink satellite.

Of course there is a markup.

They can build their own rockets

In 2040 after billions in investments.

They could create an electronic payment system to raise capital for their rockets.

And before we jump into this rabbit hole, yes, they can totally fund their own payment system with $4K and a small yellow pages startup. :o)

SpaceX was launching Falcon 9's 8 years after they were founded.. 2030 should be achievable.

And they can deny service to anyone.

I'm not too sure about this, given the highly limited market options and high regulation of space travel, I'm guessing they'd be hit with anti-trust laws pretty quickly.

Yeah but with the volume we're talking about here, SpaceX is definitely going to be extremely constrained by launch capacity. I doubt they'd be hit with any antitrust lawsuit over not cancelling their own launches in order to take orders from their competitors. In any case, there's no chance that all of those companies are going to be able to launch their constellations in anywhere close to their targeted timeframes. There's just not enough launch capacity in the world to handle that in such a short amount of time.

My impression of the Starlink project is that it's at least in part a backup project for unused launch capacity.

That is, if they happen to not find paying customers for all their reusable launch fleet, this is a way for it to not stand around idle.

Under this theory, paying customers will always get precedent over throwing up another 44 satellites.

>> I'm not too sure about this, given the highly limited market options and high regulation of space travel

The government grants monopolies (patent) on very important things (drugs and treatments for all sorts of things) and then lets companies charge different rates to different customers - even higher prices to the uninsured. It seems unlikely they'd make SpaceX launch for a competitor at anywhere near their own cost. But then politics....

They get to launch* their own stuff for only the rocket cost, that part is fine.

I mean if they have Boeing, NASA, etc. as paying customers at a standard $X they can't just go "Well Google you're launching a competitor to us so we're gonna charge you 20 * $X instead to prevent you from launching".

And Samsung can stop selling apple screens. But they don't, because they realize they need the revenue for things later on.

Ans have spacex deploy it's own satellites in the same launch for free

I think OneWeb is already in bed with

Arianespace: http://www.arianespace.com/client/oneweb/

Aren't they significantly more expensive?

Sure but would you want to rely solely on your direct and possibly only competitor?

And SpaceX can refuse to sell them. Or they can sell them expensive and that helps financing their own system.

And anyone can buy a launch on an ISRO rocket.

Could they? When you're launching such a large constellation the key metric is $/kg. All of ISRO's rockets are at least double as expensive as SpaceX's prices for customers. Plus, they're smaller rockets and it'd remain to be seen if ISRO could produce enough of them to launch an entire constellation.