[DennisP]

There's a reason we don't throw away a 747 after one flight.

Fuel is only a couple percent of the cost of a rocket launch. Rockets are expensive, and on the Falcon 9 the first stage is 80% of the cost of the rocket.

In the early days of SpaceX I saw claims that expendables were more economical because you could build them cheaper, but then SpaceX went and build reusable rockets that are also cheap; their rockets cost less than competitors even without reuse.

They're getting government business because they charge the government less than their competitors charge. The subsidy argument doesn't make sense unless it's only the commercial launches that are cheap.

There are a lot of vested interests in this industry that are strongly incentivized to say reusable rockets are no big deal. I think they'll continue to say that until the BFR flies and SpaceX drops their rates to levels far below what disposables could ever achieve.

[hef19898]

Well, one point made in the article I mentioned above was that thermal stress limits the life time of a reusable rocket. Plus the lost payload as you need fuel to land the rocket again. Also, and at least that part I can confirm, a missile is tube is not a lot more than a aluminium tube. So, there is my home work for the weekend: rocket science and LEO launch economics!

[DennisP]

There's also the rocket engine, which is a lot more than a tube. Disposables throw that away too. You can just look at the cost of the rocket to see how much expensive stuff is getting thrown away.

SpaceX does still use the Falcon 9 as a disposable for extra-heavy payloads. That doesn't mean you have to throw your rocket away for all payloads, and the capacity difference isn't dramatic. And there won't be that many payloads too big for the BFR in reusable mode.

It's true that the rocket wears out eventually. SpaceX is aiming for about ten reuses initially, thus lowering hardware costs by 90%. They hope to do much better than that later.

[hef19898]

We agree on the hardware cost side. The missing step now is to compare these saved hardware costs against the lost payload and the revenue value of that payload. Averages will be fine I think.

Just looking at the hardware side without opportunity cost aonewhat misses the point. A really thorough cost analysis would need to factor in the impact on reliability, the risk of failed launches due to reused rocket stages. I assume these numbers are impossible to come by.

One other data point would be, if that's the case, how much premium SpaceX is asking for the use of virgin rockets as compared to reused ones. This would allow a rough estimation of the increase risk for failure and other costs. Again, I don't expect these numbers to be available.