Falcon 9 - Has Reusability Affected Customer Launch Prices?

[butters]

Another factor in this discussion is the external inflationary pressure, especially in the aerospace industry but across the global economy in general. Most product categories are experiencing pronounced price inflation, but orbital launch prices are relatively stable. So, there's an extent to which launch is getting cheaper relative to a lot of comparable goods and services. Just going by the CPI deflator, which probably understates aerospace sector inflation, $62M in 2010 is about $90M today.

[envy887]

elastic is the demand for launch services?

Clearly, it's very elastic, as we are celebrating the 250th Starlink launch in 6 years.

It's quite clear that F9 cost is a major part of the Starlink business model, but it's not a good example of theoretical price elasticity. The theory assumes a large number of independent customers making independent decisions based on price. SpaceX' F9 launch business cannot induce SpaceX' Starlink business to launch more satellites by reducing the internal launch price, and SpaceX cannot induce some other constellation company to launch more satellites on F9 by reducing the launch price. If Kuiper comes to their senses and suddenly starts launching on F9, SpaceX won't induce them to launch more satellites by reducing the launch price.

Why do they have to be independent?

I wouldn't call it inducement so much as enabling. SpaceX has some more or less fixed amount of capital (and other resources) they plan to spend on Starlink launches, and if costs are decreased, that capital will cover more launches. Higher internal costs necessarily mean fewer launches. All the same forces as external customers are at play here, just with different efficiencies.

The main difference between Starlink and external demand are that Starlink payloads are relatively cheap, so launch is a higher fraction (~50% instead of ~10% for NASA/DoD) of costs, so saving on launch lets them build a lot more payloads, which in turn means more launches. Starlink can be cheap in large part because it is tightly integrated by design to the launch vehicle, since both are under one roof.

Conversely for NASA, a halving in launch cost only nets them about 5% lower mission cost. That's not even enough for budget creep, much less a bunch of new missions. Cheaper launch justifies having a bunch of heavy weight, low budget missions. But NASA just doesn't do things that way.

[matthewkantar]

elastic is the demand for launch services?

Clearly, it's very elastic, as we are celebrating the 250th Starlink launch in 6 years.

It's quite clear that F9 cost is a major part of the Starlink business model, but it's not a good example of theoretical price elasticity. The theory assumes a large number of independent customers making independent decisions based on price. SpaceX' F9 launch business cannot induce SpaceX' Starlink business to launch more satellites by reducing the internal launch price, and SpaceX cannot induce some other constellation company to launch more satellites on F9 by reducing the launch price. If Kuiper comes to their senses and suddenly starts launching on F9, SpaceX won't induce them to launch more satellites by reducing the launch price.

Why do they have to be independent?

I wouldn't call it inducement so much as enabling. SpaceX has some more or less fixed amount of capital (and other resources) they plan to spend on Starlink launches, and if costs are decreased, that capital will cover more launches. Higher internal costs necessarily mean fewer launches. All the same forces as external customers are at play here, just with different efficiencies.

The main difference between Starlink and external demand are that Starlink payloads are relatively cheap, so launch is a higher fraction (~50% instead of ~10% for NASA/DoD) of costs, so saving on launch lets them build a lot more payloads, which in turn means more launches. Starlink can be cheap in large part because it is tightly integrated by design to the launch vehicle, since both are under one roof.

Conversely for NASA, a halving in launch cost only nets them about 5% lower mission cost. That's not even enough for budget creep, much less a bunch of new missions. Cheaper launch justifies having a bunch of heavy weight, low budget missions. But NASA just doesn't do things that way.

Bolded: Yet.

[RedLineTrain]

elastic is the demand for launch services?

Clearly, it's very elastic, as we are celebrating the 250th Starlink launch in 6 years.

It's quite clear that F9 cost is a major part of the Starlink business model, but it's not a good example of theoretical price elasticity. The theory assumes a large number of independent customers making independent decisions based on price. SpaceX' F9 launch business cannot induce SpaceX' Starlink business to launch more satellites by reducing the internal launch price, and SpaceX cannot induce some other constellation company to launch more satellites on F9 by reducing the launch price. If Kuiper comes to their senses and suddenly starts launching on F9, SpaceX won't induce them to launch more satellites by reducing the launch price.

I'm not sure that I understand fully what you are stating, but the reality here is that at $15 million per launch, there is evidently much more internal demand than SpaceX is able to satisfy.

[Robotbeat]

elastic is the demand for launch services?

Clearly, it's very elastic, as we are celebrating the 250th Starlink launch in 6 years.

It's quite clear that F9 cost is a major part of the Starlink business model, but it's not a good example of theoretical price elasticity. The theory assumes a large number of independent customers making independent decisions based on price. SpaceX' F9 launch business cannot induce SpaceX' Starlink business to launch more satellites by reducing the internal launch price, and SpaceX cannot induce some other constellation company to launch more satellites on F9 by reducing the launch price. If Kuiper comes to their senses and suddenly starts launching on F9, SpaceX won't induce them to launch more satellites by reducing the launch price.

I'm not sure that I understand fully what you are stating, but the reality here is that at $15 million per launch, there is evidently much more internal demand than SpaceX is able to satisfy.

Theyre still trying to ramp launch rate

[jebbo]

If you look at worldwide launch rates from ~2000, numbers were slowly and (very approximately) linearly climbing until 2019/2020 where there is a clear break.

If you subtract out China (i.e. the green line), it's obvious that US launches increased massively after that, but the SpaceX effect is more subtle, I think.

Ignoring the early days of F9 (no reuse etc), you can argue that between 2015 and 2019 (in the red oval), SpaceX cannibalised much of the US launch market - due to either lower prices or just ability to launch. Then, after 2019, and also ignoring Starlink, there is a clear increase in demand (green oval).

Whether this non-Starlink increase is down to lower prices or just Spacex's ability to increase cadence is unclear (at least to me). Still, it does demonstrate there is *latent* demand for launches, though I don't think you can categorically measure demand elasticity yet.

That SpaceX can maintain its prices to the government suggests this part of the demand is inelastic.

--- Tony

[RedLineTrain]

elastic is the demand for launch services?

Clearly, it's very elastic, as we are celebrating the 250th Starlink launch in 6 years.

It's quite clear that F9 cost is a major part of the Starlink business model, but it's not a good example of theoretical price elasticity. The theory assumes a large number of independent customers making independent decisions based on price. SpaceX' F9 launch business cannot induce SpaceX' Starlink business to launch more satellites by reducing the internal launch price, and SpaceX cannot induce some other constellation company to launch more satellites on F9 by reducing the launch price. If Kuiper comes to their senses and suddenly starts launching on F9, SpaceX won't induce them to launch more satellites by reducing the launch price.

I'm not sure that I understand fully what you are stating, but the reality here is that at $15 million per launch, there is evidently much more internal demand than SpaceX is able to satisfy.

Theyre still trying to ramp launch rate

And if they could readily increase their launch cadence even faster without breaking the company, I believe that they would do so.

Demand isn't limitless at $15 million per launch, but it doesn't seem like we have come close to knowing the limit.

[Robotbeat]

I agree. In SpaceX case, I think they are also trying to balance the opportunity cost of adding another drone ship versus funding to accelerate starship. If they knew that starship would never exist, and they would be stuck with falcon for decades, then there�s a bunch of stuff they could do to increase launch rate and reduce cost.