Who will compete with SpaceX? The last two and next two years.

[Robotbeat]

ULA will continue to (effectively) compete against SpaceX for a while for natsec payloads. Until another challenger arises, perhaps Blue Origin.

[joek]

ULA will continue to (effectively) compete against SpaceX for a while for natsec payloads. Until another challenger arises, perhaps Blue Origin.

Depends on your definition of "compete".  Only if "assured access" is part of the "compete" equation--assuming SpaceX does not FU.  That is, one provider is subsidized either explicitly or implicitly to ensure assured access; e.g., guaranteed X launches or via ELC.

[envy887]

With 5000 (or closer to 6000) employees - and still growing, they have quite a bit of overhead they need income for in order to be cash flow positive. Their payroll alone is likely over half a billion per year! They also need profit to roll into development of the Mars system.

To be fair they have more than Falcon 9/Heavy launch services to provide money for future projects:
[snip]

True, but they also need to stay in business. This simple math can be eye opening: (please excuse very simplistic lowballed numbers) 5000 employees * $100k/year pay = $0.5 billion dollars. If they do 20 launches - $25 million from every launch goes straight to paying the employees. (If only 10 launches, that number goes up to $50 million per launch) Now add rocket raw materials, external contracts, facility rent & upkeep, taxes, debt, and the absolute minimum price starts to get quite a bit higher. Oh, and you want some profit on top of that too. 

Therefore one could actually argue that flight rate is just as important as reusability (or even more so) for lowering costs. And for SpaceX to succeed, they need to do both. Launch often AND reuse.

Anyway, sorry for the side tracking, just trying to illustrate from a different point of view how difficult it is to bring down costs.

A lot of those employees are doing things other than building and launching Falcon 9. Some of those things make money (e.g. Dragon), and some are development (e.g. D2, FH) and capital investment (e.g. mfg, test, launch facilities) that's amortized over FAR more than 10 or 20 launches.

And most of the cost of those employees is represented by the cost of recovered hardware. That cost doesn't simply disappear when they recover the hardware: they either launch more (and thus spread out the cost to more launches), or they can launch as often but require fewer employees and thus have lower operating costs.

So just citing current total payroll is not at all a good indication of cost per launch in 2-3 years.

[rst]

Depends on your definition of "compete".  Only if "assured access" is part of the "compete" equation--assuming SpaceX does not FU.  That is, one provider is subsidized either explicitly or implicitly to ensure assured access; e.g., guaranteed X launches or via ELC.

The definition of "assured access" in the law is simpler --

(1) the availability of at least two space launch vehicles (or families of space launch vehicles) capable of delivering into space any payload designated by the Secretary of Defense or the Director of National Intelligence as a national security payload; and
(2) a robust space launch infrastructure and industrial base.

Nothing about financial arrangements, other than that whatever they are, they must keep two providers in business. So if, say, Blue and SpaceX are both certified, and getting by without subsidies or guarantees, there's no reason to provide either. (See https://www.law.cornell.edu/uscode/text/10/2273. And note also that ULA is no longer getting ELC, though it still has some guaranteed business over the next few years through the somewhat controversial block buy.)

[Lars-J]

So just citing current total payroll is not at all a good indication of cost per launch in 2-3 years.

True, but it is a still a very useful macro overview of the cost situation. Even if there was no refurbishment cost needed at all, and the system was completely reusable, flight rate would become the dominant factor in reducing costs. If you have overall expenses totalling X, you've got to cover that over Y launches.

I'm primarily pushing back against the idea that SpaceX's overhead is only represented by 10% of the current cost. It is far more. Even if reuse was complete and refurbishment minimal.

[AncientU]

Depends on your definition of "compete".  Only if "assured access" is part of the "compete" equation--assuming SpaceX does not FU.  That is, one provider is subsidized either explicitly or implicitly to ensure assured access; e.g., guaranteed X launches or via ELC.

The definition of "assured access" in the law is simpler --

(1) the availability of at least two space launch vehicles (or families of space launch vehicles) capable of delivering into space any payload designated by the Secretary of Defense or the Director of National Intelligence as a national security payload; and
(2) a robust space launch infrastructure and industrial base.

Nothing about financial arrangements, other than that whatever they are, they must keep two providers in business. So if, say, Blue and SpaceX are both certified, and getting by without subsidies or guarantees, there's no reason to provide either. (See https://www.law.cornell.edu/uscode/text/10/2273. And note also that ULA is no longer getting ELC, though it still has some guaranteed business over the next few years through the somewhat controversial block buy.)

ULA is still getting ELC.  For Phase 1A competitions, though, they have to bid without it. Same with Phase 2 competitions.

The USAF catch phrase for keeping one particular provider in business is 'Managed Competition' -- better known as allocation -- wouldn't be needed (and it is needed) if free and open competition would produce the same result.

[MP99]

With 5000 (or closer to 6000) employees - and still growing, they have quite a bit of overhead they need income for in order to be cash flow positive. Their payroll alone is likely over half a billion per year! They also need profit to roll into development of the Mars system.

To be fair they have more than Falcon 9/Heavy launch services to provide money for future projects:
[snip]

True, but they also need to stay in business. This simple math can be eye opening: (please excuse very simplistic lowballed numbers) 5000 employees * $100k/year pay = $0.5 billion dollars. If they do 20 launches - $25 million from every launch goes straight to paying the employees. (If only 10 launches, that number goes up to $50 million per launch) Now add rocket raw materials, external contracts, facility rent & upkeep, taxes, debt, and the absolute minimum price starts to get quite a bit higher. Oh, and you want some profit on top of that too. 

Therefore one could actually argue that flight rate is just as important as reusability (or even more so) for lowering costs. And for SpaceX to succeed, they need to do both. Launch often AND reuse.

Anyway, sorry for the side tracking, just trying to illustrate from a different point of view how difficult it is to bring down costs.

You are including the employees who are employed in the contract for Dragon 2 and crewed launch, and Dragon 1 / cargo launch services.

Cheers, Martin

[MP99]

With 5000 (or closer to 6000) employees - and still growing, they have quite a bit of overhead they need income for in order to be cash flow positive. Their payroll alone is likely over half a billion per year! They also need profit to roll into development of the Mars system.

To be fair they have more than Falcon 9/Heavy launch services to provide money for future projects:
[snip]

True, but they also need to stay in business. This simple math can be eye opening: (please excuse very simplistic lowballed numbers) 5000 employees * $100k/year pay = $0.5 billion dollars. If they do 20 launches - $25 million from every launch goes straight to paying the employees. (If only 10 launches, that number goes up to $50 million per launch) Now add rocket raw materials, external contracts, facility rent & upkeep, taxes, debt, and the absolute minimum price starts to get quite a bit higher. Oh, and you want some profit on top of that too. 

Therefore one could actually argue that flight rate is just as important as reusability (or even more so) for lowering costs. And for SpaceX to succeed, they need to do both. Launch often AND reuse.

Anyway, sorry for the side tracking, just trying to illustrate from a different point of view how difficult it is to bring down costs.

A lot of those employees are doing things other than building and launching Falcon 9. Some of those things make money (e.g. Dragon), and some are development (e.g. D2, FH) and capital investment (e.g. mfg, test, launch facilities) that's amortized over FAR more than 10 or 20 launches.

And most of the cost of those employees is represented by the cost of recovered hardware. That cost doesn't simply disappear when they recover the hardware: they either launch more (and thus spread out the cost to more launches), or they can launch as often but require fewer employees and thus have lower operating costs.

So just citing current total payroll is not at all a good indication of cost per launch in 2-3 years.

Agreed on other projects (as you'll see above).

Once block 5 development is complete, it seems the plan is for SpaceX to convert those costs into launch profits (rather different from them being launch costs), then spend those profits on BFR / BFS / Mars.

Of course, this is just another way of saying that they will not reduce launch prices in line with costs, and they'll pocket the profit to use as they please.

Cheers, Martin

[ZachF]

ULA will continue to (effectively) compete against SpaceX for a while for natsec payloads. Until another challenger arises, perhaps Blue Origin.

Depends on your definition of "compete".  Only if "assured access" is part of the "compete" equation--assuming SpaceX does not FU.  That is, one provider is subsidized either explicitly or implicitly to ensure assured access; e.g., guaranteed X launches or via ELC.

This is true, and why I think Blue Origin will be the "killer" of Arianespace and ULA. As per the chart earlier in the thread SpaceX has eaten mostly Proton payloads, and a few USG payloads from ULA, but the market still likes a backup provider. When the commercial market and US government has two cheap launch providers to chose from, ULA and Arianespace are going to get pushed to the wayside quickly.

I personally think Boeing and Lockheed will eventually pull the plug on ULA. The launch business when they were a monopoly was an easy money-printing business using old proven technology and cost-plus contracts. It will become a lower cost, lower margin business going against two of the best businessmen/innovators (Musk and Bezos) alive.

Arianespace will probably get 3-6 EU charity launches per year, and maybe a couple launches here and there that BO/SX couldn't fit into their manifest.

[AncientU]

This exchange in Iridium thread was interesting:

Many thanks for the interesting article. It’s nice to get some perspective on what cheaper launch capability enables.

Reminded me of a recent article where Iridium made clear how important SpaceX is to their plans:

"We tried to work with other launch vehicle companies but all were at least twice the cost of SpaceX and unaffordable based on the scope of the network we needed to launch," Desch said

http://www.cnbc.com/2017/06/25/spacex-is-critical-to-iridiums-future-says-ceo-matt-desch.html

And:

If it is true, market elasticity is greater than some people here predicted. There are already payloads that would not be launched without SpaceX.

Makes sense that elasticity will reveal at low-cost end of market... so SpaceX will 'compete' with no one (or everyone, depending on your perspective) while picking up these new entrants.  BulgariaSat was similar case.


Edit/Add Note: Iridium's original constellation was launched in the 1990s.
From Wikipedia:

Motorola used launch vehicles from three companies from three different countries—the Delta II from McDonnell Douglas; the Proton K from Khrunichev in Russia; and the Long March IIC from China Aerospace Science and Technology Corporation.

[Lars-J]

With 5000 (or closer to 6000) employees - and still growing, they have quite a bit of overhead they need income for in order to be cash flow positive. Their payroll alone is likely over half a billion per year! They also need profit to roll into development of the Mars system.

To be fair they have more than Falcon 9/Heavy launch services to provide money for future projects:
[snip]

True, but they also need to stay in business. This simple math can be eye opening: (please excuse very simplistic lowballed numbers) 5000 employees * $100k/year pay = $0.5 billion dollars. If they do 20 launches - $25 million from every launch goes straight to paying the employees. (If only 10 launches, that number goes up to $50 million per launch) Now add rocket raw materials, external contracts, facility rent & upkeep, taxes, debt, and the absolute minimum price starts to get quite a bit higher. Oh, and you want some profit on top of that too. 

Therefore one could actually argue that flight rate is just as important as reusability (or even more so) for lowering costs. And for SpaceX to succeed, they need to do both. Launch often AND reuse.

Anyway, sorry for the side tracking, just trying to illustrate from a different point of view how difficult it is to bring down costs.

A lot of those employees are doing things other than building and launching Falcon 9. Some of those things make money (e.g. Dragon), and some are development (e.g. D2, FH) and capital investment (e.g. mfg, test, launch facilities) that's amortized over FAR more than 10 or 20 launches.

And most of the cost of those employees is represented by the cost of recovered hardware. That cost doesn't simply disappear when they recover the hardware: they either launch more (and thus spread out the cost to more launches), or they can launch as often but require fewer employees and thus have lower operating costs.

So just citing current total payroll is not at all a good indication of cost per launch in 2-3 years.

Sigh. You are completely missing my larger economic point. But I get it, it is not a popular message.

[envy887]

Sigh. You are completely missing my larger economic point. But I get it, it is not a popular message.

No your point is just wrong. SpaceX does not need 5000 people to launch 20 Block 5 F9s per year. If that was all they wanted to do, they could probably do it with 500 or fewer people: they would only need to build and test 2 boosters per year.

[Lars-J]

Sigh. You are completely missing my larger economic point. But I get it, it is not a popular message.

No your point is just wrong. SpaceX does not need 5000 people to launch 20 Block 5 F9s per year. If that was all they wanted to do, they could probably do it with 500 or fewer people: they would only need to build and test 2 boosters per year.

But they still do have those employees. And even more than 5000 employees. Probably closer to 6000 now. And not all of them are doing hands-on work to launch rockets. But those employees still exist, and they still need to get paid. This is obvious and beyond dispute. So what if some work on Dragon and future business? They still need to get paid now. All combined income needs to be brought together, and whatever remains after paying your employees and all other expenses is profit. If you can't meet your expenses you are operating at a loss and will soon go out of business.

So I'm going to try this again, real slow this time. Even if only 100 people are needed in the future to operate existing LV fleet, they would never lay all the other people off. There will always be a significant number of employees who are NOT involved in direct touch on labor. You seem to not want to count them. But I do, and any person running a company would agree.

You may like to call the other employees as people who are working on future development, and that they are "amortized" as such and don't count. But how do you think that amortization works? That's right... by paying it off over a longer period. With current income. It is still a current expense.

[envy887]

Sigh. You are completely missing my larger economic point. But I get it, it is not a popular message.

No your point is just wrong. SpaceX does not need 5000 people to launch 20 Block 5 F9s per year. If that was all they wanted to do, they could probably do it with 500 or fewer people: they would only need to build and test 2 boosters per year.

But they still do have those employees. And even more than 5000 employees. Probably closer to 6000 now. And not all of them are doing hands-on work to launch rockets. But those employees still exist, and they still need to get paid. This is obvious and beyond dispute. So what if some work on Dragon and future business? They still need to get paid now. All combined income needs to be brought together, and whatever remains after paying your employees and all other expenses is profit. If you can't meet your expenses you are operating at a loss and will soon go out of business.

So I'm going to try this again, real slow this time. Even if only 100 people are needed in the future to operate existing LV fleet, they would never lay all the other people off. There will always be a significant number of employees who are NOT involved in direct touch on labor. You seem to not want to count them. But I do, and any person running a company would agree.

I'm not saying they will lay them off, obviously. Because SpaceX has MUCH bigger plans than just launching 20 rockets a year. Try more like 200.

Most of SpaceX expenses, including much of that payroll, is development and not operations. Development CAN be run at a loss, because it is an investment in your future, and is paid back over a long period of time (thousands of launches). And a lot of it is being paid for by NASA anyways. Plus Dragon pays a lot of payroll, you can't count that part against commsat launches.

Operations cannot be run at a loss for long, but their operations are sized for far more than just 20 launches per year. If 20 is all that they get long term, then they will definitely downsize their operational staff.

You cannot reconcile 20 launches per year long term and 5000 employees. Using those numbers together is based on fundamentally flawed assumptions.

[groundbound]

Back to a reply to the original question; I had to mention this because no one else has.

One possibility that occurs to me, and it is just a possibility really, is that the short to medium term competition will come from an old rocket with a new subsidy.

There are lots of boosters in the world which are at least partly backed by sovereign interests. There are various amounts of wishful thinking or denial present in the management of various launchers. So if the launch market truly goes into oversupply, there are quite a few entities that might be able to convince their government to intervene during the "temporary" (they think) disruption. All it would really take is one of the various possible launchers and they would be SpaceX's de-facto competition for a few years.

[AncientU]

Sigh. You are completely missing my larger economic point. But I get it, it is not a popular message.

No your point is just wrong. SpaceX does not need 5000 people to launch 20 Block 5 F9s per year. If that was all they wanted to do, they could probably do it with 500 or fewer people: they would only need to build and test 2 boosters per year.

But they still do have those employees. And even more than 5000 employees. Probably closer to 6000 now. And not all of them are doing hands-on work to launch rockets. But those employees still exist, and they still need to get paid. This is obvious and beyond dispute. So what if some work on Dragon and future business? They still need to get paid now. All combined income needs to be brought together, and whatever remains after paying your employees and all other expenses is profit. If you can't meet your expenses you are operating at a loss and will soon go out of business.

So I'm going to try this again, real slow this time. Even if only 100 people are needed in the future to operate existing LV fleet, they would never lay all the other people off. There will always be a significant number of employees who are NOT involved in direct touch on labor. You seem to not want to count them. But I do, and any person running a company would agree.

You may like to call the other employees as people who are working on future development, and that they are "amortized" as such and don't count. But how do you think that amortization works? That's right... by paying it off over a longer period. With current income. It is still a current expense.

Three CRS flights a year yield $400-450M alone.  CCtCap yields milestone payments of several $100M.  That generously covers payroll.  Add in 8-10 other launches -- basically one per month when added to CRS flights -- and they are healthy financially.

When launch interruptions occurred and flight rate was half a year instead of full year, they went cash flow negative, so you are correct that they need to sustain a significant launch rate to make payroll... but it is interrupting CRS flights that hurts cash flow the most.  Rebuilding a launch complex for several $100M doesn't help either.

Popular message or no, SpaceX isn't running aground financially -- if they keep launching.

[joek]

Nothing about financial arrangements, other than that whatever they are, they must keep two providers in business. So if, say, Blue and SpaceX are both certified, and getting by without subsidies or guarantees, there's no reason to provide either. (See https://www.law.cornell.edu/uscode/text/10/2273. And note also that ULA is no longer getting ELC, though it still has some guaranteed business over the next few years through the somewhat controversial block buy.)

AncientU covered the basics.  Also worth noting that block buy and ELC were negotiated together--block buy is FFP/cap for hardware and some basic services; ELC is cost+incentive negotiated annually based on expected number of launches--all assuming a certain number of launches by ULA (that is, total is same as block buy total).[1]

Last I looked  (FY2016) had ELC budgeted through block buy fly-out (FY2019-2020? IIRC).  So current conditions give ULA 2-3 FY years of runway.  *But* phase 2 acquisitions start FY2018, at which point we'll see if ULA can hold their own without ELC, or will require "allocation" to remain viable (almost certainly with respect to DIV-H in out-years).

If you're interested in details, go check the USAF budget requests.


[1] edit: p.s.  So likely USAF is already practicing "allocation" to ensure ULA  block buy contract minimums are satisfied in the face of SpaceX competition-protests-etc.

[edkyle99]

Because SpaceX has MUCH bigger plans than just launching 20 rockets a year. Try more like 200.

200 per year is simply fantasy.  Even 20 annually for more than a few years (as SpaceX clears its delayed backlog) is optimistic.

In all history only two launch vehicle families have managed to fly more than 20 times per year for any length of time:  R-7 (1960s-1990s) and Thor (1960s).  R-7 peaked at roughly 55 per year during the 1980s, but the payloads that drove that cadence (Cold War film return recon) are gone.  During this decade, all of the rockets in all of the world's countries combined have only managed an average of a bit less than 83 orbital attempts per year.

 - Ed Kyle

[gospacex]

Because SpaceX has MUCH bigger plans than just launching 20 rockets a year. Try more like 200.

200 per year is simply fantasy.  Even 20 annually for more than a few years (as SpaceX clears its delayed backlog) is optimistic

Tell this to ESA, which recently literally begged European customers to not desert to SpaceX.

[joek]

In all history only two launch vehicle families have managed to fly more than 20 times per year for any length of time:  R-7 (1960s-1990s) and Thor (1960s).  R-7 peaked at roughly 55 per year during the 1980s, but the payloads that drove that cadence (Cold War film return recon) are gone.  During this decade, all of the rockets in all of the world's countries combined have only managed an average of a bit less than 83 orbital attempts per year.

Those launch rates are based on historical demand.  So when in all history has there been demand for constellations of several thousand LEO-MEO satellites (along with annual replenishment rates in the hundreds+)?  Never.  If the demand (money) is there, the launches will proceed apace.