See Bill Whittle re: credibility in his latest. via Glenn Reynolds
Plus he describes operating costs as a tax… unbelievable. You allocate operating cost first in a budget. Everything else gets what’s left. If you are going to use a building, the lights stay on and are part of the fixed costs of a budget. Actually, operating costs are second to determining what is or is not going to be part of your budget. Once you decide that, then fixed costs come first.
To look at it any other way is deranged.
It is the operating costs of the SLS that make it catastrophically bad policy. Whether the development cost is $20B, $30B, or $40B will, in the long run, not matter so much (though why we would choose that over $1B or $3B options like Falcon Heavy and Atlas Phase 2 would be hard to defend). But SLS will keep the operating costs of NASA high in to the indefinite future, until is is canceled.
For Mike Griffin to rail against high operating costs, and then to advocate a strategy designed to MAXIMIZE future operating costs, is indefensible.
Ah, but that’s part of the magic of budgets, future operating cost beyond the budget window can be completely ignored if it helps make a case. You are absolutely right that it is totally indefensible.
Which goes to credibility. This country has a serious problem about not taking credibility into account.
The other question I would like to ask Mr. Grifffin is this: if the SLS budget is to be increased by $400 million a year for the next two decades, what other NASA programs would he raid to find the money?
AIAA elected Griffin president – again – so I’m wondering whether to renew my membership…
Operating cost for the SLS, like any other rocket system, will depend on how frequently its used. Under the Obama plan of limited SLS launches punctuated by several years of no launches at all, SLS cost will be appallingly high.
However, if the SLS is used for a lunar base program then probably at least four to six launches will be required per year. Under that scenario, SLS cost should be close to that of the space shuttle– not including the additional cost of the lunar lander and payload.
If the SLS is also used to launch the largest Bigelow space stations (BA-2100) to LEO, smaller storm shelter stations (BA-330) to L1, space depots to LEO and L1, and space solar power plants to GEO then cost per launch should be even lower.
Marcel, the numbers are in on Shuttle.. it cost about $1.5B/launch. Four to six launches per year would be $6B to $9B per year. There’s no budget for that. So either you meant the fictional numbers of the shuttle that have been thrown around for years or you’re continue to labor under the delusion that the NASA budget is going to go up sometime in the future.
“However, if the SLS is used for a lunar base program then probably at least four to six launches will be required per year. ”
If. Congress shows no inclination to fund such a program, which would require more funding for launching SLS at a higher flight rate, even more for the payloads to be launched, and a substantial R&D program for lunar landers that do not currently exist and the same for Lunar surface systems, ditto.
The appallingly low flight rate for SLS is a direct result of the funding that is likely to available for the payloads and launcher.
Absent a credible plan to increase the total funding, the desirability of a NASA unique launcher with high fixed costs should be reexamined.
Do the SLS cost estimates include the cost of money?
Most of the discussions I see of NASA vehicle development costs appear to calculate the total cost as the sum of the costs for Years 1 to N.
That analysis ignores Net Present Value, however. The NPV of dollar spent in year 1 is not the same as a dollar spent in year N. Similarly with benefits received.
An accurate analysis would account for the cost of money by applying an internal discount rate. Such an analysis assumes that development dollars are “borrowed” from the Treasury at the prime rate and not “paid back” until the benefits are received (i.e., the launches actually occur).
If the estimates that are being bandied about do not include the cost of money, then SLS is an even worse investment than it appears.
@Trent Waddington: “Marcel, the numbers are in on Shuttle.. it cost about $1.5B/launch. Four to six launches per year would be $6B to $9B per year.”
$1.5 billion is the cost of the entire program over the total number of launches. I’ve yet to see an authoritative cite for the marginal cost of an additional launch. NASA’s FAQ cites $450 million for “launch,” though it’s unclear as to whether that includes every dollar increment over annual fixed costs.
@Marcel: “Operating cost for the SLS, like any other rocket system, will depend on how frequently its used.”
Obviously not, or you’d end up with a rocket costing infinitely close to $0 to fly as you schedule more and more launches. Each mission will have some incremental cost in addition to the fixed baseline.
@Edward: “Do the SLS cost estimates include the cost of money?”
@McLean: “Absent a credible plan to increase the total funding, the desirability of a NASA unique launcher with high fixed costs should be reexamined.”
An increase in total funding isn’t strictly necessary, but reconfiguring the budget to support SLS and manned lunar exploration R&D leads to politically untenable choices. Ironically, one of those sacred cows, the ISS, is a key crowbar for NASA to pry funding for commercial crew.
@Trent Waddington
“Marcel, the numbers are in on Shuttle.. it cost about $1.5B/launch. Four to six launches per year would be $6B to $9B per year. There’s no budget for that. So either you meant the fictional numbers of the shuttle that have been thrown around for years or you’re continue to labor under the delusion that the NASA budget is going to go up sometime in the future.”
Shuttle cost were about $450 million per launch. Check the NASA website.
The $1.5 billion per launch is only if you include the development cost from the 1970s (money already spent). But NASA was not spending $7.5 billion a year ($1.5 billion per flight) to launch five shuttles in 2009.
Obama inherited an $8.4 billion a year budget related to the manned space program (space shuttle, ISS, and Constellation program). That’s $84 billion over a ten year period: plenty of money to develop a $38 billion dollar (worst case scenario) SLS/MPCV since there is no $3 billion a year shuttle program nor any $3.4 billion a year Constellation program even with $30 billion spent on the ISS pork program over the next ten years since Obama raised spending on the ISS from $2 billion to about $3 billion per year.
@Marcel: “The $1.5 billion per launch is only if you include the development cost from the 1970s (money already spent). But NASA was not spending $7.5 billion a year ($1.5 billion per flight) to launch five shuttles in 2009.”
I’m not sure what the $450 million entails, but given that the line item for Shuttle spending frequently amounts to $3-3.5 billion annually, I’m uncertain how to break that down as recurring costs do to a) scheduled amortization, b) capitalization per launch, and c) actual cost of putting the Shuttle on the pad and lighting a match.
plenty of money to develop a $38 billion dollar (worst case scenario) SLS/MPCV since there is no $3 billion a year shuttle program nor any $3.4 billion a year Constellation program
So, replacing the Shuttle and Constellation with an even bigger, more expensive rocket will somehow save money?
That seems, um, rather improbable. 🙂
Where do so many space cadets get the idea that the way to make things less expensive is to spend money?
Where do so many space cadets get the idea that the way to make things less expensive is to spend more money?
What I meant to say.
“Shuttle cost were about $450 million per launch. Check the NASA website.”
That is just NASA’s marginal cost estimate. It doesn’t include the fixed costs/overheard to run the Shuttle program.The average Shuttle flight cost was $1.5B if you include all spending on the Shuttle program from 1971-2011. If only the Shuttle program spending from 1982-2011 is used, i.e. leave out development and the initial test flights, the average was $1.2B. See tinyurl.com/3rqawwh
Griffin wants to pretend that the fixed costs/overhead are somehow separate and would be there even if NASA flew no manned flights. That is nonsense.
If, say, Boeing charged NASA $300M for each CST-100/Atlas V flight, people would laugh if they said that the real cost was just $100M because only the marginal cost counts. NASA and Griffin’s use of marginal costs are just as ridiculous.
@Clark: “That is just NASA’s marginal cost estimate. It doesn’t include the fixed costs/overheard to run the Shuttle program.The average Shuttle flight cost was $1.5B if you include all spending on the Shuttle program from 1971-2011.”
Marcel was responding to a comment by Trent in which he used $1.5 billion to calculate the cost of an additional flight.
“Marcel was responding to a comment by Trent in which he used $1.5 billion to calculate the cost of an additional flight.”
Marcel was talking about “four to six” additional launches. NASA struggled mightily to achieve its annual launch rate. Even if one believes that NASA might have squeezed in one additional launch each year with an extra $450M, to go beyond that would have required a big expansion in facilities, manpower, and additional Shuttles, and thus a big increase in its fixed costs and overhead.
There may have been some economies of scale in ET production and a few other items but the number of facilities & man-hours per Shuttle turnaround was essentially incompressible. For NASA to have doubled its Shuttle flight production, it would have had to nearly double its “factory” size.
I’ll also note that marginal cost calculations are informative when a factory is producing large numbers of widgets and one wants to know how much in materials, power, etc costs there are for each additional widget. For small unit production like shuttle flights, it is the average cost that is the significant number. The average cost of an additional Shuttle flight would have been more than $450M. If, e.g., there were 4 flights per year with a $4B annual budget and 5 flights for $4.45B, then the average cost goes from $1B to $890M.
@Clark:
Marcel was talking about ‘four to six’ additional launches.
To which Trent replied: “[Shuttle] cost about $1.5B/launch. Four to six launches per year would be $6B to $9B per year.” Marcel was responding to this point.
Four to six additional shuttle launches is 7 to 9 launches a year. In 1994 and 1995, the US managed 8 launches under a total HSF budget of about $7.3 billion (2011 USD). This is $2 billion or so more than space flight budget for 2010.
NASA struggled mightily to achieve its annual launch rate. Even if one believes that NASA might have squeezed in one additional launch each year with an extra $450M, to go beyond that would have required a big expansion in facilities, manpower, and additional Shuttles, and thus a big increase in its fixed costs and overhead.
Yes, with diminishing fleet and facilities that’s undoubtedly true. Hell, you can say with all honesty today that we’d have to expand the fleet and facilities to launch an additional shuttle; we’ve shut them all down.
There may have been some economies of scale in ET production and a few other items but the number of facilities & man-hours per Shuttle turnaround was essentially incompressible. For NASA to have doubled its Shuttle flight production, it would have had to nearly double its “factory” size.
Or replaced Columbia and retained the fleet. The point is that $1.5 billion is not the marginal cost of launch; it’s the average cost per launch over the life of the program. Still an important figure, but one not driven solely by the architecture’s built in costs but also by the customer’s use. This was the point I believe Marcel was trying to make. I don’t buy that $450 million funds a launch de novo (the appropriation for an additional flight in 2011 was $620 million). I also don’t believe that its a fixed, known cost. I also don’t buy that $38 billion for a lower end super-heavy lifter is going to overcome two orders of magnitude difference in launch costs by increasing flight rates. But on that last point, I could be wrong.
I’ll also note that marginal cost calculations are informative when a factory is producing large numbers of widgets and one wants to know how much in materials, power, etc costs there are for each additional widget.
Specifically, it’s fundamental to calculating your minimum recurring cost.
For small unit production like shuttle flights, it is the average cost that is the significant number. The average cost of an additional Shuttle flight would have been more than $450M. If, e.g., there were 4 flights per year with a $4B annual budget and 5 flights for $4.45B, then the average cost goes from $1B to $890M.
Doesn’t seem to be that straightforward a calculation, if only because NASA could execute 8 flights per year in 1997 for $2 billion (2011 USD) more than the $5 billion spent on 3 flights this past year.
I like to think of launch costs in SpaceX years using one billion per decade. So a single shuttle launch was 50% to 120% of a ten year.
I also don’t buy that $38 billion for a lower end super-heavy lifter is going to overcome two orders of magnitude difference in launch costs by increasing flight rates.
A common fallacy says that if you “invest” in an expensive system, you will automatically get low operating costs.
In fact, the opposite tends to be true because once you buy a system, you have maintenance and upkeep.
Upkeep costs are a percentage of the original purchase cost. Expensive systems tend to have expensive upkeep. Buying a Ferrari does not mean you will have low maintenance costs.
With good design, you can reduce the unkeep percentage, but you can never eliminate it. With super-heavy lifters, a large part of the investment is going to be in launch facilities and infrastructure — vehicle assembly buildings, launch pads, etc. Those are mature technologies, and it is unlikely you will be able to reduce the upkeep costs very much.
The late Dr. Max Hunter stated that low development costs and low operating costs go hand in hand — reducing operating costs will tend to reduce development costs, and vice versa. He appeared to prove that with DC-X, and it has been proven since by Burt Rutan, XCOR, Armadillo, etc.
Still, the myth of the big expensive rocket with low operating costs persists, even though a specimen has not been seen in the real world. I suggest calling it “Bigfoot.”
The $1.2 Billion cost per Space Shuttle flight is probably comparable to the per flight costs of the Saturn V rocket.
NASA-HQ asked NASA-Marshall to design SLS for a per launch cost of $900-Million (i.e. the SLS budget of $1.8-Billion per year divided by 2 flights per year when SLS becomes operational) , and NASA-Marshall supposedly found this goal difficult to meet.
I remember NASA contractors telling me a decade ago that their proposals for a shuttle-derived heavy lifter had per launch costs around $1.2-Billion.
It is likely that the Space Shuttle was launched at over $1.2-Billion per mission and over $200-Million per Astronaut.
NASA will be getting a bargain at $65-Million per Astronaut from the Russians and possibly $20-Million per Astronaut from SpaceX. NASA is already saying that it will double ISS utilization to 4 American Astronauts rotating every 3 to 6 months when the new commercial crew vehicles are available in 2016.
NASA needs to turn all spaceflight over to the commercial sector, and NASA needs to re-structure itself by reinforcing its successes within COTS, CRS, and CCDEV 1 & 2.
@Edward Wright:
The late Dr. Max Hunter stated that low development costs and low operating costs go hand in hand — reducing operating costs will tend to reduce development costs, and vice versa. He appeared to prove that with DC-X, and it has been proven since by Burt Rutan, XCOR, Armadillo, etc
I’d also argue that a low operating cost generates a high flight rate, because more missions become affordable, as opposed to the thinking during Shuttle development, that a high flight rate would produce low operating costs.
It is a moot point now the shuttles are gone but the sunk costs of shuttle development should not be used when evaluating alternatives.
The costs associated with operating the shuttles in the past or what they would be if we could magically have the production lines in place to continue launches next month, are hard to nail down. Which is why it would be nice to deal with price instead of cost.
Sunk costs are sunk, yes. SLS development costs not yet spent are not yet sunk, so they absolutely should be considered in the life cycle costs of the SLS.
People who’ve destroyed wonderful dreams about traveing through outer space:
Rick Berman, Brannon Braga, George Lucas, and Mike Griffin.
Do ya think Joss Whedon would accept a post as NASA administrator? Of course, that might have a downside… vampires in space.
As opposed to the blood-suckers here on earth?
“Sunk costs are sunk, yes. SLS development costs not yet spent are not yet sunk, so they absolutely should be considered in the life cycle costs of the SLS.”
Yup. People were talking about the cost of a shuttle launch and using sunk costs as part of the cost of a shuttle launch. Which in the bizzaro world of the federal government might make sense.
Figuring out the cost of an SLS launch will be just as convoluted as figuring out the cost of a shuttle launch. Maybe even more so. At least with cots and ccdev, we will be able to deal with the price of a contract instead of the costs, which depending on accounting gimmickry can come up with conflicting numbers.
I am not really sold on the SLS. We don’t know where we are going or what we will be doing there. We also don’t know what an alternative would look like. So while we know that the SLS will be expensive, we really don’t know how expensive an alternative would be to accomplish the same goals.
The various “constructed” numbers on Shuttle or SLS costs don’t matter. Total program cost divided by number of flights tells you nothing about next year’s budget requirements. “Incremental” costs, such as “it would only cost $100M to add a shuttle launch to the manifest, except that we are already launching as many as we can, and we’d have to hire more people or upgrade facilities to launch another one, so we can’t” also tell you nothing.
What matter is the ANNUAL FIXED COST to keep the program alive. For Shuttle, that was between $2B and $3B a year depending on how you account for the infrastructure. At first glance, SLS seems to shift costs (no orbiter processing, but a lot more in engine manufacturing), but not reduce them. Therefore, there is no reason to expect it to be any cheaper.
When people start arguing over “total cost”, “life cycle cost”, or “marginal cost”, they are hiding the ball.
not really sold?
Hey, I’ve got a bridge for ya.
ken anthony Says:
“Hey, I’ve got a bridge for ya.”
I am not really sold on your bridge selling pitch.
But the bridge is one of Obama’s infrastructure/jobs projects and he’s going to bill you for it anyway. That’s why you might as well list “sucker” as your job title when you fill out your 1040.
Not sure how anything I wrote could be interpreted as being suckered or in favor of the SLS.
I am not really sold on the SLS.
Is a pretty weak negative. You could do better than that. Let the real Wodun out. Vikings weren’t know for being wimps (got some Swedish in me.)
See Bill Whittle re: credibility in his latest. via Glenn Reynolds
Plus he describes operating costs as a tax… unbelievable. You allocate operating cost first in a budget. Everything else gets what’s left. If you are going to use a building, the lights stay on and are part of the fixed costs of a budget. Actually, operating costs are second to determining what is or is not going to be part of your budget. Once you decide that, then fixed costs come first.
To look at it any other way is deranged.
It is the operating costs of the SLS that make it catastrophically bad policy. Whether the development cost is $20B, $30B, or $40B will, in the long run, not matter so much (though why we would choose that over $1B or $3B options like Falcon Heavy and Atlas Phase 2 would be hard to defend). But SLS will keep the operating costs of NASA high in to the indefinite future, until is is canceled.
For Mike Griffin to rail against high operating costs, and then to advocate a strategy designed to MAXIMIZE future operating costs, is indefensible.
Ah, but that’s part of the magic of budgets, future operating cost beyond the budget window can be completely ignored if it helps make a case. You are absolutely right that it is totally indefensible.
Which goes to credibility. This country has a serious problem about not taking credibility into account.
The other question I would like to ask Mr. Grifffin is this: if the SLS budget is to be increased by $400 million a year for the next two decades, what other NASA programs would he raid to find the money?
AIAA elected Griffin president – again – so I’m wondering whether to renew my membership…
Operating cost for the SLS, like any other rocket system, will depend on how frequently its used. Under the Obama plan of limited SLS launches punctuated by several years of no launches at all, SLS cost will be appallingly high.
However, if the SLS is used for a lunar base program then probably at least four to six launches will be required per year. Under that scenario, SLS cost should be close to that of the space shuttle– not including the additional cost of the lunar lander and payload.
If the SLS is also used to launch the largest Bigelow space stations (BA-2100) to LEO, smaller storm shelter stations (BA-330) to L1, space depots to LEO and L1, and space solar power plants to GEO then cost per launch should be even lower.
Marcel, the numbers are in on Shuttle.. it cost about $1.5B/launch. Four to six launches per year would be $6B to $9B per year. There’s no budget for that. So either you meant the fictional numbers of the shuttle that have been thrown around for years or you’re continue to labor under the delusion that the NASA budget is going to go up sometime in the future.
“However, if the SLS is used for a lunar base program then probably at least four to six launches will be required per year. ”
If. Congress shows no inclination to fund such a program, which would require more funding for launching SLS at a higher flight rate, even more for the payloads to be launched, and a substantial R&D program for lunar landers that do not currently exist and the same for Lunar surface systems, ditto.
The appallingly low flight rate for SLS is a direct result of the funding that is likely to available for the payloads and launcher.
Absent a credible plan to increase the total funding, the desirability of a NASA unique launcher with high fixed costs should be reexamined.
Do the SLS cost estimates include the cost of money?
Most of the discussions I see of NASA vehicle development costs appear to calculate the total cost as the sum of the costs for Years 1 to N.
That analysis ignores Net Present Value, however. The NPV of dollar spent in year 1 is not the same as a dollar spent in year N. Similarly with benefits received.
An accurate analysis would account for the cost of money by applying an internal discount rate. Such an analysis assumes that development dollars are “borrowed” from the Treasury at the prime rate and not “paid back” until the benefits are received (i.e., the launches actually occur).
If the estimates that are being bandied about do not include the cost of money, then SLS is an even worse investment than it appears.
@Trent Waddington: “Marcel, the numbers are in on Shuttle.. it cost about $1.5B/launch. Four to six launches per year would be $6B to $9B per year.”
$1.5 billion is the cost of the entire program over the total number of launches. I’ve yet to see an authoritative cite for the marginal cost of an additional launch. NASA’s FAQ cites $450 million for “launch,” though it’s unclear as to whether that includes every dollar increment over annual fixed costs.
@Marcel: “Operating cost for the SLS, like any other rocket system, will depend on how frequently its used.”
Obviously not, or you’d end up with a rocket costing infinitely close to $0 to fly as you schedule more and more launches. Each mission will have some incremental cost in addition to the fixed baseline.
@Edward: “Do the SLS cost estimates include the cost of money?”
If all goes according to the Book, then yes.
@McLean: “Absent a credible plan to increase the total funding, the desirability of a NASA unique launcher with high fixed costs should be reexamined.”
An increase in total funding isn’t strictly necessary, but reconfiguring the budget to support SLS and manned lunar exploration R&D leads to politically untenable choices. Ironically, one of those sacred cows, the ISS, is a key crowbar for NASA to pry funding for commercial crew.
@Trent Waddington
“Marcel, the numbers are in on Shuttle.. it cost about $1.5B/launch. Four to six launches per year would be $6B to $9B per year. There’s no budget for that. So either you meant the fictional numbers of the shuttle that have been thrown around for years or you’re continue to labor under the delusion that the NASA budget is going to go up sometime in the future.”
Shuttle cost were about $450 million per launch. Check the NASA website.
The $1.5 billion per launch is only if you include the development cost from the 1970s (money already spent). But NASA was not spending $7.5 billion a year ($1.5 billion per flight) to launch five shuttles in 2009.
Obama inherited an $8.4 billion a year budget related to the manned space program (space shuttle, ISS, and Constellation program). That’s $84 billion over a ten year period: plenty of money to develop a $38 billion dollar (worst case scenario) SLS/MPCV since there is no $3 billion a year shuttle program nor any $3.4 billion a year Constellation program even with $30 billion spent on the ISS pork program over the next ten years since Obama raised spending on the ISS from $2 billion to about $3 billion per year.
@Marcel: “The $1.5 billion per launch is only if you include the development cost from the 1970s (money already spent). But NASA was not spending $7.5 billion a year ($1.5 billion per flight) to launch five shuttles in 2009.”
I’m not sure what the $450 million entails, but given that the line item for Shuttle spending frequently amounts to $3-3.5 billion annually, I’m uncertain how to break that down as recurring costs do to a) scheduled amortization, b) capitalization per launch, and c) actual cost of putting the Shuttle on the pad and lighting a match.
plenty of money to develop a $38 billion dollar (worst case scenario) SLS/MPCV since there is no $3 billion a year shuttle program nor any $3.4 billion a year Constellation program
So, replacing the Shuttle and Constellation with an even bigger, more expensive rocket will somehow save money?
That seems, um, rather improbable. 🙂
Where do so many space cadets get the idea that the way to make things less expensive is to spend money?
Where do so many space cadets get the idea that the way to make things less expensive is to spend more money?
What I meant to say.
“Shuttle cost were about $450 million per launch. Check the NASA website.”
That is just NASA’s marginal cost estimate. It doesn’t include the fixed costs/overheard to run the Shuttle program.The average Shuttle flight cost was $1.5B if you include all spending on the Shuttle program from 1971-2011. If only the Shuttle program spending from 1982-2011 is used, i.e. leave out development and the initial test flights, the average was $1.2B. See tinyurl.com/3rqawwh
Griffin wants to pretend that the fixed costs/overhead are somehow separate and would be there even if NASA flew no manned flights. That is nonsense.
If, say, Boeing charged NASA $300M for each CST-100/Atlas V flight, people would laugh if they said that the real cost was just $100M because only the marginal cost counts. NASA and Griffin’s use of marginal costs are just as ridiculous.
@Clark: “That is just NASA’s marginal cost estimate. It doesn’t include the fixed costs/overheard to run the Shuttle program.The average Shuttle flight cost was $1.5B if you include all spending on the Shuttle program from 1971-2011.”
Marcel was responding to a comment by Trent in which he used $1.5 billion to calculate the cost of an additional flight.
“Marcel was responding to a comment by Trent in which he used $1.5 billion to calculate the cost of an additional flight.”
Marcel was talking about “four to six” additional launches. NASA struggled mightily to achieve its annual launch rate. Even if one believes that NASA might have squeezed in one additional launch each year with an extra $450M, to go beyond that would have required a big expansion in facilities, manpower, and additional Shuttles, and thus a big increase in its fixed costs and overhead.
There may have been some economies of scale in ET production and a few other items but the number of facilities & man-hours per Shuttle turnaround was essentially incompressible. For NASA to have doubled its Shuttle flight production, it would have had to nearly double its “factory” size.
I’ll also note that marginal cost calculations are informative when a factory is producing large numbers of widgets and one wants to know how much in materials, power, etc costs there are for each additional widget. For small unit production like shuttle flights, it is the average cost that is the significant number. The average cost of an additional Shuttle flight would have been more than $450M. If, e.g., there were 4 flights per year with a $4B annual budget and 5 flights for $4.45B, then the average cost goes from $1B to $890M.
@Clark:
To which Trent replied: “[Shuttle] cost about $1.5B/launch. Four to six launches per year would be $6B to $9B per year.” Marcel was responding to this point.
Four to six additional shuttle launches is 7 to 9 launches a year. In 1994 and 1995, the US managed 8 launches under a total HSF budget of about $7.3 billion (2011 USD). This is $2 billion or so more than space flight budget for 2010.
Yes, with diminishing fleet and facilities that’s undoubtedly true. Hell, you can say with all honesty today that we’d have to expand the fleet and facilities to launch an additional shuttle; we’ve shut them all down.
Or replaced Columbia and retained the fleet. The point is that $1.5 billion is not the marginal cost of launch; it’s the average cost per launch over the life of the program. Still an important figure, but one not driven solely by the architecture’s built in costs but also by the customer’s use. This was the point I believe Marcel was trying to make. I don’t buy that $450 million funds a launch de novo (the appropriation for an additional flight in 2011 was $620 million). I also don’t believe that its a fixed, known cost. I also don’t buy that $38 billion for a lower end super-heavy lifter is going to overcome two orders of magnitude difference in launch costs by increasing flight rates. But on that last point, I could be wrong.
Specifically, it’s fundamental to calculating your minimum recurring cost.
Doesn’t seem to be that straightforward a calculation, if only because NASA could execute 8 flights per year in 1997 for $2 billion (2011 USD) more than the $5 billion spent on 3 flights this past year.
I like to think of launch costs in SpaceX years using one billion per decade. So a single shuttle launch was 50% to 120% of a ten year.
I also don’t buy that $38 billion for a lower end super-heavy lifter is going to overcome two orders of magnitude difference in launch costs by increasing flight rates.
A common fallacy says that if you “invest” in an expensive system, you will automatically get low operating costs.
In fact, the opposite tends to be true because once you buy a system, you have maintenance and upkeep.
Upkeep costs are a percentage of the original purchase cost. Expensive systems tend to have expensive upkeep. Buying a Ferrari does not mean you will have low maintenance costs.
With good design, you can reduce the unkeep percentage, but you can never eliminate it. With super-heavy lifters, a large part of the investment is going to be in launch facilities and infrastructure — vehicle assembly buildings, launch pads, etc. Those are mature technologies, and it is unlikely you will be able to reduce the upkeep costs very much.
The late Dr. Max Hunter stated that low development costs and low operating costs go hand in hand — reducing operating costs will tend to reduce development costs, and vice versa. He appeared to prove that with DC-X, and it has been proven since by Burt Rutan, XCOR, Armadillo, etc.
Still, the myth of the big expensive rocket with low operating costs persists, even though a specimen has not been seen in the real world. I suggest calling it “Bigfoot.”
The $1.2 Billion cost per Space Shuttle flight is probably comparable to the per flight costs of the Saturn V rocket.
NASA-HQ asked NASA-Marshall to design SLS for a per launch cost of $900-Million (i.e. the SLS budget of $1.8-Billion per year divided by 2 flights per year when SLS becomes operational) , and NASA-Marshall supposedly found this goal difficult to meet.
I remember NASA contractors telling me a decade ago that their proposals for a shuttle-derived heavy lifter had per launch costs around $1.2-Billion.
It is likely that the Space Shuttle was launched at over $1.2-Billion per mission and over $200-Million per Astronaut.
NASA will be getting a bargain at $65-Million per Astronaut from the Russians and possibly $20-Million per Astronaut from SpaceX. NASA is already saying that it will double ISS utilization to 4 American Astronauts rotating every 3 to 6 months when the new commercial crew vehicles are available in 2016.
NASA needs to turn all spaceflight over to the commercial sector, and NASA needs to re-structure itself by reinforcing its successes within COTS, CRS, and CCDEV 1 & 2.
@Edward Wright:
I’d also argue that a low operating cost generates a high flight rate, because more missions become affordable, as opposed to the thinking during Shuttle development, that a high flight rate would produce low operating costs.
It is a moot point now the shuttles are gone but the sunk costs of shuttle development should not be used when evaluating alternatives.
The costs associated with operating the shuttles in the past or what they would be if we could magically have the production lines in place to continue launches next month, are hard to nail down. Which is why it would be nice to deal with price instead of cost.
Sunk costs are sunk, yes. SLS development costs not yet spent are not yet sunk, so they absolutely should be considered in the life cycle costs of the SLS.
People who’ve destroyed wonderful dreams about traveing through outer space:
Rick Berman, Brannon Braga, George Lucas, and Mike Griffin.
Do ya think Joss Whedon would accept a post as NASA administrator? Of course, that might have a downside… vampires in space.
As opposed to the blood-suckers here on earth?
“Sunk costs are sunk, yes. SLS development costs not yet spent are not yet sunk, so they absolutely should be considered in the life cycle costs of the SLS.”
Yup. People were talking about the cost of a shuttle launch and using sunk costs as part of the cost of a shuttle launch. Which in the bizzaro world of the federal government might make sense.
Figuring out the cost of an SLS launch will be just as convoluted as figuring out the cost of a shuttle launch. Maybe even more so. At least with cots and ccdev, we will be able to deal with the price of a contract instead of the costs, which depending on accounting gimmickry can come up with conflicting numbers.
I am not really sold on the SLS. We don’t know where we are going or what we will be doing there. We also don’t know what an alternative would look like. So while we know that the SLS will be expensive, we really don’t know how expensive an alternative would be to accomplish the same goals.
The various “constructed” numbers on Shuttle or SLS costs don’t matter. Total program cost divided by number of flights tells you nothing about next year’s budget requirements. “Incremental” costs, such as “it would only cost $100M to add a shuttle launch to the manifest, except that we are already launching as many as we can, and we’d have to hire more people or upgrade facilities to launch another one, so we can’t” also tell you nothing.
What matter is the ANNUAL FIXED COST to keep the program alive. For Shuttle, that was between $2B and $3B a year depending on how you account for the infrastructure. At first glance, SLS seems to shift costs (no orbiter processing, but a lot more in engine manufacturing), but not reduce them. Therefore, there is no reason to expect it to be any cheaper.
When people start arguing over “total cost”, “life cycle cost”, or “marginal cost”, they are hiding the ball.
not really sold?
Hey, I’ve got a bridge for ya.
ken anthony Says:
“Hey, I’ve got a bridge for ya.”
I am not really sold on your bridge selling pitch.
But the bridge is one of Obama’s infrastructure/jobs projects and he’s going to bill you for it anyway. That’s why you might as well list “sucker” as your job title when you fill out your 1040.
Not sure how anything I wrote could be interpreted as being suckered or in favor of the SLS.
I am not really sold on the SLS.
Is a pretty weak negative. You could do better than that. Let the real Wodun out. Vikings weren’t know for being wimps (got some Swedish in me.)
Ya, it was weakly negative. Fair enough.