Congress, NASA, White House

How do you pay for JWST?

The cost increases and schedule delays associated with the James Webb Space Telescope (JWST) have become a major concern in the scientific community and beyond, as best illustrated when the House Appropriations Committee offered no funding for the program in its FY2012 appropriations bill, which is currently pending consideration by the full House. Although the Senate has proposed $530 million for JWST in its FY12 appropriations bill, there is still the open question of how NASA proposes to cover the costs of JWST over the long haul, through its planned launch in 2018. One member of Congress is now openly looking for answers.

In a letter Wednesday to Jacob Lew, director of the Office of Management and Budget, Rep. Frank Wolf (R-VA), chairman of the Commerce, Justice, and Science subcommittee of the House Appropriations Committee, wants answers from the administration on its plans for paying for JWST. “While acknowledging that substantial cuts will be necessary, the Administration has so far failed to identify a single specific proposal to offset the increase in JWST spending above the levels contained in the President’s fiscal year 2012 request,” he wrote, referring to cots in other NASA programs to cover the costs of JWST. “Either no offsets have been proposed because JWST really isn’t a top priority, or the Administration is hoping that remaining silent will force Congress to act unilaterally and thereby take sole ownership of the cuts necessitated by the Administration’s actions.” To emphasize his concern, Wolf scrawled below his signature on the copy of the letter the words, “This is very important.”

NASA has started to identify where those funds would come from. In a webinar last week by the Space Telescope Science Institute that covered both the scientific potential of the program as well as its management, project officials discussed their current plans for covering the project’s costs in 2012 and beyond. “The replan is on track to support the ’13 budget process,” Rick Howard, program director for JWST at NASA Headquarters, said. “All the details will be rolled out in February [2012] when the president’s FY13 budget is released.”

Howard said that JWST needs $1.223 billion above the administration’s budget projections for the program (about $355-370 million per year) from 2012 through launch (now planned for October 2018, or the beginning of fiscal year 2019). That includes an additional $156 million in FY12. Half of that money, he said, would come from NASA’s Science Mission Directorate (SMD), although earth science programs would be exempt from any cuts. The other half comes from “agency institutional support”, a reference to the Cross-Agency Support line in the budget, which is about $3.2 billion in FY12. The specific programs in SMD and Cross-Agency Support that would lose money “is still being worked,” he said. Likewise, he added, finding the additional money needed in FY2013 and beyond hasn’t been determined.

In the meantime, JWST is still something of a punching bag for people who want to criticize cost overruns on NASA programs or government programs in general. “Right now, as all of you are probably aware, there is considerable pressure on Congress to be a better steward of the people’s money,” Rep. Ken Calvert (R-CA) said in a luncheon speech Wednesday at the AIAA Space 2011 conference in Long Beach, California. “This means we need to figure out a way to end the days of overbudget and underdelivery. A perfect example of that is the James Webb Space Telescope.” Citing the cost growth in the program, he blamed “a government-wide acquisition problem that couples unrealistic government specifications with overpromising by industry.” Ironically, Calvert made the comments at a luncheon sponsored by Northrop Grumman, the prime contractor for JWST; all the luncheon attendees received a tote bag emblazoned with an illustration of JWST.

40 comments to How do you pay for JWST?

  • Robert G. Oler

    How do you pay for JWST?

    Just say no

    RGO

  • vulture4

    I know where they could get $3B a year for JWST without losing anything useful….

  • pr

    “”Ironically, Calvert made the comments at a luncheon sponsored by Northrop Grumman,”

    Courageously would be a more appropriate word there.

    he blamed “a government-wide acquisition problem that couples unrealistic government specifications with overpromising by industry.”

    That hits the nail right on the head. The government wants everything in every program, and the contractors know if they low ball it they might win, but if they’re honest they’re sure to lose. Once the contract is awarded they have every incentive to drag it out and pad bills.

    My proposal: every line in the spec needs a dollar value next to it, preferably with a linear function that punishes for for going under and rewards for going over, up to the point where nobody cares anymore. When the contractor delivers product and it meets that spec, they get paid per the contract. No more, no less.

    That way the government has to figure out what’s really valuable before the RFP goes out, and when the contractor figures out that it costs more than it’s worth to meet a spec (no matter how sincerely they wanted to) thye can drop it. Makes it easy to de-scope programs long before it gets out of hand. the contractor is on the hook for final performance and that’s what they get paid for, rather than for how much money they spent.

  • vulture4

    It does no good to get a precise estimate, when the estimate is biased. Time and again we go through budgets in extreme detain, spending a lot of time and money. But if there is a reward for making it low (in bidding) or high (once the contract is signed) then you can find a thousand reasons to do it.

    The best you can do is to get an _unbiased_ estimate. Often someone with experience can listen to a 10-minute description of a major project and give a cost accurate within 20%… if he/she has no incentive except to see science progress and the country benefit. This is the job of the agency manager, who has to have judgement and integrity.

    The vast array of rules surrounding contracting work as long as you are just buying cement or gasoline. When you are buying R&D they are expensive and useless.

  • Rhyolite

    “How do you pay for JWST?”

    Cancel the present design and replace it with a smaller, simpler, less risky one. In other words, keep the name plate and change the telescope.

  • Coastal Ron

    pr wrote @ September 29th, 2011 at 11:38 am

    Courageously would be a more appropriate word there.

    Northrop Grumman only got 2.27% of NASA’s contractor money last year, so they don’t have much to lose by providing a forum for someone to point the finger at their competitors (LM, Boeing & ATK).

    My proposal: every line in the spec needs a dollar value next to it…

    Essentially you’re proposing Firm Fixed Price, which has always been an option for NASA. But FFP doesn’t work for products that are new or evolving, as NASA doesn’t know what it doesn’t know for what it needs in cutting-edge vehicles.

    And don’t think that FFP is a magic bullet either, since government contractors excel at add-ons and change orders, both warranted and un-warrented.

    Whatever the solution, without the threat of consequences there can be no fear of cancellation, and that is the single largest leverage you can have with a contractor.

  • Rhyolite

    “I know where they could get $3B a year for JWST without losing anything useful….”

    JWST is certainly a better use of money than SLS but it is not clear to me that the current form of JWST is a good use of our limited space science dollars on its own merits. The concept of an origami space telescope is really neat but it is also very high risk. It would seem better suited for a technology demonstration mission than a flagship observatory.

    The continually rising price underscores that. Spending $6.5 billion on a single mission is bad. But having the mission rendered useless by the failure of a previously untried deployment mechanism would be much, much worse. It’s a very real possibility.

    I would rather see us build a 4 m, monolithic telescope that is low risk than potentially lose a whole generation of research opportunities by trying to get a little larger aperture. By all means, let’s test deployable telescope apertures but not on the critical path.

  • Vladislaw

    Coastal Ron wrote:

    “Essentially you’re proposing Firm Fixed Price, which has always been an option for NASA. But FFP doesn’t work for products that are new or evolving, as NASA doesn’t know what it doesn’t know for what it needs in cutting-edge vehicles.”

    Can you also break the project up into milestones? Where the contractor is on the hook for completing the milestone first with their money, rather than the taxpayer’s money? I know with R&D you can not always do this because there will be times you can not break it down to a single milestone. But requiring that the contractor has to get some skin into the game on the front end gives them some more incentive to actually produce some hardware.

    If they have absolutely no risk for failure and every incentive to not care if something goes incredibly wrong all we will get is more of the same?

  • DCSCA

    “How do you pay for JWST?”

    You don’t.

    Memo to the Super Committee: kill it.
    NASA’s contribution made.

  • Das Boese

    Rhyolite wrote @ September 29th, 2011 at 12:43 pm

    I would rather see us build a 4 m, monolithic telescope that is low risk than potentially lose a whole generation of research opportunities by trying to get a little larger aperture.

    Or an array of them, which gets you that larger aperture while still being useful during construction or if one unit fails.

  • pr

    “My proposal: every line in the spec needs a dollar value next to it…”

    “Essentially you’re proposing Firm Fixed Price, which has always been an option for NASA. But FFP doesn’t work for products that are new or evolving, as NASA doesn’t know what it doesn’t know for what it needs in cutting-edge vehicles.”

    No, I’m not proposing firm fixed price. FFP says “meet absolutely every one of these specs, or get zero.” If meeting ONE spec costs a billion dollars, you have to do it. Or get nothing.

    Or more likely, go beg for a waiver. The government wants the product, so you probably get it, but only after much torture to prove that it can’t be done.

    I’m proposing something that has a built-in de-scope method. If you’ve just figured out that meeting spec 45682 is going to cost $10 million, but the government has already told you that it’s worth $2 million to them, then you deliver what you have and take the $2 mil hit in payment, rather than the $10 mil hit in cost.

    As important, it forces the upstream users (scientists, or whatever) to prioritize capabilities ahead of time, and figure out what’s really valuable and what’s not.

    Whether the details of this woolly notion stand up, the bottom line is that there has to be a reasonable way to cap programs other than just canceling them and taking delivery of a bucket of bolts. And the contractors need to get paid for what they deliver, rather than for what they spend.

  • amightywind

    JWST is certainly a better use of money than SLS but it is not clear to me that the current form of JWST is a good use of our limited space science dollars on its own merits.

    Yeah, I’d much rather continue to spend $3 billion a year doing rat experiments the ISS. Finding money to fund JWST should be easy.

  • amightywind

    Musk is hallucinating again. Here is a video of a flyback F9 booster, laws of physics be damned. I found it on Breitbart.

    http://www.spacex.com/assets/video/spacex-rtls-green.mp4

    Not sure how a booster would fly 500 mi retrograde to the launch site after delivering the second stage to speed and altitude. It is surprising what you can do with some of the lowest performing engines in the industry. The video graphics are first rate, however.

  • KDP

    As Rhyolite said, “. . . having the mission rendered useless by the failure of a previously untried deployment mechanism . . .”

    A wise man once said, “We are going to build a little, fly a little . . . etc”

    In contrast there is something very wrong with an organization run by people who sit at CAD stations and design things and then expect them to work as they “decreed” without constantly testing every incremental change. The unproven complexity in the JWST or that Rube Goldberg design for the Curiosity rover’s landing crane has me holding my breath.

    “In theory” these things could work, but anybody who has built anything more complex than a doghouse knows that the real world has a habit of taking academic arrogance and spitting it back out.

    All this from people who mixed up imperial and metric units and didn’t check the Hubble’s mirror before they launched it.

    I wonder how many of them get out of the office and get their hands dirty!

    I truly hope I am wrong, but I am expecting to see more expensive failures.

    Ken

  • Marc Trolinger

    I would have to agree with RGO, just say no. However, there has to be a way forward that makes use of the work already done in a fashion that yields something of value.

  • Robert G. Oler

    amightywind wrote @ September 29th, 2011 at 4:56 pm

    “Not sure how a booster would fly 500 mi retrograde to the launch site after delivering the second stage to speed and altitude.”

    for someone whose “expertise” had the Falcon9 second stage not making orbit…well anyway…think outside the box. tick tock RGO

  • common sense

    @ amightywind wrote @ September 29th, 2011 at 4:56 pm

    “Not sure how a booster would fly 500 mi retrograde to the launch site after delivering the second stage to speed and altitude.”

    There are way more problems than just that.

    Yet you can use a little SLS technology, i.e. faith. I guess.

    Oh well…

  • Coastal Ron

    pr wrote @ September 29th, 2011 at 2:43 pm

    Whether the details of this woolly notion stand up, the bottom line is that there has to be a reasonable way to cap programs other than just canceling them and taking delivery of a bucket of bolts. And the contractors need to get paid for what they deliver, rather than for what they spend.

    I agree with your sentiment, though not necessarily the method.

    Having spent part of my career working for government contractors, mainly on new programs, the challenge is that rarely does the government ever stick to their initial set of specs. This is both a blessing and a curse, since change orders can add profit, but making changes after something is designed can be messy. That’s why Cost-Plus Contracting is still around, despite the abuse challenges.

    If we look back on both Constellation and JWST, part of the problems could have been addressed if there was proper oversight along the way. Griffin certainly didn’t mind the cost growth of Ares I and Orion, nor the JWST, so that was certainly one of the areas where management failed.

    One possible solution would be to have an auditor group that releases public scorecards of how the programs are doing. The GAO only does this on request, so this would be more of a permanent group. That way there is no excuse for “we didn’t know” situations where suddenly the program has used up it’s reserve and needs to beg for more budget. And you really only need to do this at a high level, since it’s the trends you’re most concerned about.

    The reality is that sometimes the politics of the situation are that no one in the Administration and Congress cares if something goes over budget. We, the U.S. Taxpayers care, so how do we shine a light on these problems and keep our elected officials accountable?

    Tough problem to solve.

  • vulture4

    All the talk about JWST seems to reference total cost, and include five years of operations. Any economist will tell you that sunk cost is irrelevant in making a decision. What is the _difference_ in cost between shuttling the program down now and continuing to operational status, exclusive of operations after launch? That’s what we should consider now.

    As to reuse of the Falcon, the concept of the “rocketback” maneuver has been around for decades and the theory, at least, is solid. Throughout the Shuttle program the RTLS, or Return to Launch Site, maneuver has been part of the potential contingency response. After dropping the SRBs the Shuttle would do a pitchover and utilize main engine thrust to reverse the velocity vector. Obviously it will be much safer to evolve the maneuver with unmanned systems; there are many vehicle and trajectory variables to optimize. On the other hand, the Falcon is a long ways downrange at staging. The DOD study on this advocates recovering only the booster, staging early to minimize thermal stresses, and using an expendable second stage.

  • Mr Earl

    Hold on Oler and CS, the only thing more ridiculous than windy is that fantasy put out by SpaceX. You both know it, so you have to attack the source who happened to get it right this time.
    If Musk is serious about this, I think this brings into question most of the other “financial sunshine” they’ve been feeding us over the years.

  • Jeff Foust

    A reminder that this post is about JWST, not SpaceX. Please keep your comments on topic. Thank you for your cooperation.

  • aberwyse

    Does anyone else notice that people are fighting over both SLS and JWST at the same time? What if both are killed by politics…what happens to NASA then?

    Just sayin’….

  • stargazer

    KDP wrote: “In contrast there is something very wrong with an organization run by people who sit at CAD stations and design things and then expect them to work as they ‘decreed’ without constantly testing every incremental change. The unproven complexity in the JWST or that Rube Goldberg design for the Curiosity rover’s landing crane has me holding my breath.”

    You do realize that most of the cost increase for JWST is in support of testing, right? There’s literally years of testing planned before launch, precisely because it is a complex system.

    KDP also wrote: “All this from people who mixed up imperial and metric units and didn’t check the Hubble’s mirror before they launched it.”

    And why wasn’t the Hubble mirror tested fully before launch? Because the program management back then thought that test would be too expensive and wasn’t worth doing. JWST is trying pretty hard to avoid that mistake – look at their plans for the Chamber A tests in JSC – but testing a 6.5m cryogenic telescope ain’t cheap.

    Look, either complain that you don’t think JWST will work because it’s insufficiently tested or complain about the cost, but don’t do both at the same time…

  • Rhyolite

    “What if both are killed by politics…what happens to NASA then?”

    NASA would be in a lot better position then to proceed with a rational science and exploration program. That’s probably true even if the overall budget were cut. At least then the rest of NASA would not be under constant threat of being gobbled up cost overruns.

  • Coastal Ron

    Rhyolite wrote @ September 29th, 2011 at 7:27 pm

    At least then the rest of NASA would not be under constant threat of being gobbled up cost overruns.

    I agree. Regardless the merit of the JWST, the money to finish it has to come from somewhere, which means other planned programs may have to be cut or de-scoped.

    Constellation was doing the same on the HSF side, and SLS will do the same unless it is cancelled. NASA without the SLS, even if that means the corresponding budget goes with it too, is better off in the long run. NASA will be able to focus on more sustainable HSF, which will also have the advantage of getting going a lot faster too. I see no downside to canceling the SLS.

  • Robert G. Oler

    stargazer wrote @ September 29th, 2011 at 7:19 pm

    “You do realize that most of the cost increase for JWST is in support of testing, right? There’s literally years of testing planned before launch, precisely because it is a complex system.”

    Well that is the kind way to think of it.

    In reality KDP is in my view more correct then wrong.

    First off it is accurate that the flaw in Hubble would have been found by testing, but this kind of flaw was precisely the kind of mistake that should have been found by some fool sitting behind a computer screen doing some basic calculations. Basic calculations that are known and done by anyone building a “tube” at their house.

    Who knows what kind of goofiness is being missed at the JWST level because the entire program is caught up in the complexity of the thing….

    Plus most of the vehicle cannot be “tested” in any real form on Earth, we are talking about deployments of mirrors in a microgee environment with massive thermal issues…and while they can be “small scale” tested and parts here and there tested a full up test is near impossible…yet this is what we would commit the bulk of space astronomy science to.

    The problem at NASA and now it is creeping over into the uncrewed stuff (go look at the turd buckets who have left shuttle/station and moved to Webb) is that we have a bunch of people who have never really executed an engineering plan successfully (meaning achieving program objectives at a reasonable estimated cost) that is more complicated then where to do lunch.

    They are use to getting away with it because the federal till is almost (until now) unlimited but in the end we have come to the end of the till…and we should come to the end of programs that are far to ambitious and lack reasonable cost controls.

    Robert G. Oler

  • Beancounter from Downunder

    The really tricky part about the cost overrun on the JWST is about what you get once it’s flying. How do you quantify the extra cost or even total cost of the craft compared to the missions that might be cut to pay for it?

    I for one, don’t think anyone on this blog can possibly do that. It’s a highly technical issue and one that the scientific body are going to have to resolve. It seems that the Planetary boys and girls are now, belately, getting involved, and I predict it will turn pretty ugly before much longer.

  • Robert G. Oler

    Beancounter from Downunder wrote @ September 29th, 2011 at 9:26 pm

    “and I predict it will turn pretty ugly before much longer.”

    there is nothing more ugly then a disagreement concerning funding among “colleagues” (grin).

    What is hard to see right now is how the overall politics play out in terms of the deficit reduction drive by the GOP, which I think is losing steam almost everywhere….but in my view what is for certain is that the projects that were once sacred cows but now have gotten so large that they are eating everything are now starting to meet resistance among those they use to eat. The planetary folks have been here a few times.

    Robert G. Oler

  • stargazer

    Robert G. Oler wrote: most of the vehicle cannot be “tested” in any real form on Earth, we are talking about deployments of mirrors in a microgee environment with massive thermal issues…and while they can be “small scale” tested and parts here and there tested a full up test is near impossible
    [...]
    we have a bunch of people who have never really executed an engineering plan successfully (meaning achieving program objectives at a reasonable estimated cost) that is more complicated then where to do lunch

    I’ll grant you that they’ve surely not gotten the cost estimates right thus far, but Northrup Grumman does have a 100% perfect success rate for mechanism deployments on orbit. That includes radio spy sats with antennae 10x larger than any structure on JWST (though admittedly with coarser tolerances). Everybody thinks the deployment of the telescope is the scary part, but really it’s not.

    Even the phasing up of the mirror is something the ground telescope guys have been doing for 20+ years now at Keck. JWST is in some ways harder than Keck ’cause it’s cryo and the mechanisms are lightweighted, but in other ways much easier since it’s zero g and so much more stable.

    The thermal modeling is way more worrisome from a technical perspective, much harder to get right. And that’s why those tests at JSC are going to be so important. That’s the only way for the thermal teams to demonstrate they’ve got their act together. You’re right that a full scale test is near impossible, but by god they appear to be trying anyway. I got to tour that facility a couple years ago and it’s one bloody enormous chamber.

  • KDP

    Hi Stargazer,

    Thanks for the considered response. I hope you are right.

    The theme of my comment was about not about costs, but rather, not wasting money by failing — specifically, failing due to a lack of practical experience. I simply prefer that design be founded upon robust incremental experience.

    I note your response to Robert Oler addresses some of the experience issue.

    Thanks,

    Ken

  • Space news is reporting ExoMars is finished? Another JWST casualty?

    Gary Anderson
    New England Reg Coordinator
    TPIS

  • Robert G. Oler

    Gary. I have lost the bubble on ExoMars…last I heard we were providing the launch vehicle? RGO

  • Rhyolite

    “You do realize that most of the cost increase for JWST is in support of testing, right? There’s literally years of testing planned before launch, precisely because it is a complex system.”

    Testing is the right way to retire risk in a complex system but it still does not excuse the cost over runs. We knew this was a complex system from the beginning and the testing should have been included in the original estimates. We just didn’t discover that JWST was complex ten years into the project. If the full costs had been fully described upfront, a lower risk alternative may have been selected instead.

    Testing also only gets you so far. Aside from the Rumsfeldian unknown unknowns, launch vehicles fail and so do satellite buses that have been built in significant numbers. That is the nature of the business and nothing we can do will change it in the foreseeable future. Replacing a JWST failure would take the better part of a decade and cost several billion dollars. It would likely suck the life out of our space sciences community for an entire generation. Putting that many eggs in an uncertain basket is unwise.

  • David Davenport

    Let’s not let a good crisis be wasted.

    It’s time to start talking up a service mission to L2. … Astronauts Curly, Larry and Moe equipped with pry bars and hammers.

  • vulture4

    It would be better to keep the production line running. The second in a series of duplicate spacecraft would be about 10% of the cost of the first, cheaper than a servicing mission. If we start now, it should be ready for final assembly just about the time when we find out what needs to be changed.

  • E.P. Grondine

    If past experience is any guide, the neocons will stop JWST, move the work from Mikulsky’s district to a “red” one, and then start it back up again.

    Or they’ll label JWST as pork for Mikulsky.

    In the meantime, Ed Weiler will go off somewhere and work as a mouthpiece, as did Griffin.

    And of course, the neocons will blame Obama.

  • David Davenport

    It would be better to keep the production line running. The second in a series of duplicate spacecraft would be about 10% of the cost of the first,…

    That’s a splendid idea. Double up and double down on the “fancy folder” WEbb Telescope design. Just great.

    Why not re-start manufacture of Space Shuttles? Following your logic, Shuttle costs will be cheaper with each iteration, correct?

    Hey, we could save money by re-starting Saturn V production. Wait, NASA’s already trying to do that. Except they’ve changed the name.

  • If past experience is any guide, the neocons will stop JWST, move the work from Mikulsky’s district to a “red” one, and then start it back up again.

    If past experience is any guide, you’ll continue to make meaningless references to “neocons” without explaining what “neocons” are, or what they have to do with space policy.

  • David Davenport

    And of course, the neocons will blame Obama. …

    My understanding is that the Webb started getting tangled — geddit? — when Dan Goldin was the boss. Yes, the Webb gies back a ways.

    The Euro. Space Agency agreed to launch the thing on Ariane V, which can accommodate a four meter diameter shroud. The original JWST design called for a four m. nominal diameter main telescope mirror.

    Goldin allowed the mirror size to be increased to 6.5 m, which motivated the fancy folder design. Why the French haven’t tried during the past fifteen or so years to enlarge Ariane to handle a larger diameter shroud, I can’t say.

    The question is, is a 6.5 m. main mirror really necessary? Yes, I agree that bigger telescopes are better as a general principle. But would infrared astronomy keel over and die if the Webb main mirror were, say, 5 m. in diameter? ( 5 m. — OK for Delta IV or Shuttle or upsized Ariane. )

  • Rhyolite

    “Goldin allowed the mirror size to be increased to 6.5 m, which motivated the fancy folder design.”

    I think Goldin actually pushed it up to 8 m at one point but it was later descoped to 6.5 m. Incidentally, Lockheed proposed doing a 6 m monolithic telescope by enlarging the Atlas fairing to 7 m but they lost out to TRW (now NG) who proposed a folding design. In hindsight, that probably would have been a better alternative.

Leave a Reply

  

  

  

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>