United States Department of Defense United States Department of Defense

News Transcript

Press Operations Bookmark and Share


DOD Background Briefing with Senior Defense Officials from the Pentagon

Presenter: Senior Defense Officials
June 01, 2010

                MODERATOR:  Just before we get it started, let me remind everybody of the ground rules.  Here we are on background today.  You can refer to the two individuals up here -- who you have their bona fides -- as Senior Defense Official 1 to my immediate left, and further over is Senior Defense Official Number 2. 
                We are here today to talk about the Nunn-McCurdy certifications, and so we will keep it confined to that topic today.  And with that, I think Senior Defense Official 1 is going to give you a brief overview as to why we're here today and then open it up for your questions.
                SR. DEFENSE OFFICIAL 1:  Okay, well, the reason to have a Q&A session is just that this is very technical and mind-numbing.  There are press briefing packets here that you may have had time to go through, but we're making ourselves available principally to answer your questions. 
                Let me just take it from the top.  Nunn-McCurdy is a statutorily required process -- statutorily required for some decades now, the critical feature of which is the following:  We are required -- every program in the Department of Defense is required -- the program manager is required constantly to estimate the total cost of his or her program.  That's an estimate.
                So, the Joint Strike Fighter program -- for example is probably the one you're most interested in in this stack -- at the beginning of its program has produced around about 1 percent of the aircraft it will finally produce, and the program and this department estimate how much that will eventually cost and we're required constantly to make those estimates and that's one of the managerial tools that the department uses.
                When that program -- when a program manager finds that his or her estimate is half again as large in -- that is, 50 percent overrun in inflation-adjusted dollars relative to when the program started, he or she has to report that to their service secretary, say my estimates show that the unit cost, the total program cost divided by the number of things delivered, is half again as large as it was estimated to be at program onset.
                That is the so-called critical Nunn-McCurdy threshold.  And then when that happens, they report it to the service secretary, who reports it to Congress -- and these were all reported to Congress a couple of months ago.  And at that point, I need to do several things.  Obviously -- now, this is never a surprise -- should not be a surprise to the department that if a program is executing poorly enough that its cost is growing that much, we should know that before this legal technicality happens.
                And also, if we have decided to buy fewer, then the unit cost goes up, we should know that in advance also, so this should not be a surprise, any of this a surprise to the department, but the law requires us to take note of it and requires me, as the acquisition executive, to certify that the continuation of the program is essential to the national security, that there are no alternatives to the program which will provide acceptable capability to meet the joint military requirement at less cost -- I assume you guys have all this language -- that the new estimates -- new estimates of the program acquisition cost or program unit cost have been determined by the director of Cost Assessment and Program Evaluation Office (CAPE) to be reasonable.
                In other words, you have to do another assessment.  The program manager already did one, right?  That's why the program manager reported the overrun.  Now we're supposed to, department-wide, re-do the re-estimate -- check it, adjust it, and CAPE is the department's cost estimator and does that.
                The program is a higher priority than programs whose funding must be reduced to accommodate the growth in the cost of the program.  And the management structure for the program is probably the most important one, as a practical matter -- the management structure for the program is adequate to manage and control program acquisition unit cost or procurement unit cost.
                That's the process and it began some months ago and now today the documentation of those certifications is required to be on the Hill.  There are six of them.  You have each one.  They all tell a somewhat different story.  I'll just go through a few of them. 
                The simplest ones are ones like the Apache Block III program, wherein the type of aircraft -- the new aircraft were being added to a program that was refurbishing existing aircraft, and that was the principal source of the cause of the breach. 
                The Wideband Gapfiller System (WGS) satellite system was a situation in which the department, early in the decade, set out to buy a family of satellites, changed its mind in the middle, interrupted the buy, interrupted production of the satellites.  Therefore, if there is a production break, you drive up cost.  And you can read the history of all these; I won't go through them.
                I think the ones that are most interesting because their largest programs are the DDG-1000, the Joint Strike Fighter program, so let me address them and then we'll take your questions. 
                The DDG-1000 -- the principal factor which caused the DDG-1000 destroyer to trip the Nunn-McCurdy threshold was that the buy was reduced from 10 to three last year.  And remember what determines the Nunn-McCurdy breach is the unit cost.  So the costs of the program used to be divided by 10 and are now divided by three, with the predictable result.  And of course this was known last year but the paperwork is just catching up now to that decision on the DDG-1000, which was made last spring.
                The Joint Strike Fighter is a case in which the program began early in this decade.  The baseline -- it was baselined in 2001.  Last fall, when the Joint Estimating Team (JET) II report was done -- which I'm sure you're familiar with if you follow the Joint Strike Fighter program -- which led to the restructuring of the program, that JET II estimate said the Joint Strike Fighter program is going to cost more than the program office and the contractors say it is going to cost.
                I ascertained and the secretary ascertained -- or I'll put it this way:  The department ascertained that that cost estimate was credible -- more credible.  It, as we said at the time, would put the program in Nunn-McCurdy breach.  In other words, if you adopted this estimate and said, yes, I think that estimate is right, the consequence was that the program would have a critical Nunn-McCurdy breach.  So this is, again, the paperwork sort of catching up with that. 
                There are two things that are important that came out of this process.  One is if you remember several months ago, we said that the department was refining the cost estimate and would -- now, this is a matter of forecasting, so I always say it's not weather forecasting but it's forecasting -- that we would refine the overall cost estimate, and we have.
                Now, this is work done by CAPE, the cost estimators at CAPE, and they reported to you that the estimated unit cost would be within a certain range and that was the best we could tell you at that time.  Well, that range has now been refined, and so -- I don't know if you will remember these numbers, but in '02 dollars, a Joint Strike Fighter program -- a Joint Strike Fighter, over the life of the entire program, was, when the program began a decade ago, supposed to cost $50 million, and we said it was going to be significantly higher than that.
                And the CAPE has completed their estimate and, again in '92 dollars, the estimate is --
                MS.    :  '02.
                SR. DEFENSE OFFICIAL 1:  Sorry, what did I say -- '02 --
                MS.    :  '02.
                SR. DEFENSE OFFICIAL 1:  -- is 92.4 (million dollars), which is in the range we reported a few months ago, but at any rate is more precise number.
                MR.    :  (Off mike.)
                SR. DEFENSE OFFICIAL 1:  Yes. 
                I will say that there is a breakdown of the sources of that cost growth, and I will give you the principal features.  They're in the press release.  The first is that the program's development phase, the SDD phase, was, back when the program began, planned to end in 2012.  Well, it's not going to.  It's going to end in 2016, as you all know, and it's fallen behind for the reasons you know, which we reported in the course of the JET estimate and so forth.
                A second factor was that the Navy reduced their planned procurement quantity by 409 aircraft.  This was shortly after Milestone B some years ago.  So the quantities went down and therefore the cost growth was incurred as a result of that in the unit cost. 
                And the third is actual and forecast contractor labor and overhead rates and fees have increased significantly, and this is the single-largest contributor to cost growth, according to CAPE.
                And I am directing, in an Acquisition Decision Memorandum (ADM) that accompanies this, the new Program Executive Officer (PEO) managing the Joint Strike Fighter program, Adm. Venlet, that this part especially of the cost growth that contractor labor and overhead rates and fees, as forecast in the early part of the decade, are now higher than they were forecast to be.  That is a part of the cost structure of the Joint Strike Fighter that has to be worked to reduce so that affordability is restored to this program. 
                This is something I've said repeatedly and this gives you some of the sources of that cost growth.  We don't want to experience that cost growth.  I believe this estimate is credible based upon how these rates have grown in the past decade, but I don't believe that that's acceptable or that the department should have to incur those costs.  So they need to be reduced.
                So this is just sort of more precision on something that we told you several months ago, but it does -- they have refined the cost estimate, and the department is adopting that cost estimate for purposes of the Nunn-McCurdy breach but not for managerial purposes.
                Another thing that you should note is that the cost estimate contains some changes that are refinements but important to note -- important to me, anyway, that you note.  One is that some things were identified in the process of this more exhaustive cost estimate that were not taken into account previously in the development program.
                Those are listed here, but with the language says that several important aspects of the program were not fully defined and captured in previous cost estimates, which means that they were forgotten and left out.  And these include something called the verification simulation capability, which is an important part of the operational test capability, and that's a necessary -- and it's necessary to pay for it, so we had to add that to the cost estimate. 
                The extent of the tooling required to produce the aircraft, which it was not fully taken into account by the previous cost estimates, CAPE cost estimates, and some of the military construction requirements required to support introduction of the aircraft.  So these are things that the cost estimators have discovered that were omitted earlier.
                And those things account for much, but not all, of the cost growth in the -- growth in the estimate of the System Design and Development (SDD) phase.  Another part is the continuing -- and it is continuing -- falling behind of the SDD program, which we reported to you first back in November, and this is something the program is going to have to continue to struggle with, and Adm. Vanlet and we will be continuing to struggle with. 
                But this is the dynamic that has been seen over the last few months of the difficulty, especially of getting the Short Takeoff Vertical Landing (STOVL) version, which is the most complicated, all set for a flight test.  And so that's reflected also in these cost estimates.
                So that's the Joint Strike Fighter program and it's all laid out here, and I hope that clarifies things for you.  Let me ask my colleague here if I've left anything out, and if not, we'll get to questions.
                SR. DEFENSE OFFICIAL 2:  No, I think you covered it.
                SR. DEFENSE OFFICIAL 1:  Unless your eyes are glazed over, if anybody has any question about the Nunn-McCurdy process -- nothing else, please.
                Q     Sir, a few things -- apples and apples.  The total program, a few years ago when it started, then your estimate was $232 billion.  It grew to 298 (billion dollars.)
                SR. DEFENSE OFFICIAL 1:  Tony -- Tony, I'm not going to have all these numbers in my head.
                Q     Okay, no, I'm just --
                SR. DEFENSE OFFICIAL 1:  I keep track of one thing and then you can do the math and we'll get you the appropriate numbers.  Let me just go over, in my mind's eye, the math is the following:  $50 million in '02 dollars, now, so apples to apples, is what was said a Joint Strike Fighter unit cost would be.  Now it is 92 (million dollars) in '02 dollars. 
                Now, you can take that and you can turn that into today's dollars, because I think inflation has been about -- I think it was at 18 percent, something like that.  Anyway, we can get you those numbers.  You can multiply by the number of airplanes and get a total program cost by taking the unit cost times the number of airplanes, and we can do that too.
                So, you can multiply, divide, escalate, de-escalate, but there are basically only two things to keep in your mind, which is --
                Q     (Off mike.)
                SR. DEFENSE OFFICIAL 1:  No, no, no, no, no, I'm trying to simplify things --
                Q     Okay.
                SR. DEFENSE OFFICIAL 1:  -- because I try to simplify it in my own mind -- 50 million (dollars), 92 million (dollars), and then you can massage those numbers however you want but that's the --
                Q     Well, at first, though, sir, the total cost was $328 billion -- the then-year total program cost.  My understanding now, what you sent to the Hill, it's now $382 billion.  That's about $54 billion of increase since April 1st.  I just want to get you to elaborate and, you know, confirm why that's happened.
                SR. DEFENSE OFFICIAL 1:  I wouldn't focus on the first number because that was the Selective Acquisition Report (SAR) number.  Well, Tony, I know you want to go from one month to the next, but let me just tell you that the SAR -- we have to do the paperwork and the paperwork -- we told you then not to focus on that number because it was not a complete cost estimate.  It was two different cost estimates stitched together.
                You may have even written at the time that it was two cost estimates stitched together, which we told you.  So that is -- that's just not a baseline that -- we had to send in a SAR because you have to send in a SAR, but we told you at the time, don't pay any attention to this; wait until we do this estimate.
                So I think the thing to focus on is the estimate has been refined from -- it was this wide range -- and I don't do these cost estimates -- from this range to a number, which is now $92 million.
                Q     Wasn't it 382 (billion dollars), though, for the total program, the refined program estimate?
                SR. DEFENSE OFFICIAL 1:  I'll get you the -- I'm not keeping the number -- yes, I'm sure that's right.  If you've taken the unit cost and multiplied by the number of airplanes, you've almost certainly got it right.
                Q     (Off mike) -- 382.4 (billion dollars).
                SR. DEFENSE OFFICIAL 1:  Okay, I'm not trying to be difficult; I'm just saying I don't keep that number.
                Q     Sir, I had a question about the basis for the cost estimate.  Lockheed has made a big to-do about saying that these -- you know, or sort of has expressed a concern that these estimates are not based on actuals.  Do they include actual programs?
                SR. DEFENSE OFFICIAL 1:  They do.  That's --
                Q:  Do they reflect --
                SR. DEFENSE OFFICIAL 1:  They do.  They do.  They include actuals up until some time ago, which is the best that can be had, and they also compare the Joint Strike Fighter program to other historical programs and say, how much did it cost to do this, that or the other thing?  Obviously, it's not a perfect science.  I hope that the taxpayer never has to pay this bill.  The contractors who work on the program say that it's too high and I agree.  So it should come down.
                This process underscores the managerial dilemma caused by independent cost estimates, which we are required to do, but the dilemma is this:  The cost estimate is what the airplane will cost if things keep going the way they are.  I ask not what the program will cost only, but what it should cost. 
                So the department does will-cost analysis; the department is also doing should-cost analysis.  Those are two different things.  I don't want to just take these estimates and say, well, let's put all that money in the budget.  That's the taxpayers' money.  We should do better than that when we have a Nunn-McCurdy breach.
                Nunn-McCurdy breach is a signal that the value that was promised to the taxpayer isn't being delivered, and we should try to restore that value and restore affordability, not just lay in the dollars that are estimated by cost estimators.  That would be an abrogation of managerial responsibility by the department, but most importantly by those who are performing the work -- the contractors that are performing the work.
                Q     So the logical question, then, is what should this program cost?  If, you know, your best estimate now is that it will -- you know, it will cost 382 billion (dollars) if it continues the way it does, what are you aiming for?
                SR. DEFENSE OFFICIAL 1:  Well, we're going to be the lowest possible.
                Q     But what kind of a discount to that -- I mean, you know, between the 50 (million dollars) and the 92 (million dollars)? 
                SR. DEFENSE OFFICIAL 1:  The lowest possible.  We just need to scrub -- I don't want to give you a number.  We're going to scrub every column of cost in this program and try to restore it as a close as possible to what a decade ago the department said it was going to cost.
                You know, some things will be facts of life.  Some things are inflation adjustments.  Some things are the cost of materials going up.  Some things are the cost of labor going up, for perfectly good reasons.  Some things are that programs are complicated and run into unforeseen issues.  The STOVL version of Joint Strike Fighter is an example of that.
                But other aspects of cost we want to look at and say, is that warranted?  Is that necessary?  Why should we pay that bill?  And I don't doubt the cost estimators who are saying this is what it will cost if you keep doing what you're doing, but the answer to that isn't just lay in the money.
                Q     So you said you're pressing the PEO to cut overhead costs and fees.  How much is enough?  How far do they have to come down?
                SR. DEFENSE OFFICIAL 1:  Well, that's the question that Andrea (sp) just named and I said we're going to try to get as far as we can.  She just asked -- we're going to get as far as we can.
                Q     So there's no metrics to judge success right now.
                SR. DEFENSE OFFICIAL 1:  Yes.  The less we pay -- that's a very good metric -- is how much less we pay than the cost estimators say we're going to pay.
                Q     And do you plan to meet with them again and assess their progress at a certain point in time?
                SR. DEFENSE OFFICIAL 1:  Of course.  This is a continuing thing.  This is just a statutorily -- a very important, statutorily required milestone.  The program goes on.  The management of it goes on, and we'll be every month trying to get the cost under control, get it down, keep the program on schedule, all that stuff.  That's what programs are -- you don't just forecast and then sit back.
                Q     Right, but, I mean, you're just saying do better. 
                SR. DEFENSE OFFICIAL 1:  No, we're going -- specifically how to do better.
                Q     Okay.
                SR. DEFENSE OFFICIAL 2:  Can I just add to that?  This is about as intensely managed a program as the department has.  I would say it is probably more intensely managed than any other program just because of the magnitude of it, so we're going to be paying very close attention to getting costs down.
                Q     Can I just ask you to step back just a little bit and look at the whole Nunn-McCurdy process from, like, orbit?  Is it doing what it was supposed to do?  When it first came up --
                SR. DEFENSE OFFICIAL 1:  That's a very good question.  It certainly is something the department should pay attention to when we come to believe, through these cost estimates, that something is costing half again as much.  Fine. 
                That said, we don't need to wait for that.  We should know, do know, and are -- the two of us are trying to make sure that we will know in the future well before it gets to 50 percent cost growth, that -- so I want to know when it's 5 percent.  I want to know when it's 10 percent, why?  Is there anything we can do to reverse that so it doesn't get up to 50 percent? 
                So that's the first thing is that the Nunn-McCurdy bill rings well after the managers of the enterprise should know about what is happening and be acting.  That's not a criticism of it; it's just that it comes along late. 
                Secondly, it frequently rings not because the program's cost has actually increased that much but because the numbers have changed or we have changed our mind.  That may be the dominant factor, in fact, in some of these breaches, which is fine but it's not actually indicative of poor performance by the program; it's indicative of the fact that we've changed the buy.  So there can be false alarms with it. 
                The third is that it is extremely cumbersome and time-consuming as a matter of paperwork, and we are actually doing a cost estimate of how much it costs to do Nunn-McCurdy's -- now, I'm serious -- because it is so burdensome. 
                Q     Because -- (off mike)?
                SR. DEFENSE OFFICIAL 1:  No, but they're participating, because it's so expensive and you really have to ask, what is the value added to all of this process because -- I mean, value added to the actual delivery of value to the program.  We're all working on these programs all the time anyway and it's good if this adds to the management. 
                It certainly is good for transparency to the taxpayer.  There is definite value in that.  But there are other ways that we can and should be doing that than Nunn-McCurdy, and it has grown very expensive and consuming of time relative to the actual value it adds to the management of the program.
                So that, in addition to the fact that it comes along late and has a high false alarm rate, mean that it is not the best mechanism for us.  We need and are building better mechanisms.
                Q     Do you have a preliminary estimate on those numbers, by the way?  Do you have a preliminary estimate on those costs?
                SR. DEFENSE OFFICIAL 1:  No, not yet -- just beginning -- we've just finished -- we've just done it today, but I'm booking this hour.  (Laughter.)
                Q     When do you think that you --
                 MODERATOR:  Actually like 30 minutes.  We probably have time for one, maybe even two more if they're not --
                Q     When do you think you'll be done with that cost estimate of the Nunn-McCurdy work that you're doing?  And also, I have a question on the DDG-1000 real quick after they answer that.
                SR. DEFENSE OFFICIAL 2:  In a week or so.  It's just a question of adding up the hours that people have put into this.  So it won't take very long.
                Q     And the DDG --
                SR. DEFENSE OFFICIAL 1:  We do wonder whether the Nunn-McCurdy process is itself a Nunn-McCurdy breach, though.  (Laughter.) 
                SR. DEFENSE OFFICIAL 2:  I want to add something about the process.  An artifact of the process is it forces us to reexamine decisions which we have generally already made.  So we already know that the cost increase is incurred.  We've often already taken action to either terminate the program or cut the quantities or something else.
                In some cases it's an artificial Nunn-McCurdy where we've added production, in this case with Apache Block III.  We just added some new helicopters.  I'm very happy with the program we had for remanufacturing so we added some new manufacturing as well.  New helicopters are much more expensive than remanufactured ones, and that forced us into a Nunn-McCurdy, which we had to drag the Army kicking and screaming into declaring -- and went through this process.
                I mean, having gone through it now, I kind of regret that we forced the Army to do that, but we really didn't have much choice under the law. 
                Q     And that was one of my questions on the DDG-1000 -- I'm sorry if this is obvious or you addressed this -- but how does reducing the buy from 10 to three then increase the costs?
                SR. DEFENSE OFFICIAL 1:  Because what triggers Nunn-McCurdy is the unit cost.  So when you're buying fewer, you're buying them uneconomically.  You're only spreading the development cost, the cost of initial tooling and all of that, over the first three ships, whereas if you're buying 10, all those fixed costs are spread out over a buy of 10.
                Q     The developmental costs and so forth?
                SR. DEFENSE OFFICIAL 1:  Yeah, the developmental costs -- tooling and the so-called nonrecurring costs.  So the more of something you make, the further you get down the learning curve and the incremental cost of each additional ship gets less and less.  
                Q     Well, why not keep 10 ships if it's all going to cost the same?
                SR. DEFENSE OFFICIAL 1:  Well, it does cost more to have 10 ships that three; it just doesn't cost as much for each one.
                SR. DEFENSE OFFICIAL 2:  Yeah, we have made a decision to --
                SR. DEFENSE OFFICIAL 1:  And the decision was made last year --
                SR. DEFENSE OFFICIAL 2:  Right.
                SR. DEFENSE OFFICIAL 1:  -- to only have three and not 10. 
                SR. DEFENSE OFFICIAL 2:  So once that decision was made, you then have to recalculate the unit cost, and you do it both just the production costs and also including in all the research and development.  So, when you have fewer units, you get a higher cost.
                SR. DEFENSE OFFICIAL 1:  Okay.
                Q     Just, so what would you recommend then -- could you provide a little bit more detail about what you'd recommend to replace Nunn-McCurdy and a better way to monitor this cost --
                SR. DEFENSE OFFICIAL 1:  We're working on a mechanism that would use, in part, the performance assessment and root cause analysis mechanism which was created by statute last year, and the statutory language suggest that this would be a perfect purpose for that organization. 
                And so, we want to make that a part of an early warning system and a root cause analysis organization.  I really have to tell you that whether it's legislated or not, Nunn-McCurdy process, Performance Assessment and Root Cause Analysis (PARCA), that is the job of the acquisition oversight function in the Department of Defense.
                So if we're not doing performance assessment, what are we doing?  If we're not doing root cause assessment, what are we doing?  And so, it's important that the legislation recognizes the necessity for doing this, and we have programs -- it's certainly true -- that illustrate the need to do it better, but whether we call it PARCA or whatever, it's our responsibility to be looking at our programs and saying, is the taxpayer getting the deal they want and that they signed up for?
                And if not, why not, and why is the program not being managed to whatever cost savings we can get, even if we don't know how far we'll get when we squeeze?
                MODERATOR:  I want to thank you for your indulgence, late on a Tuesday afternoon, and remind you again of the ground rules on background, SR. DEFENSE OFFICIALs.  Thank you.





Additional Links

Stay Connected