A Dose of Reality: Rising Costs for Nuclear Weapons

September 16, 2015 | 11:18 am
Eryn MacDonald

In a previous post Stephen Young and I looked at the overall changes in cost estimates for the NNSA’s 3+2 program to replace the entire nuclear weapons stockpile. As we noted, the FY16 Stockpile Stewardship and Management Plan (SSMP) shows a significant increase in cost estimates for most life extension programs (LEPs) when compared with the FY15 SSMP, which showed a largely unexplained drop in cost estimates from those in the FY14 report. The newer cost estimates for individual programs have now increased to what may be a more realistic level, although don’t be surprised if there are further increases to come. Below is a look at some of the changes to individual LEPs from FY15 to FY16.

W88 Alt 370

W88 Alt 370 total cost (FY12 to FY24), billions of FY15 dollars

  FY15 SSMP Estimates FY16 SSMP Estimates
Low Estimate 1.4 1 2.4 1.9 0.8 2.7
High Estimate 1.8 1 2.8 2.7 1 3.7


The W88 Alt 370 cost estimate increased from FY15 to FY16, but the increase is easily explained. In November 2014 the Nuclear Weapons Council—a joint Department of Defense-Department of Energy group that oversees planning for nuclear weapons life extensions—decided to include a refresh of the weapon’s conventional high explosive (CHE) as part of the Alt.

CHE degrades over time and must be replaced at regular intervals. Normally, an Alt would not involve replacement of a warhead’s high explosive charge—this would take place during a full refurbishment of the warhead—but replacing the CHE during the current Alt would push out the date when a more substantial refurbishment would be needed. According to Navy Strategic Systems Program Director Vice Admiral Terry Benedict, including the CHE refresh could sustain the W88 until the late 2030s; without it, the warhead could need to be refurbished again as soon as 2030.

Originally, the CHE refresh was not part of the plan because the first planned interoperable warhead, IW-1, was going to replace the W88 before a CHE refresh would be needed. In last year’s SSMP, however, the IW-1 program was postponed for five years, giving it a new first production unit date of FY30. That postponement was maintained in the FY16 SSMP. Given historical experience, and the fact that IW-1 will be the first interoperable warhead, there is a good chance that the program will run into delays, which would risk the CHE in the W88 warheads being past its lifetime before they could be replaced with the IW-1. So, the new plan is to replace the CHE while carrying out the rest of the previous plan for Alt 370, allowing more wiggle room for timing of the IW-1.

And, of course, the fact that the Navy has never been particularly excited about the interoperable warhead plan only reinforces their interest in ensuring that the expected life of the current version should be extended as long as possible.

Finding the money to pay for the increase in the Alt 370 program has required the NNSA to reduce or delay other activities. The FY16 SSMP notes that one such tradeoff is a reduction in surveillance for legacy B61 and B83 warheads (which will be phased out over time). This, however, only gets them part way—the document states elsewhere that “To cover remaining shortfalls for the CHE replacement, NNSA will annually evaluate its budget submissions to look for ways to fund the remaining CHE replacement costs.” Added together with the already large price tags for other LEPs and infrastructure projects, such savings are unlikely to be easy to come by.

Interoperable Warheads

IW-1 total cost (FY13 to FY14, FY20 to FY43), billions of FY15 dollars

  FY15 SSMP Estimates FY16 SSMP Estimates
Low Estimate 7.2 0.6 7.8 8.8 1.1 9.9
High Estimate 9 0.8 9.8 11.7 3.2 14.9


IW-2 total cost, billions of FY15 dollars

FY15 SSMP Estimates* FY16 SSMP Estimates*
Low Estimate 8.3 0.6 8.9 9.2 1.1 10.3
High Estimate 9.9 0.8 10.7 12.1 3.2 15.3

*In the FY15 SSMP, the program is assumed to run from FY2023 to FY2051. In the FY16 SSMP, the program is assumed to run from FY2023 to FY2049.


IW-3 total cost, billions of FY15 dollars

  FY15 SSMP Estimates* FY16 SSMP Estimates*
Low Estimate 9.4 0.6 10 8.5 1.1 9.6
High Estimate 11.9 0.8 12.7 11 3.2 14.2

*In the FY15 SSMP, the program is assumed to run from FY2023 to FY2051. In the FY16 SSMP, the program is assumed to run from FY2030 to FY2057.


The estimated costs for the three planned interoperable warheads that make up the heart of the 3+2 plan—IW-1, IW-2, and IW-3—increased significantly from the FY15 SSMP numbers. This is after the FY15 SSMP estimates fell far below the original estimates for the programs in the FY14 SSMP. According to the NNSA, part of this increase is due to changes in the cost estimation methods for the FY16 SSMP that bring it more in line with accepted DOD cost estimating practices.

Part of the increase is also due to a change in the plans for interoperability, which increased the DOD costs for these programs.

The new interoperability plan calls for the Air Force and Navy versions of the warheads to each have their own separate fuzes, which adds significantly to the cost for each of the services. Interoperability will be limited to the nuclear explosive package (NEP)—the rest of the warhead will be designed either to go on an ICBM or an SLBM reentry vehicle, so it will not be possible to simply swap it from one type of missile to the other.

For the IW-1, the current plan calls for producing the Air Force version first, before moving on to produce the Navy version. This is because the W78 warheads that the IW-1 will replace in the Air Force leg of the triad are older than the W88 warheads in the Navy leg, and the W88 Alt 370 also gives the Navy an additional cushion. In fact, the Alt 370 gives enough cushion to raise questions about the timing of replacing the W88 as part of the IW-1, but that is still the official plan.

Air-Launched Cruise Missile

In October 2014 the NNSA announced that it had selected the W80-4 warhead to be used for the new air-launched cruise missile known as the Long-Range Stand Off (LRSO) weapon. This program is another for which last year’s SSMP showed a significant decrease in costs, from an estimate of $8.3 billion in the FY14 SSMP to a range of $4.6 to $6.1 billion in the FY15 SSMP (all in FY15 dollars).# According to an August 2015 Government Accountability Office review of the NNSA’s budget, the NNSA said this lower cost was due in part to the selection of the W80-4 warhead, which lowered costs by removing the need to research three different warheads for potential use in the program and also “allowed the agency to eliminate uncertainties related to component design, technology development efforts, and certification requirements.” If this is the case, then the decision to choose the W80-4 warhead must have been made long before it was announced, since the FY15 SSMP was released in April 2014.

The current FY16 SSMP predicts a range of $5.8 to $7.8 billion (in FY15 dollars)—not quite a return to the FY14 estimates, but a notable increase over the FY15 numbers.

The timing of the program has changed as well. In the FY15 plan, the NNSA decided to delay the start of production for two years, to FY27. The FY16 plan accelerates the program, with the beginning of production shifting back to its original date in FY25. While the FY16 SSMP notes this, calling it a “significant profile change,” it also goes on to say that the change in timing is “not a change to the overall W80-4 program cost in and of itself.”

But if the shift in timing is not the explanation for the program’s fluctuating costs, then it is not clear what is—there is no major change to the substance of the program, and the costs for researching multiple warheads and related uncertainties in design and certification have presumably not reappeared. Instead, it seems to be simply a correction of last year’s overly optimistic cost estimate.

As the program goes forward, these costs may increase even further. Right now, the W80-4 LEP plan assumes only moderate refurbishment of the nuclear explosive package. Should this plan change—as it has for several other LEPs—costs would be likely to increase.

What Does the Future Hold?

The return to higher cost estimates in the FY16 SSMP is likely to continue in next year’s document. If anything, history shows that these programs will increase in cost as time goes on and the uncertainty of early cost estimates is worked out. It is still not clear how the FY15 estimates came to be so low for many programs—the NNSA’s explanation that different cost estimating models and methodologies were used isn’t very helpful—but this year’s report takes a step toward correcting the record. Hopefully it will help to inspire a more realistic and vigorous debate over the affordability of the NNSA’s ambitious modernization programs and the significant tradeoffs that they would require if all are pursued as proposed.

#While the FY14 SSMP does not provide information in constant dollars, using information from Figure 8-21 of that report and an escalation factor of 3.4% gives an estimate of $8.3 billion in FY14 dollars for the ALCM. Note that FY14 and FY15 dollars are equivalent because inflation has been so low.