Less Expensive Alternatives

The United States is planning to spend hundreds of billions of dollars over the next two decades to rebuild a nuclear arsenal like the one it has today and to last another 50 years. The current and planned U.S. financial investment in nuclear forces is unrivaled by any other nuclear power. But the spending plans face significant budgetary, programmatic, and political challenges. The question, then, is not whether the United States is falling behind its competitors—it is not—but whether the size and configuration of the current arsenal and the Trump recapitalization plans are necessary, sustainable, and safe. The answer is that the current course is unnecessary, unsustainable, and unsafe—and must be rethought. It is not too late to pursue a different path. Now is the time to re-evaluate nuclear weapons spending plans before the largest investments are made.

A common argument made in support of the approach proposed in the Trump NPR is that the only alternative is to allow the U.S. nuclear deterrent to waste away. But this is a false choice. The October 2017 CBO report evaluated numerous alternatives to the current sustainment and recapitalization program that, if pursued, would reduce nuclear weapons spending while still maintaining a reliable and credible nuclear deterrent. The options range from blended reductions to each or several legs of the triad to moving to a dyad. The report measured the capability of the alternatives relative to that of the current program across four metrics: the number of warheads, crisis management, limited nuclear strikes, and large-scale nuclear exchanges.

Of course, pressure on the defense budget cannot be relieved solely by reducing nuclear weapons spending. A significant portion of the overall cost of nuclear weapons is fixed. Key components of the supporting infrastructure, such as the command-and-control systems and nuclear laboratories, would remain whether the United States possessed 10 nuclear weapons or 10,000. That said, changes to the nuclear replacement program could make it easier to execute and ease some of the hard choices facing the overall defense enterprise, while still leaving a force more than capable of deterring nuclear attacks against the United States or its alliance partners.

The following analysis describes three realistic options for reducing U.S. spending on nuclear weapons that would save at least an estimated $29 billion to $282 billion from fiscal year 2017 to 2046. The bulk of these savings would occur over the first 20 years of the 30-year period. Unlike the Trump NPR, the second and third options in particular would reflect a nuclear strategy that reduces reliance on nuclear weapons, emphasizes stability and survivability, de-emphasizes nuclear warfighting, reduces the risk of miscalculation, and is more affordable and executable.

The baseline for these estimates are the October 2017 CBO estimate of the Obama administration’s plans to maintain and replace U.S. nuclear forces and their supporting infrastructure and our projection of the costs of the additions proposed by the Trump NPR.134 With the exception of the first option, which is in fiscal year 2018 dollars, all estimates are in fiscal year 2017 dollars unless otherwise noted. The estimated savings from each option includes savings from research and development, procurement, and operations and sustainment unless otherwise noted.

The first option would eliminate the additions to the Obama-era recapitalization program proposed in the Trump NPR. This option would avoid an estimated $28.8 billion in additional costs above the CBO baseline over the next 30 years

The second option would reduce costs by more cost-effectively deploying 1,550 New START-accountable strategic warheads. This option would save an estimated $120.5 billion relative to the CBO baseline and $149.3 billion when combined with eliminating the Trump additions over the next 30 years.

The third option would eliminate the ICBM leg of the triad and decrease the number of New START-accountable strategic warheads to 1,000. This option would save an estimated $253 billion relative to the CBO baseline and $281.8 billion when combined with eliminating the Trump additions over the next 30 years. Reductions below 1,000 deployed warheads could result in even bigger savings

Each option would involve the purchase of new fleets of ballistic missile submarines, SLBMs, and long-range penetrating bombers. All of the options would also retain at least one low-yield nuclear delivery option. None of the options would alter current plans to upgrade nuclear command, control, communications, and early-warning capabilities. Upgrading these capabilities and reducing their vulnerability to attack should be a top priority and will likely require additional funding beyond what the Pentagon has identified to date. All of the options also assume the continued maintenance of the three national nuclear laboratories: Los Alamos, Livermore, and Sandia.135 In fact, even under the most ambitious cost-saving option, the United States would still be poised to spend roughly $1 trillion on nuclear weapons over the next 30 years due in large part to the fixed costs of the nuclear weapons enterprise.

This option would eliminate the additional investments proposed in the Trump NPR for the low-yield SLBM option ($125 million in savings), the new SLCM ($11 billion in savings), maintaining the B83-1 until a suitable replacement is found ($13 billion in savings), and building more plutonium pits ($4.6 billion in savings) (See figure below). This option would not make adjustments to the Obama-era plans, which the CBO estimated will cost $1.2 trillion between fiscal years 2017 and 2046. As noted earlier in this report, the Trump additions to the recapitalization program he inherited constitute unnecessary, unsustainable and unsafe overreach. Eliminating the Trump additions would still leave an arsenal that exceeds what is necessary for deterrence and poses a significant affordability challenge.

Eliminate SLBM $125 million
Eliminate SLCM $11 billion
Retire B83-1 $13 billion
Scale back pit production $4.6 billion
Total savings $28.8 billion

Savings of $28.8 billion could purchase nine Virginia-class attack submarines (estimated by the CBO at $3.1 billion per boat in fiscal year 2018 dollars), 90 “Penetrating Counter Air” aircraft to replace the F-15 and F-22 (estimated by the CBO at $317 million per aircraft in fiscal year 2018 dollars), the sustainment of an army infantry or stryker brigade combat team for 10 years (estimated by the CBO at $2.7 and $2.8 billion per year, respectively), or nearly the entire remaining acquisition cost of the Joint Air-to-Ground Missile (estimated by the GAO at $4.8 billion in fiscal year 2018 dollars), Long Range Precision Fires (estimated by the GAO at $2.9 billion in fiscal year 2018 dollars), and Arleigh Burke Class Destroyer (estimated by the GAO at $23.4 billion in fiscal year 2018 dollars) programs.136

Estimating the amount of money that could be saved under this option is difficult given that in most cases the plans for these additions have yet to be clearly defined and official cost estimates do not yet exist.

For this report we follow the CBO in assuming that the Navy would build a SLCM similar in design to the LRSO to reduce development costs and maximize the missile’s ability to penetrate adversary air defenses.137 Another option would be to leverage the Navy’s planned replacement for the Tomahawk with the Next-Generation Land Attack Weapon. But this effort is in its infancy, having recently completed an analysis of alternatives138.

The CBO assumes a SLCM and its associated warhead would cost 50 percent less to develop than the LRSO and W80-4, respectively, and the same as the LRSO and W80-4 to produce. Assuming the purchase of 350 missiles, which is the same as the number of nuclear Tomahawks that were built during the 1980s, we estimate the cost to develop and procure the SLCM and its associated warhead at roughly $11 billion. Additional funds, likely several billion dollars, would also be required through 2046 to recertify ships or attack submarines to carry SLCMs and to operate and sustain the missiles.

Estimating the additional cost of the Trump administration’s plans to expand pit production is difficult given the lack of published projections of the cost of the Obama administration’s plans to produce 50-80 pits annually. For this report we base our estimate on NNSA’s April 2018 engineering assessment of plutonium pit production and assume the agency would pursue the least expensive option to produce 80 pits annually.139 According to the assessment, producing 50 additional pits per year at the Savannah River Site would cost $13.4 billion more between 2030 and 2080 to operate than the least expensive option to produce the same number of pits at Los Alamos. Pursuing the less expensive option would thus save about $4.6 billion through fiscal year 2046. However, this estimate could be significantly understated. The CBO projects that reverting to the Obama administration’s plan to produce 50-80 pits annually would save $9 billion in then-year dollars between fiscal years 2019 and 2028.140

For the B83-1 we assume that NNSA would retain the warhead and proceed to extend its life. In 2013 the agency estimated that it would cost $4 billion in fiscal year 2012 dollars to perform an alteration of the warhead beginning in fiscal year 2019 and $7 billion to $9 billion to conduct a more comprehensive life extension program beginning in fiscal year 2035. We estimate that sticking with the Obama administration’s original plan to retire the warhead after confidence in the B61-12 is achieved would thus save about $13 billion through fiscal year 2046.

Trump’s intention to leave the INF Treaty and his uncertain commitment to New START could further increase the long-term price of the administration’s nuclear weapons spending plans, though by how much is difficult to project. The Defense Department requested, and Congress approved, $48 million in fiscal year 2019 to explore concepts and options for conventional, ground-launched, intermediate-range missile systems in response to Russia’s violation of the treaty. The Pentagon is planning to test a land-based cruise and ballistic missile system by the end of 2019. Although the current effort is focused on a conventional missile, the administration or its successor could ultimately opt to develop a nuclear-capable version.

Prior to the negotiation of the INF Treaty in 1987, the United States deployed several hundred nuclear-armed intermediate-range Pershing II ballistic missiles and ground-launched cruise missiles in Europe, the latter of which were an adaptation of the Tomahawk. The Pentagon spent $2.6 billion (in fiscal year 1987 dollars) to develop and procure 247 Pershing II missiles and associated launchers and $3.5 billion to develop and procure 442 ground-launched cruise missiles through fiscal year 1987, according to the GAO141. However, the cost today to develop a new ballistic missile system would be higher given that several decades have passed since the development of the Pershing II. In addition, the range of the new missile would likely need to be much greater than the 1,800-kilometer range of the Pershing II to have any utility against China in the Pacific.

The budget implications of withdrawing from or failing to extend New START are difficult to judge. It is plausible that the U.S. military could call for additional nuclear deployments amid greater uncertainly about Russian nuclear force posture and planning. In the near term this might include redeploying the 48 SLBMs and 50 ICBMs that were removed under the treaty. The Pentagon could also add additional warheads to SLBMs and, with greater difficulty, to ICBMs. In the longer term, the Pentagon could propose to accelerate and build more new bombers, submarines and ICBMs. The CBO estimated that building 16 instead of the current plan of 12 Columbia-class submarines and building enough new GBSD missiles to support a deployed force of 450 ICBMs instead of the current plan of 400 would cost an additional $33 billion in fiscal year 2017 dollars through fiscal year 2046.

This option would more cost-effectively deploy 1,550 New START-accountable warheads by reducing the size of the triad to 10 SSBNs and 300 ICBMs ($40 billion in savings), delaying the GBSD program in favor of extending the life of the Minuteman III and reducing the number of ICBMs ($17.5 billion in savings), eliminating the LRSO ($30 billion in savings), no longer deploying B61 gravity bombs in Europe ($17 billion in savings), pursuing simpler warhead life extension programs ($10 billion in savings), and developing a capability to produce 30 plutonium pits per year ($6 billion in savings) (See figure below). This option would save $120.5 billion relative to the CBO baseline and $149.3 billion when combined with eliminating the Trump additions. Additional savings could be achieved by reducing the triad to eight SSBNs and 150 ICBMs and delaying the B-21 program or purchasing fewer B-21s.

Reduce triad to 10 SSBNs and 300 ICBMs $40 billion
Delay GBSD program $17.5 billion
Eliminate LRSO $30 billion
Don’t deploy B61 gravity bombs in Europe $17 billion
Pursuing simpler warhead LEPs $10 billion
Reduce to 30 pits/year production capacity $6 billion
Eliminate Trump additions $28.8 billion
Total savings $149.3 billion

Savings of $149.3 billion would cover nearly the entire additional acquisition cost over the next 30 years relative to current plans to grow the Navy to 355 ships by the late 2030s (estimated by the CBO at $161 billion)142.

Under this option the United States in 2030 would deploy roughly 1,550 New START accountable warheads on 520 strategic delivery systems consisting of 200 SLBMs, 300 ICBMs, and at least 20 nuclear-capable bombers. To achieve this level of warheads with a smaller number of delivery systems, the Navy would deploy an average of six warheads on each SLBM instead of the current loading of four-to-five warheads. By 2046 the United States would deploy about 1,450 warheads on 160 SLBMs, 300 ICBMs, and 100 nuclear-capable bombers. Each SLBM would carry six-to-seven warheads. By 2046 the current fleet of Ohio-class submarines armed with 20 operational SLBM tubes under New START will be replaced by the Columbia-class boats armed with 16 SLBM tubes. In addition, the current fleet of about 40 nuclear-capable B-52 bombers would be removed from the nuclear mission due to the elimination of ALCMs and replaced by a fleet of at least 100 B-21 bombers, some or all of which will be nuclear-capable. The Air Force plans to begin retiring the B-2 bomber in the late 2030s143.

Reducing the triad to eight SSBNs and 150 ICBMs and the number of deployed warheads to 1,000 would save an additional $45 billion through fiscal year 2046, according to the CBO.

Reducing the Number of Submarines

The Navy is planning to purchase 12 new Columbia-class submarines to meet current military requirements but shifting to 10 strategic submarines would still provide a devastating sea-based nuclear deterrent. The CBO projects that purchasing two fewer new boats would save $17 billion through fiscal year 2046. The majority of the savings from reducing to 10 submarines would come during the early 2030s when the last two Columbia-class submarines are scheduled to be purchased. These savings would lessen the burden on the Navy’s shipbuilding budget as it seeks to expand the fleet to roughly 350 ships.

The Navy is seeking 12 new boats rather than 14 because the new submarine will not need a four-year mid-life refueling, but only a two-year overhaul.144 The Navy originally planned to start deploying the replacement boats in 2029, but in 2012 the Pentagon announced a two-year delay to the program, pushing back the fielding of the first new submarine to 2031. As a result, the Navy will field only 10 ballistic missile submarines in the 2030s. Under New START the Navy plans to deploy about 1,000 warheads on 240 SLBMs.

Current military requirements call for 10 strategic submarines to be operational at all times to ensure that five submarines are “on station” within range of their targets so SLBMs can be launched promptly, as quickly as within 15 minutes of an order to do so. For the Navy to operate five submarines on station, it would need 12 submarines in total: five in the Atlantic, with two on station and the rest in transit or in port (such as for maintenance), and seven in the Pacific with three on station and the rest in rotation. Initially, only 10 submarines are needed to meet these goals.145

The need for 12 submarines, then, has as much to do with where the warheads are deployed and how promptly they could reach their targets as it does with the number of warheads. A fleet of 10 Columbia-class submarines can carry nearly 1,300 warheads, but it cannot support five submarines that are deployed close enough to their targets in Russia and China, ready for quick launch. In addition, reducing the number of submarines to 10 and increasing the number of warheads per SLBM would reduce, though not eliminate, the Navy’s flexibility to deploy SLBMs with one or two warheads and to upload nondeployed warheads in the event of a need to significantly increase the deployed arsenal. Reducing to eight boats would further reduce that flexibility. Increasing the number of warheads per SLBM could also reduce the range of the missile.

However, relaxing the requirements for prompt launch, especially against counterforce targets, would allow fewer than five submarines to be on station, thereby eliminating the need for 12 new submarines. So too would reducing the required number of deployed warheads. With a smaller arsenal, the United States also would not need such a large upload capability. Carrying extra spaces on SLBMs for warheads is an expensive hedge.

Refurbishing the Minuteman III and Reducing Their Number

The ICBM leg of the triad is the least valuable leg of the triad and plans to sustain it should reflect this reality. Due to a limited range of flight trajectories, the Minuteman III is essentially unusable outside of a nuclear conflict with Russia. The Trident D-5 SLBM is mobile, highly accurate, and capable of prompt launch. In addition, continuing to maintain ICBMs in a “launch under attack” mode is unnecessary and risky. The primary mission of the land-based leg of the triad is to deter an adversary nuclear first strike attack and serve as a backstop to an unforeseen and unlikely future vulnerability in the SLBM force. But these functions can continue to be performed at lower numbers of ICBMs and by deferring the development of a new ICBM.

The CBO projects that $17.5 billion could be saved over the next 30 years by delaying development of a new ICBM by 20 years and instead extending the life of the Minuteman III by buying new engines and new guidance systems for the missiles. Crucially, however, this approach would save $37 billion through fiscal year 2036 when the vast majority of nuclear recapitalization spending is scheduled to take place. The Air Force has to contend with the high cost of several other priorities during this period in addition to GBSD, including the F-35, B-21, and new tanker programs.

The U.S. Air Force currently deploys about 400 single warhead Minuteman III ICBMs located at F.E. Warren Air Force Base, Wyoming; Malmstrom Air Force Base, Montana; and Minot Air Force Base, North Dakota. Under New START, the Air Force maintains 50 extra missile silos in a “warm” reserve status. The Minuteman III was designed in the 1960s and entered service throughout the 1960s and 1970s. Today’s Minuteman weapon system is the product of almost 40 years of continuous enhancement. The Pentagon spent over $7 billion in the early 2000s to keep the ICBMs safe, secure and reliable through 2030146. This modernization program has resulted in an essentially “new” missile, expanded targeting options, and improved accuracy and survivability147.

To reduce the number of ICBMs from 400 to 300, the Air Force would have several options. It could remove 100 missiles from the deployed force and distribute the reductions across the three bases while leaving the silos in a warm status. It could also eliminate a wing of 50 silos from two of the three bases. Yet another option would be to eliminate an entire base of 150 silos, which would save roughly $200 million per year148.

The Defense Department is planning to replace the Minuteman III missile, its supporting launch control facilities, and command-and-control infrastructure. The plan is to purchase 666 new missiles, 400 of which would be operationally deployed through 2070149. The remaining missiles would be used for test flights and as spares. The Pentagon is seeking to make significant capability upgrades as part of the recapitalization program, known as the GBSD. According to the 2018 NPR, the life of the Minuteman III “cannot be extended further” and the missiles “will have increasing difficulty penetrating future adversary defenses.”

The Air Force initially estimated the cost of the GBSD program at $62 billion in then-year dollars. But the Pentagon in August 2016 set the estimated acquisition cost of the program at $85 billion and the total life-cycle cost at $238 billion in then-year dollars. The $85 billion estimate is at the lower end of an independent Pentagon cost estimate that put the acquisition price tag as high as $150 billion.150 Many ICBM proponents argue that they are the cheapest leg of the triad to maintain and modernize, but the independent estimate of approximately $150 billion exceeds the projected cost of $128 billion for the Columbia-class submarine program.

The Air Force in 2014 conducted an analysis of alternatives to sustain the ICBM leg of the triad which showed that the price to build a new missile system would be roughly the same as the cost to maintain the Minuteman III151. However, the service arrived at this conclusion by comparing the total lifecycle cost of each option through 2075. This meant that the Minuteman III life extension option included the costs of both refurbishing the existing missiles and the costs of building a new fleet of replacement missiles. The analysis of alternatives also assumed a need to deploy 450 missiles.

In contrast, the CBO evaluated the cost of the two options over a shorter period of time. In addition, a 2014 report by the RAND Corporation on the future of the ICBM force found that “any new ICBM alternative will very likely cost almost two times—and perhaps even three times—more than incremental modernization of the current Minuteman III system.”152 The report said continuing to maintain the Minuteman III through life-extension programs and “gradual upgrades is a relatively inexpensive way to retain current ICBM capabilities.”

The RAND study identified two challenges to this approach. First, the number of Minuteman III missile bodies is declining due to test launches. Based on the current testing pace of roughly 4–5 tests per year, maintaining a force of 400 missiles, as is the plan under New START, would deplete the test inventory by 2035. Second, the report said incremental modernization would be “viable” only if the capability the Minuteman III provides “is not substantially changed.”

But reducing the number of ICBMs to 300 and forgoing capability upgrades, which are unnecessary for the ICBM force to continue to serve its sponge function, would mitigate these challenges. Life-extended Minuteman III missiles can get blown up in their silos by incoming Russian ICBMs less expensively than new GBSD missiles. Moreover, the claim that the Minuteman III may not be able to overcome expected advances in adversary air and missile defenses over the next two decades merits further scrutiny given the repertoire of countermeasures the missile is already believed to contain to overcome such defenses.

Some analysts argue that refurbishment is not viable due to the aging-out of the Minuteman III’s component parts153. Extending the life of the Minuteman III could entail some technical risk. However, neither RAND nor the 2014 analysis of alternatives determined that doing so is infeasible.

Additional savings could potentially be found from keeping Minuteman III missiles past their anticipated expiration in the early 2030s, which would delay, if not obviate, the need to refurbish the missiles. For example, Todd Harrison, the director of the Aerospace Security Project at the Center for Strategic and International Studies, wrote in a 2017 report that keeping the missiles three years longer than planned, would increase the probably of failure at launch from 1.3 to 3.8 percent.154 But as Harrison notes:

“If the primary purpose of the ICBM force is to deter an attack by acting as a missile sponge, then quantity is arguably more important than reliability. At extremely high levels of unreliability, an adversary could begin to disregard the missiles altogether. At the range of failure rates discussed here (up to 3.8 percent), though, that adversary would still need to target all of the ICBMs to neutralize them.”

The Defense Department continues to use even older Minuteman II rocket motors for military space launches, which suggests the department has high confidence in the boosters. Which begs the question: If older Minuteman II motors are still functioning reliably, could the newer Minuteman III boosters reliably last longer than currently planned?155

The Air Force has yet to demonstrate that sustaining the Minuteman III beyond the missiles’ expected retirement in the 2030 timeframe is not a viable or more cost-effective nearer-term option. Former undersecretary of defense for policy Michèle Flournoy stated in 2017 that “the Defense Department should more seriously consider further extending the life of the existing Minuteman III ICBMs as a cheaper near-term alternative to the current plan to build an entirely new ICBM system.”156 Pursuing this approach would carry significant option value as it would defer a decision on whether to build an entirely new system.

Additional Savings from this Option

Eliminating the LRSO and no longer deploying B61 gravity bombs in Europe would save $47 billion through fiscal year 2046. As noted above, these weapons are militarily unnecessary and in the case of the LRSO pose underappreciated risks to stability. Under this option removing tactical nuclear weapons from Europe would not alter the scope of the B61-12 life extension program but would result in foregoing plans to make the F-35 nuclear-capable.

Jettisoning the LRSO and F-35 nuclear capability would leave B61-12s delivered by the B-2 and later the B-21 as the lone low-yield option in the U.S. nuclear arsenal. Given the enormous investment that is being made in the B-21, there should be no doubt about the bomber’s ability to penetrate even the most advanced adversary air defenses for the foreseeable future. If a second low-yield option is required, then fielding the low-yield SLBM as proposed by the Trump NPR would be a far more cost-effective option, though the risks to stability and the survivability of SSBNs highlighted earlier in this report would still remain.

Building a capability to develop 30 plutonium pits per year instead of 80 and ditching plans to build interoperable warheads would save $14.5 billion through 2046. The estimated $6 billion in fiscal year 2018 dollars in savings from building only 30 plutonium pits, which is based on forgoing the least expensive option studied by NNSA to build an additional 50 pits, is likely an underestimate given that the cost to build the necessary infrastructure to produce the extra pits is at the low end of NNSA’s projected cost range for the option.

As described earlier, the United States does not need to build at least 80 pits per year. Achieving the capability to build even 30 by 2030 would be an enormous achievement given that the Energy Department has not produced pits at such a quantity since the 1980s and had to cease major plutonium operations at Los Alamos’ Plutonium Facility-4 from 2013-2017 due nuclear criticality safety concerns. Once NNSA demonstrates a capability to manufacture 30 pits per year, it can re-evaluate the need for additional pits based on the anticipated aging of existing pits, the size of the total warhead stockpile, and the international security environment.

The need for increased pit production could also be reduced by pursuing less ambitious warhead life extension programs. The model for future life extensions should instead be the Navy’s simpler, $4 billion in then-year dollars W76 SLBM warhead life extension program.157 In addition, instead of rebuilding the W78 warhead, it should be retired. A smaller ICBM force means there is no need to keep two different ICBM warheads. The W87 is newer and has modern safety features. Enough W87 warheads have been produced (more than 500) to arm the entire ICBM fleet. The estimated savings of $10 billion in this option from forgoing the development of interoperable warheads assumes a simpler life extension program for the W78 warhead, not its retirement.

Delaying the B-21 bomber program or purchasing fewer B-21s would result in additional savings. According to the CBO, deferring development of the B-21 would save $34.5 billion through fiscal year 2046. The CBO projects the cost of each B-21 at $690 billion. Reducing the planned buy from 100 to 70 bombers could thus save $20.7 billion, though the actual savings might be less than that amount due to the loss of economies of scale.

The primary mission of the B-21 is to allow the Air Force to continue to provide a conventional long-range penetrating bomber. The B-21 would not be certified to carry nuclear weapons until two years after it is first deployed. According to the Pentagon, only about five percent of the bombers’ acquisition cost would go directly to making the bombers capable of carrying nuclear weapons158. However, if the bomber did not have a nuclear mission, the overall program could be less expensive. For example, the bombers might not need to use pilots, but could be operated remotely.

Even with a 10-year delay, a new bomber would still be ready by about the time current bombers are reaching the end of their service life and the delay would allow the new bomber to incorporate technological advances made during that time. “Taking advantage of future technological developments can be particularly valuable for weapon systems that are expected to be in use for several decades,” the CBO states159. In addition, by moving B-21 funding into the future, the Air Force would free up resources for other priorities, such as buying KC46A tankers and F-35A fighters. Buying fewer bombers would also save money for the “Penetrating Counter Air” aircraft the Air Force is seeking to replace the F-22 and part of the F-15 fleet beginning in 2030.160

The B-2, the last U.S. bomber built, provides a cautionary tale. In the 1980s, plans called for 132 B-2s, and then 75, but the dissolution of the Soviet Union led to growing congressional opposition. In 1992, President George H.W. Bush announced that production would be limited to 20 aircraft. Twenty-one B-2s were ultimately built, at a cost of more than $2 billion each, far above initial estimates. Its predecessor, the B-1, also was never built in the numbers envisioned.161

However, delaying fielding of the B-21 until the late 2030s would limit the Air Force’s inventory of stealthy bombers able to fly in defended airspace to the 20 B-2s in today’s bomber force. While this disadvantage would be less pronounced if the B-21 only had a nuclear mission, the main purpose of the bomber is conventional. Delaying the bomber might also weaken the rationale for forgoing the LRSO in favor of the B61-12.

This option would reduce the number of deployed strategic warheads to 1,000 New START-accountable warheads by eliminating the ICBM leg of the triad and reducing the number of SSBNs to eight ($200 billion in savings), eliminating the LRSO ($30 billion in savings), no longer deploying B61 gravity bombs in Europe ($17 billion in savings), and developing a capability to produce 30 plutonium pits per year ($6 billion in savings) (See Figure below). This option would save $253 billion relative to the CBO baseline and $281.8 billion when combined with eliminating the Trump additions. Additional savings could be found by delaying the B-21 program or purchasing fewer B-21s.

Eliminate the ICBM leg and reduce to 8 SSBNs $200 billion
Eliminate the LRSO $30 billion
Don’t deploy the B61 gravity bomb in Europe $17 billion
Reduce to 30 pits/year production capacity $6 billion
Eliminate Trump additions $28.8 billion
Total savings $281.8 billion

Savings of $281.8 billion would nearly equal the combined price of the fiscal year 2019 budget request for the Department of Veterans Affairs ($199 billion), Department of State ($37.8 billion), and Department of Homeland Security ($47.5 billion). Such savings could also cover nearly the entire projected cost to cleanup legacy nuclear weapons facilities and waste (estimated by the Energy Department at $377 billion in then-year dollars).162

Under this option the United States would by 2030 deploy roughly 1,000 New START accountable warheads on 180 strategic delivery systems consisting of 160 SLBMs and at least 20 nuclear-capable bombers. By 2046 the United States would deploy the same number of warheads on 128 SLBMs and 100 nuclear-capable bombers.

The main difference between Option 2 and Option 3 is that this option would eliminate the ICBM leg of the triad and its associated warheads and supporting infrastructure and retain eight submarines instead of 10. In order to deploy 1,000 warheads under this option, each SLBM would carry an average of six warheads.

The case for eliminating ICBMs is that they are a redundant backup to SLBMs and because they are primed for “launch-under-attack,” could increase the potential of an accidental nuclear war. Discarding ICBMs would also remove the rationale for those Russian nuclear weapons targeted at the missiles. Phasing-out the ICBM force would save $149 billion through fiscal year 2046, according to the CBO.

Eliminating the missiles would drastically reduce the number of aim-points an adversary would have to strike to destroy delivery systems and warheads based on U.S. soil. But as the CBO notes, so long as submarines “remained undetectable, the United States would still have several hundred warheads on SLBMs available for a retaliatory strike.”

For the foreseeable future there does not appear to be any threat to submarines that would allow an adversary to prevent massive retaliation by the United States. As James Miller, a former undersecretary of defense for policy, and Richard Fontaine, the executive director of the Center for a New American Security, write “The United States is proceeding with its Columbia-class replacement SSBN [ballistic missile submarine], which will continue to serve as the backbone of the U.S. strategic deterrent, indicating that the United States does not regard threats to its strategic missile submarines in the coming generation as unmanageable.”163 They note that “it is likely that the measure-countermeasure interaction between offensive strike capabilities…and defensive and survival capabilities…will continue both in undersea warfare and for time-critical targeting of mobile missiles without a fundamental shift in the basic strategic reality of the nuclear era.”

Even if an unforeseen breakthrough in antisubmarine warfare were to emerge, retaining ICBMs might not provide a reliable hedge given their vulnerability. This has led some to suggest the development of a mobile basing mode to enhance the survivability of ICBMs. In fact, the Trump NPR suggests the Pentagon might consider a mobile basing mode. But such an approach would be far more expensive than even the already pricey GBSD program, which will retain silo-basing, and would likely be politically infeasible in any event164.

Under this option the United States would still retain bombers, which could be put on alert during a crisis or major conflict, as a hedge against submarine vulnerability. If a future U.S. president believes that an additional hedge against strategic submarine vulnerability is required, there would be the option of developing and fielding a nuclear-armed SLCM on attack submarines. This would be more cost-effective than retaining ICBMs or even building more Columbiaclass boats. But development of a SLCM would not need to begin for some time given current confidence in the invulnerability of ballistic missiles submarines. A SLCM would increase the number of nuclear-capable submarines by a factor of five, though as noted earlier the capability would not come without risks165.



  1. Though the October 2017 CBO report covers the period between fiscal year 2017 and 2046, fiscal years 2017 and 2018 have already concluded and fiscal year 2019 will be nearly halfway complete at the time of publication of this report. Thus, the savings from the options described in this report would be different, though not substantially so, if implemented starting in fiscal year 2019.
  2. A smaller arsenal could substantially reduce the rationale for maintaining three nuclear weapons laboratories.

Eliminate Trump Additions:

  1. U.S. Congressional Budget Office, “An Analysis of the Navy’s Fiscal Year 2019 Shipbuilding Plan,” October 2018. Located at: https:// www.cbo.gov/system/files?file=2019-01/54564-FY19Shipbuilding.pdf; U.S. Congressional Budget Office, “The Cost of Replacing Today’s Air Force Fleet,” December 2018. Located at: https://www.cbo.gov/ system/files?file=2018-12/54657-AirForceAviationFunding.pdf; U.S. Congressional Budget Office, “CBO’s Interactive Force Structure Tool,” last updated August 15, 2018. Located at: https://www.cbo.gov/ publication/54351; U.S. Government Accountability Office, “Weapons Systems Annual Assessment: Knowledge Gaps Pose Risks to Sustaining Recent Positive Trends,” April 2018. Located at: https://www.gao.gov/ assets/700/691473.pdf.
  2. U.S. Congressional Budget Office, “Projected Costs of U.S. Nuclear Forces, 2019 to 2028.”
  3. Jason Sherman, “Navy extending effort to define potential Tomahawk to follow-on program,” Inside Defense, January 23, 2018. Located at: https://insidedefense.com/daily-news/navy-extending-effortdefine-potential-tomahawk-follow-program.
  4. U.S. Department of Energy, National Nuclear Security Administration, “Engineering Assessment Report: Pu Pit Production Engineering Assessment.”
  5. U.S. Congressional Budget Office, “Projected Costs of U.S. Nuclear Forces, 2019 to 2028.”
  6. U.S. Government Accountability Office, “INF Treaty: Cost of Weapon Systems To Be Destroyed and Possible Saving,” March 1988. Located at: https://www.gao.gov/assets/90/87960.pdf.

Deploy a 1,550-Warhead Triad:

  1. U.S. Congressional Budget Office, “Costs of Building a 355-Ship Navy,” April 2017. Located at: https://www.cbo.gov/system/ files?file=115th-congress-2017-2018/reports/52632-355shipnavy.pdf.
  2. Leigh Giangreco, “USAF not settled on number of B-21s,” FlightGlobal, July 21, 2016. Located at: https://www.flightglobal.com/ news/articles/usaf-not-settled-on-number-of-b-21s-427700/.
  3. O’Rourke, “Navy Columbia (SSBN-826) Class Ballistic Missile Submarine Program: Background and Issues for Congress.”
  4. Collina, “The Unaffordable Arsenal: Reducing the Costs of the Bloated U.S. Nuclear Stockpile.”
  5. Amy Woolf, U.S. Strategic Nuclear Forces: Background, Development, and Issues, November 21, 2018, Located at: https://fas.org/sgp/crs/nuke/ RL33640.pdf.
  6. “Life Extension Programs send missiles into the future,” U.S. Air Force, October 26, 2012. Located at: https://www.20af.af.mil/News/ Article-Display/Article/457746/life-extension-programs-send-missilesinto-the-future/.
  7. Lauren Caston et al., “The Future of the U.S. Intercontinental Ballistic Missile Force,” Rand Corporation, 2014, Located at: https:// www.rand.org/pubs/monographs/MG1210.html.
  8. Kingston Reif, “Price Tag Rising for Planned ICBMs,” Arms Control Today, October 2016. Located at: https://www.armscontrol.org/ ACT/2016_10/News/Price-Tag-Rising-for-Planned-ICBMs.
  9. Reif, “New ICBM Replacement Cost Revealed.”
  10. Todd Harrison, “Options for the Ground-Based Leg of the Nuclear Triad,” Center for Strategic and International Studies, September 21, 2017. Located at: https://www.csis.org/analysis/options-ground-basedleg-nuclear-triad.
  11. Lauren Caston et al., “The Future of the U.S. Intercontinental Ballistic Missile Force.”
  12. Mark Gunzinger et al., “America’s Endangered Nuclear Deterrent: The Case for Funding Two Critical Capabilities,” War on the Rocks, April 23, 2018. Located at: https://warontherocks.com/2018/04/ americas-endangered-nuclear-deterrent-the-case-for-funding-two-criticalcapabilities/.
  13. Todd Harrison, “Options for the Ground-Based Leg of the Nuclear Triad.”
  14. U.S. Government Accountability Office, “Surplus Missile Motors: Sale Price Drives Potential Effects on DoD and Commercial Launch Providers,” August 2017. Located at: https://www.gao.gov/ assets/690/687571.pdf.
  15. “U.S. Nuclear Weapons Policies: A Conversation with Michèle Flournoy,” Kingston Reif and Maggie Tennis interview with Michèle Flournoy, Arms Control Today, July/August 2017. Located at: https:// www.armscontrol.org/act/2017-07/interviews/us-nuclear-weaponspolicies-conversation-michele-flournoy.
  16. Collina, “The Unaffordable Arsenal: Reducing the Costs of the Bloated U.S. Nuclear Stockpile.”
  17. U.S. Congress, House of Representatives, Committee on Armed Services, “Military Assessment of Nuclear Deterrence Requirements,” March 8, 2017. Located at: https://docs.house.gov/Committee/Calendar/ ByEvent.aspx?EventID=105640.
  18. U.S. Congressional Budget Office, “Options for Reducing the Deficit: 2019-2028,” December 2018. Located at: https://www.cbo.gov/ system/files?file=2018-12/54667-budgetoptions.pdf.
  19. U.S. Congressional Budget Office, “The Cost of Replacing Today’s Air Force Fleet.”
  20. Collina, “The Unaffordable Arsenal: Reducing the Costs of the Bloated U.S. Nuclear Stockpile.”

Deploy a 1,000-Warhead Dyad Without ICBMs:

  1. U.S. Government Accountability Office, “Program-Wide Strategy and Better Reporting Needed to Address Growing Environmental Cleanup Liability.”
  2. James N. Miller Jr. and Richard Fontaine, “A New Era in U.S. -Russian Strategic Stability,” Center for a New American Security, September 19, 2017. Located at: https://www.cnas.org/publications/ reports/a-new-era-in-u-s-russian-strategic-stability.
  3. Kingston Reif, “Air Force Seeks Mobile ICBM Option,” Arms Control Today, April 2016. Located at: https://www.armscontrol.org/ ACT/2016_04/News/Air-Force-Seeks-Mobile-ICBM-Option.
  4. James N. Miller Jr. and Richard Fontaine, “Navigating Dangerous Pathways,” Center for a New American Security, January 30, 2018. Located at: https://www.cnas.org/publications/reports/navigatingdangerous-pathways.

U.S. Nuclear Excess

The projected cost of the proposed U.S. nuclear spending spree is staggering and it is growing. The United States currently plans to spend nearly $500 billion, after including the effects of inflation, to maintain and replace its nuclear arsenal over the next decade. Over the next 30 years, the price tag is likely to top $1.5 trillion and could even approach $2 trillion. This report describes the ways in which this level of spending is unnecessary, unsustainable, and unsafe. It outlines three realistic options to reduce spending on nuclear weapons while still maintaining a devastating nuclear deterrent. The report also recommends key steps Congress can take to enhance affordability and improve its understanding of the underlying policy assumptions and long-term budget challenges.

© Copyright 2019 - Arms Control Association