Nuclear weapons are expensive, but how much is too much? 

America is planning to spend more than $1 trillion over the next 30 years to upgrade all three legs of its nuclear triad. This hefty price tag is expected to grow even more now that, as the Air Force notified Congress on Jan. 18, the new Sentinel intercontinental ballistic missile (ICBM) will cost 37 percent more than projected just four years ago.  

This dramatic increase has triggered a multifaceted, several-month review designed to keep expensive programs from becoming unaffordable. Ensuring this review is transparent to Congress and the American people is crucial to understand the causes and consequences and enforce oversight of one of the largest nuclear weapons programs in U.S. history. 

To compound concerns, the Sentinel program is also at least two years behind schedule and has been plagued by a host of problems that can be linked directly to flawed assumptions, overly optimistic planning, and rushed decision-making from the program’s beginning. Yet, instead of returning to the drawing board and pursuing a viable, affordable option, the Pentagon is likely to argue that the Sentinel program must continue.   

What caused the overrun 

We are talking about a complex system that is literally rocket science, but the cost overrun has little to do with the missile itself. According to Andrew Hunter, assistant secretary of the Air Force for Acquisition, Technology, and Logistics, there has been a slight increase in cost of the missile due to inflation, but the cost and schedule growth for the project is largely due to unforeseen supporting infrastructure costs. The larger, more modern missiles require refurbished silos, larger launch facilities, new command and control centers, and complete replacement of cabling for communications infrastructure. Kristyn Jones, acting undersecretary of the Air Force, similarly cited the “massive ‘civil works’ project” as the primary overrun cause.  

The Government Accountability Office’s annual evaluation of Pentagon weapons programs in June 2023 additionally revealed that the Sentinel program was delayed because Northrop Grumman — which was issued an unprecedented sole-source contract for a program of this size in September 2020 — is experiencing staffing shortfalls, IT infrastructure challenges, and supply chain disruptions. 

Although infrastructure and industry costs are certainly a contributing factor, cost estimates for the Sentinel program were overly optimistic from the start. Indeed, this is not the first time the estimates have gone up.  

The Air Force’s cost estimate for the Sentinel program has increased from $62.3 billion (in then-year dollars) in February 2015 to between $85 billion and $100 billion just nine months later. Then, after the Air Force set the estimate at $85 billion in 2016, it increased again in 2020 to $95.8 billion. The next cost estimate for the program, scheduled for later this year, could jump to more than $130 billion.  

In short, the cost of the project has doubled in less than a decade. 

It is far from clear the Air Force ever did a thorough assessment of alternatives before embarking on the Sentinel program. The Air Force had multiple options, including building fewer missiles, making current missiles last longer through a life-extension program, or perhaps forgoing the ICBM program altogether and buying more bombers instead. However, there is no evidence that other options were ever seriously considered, since the Air Force never publicly released its internal Analysis of Alternatives for the ICBM replacement program.  

What we do know, from a 2016 Air Force report to Congress, is that the Analysis of Alternatives relied on arguably arbitrary requirements, such as keeping a force level of 400 ICBMs and maintaining the force until 2075. This helped ensure that Sentinel appeared as the cheaper option over a life-extension program for the current arsenal of Minuteman III ICBMs.  

As Matt Korda, senior research fellow at the Federation of American Scientists, argued in a 2021 report: slight changes in the requirements would have revealed a cheaper cost estimate in favor of life-extension. It is also clear from Hunter and Jones’s statements that the Air Force failed to thoroughly assess infrastructure needs in its original cost estimate: such large-scale and integral pieces of the program as new launch facilities and cabling should never have been unforeseen.   

Nuclear weapons, it seems, are not without a sense of irony. In launching the new ICBM program, the Air Force and other Defense officials argued that a life-extension program for the existing ICBM force was too expensive and not feasible. However, the Air Force is now, according to senior Air Force and Northrop officials, likely forced to life-extend some Minuteman III ICBMs to maintain the force level of 400 ICBMs now required by Congress since 2017 via the National Defense Authorization Act.  

In notifying Congress of the overruns, the Air Force has now admitted that the low cost projection that was used to secure congressional approval and lock the program in was made with incomplete data. After the newest cost increase was disclosed, the Air Force acknowledged: “Some of the assumptions that were made at the beginning of the program when the initial cost estimates were made were just not particularly valid.” 

What happens next 

The Sentinel program is now in “critical” breach of the Nunn-McCurdy Act. Programs in critical breach of the act are to be terminated, unless the secretary of Defense submits a certification that the program is essential to national security, which Secretary Lloyd Austin will almost certainly do. Lt. Gen. Richard Moore, deputy chief of staff for Plans and Programs, insisted the Sentinel program will not be canceled: “Sentinel will be funded. We’ll make the trades that it takes to make that happen.”  

A Congress and Defense Department that seem to care about spending and budget priorities ought to take a close look at the requirements that must be met in a certification by Secretary Austin. There are cheaper and more efficient alternatives to the Sentinel program. The Air Force’s refusal to consider and pursue these begs the question of whether and how Secretary Austin will certify the necessity of the program for national security. The continuous delays and cost growth of the Sentinel program reveal a persistent failure in program management by the Air Force. Any certification presented by Secretary Austin must address this failure. 

Incomplete data, rosy cost projections, and excessive secrecy appear to have combined to push the Sentinel program deep into the red. At a minimum, the Pentagon and/or Congress should make all steps and results of this Sentinel review process open to the public to ensure maximum transparency, scrutiny and oversight. Secretary Austin’s likely certification of the Sentinel program should be open to public interrogation, and Congress must thoroughly examine whether every requirement is met before allowing the program to continue. Congress should ask the Government Accountability Office and Congressional Budget Office to conduct independent reviews to interrogate the Pentagon’s justification for Sentinel and ensure hawk-eyed scrutiny of the program’s next steps. 

Mackenzie Knight is a Herbert Scoville Jr. Peace Fellow on the Nuclear Information Project at the Federation of American Scientists.