Congress should save the Sentinel ICBM—its true value is more than simply its cost

In January, the US Air Force notified Congress that the new Sentinel intercontinental ballistic missile (ICBM) was costing more than expected to build. At 37 percent, the projected cost overruns were considered “critical” and triggered the Nunn-McCurdy act, a law designed to curtail cost growth in major military procurements. Now, as both supporters and critics of the program gear up for the required congressional testimony this summer, it’s important to understand not only what the program costs, but also its true value.

A thorough review reveals that some of the cost estimate increases seem understandable given expensive but worthwhile new requirements. These include replacing thousands of miles of communication cables, improving security, and rebuilding hundreds of silos. Understood in full context, Congress should insist on continuing this vital program—indeed, most Nunn-McCurdy-breaching programs are continued. At the same time, supporters in Congress and the Pentagon need a better argument regarding costs and value propositions. They need to make it clear, especially in the event that costs rise further still, that the value of program should be measured in decades, over its full lifetime, and not simply by its current price tag.

What is a Nunn-McCurdy Review?

Effectively fighting to retain Sentinel requires first understanding when a Nunn-McCurdy Review (NMR) becomes necessary. There are two kinds of cost that trip an NMR: either a sharp rise in the estimated program cost (research and development, construction costs, and procurement of each weapon) or a rise within single unit cost (that is, the cost of just one bomb, aircraft, or satellite). Both numbers are looked at on a per-unit basis to make it easier to weigh the cost of a particular capability separated from order quantity. Large acquisition programs take years to complete and so undergo checks at key milestones. A program’s cost estimate is updated with each milestone and so called its ‘current’ estimate, but the original can only be updated during an NMR. Finally, there are two types of review under the NMR based on the degree of cost overrun. The first is a “significant” breach, which entails a cost increase of either 15 percent above the current projection or 30 percent over the original, while a “critical” breach would be 25 percent or 50 percent respectively. A critical review is performed under the presumption of program cancellation—and such is the case for the Sentinel program now. Therefore, continuing the program will require a root-cause analysis, program restructuring, and the personal sign-off of the US secretary of defense. So, in total, there are eight ways to trigger an NMR, as indicated below.

The Nunn-McCurdy Act passed more than four decades ago to curtail large expenditures for significant defense projects, such as the Patriot missile defense system. From 1997 through 2015, there were ninety-four NMR breaches in total, with nearly half being major programs. Additionally, almost all of these large projects breached the rule twice before completing initial purchases.

Why is Sentinel so expensive?

It is no surprise that the Sentinel program—which will rebuild more than six hundred silos and other facilities, including creating hundreds of new cutting-edge modular missiles, all while consolidating and updating several complicated command-and-control systems—is expensive. For too many years, the Department of Defense waited to carry out missile, communications, silo, and other updates, and all those areas must now be updated all at once. This necessitates a much larger project with more variables, making it even harder to predict expenses.

With the Sentinel program, the Air Force is purchasing much more than just replacements of four-hundred operational Minuteman missiles. The program will also refurbish or rebuild more than six hundred facilities and purchase hundreds of maintenance and other support vehicles, as well as more than one hundred missiles for ongoing testing.

So what, then, explains the unanticipated costs? Pushing new technology too fast often leads to unexpected costs. But seventeen of Sentinel’s eighteen major subsystems are out of the danger zone of being considered “immature,” a period in which cost overruns often originate. Sentinel’s challenges have a blend of sources, including changed requirements, lingering aftereffects of economic realities driven by the COVID-19 pandemic, and most importantly, sheer scale.

The Air Force has never attempted an infrastructure build at this scale and the initial plan was overly optimistic. That initial rosy cost estimate was reconciled with the broader Pentagon’s cost estimate, which was around $95 billion. But the initial approved budget still supposed it could avoid rewiring and major rebuilds. This assumption might have held, but by early summer 2023, the scope of the requirements for the program was changed, in part, to deal with new threats. Though costly, the altered requirements improve Sentinel by leveling up to modern communications capabilities and by adding “security enhancement of both missile hardware and facilities.” A new fiber-based underground communications structure and sixty-two towers will add redundancy while requiring fewer people to operate, given the planned halving of launch control centers. The personnel savings may also impact security forces as well, with reduced on-site maintenance activity and lower false-alarm response rates. These design changes ensure US preeminence in nuclear weapon safety and security while suggesting relief options for recruitment challenges.

Another reason for the increase in the Sentinel program’s cost has been economy-wide increases in costs of construction above background inflation rates. Difficulties in finding and retaining engineers, including getting them timely security clearances, were also cited for increasing cost projections. Manufacturing and construction have been disproportionately impacted by supply-chain interruptions and labor scarcity. In 2019, when the most recent Sentinel cost estimates were being developed, no one was predicting the enduring economic impact of a pandemic.

But the simplest and biggest factor is the sheer size and scope of the project, and with that comes added uncertainty. This is why even post-NMR, the cost estimate has the potential to rise again.

Anyone who has renovated a house or even just watched a few HGTV episodes will know that, even if a building has “good bones,” surprises happen. The fine print on home renovation contracts accounts for the possibility that, after construction begins, new problems will be discovered. Half a century ago, Minuteman siloes in Montana were “rushed to completion,” with workers pouring concrete amid the Cuban missile crisis. These Cold War–era silos remain in some of the harshest environments that exist within the contiguous United States. Sentinel’s NMR should include comprehensive assessments of silo foundations to confirm that only the planned partial silo rebuilds is needed. Likewise, renovation can often be more costly than a new build, so there is merit in looking at so-called “green pasture” build options.

A better value proposition

Sentinel is a large and complex program, but it is worth more—and is a better value—than most realize. The Sentinel missile is slated for up to a sixty-year lifespan, now with a $131.5 billion projected cost for four hundred operational missiles. This amounts to an annualized cost of $2.2 billion for the fleet and facilities, or about $2.7 million per year per missile. For comparison, the F-35A has a flyaway cost of $82.5 million and a fifty-year planned lifespan, giving a per-unit-design-year cost of just under $1.6 million. Taking into account the discrepant availability rates for these two systems, the design year costs of putting each on alert twenty-four hours a day nears parity. This designed lifetime on-mission cost comparison is admittedly incomplete and inexact. But it offers a better way to compare costs across two headline military systems, both of which have faced an NMR.

An upside of Sentinel’s longevity by design is lowered maintenance. Over the past five years, Air Force Global Strike Command logged 2.5 million maintenance hours to keep Minuteman operational, a 30 percent increase from the previous five years. In March of this year, Lt. Col. Anthony Santino, the head of ICBM flight testing, said that he expects a further 25 percent increase over the rest of this decade. That much maintenance translates to needing at least sixty more full-time mechanics at a time when it’s getting harder to meet basic recruiting goals. Sentinel will allow personnel reductions and retain that savings for at least half a century. The nature of the nuclear mission does not lend itself to just-in-time production, or maintenance, to make up for weapons that fall offline during a crisis. Sentinel will invert the manpower needs curve while improving security, reliability, effectiveness, and upgradability of the ICBM force.

Seeing how the Sentinel program adds unique value to national defense requires thinking one step past the eight ways to trip an NMR. Using the annualized design-life cost metric is perhaps one way to show that Sentinel is no more expensive than other recent major military purchases. This framing helps show the sensibility in sustaining the cheapest means of securing strategic deterrence for future generations.


Lieutenant Colonel James McCue is a senior US Air Force SkillBridge fellow with the Atlantic Council’s Scowcroft Center for Strategy and Security in the Forward Defense program.

Note: The Atlantic Council’s Scowcroft Center for Strategy and Security conducts work on nuclear and strategic forces that is sponsored by donors including Los Alamos National Laboratory, Northrop Grumman Corporation (which has the sole contract from the US Air Force to engineer and manufacture Sentinel ICBMs), the Norwegian Ministry of Defense, the Swedish Ministry of Foreign Affairs, the United States Department of Defense, the United States Department of Energy, the United States Department of State, as well as general support to the Scowcroft Center. This article did not involve any of these donors and reflects only the author’s views. The positions expressed do not reflect the official position of the United States Air Force or the Department of Defense.

Further reading

Image: An unarmed Minuteman III intercontinental ballistic missile launches during a developmental test at 11:01 Pacific Standard Time Tuesday, Feb. 5, 2019, at Vandenberg Air Force Base, Calif. (U.S. Air Force photo by Tech. Sgt. Patrick Harrower/Released)