It’s no secret that America has a weapons production problem.
On August 5, Deputy Secretary of War Steve Feinberg lit a fire under industry to solve this problem. He sent a memo to industry leaders giving them 21 days to submit plans to “drive significantly faster, more aggressive delivery schedules and/or increased production for critical capabilities. . . .Years-long development cycles are not acceptable. We must dramatically accelerate our program schedules and expand our production capacity now.”
Here at First Breakfast, we put the question to some of our favorite doers in the industry: If you had free reign to fix America’s munitions bottleneck, what would you propose?
It’s been three weeks, and we’re sharing the best—and most creative—proposals we received.
Dear DepSecWar: may you find additional inspiration in these hot takes. 🫡
Open the Magazine: Compulsory Licensing for Mass Production
Shyam Sankar, Chief Technology Officer, Palantir Technologies
The problem is not that we lack designs. We have exquisite weapons that work. We lack the ability to build them at rate, because the technical data package is a monopoly asset and the monopolist’s incentive is margin per unit, not units per month.
Fix the incentive.
Mandate that every prime and every supplier license the complete Technical Data Package (TDP) for any weapon in production or sustainment whose R&D was all taxpayer funded: drawings, materials, process specs, source code, test procedures. The price: a royalty at last year’s realized cost-plus margin per unit produced. The prime keeps the economics. It loses the exclusivity it did not earn given taxpayer-funded development.
Any firm that puts a $100M balance sheet behind the work may take the license. Capitalization is the credibility test, not past performance narratives or a facility clearance dating to the Cold War.
The licensee’s job is to optimize the weapon for manufacturing. Substitute materials. Delete features. Loosen the tolerances physics does not require. Then test the variant to characterize what it actually does and publish those numbers as the new spec. A missile at 70 percent of the original range that can be built in a tenth of the time at a third of the cost is not a degraded weapon. It is a different weapon, and it is the one that wins.
Award on one metric: qualified units per month per dollar, delivered from a line that is instrumented, automated, and designed to surge. The time for artisanal cells and high-touch labor as a business model has passed.
Congress should treat the TDP the way it treats spectrum: a public asset licensed for public use, not a private toll booth on national survival.
We are one contingency away from learning our magazine depth the hard way. The designs already exist. Free them.
The Real Munitions Bottleneck Is Qualification, Not Capacity
Kevin Capozzoli, CEO, Critical Materials Group
America’s munitions shortfall is usually framed as a capacity problem. But much of the delay is administrative (Qualification and Permitting) rather than physical. The same explosive material is qualified from scratch program after program, and even proven material waits months for a Department of Transportation EX number classification to simply move the material. The EX-number is required the first time a material is manufactured, but if the production method changes, another EX number is required.
Federal IT once faced the same problem. The FedRAMP 20x reform replaced from-scratch review with control inheritance, raising cloud authorizations from under 50 per year to more than 100 in six months.
Energetics can adopt the same model: qualify a material once, authorize a formulation once, and let end articles inherit their constituent authorization, assessing only the integration delta. Fold EX-number classification into that package, fast-track process improvements without full recertification, and create a second-source equivalence pathway that certifies conformance to an approved reference rather than re-qualifying from zero.
Stop Lighting Money on Fire
Mike Gallagher, Head of Defense, Palantir Technologies
The Pentagon is rightly rushing to spend every single one of its Congressionally appropriated dollars before the end of the fiscal year. If it succeeds, it will avoid a self-inflicted problem that has had a disastrous impact on our deterrent posture.
In the last decade, the Pentagon has lost more than $125 billion because of an outdated budgeting process that requires any unspent, Congressionally appropriated dollars to be returned to the Treasury. In other words, billions of budgeted dollars have been allowed to simply vanish instead of being used to replenish munitions stockpiles, procure new technology, or invest into our defense industrial base.
The FIRES Act I introduced as a member of Congress fixes that.
Rather than let this appropriated but unspent money disappear, the FIRES Act puts these dollars into a special fund that the Secretary of War could use to fund Congressionally mandated priorities. This could go a long way toward supporting priorities that may prove decisive in a Pacific conflict. The first of these: Deputy Secretary Feinberg’s efforts to ramp munitions production via multi-year purchases. The second: scaling the manufacturing and deployment of autonomous capabilities that can turn China’s coastal waters into a hellscape. The third: funding advanced energetics such as CL-20 in existing systems to extend both their range and destructive power.
The Pentagon has rightly identified munitions production and autonomy as key to preventing or winning war in the Pacific. But funding it sufficiently and in time remains a challenge. The FIRES Act gives the Secretary of War the flexibility and authority needed to fund enduring priorities at the speed of relevance, making us safer without budgeting another dime of taxpayer dollars.
Let ‘Em Build
Will Thibeau, Founder, Escoreal Solutions
Aggressive, stimulating industrial policy is necessary to harness the capacity of small and medium sized machine shops so they can build enough components for large-scale munition production. These shops run manual sales and operations processes disconnected from Prime customers. The Pentagon should extend an existing software platform to any component supplier with even the smallest production record for free.
The government must also radically accelerate the process by which suppliers become qualified to build and sell solid rocket motors, the critical bottleneck of the munitions industrial base. Long, cumbersome processes exist for validating safety and earning the right to sell explosives. The ATF and DOW should near-recklessly streamline a vendor’s ability to earn any of the multiple licenses (FFL, AS9100, etc.) required for participation in this market.
Relatedly, the government should move to a generic supplier qualification threshold. Instead of mandating suppliers validate components for each munition, there should only be a single regulatory gate for entry into the market. Allow program offices to ensure proper integration at the end-item level.
We do not lack machines, machinists, or money, but instead lack the will to let them build, and that is the one shortage no appropriation can fix.
It’s Not Easy, But It’s Not Rocket Science
Nadia Schadlow, Senior Fellow, Hudson Institute
In the category of “kind of simple” (recognizing that simple and the Pentagon are something of an oxymoron), there are some basic steps that could help avoid repeating past mistakes. Start by taking stock of recent efforts to support new entrants and expand capacity in the energetics space: What was funded? What actually got built? What new capacity came online? What worked, what didn’t, and why? This is not a “gotcha” exercise. It is a way to learn from recent investments and inform the next round.
For example, in 2024 the Department of Defense awarded $192.5 million to seven companies under the Defense Production Act to expand domestic production of 22 chemicals used in defense systems, including critical energetic precursors and other materials such as oxidizers and magnesium. What resulted? Which projects delivered the capacity they promised? Which did not? And why?
We Need a Materials Manhattan Project
Jake Chapman, Managing Director, Marque Ventures
America’s munitions shortage is real, but we’re in danger of solving the wrong problem.
Expanding production capacity, signing multi-year procurement contracts, licensing production to allies, and accelerating acquisition are all necessary if conflict comes tomorrow, but these are surge measures, not long-term national security investment strategy. They help us build more of today’s weapons tomorrow instead of fundamentally better and cheaper ones.
The defense-tech ecosystem has demonstrated that Group 1-3 drones and low-cost cruise missiles can be fielded faster and at lower cost than traditional programs. That is an important step forward. But even these companies remain constrained by the same physics and many of the same supply chains. Whether a missile is built by Lockheed Martin, a neo-prime, or a startup, it still depends on the performance of its rocket motor, energetics, turbine materials, structural composites, semiconductors, and manufacturing inputs. Better business models alone cannot overcome physical limits.
TLAMs are expensive partially because of who builds them and how they’re built but also partially because they deliver singular range, precision, and punch. If America wants to solve its munitions crisis without sacrificing capability, we must invest in changing the physical constraints our engineers design around.
Imagine rocket motors with significantly greater energy density, airframes built from lighter and stronger materials, turbine blades capable of operating at higher temperatures, manufacturing processes that eliminate complex machining, and energetics that deliver more kill per kilogram. Every improvement in these dimensions compounds offering greater range, larger payloads, lower cost, reduced logistics, and faster production. Rather than accepting today’s tradeoffs, we should be working to eliminate them.
The Department of War should launch a modern “Materials Manhattan Project” focused on advanced energetics, propulsion, structural materials, manufacturing processes, and metallurgy for munitions. America should not aspire to build more 1980s missiles with 2020s factories. We must invent the 2045 missile; one that flies farther, hits harder, costs less, and can be produced in overwhelming numbers.
Outlaw the Obvious
Rosie the Riveter
Bad incentives by the monopsony have produced bad business practices at the defense primes. Here are five proposals that should already be law:
Every prime should be working five days a week (nope, not the case today) and should be limited to one offsite a year. The carpet people work four days a week and go to leadership offsites every month, while the steel-toe rank and file work to build more weapons on the factory floor.
Primes shouldn’t be allowed to design fancy custom tape if duct tape will do the job, but today they get away with doing just that. FASA—the 1994 law requiring the government to prefer commercial technology—extends to prime contractors choosing whether to buy or build capabilities for a given government contract, even after contract award. But there is neither an enforcement mechanism nor an incentive with cost plus contracting to follow the law. What we get instead is primes expending enormous resources—much of it taxpayer money—rebuilding commercial capabilities that already exist. It’s time to end the self-dealing that rewards primes for prioritizing internal development at the expense of literally everything else.
Force primes to cut their consulting budget by 90 percent and invest the money into training programs for technicians and manufacturing engineers. Why are you asking the DoW for $20 million for McKinsey when you won’t spend that amount of money on improving your operations on the floor? If you won’t invest in your business operations, you should at least invest in the country’s future workforce. No more carpet people!
For every one month late a prime is on a program delivery, the DoW should get five percent ownership of that program’s intellectual property. You shouldn’t own both the critical IP and also refuse to get it made. Delay long enough, and the government gets an ownership stake it can license to whoever has a better production plan.
Require remedial tech training for the legal counsel of every prime so they can understand what “the cloud” is and why it’s OK to use it.
Form the Army Munitions Production Corps
The Angry Hobbit
The blueprint to fix the munitions production crisis exists, and we have done it before. In WWII we leveraged the Army Corps of Engineers to build multiple munitions plants. Many of those plants are still operational and operate under the government owned, contractor operated model.
We should use the corps of engineers to build out new weapons plants for guided munitions, interceptors, and missiles instead of awarding more money to primes that know how to play the game for more money.
We should then form the Army Munitions Production Corps to surge production and staff these plants to make up for the gap in contractors operating these plants. The production corps should be either a new MOS within the Army, or an adjacent formation that can take volunteers or individuals that wish to serve but cannot pass military physical requirements and train them to be critical factory workers that can be assigned to production facilities at the needs of the government.
Running these plants as primarily government owned and government operated is expensive, but does not have to be permanent. After the initial surge of soldiers/production corps personnel, their income and taxes will boost the local economy at these rural factory locations and enable these towns to support future contractor/DA Civilian jobs and recruiting.
In peace time, these programs can be used to train reservists and guardsmen with the skills to operate factory machinery and assembly lines to boost civilian industry with skilled labor and allow for mobilization to surge production at these munitions plants in times of need.
Earning by Doing
Chase Dalton, Federal Deployment Lead, Cognition
Congress should pass a War Prize Act, creating a system of commissioned military enterprises to enable a capitalist market for wartime military effects. This would parallel traditional procurement channels. During authorized conflicts, the government would maintain continuously updated price schedules for military effects: enemy soldiers killed, aircraft destroyed, ships sunk, infrastructure disabled, equipment captured, or other strategically valuable outcomes.
This proposal carries obvious second-order risks, but the hour is late and worth the gamble. As the monopsonist, the Pentagon decides what to buy, how, and from whom, then bears the development and execution risk. This has not produced the munitions we need to defeat China or Iran.
Before war, the DoW should publish non-binding indicative schedules so investors know which outcomes will command premiums. Imagine a billion dollar prize for destroying the Fujian, China’s newest carrier, within 48 hours of hostilities over Taiwan. Private capital would finance competing solutions. China would have to worry about the known U.S. order of battle and privately financed capabilities. Lower-intensity conflicts could serve as proving grounds, refining capabilities, incentives, and capital structures before a peer war.
The government would dynamically set the commission envelope of authorized and prohibited targets, preclearance strike rules, geographic limits, legal constraints, and prize values. Commissioned firms would control the financing, development, production, inventory, force structure, operational planning, and execution within that envelope.
Prize claims should be transferable and financeable. Firms could raise equity, borrow against expected payouts, form prize funds and consortia, syndicate risk, insure campaigns, and build portfolios across military objectives.
Accountability must be severe. Commissioned firms would post insurable bonds, maintain auditable records, and face prize forfeiture, commission revocation, civil liability, and criminal prosecution for violations.
Forge the (Allied) Arsenal of Democracy
Daryl Wieland, Senior Director of Partnerships, Palantir
The United States and her closest allies should create a single, treaty-backed industrial utility called the Allied Arsenal Corporation that collaborates on defense industrial production.
While that may be provocative and require a lot of trust, the question we need to ask ourselves is can the United States and its allies create a system where innovation is rewarded, production is fungible, capacity is visible, and no single owner, factory, country, or appropriation cycle becomes the limiting factor.
That system would be a licensed, modular built-to-print production network. Original design owners would retain economic upside through transparent IP licenses, per-unit royalties, and payments for engineering support and continuous improvement.
Such an arrangement wouldn’t be entirely unprecedented. Just this week, the UK announced it will share blueprints of its Storm Shadow cruise missiles with Ukraine. In July, France similarly provided the Ukrainians with production rights for the Scalp missile. America may not be too far behind. President Trump initially announced the United States would grant the Ukrainians license to produce the Patriot missile, although he has since walked back the commitment. Trump’s stated reason was “it’s a hard thing to give away, that kind of technology.”
But that’s where the Allied Arsenal could be most effective. It would standardize the process for sharing technology via approved manufacturers, secure data standards, and clear agreements about access. Specialization would turn national strengths into allied capacity, much as markets coordinate dispersed knowledge and incentives.
The result would be more than a larger arsenal. It would be an adaptive industrial base: competitive enough to reward invention, coordinated enough to survive crisis, and distributed enough that no bottleneck can decide the future of security.