The Defense Industrial Base Is the Story of 2026: Catapults, Motor Lines, and Capital Chasing Capacity
Read the defense news of the past two months as a single sequence and one theme runs through nearly all of it. The binding constraint on Western military power right now is not doctrine or design. It is the ability to physically produce things, and every major decision of the summer has been an attempt to buy, borrow, or legislate capacity that does not currently exist.
The clearest expression came out of the White House this week. A presidential memorandum ordering the Navy to rip electromagnetic catapults out of future Ford-class carriers and revert to steam sits inside a broader shipbuilding directive that also opens the door to foreign-built surface combatants and calls for a fifth public shipyard. The catapult clause is the one that will dominate coverage, and it is the weakest part of the document, because the Ford hull has no steam distribution system to revert to and the supply chain for the old catapults closed years ago. The rest of the memo is a candid admission that the American yards cannot deliver the fleet the Navy has been told to build.
Missiles show the same pattern with a cleaner solution. L3Harris signed two seven-year framework agreements to quadruple THAAD propulsion output and nearly triple PAC-3 MSE motor production, both aimed at the solid rocket motor bottleneck that has capped interceptor deliveries regardless of how much money Congress appropriates. Long-duration frameworks are what actually unlock private capital expenditure on a motor line, and the seven-year term is the meaningful number in both deals. Europe is doing the equivalent through partnership rather than expansion, with Lockheed Martin and Rheinmetall signing a memorandum of understanding for ATACMS co-production on the continent. Allied procurement is following the same logic: NATO has opened formal negotiations with Saab for up to ten GlobalEye airborne early warning aircraft, a purchase from an existing production line rather than a development program.
Private capital has drawn the same conclusion and is pricing it aggressively. Anduril’s talks around a $100 billion valuation came only months after a $61 billion round, and whatever one thinks of the multiple, the underlying bet is on manufacturing throughput rather than on any single product. Smaller rounds tell the same story at different points in the stack. Aurelius Systems raised $40 million to scale an autonomous counter-drone platform, and Antares raised $470 million to field compact fission microreactors on U.S. bases by 2028, which is an energy-supply play dressed as a nuclear one. Installation power is becoming a procurement category in its own right as directed energy and high-density computing move onto bases.
The unglamorous enablers matter more than they look. Arkenstone Defense came out of stealth with $35 million aimed squarely at the administrative burden of becoming a government contractor, which is the real filter on commercial entrants no matter how many innovation offices the Pentagon stands up. Compliance infrastructure is following the same path, with Cloudflare for Government reaching FedRAMP High and committing to pursue DoD Impact Level 4. Connectivity is being bought the same way, through SES Space and Defense winning a five-year blanket purchase agreement with Space Systems Command for managed Ku-band services. None of these are weapons. All of them determine how quickly weapons reach the field.
At the sharp end, procurement is moving toward things that can be built in volume and lost without anguish. The Army selected Teledyne FLIR’s Rogue 1 for the LASSO loitering munition program, putting man-portable precision anti-armor strike into infantry brigade combat teams. Ondas paid roughly $125 million for Cyberhawk to extend its autonomy platform into critical infrastructure inspection, a reminder that the same drone stack sells into two very different budget lines.
The recurring lesson across all of it is that capability decisions and industrial decisions have collapsed into the same decision. That is why the carrier memo is worth reading closely even if you never think about flight decks. It treats a launch system as the problem when the actual failure was concurrency, sequencing, and a supplier base too thin to absorb a redesign. The programs succeeding this summer are the ones that bought stability first and capability second.