GAO released its space supply chain report yesterday & it's not pretty. In short, the Dept of War plans to spend $33B over the next 5 yrs on proliferated satellite constellations. Some critical parts have exactly 1 qualified supplier, germanium for solar cells comes largely from China, and circuit board lead times just helped push a missile warning launch back 8 months.
Quick roundup of the 5 things the private sector should be watching based on GAO's findings:
1. The bottlenecks are clearly named.
Space Force's integration office ranked its top 5 industrial base risks: encryption devices, satellite-to-satellite laser links, the chips that turn sensor images into digital data, radiation test facilities & germanium wafers space solar cells are built on. Radiation-hardened processors and laser communication terminals also show up as risks across multiple programs.
2. Single points of failure.
There's 1 space-qualified cover glass supplier for solar arrays. Domestic suppliers of certain focal plane arrays (the imaging sensors in missile warning satellites) are down to 2. Aerospace Corp found 12+ assemblies & components w/ only 1 supplier that meets DoW needs.
3. Jan 1, 2027 is a hard date.
EO 1445 directs DoW to stop issuing routine waivers for tungsten, tantalum & rare earth magnets from adversarial nations. Germanium for space solar cells comes largely from China. One program already switched star tracker designs bc the prior supplier didn't comply with U.S> sourcing law.
4. Shortages are already costing schedule.
Resilient Missile Warning & Tracking MEO's first launch slipped from Sept 2026 to May 2027, partly on printed circuit board lead times. The same suppliers are building for the same vendor across multiple platforms. PWSA, Space Development Agency's 300+ satellite missile warning and tracking constellation, keeps hitting delays on laser communication terminal production.
5. Money & visibility are moving below the primes.
PWSA contracted directly with 2nd & 3rd tier suppliers to fund infrastructure. DPA Title III put $18.1M into U.S. germanium refining & $15M into focal plane arrays (part of an $80M effort thru FY29). A DCMA study due Mar 2027 will map production capacity, bottlenecks, and sole source risk for low Earth orbit constellations, then identify needed investments. An Air Force memo directs Space Force programs to map their supply chains down to the component level.
Full report below:
Army had previously funded a MuniRem demonstration in 2009 ($140K), a $333K field validation task from 2013 to 2015, and a $125K neutralization order at Aberdeen in 2024
Army Contracting Command Rock Island had awarded MuniRem $124,647, sole source, for lead azide neutralization at the Iowa plant, w/ work starting March 26. The federal record shows one modification, dated Aug 13, w/ the full amount still obligated and performance listed thru March 2027. Army's environmental assessment, posted this week, says neutralization would have avoided a blast plume but required weeks of handling crystallized drums.
Another rough GAO report this week: munitions.
DoW has persistent munitions shortages. Congress appropriated at least $25B in 2025 to strengthen munitions resources. And DoW has no overarching plan linking production & funding to what it actually needs.
The public version strips out the actual shortfall numbers but what's left still tells you a lot
Here's what matters if you're building or investing in this space:
Demand is reactive. Without a plan, DoW & the services react to global events instead of total requirements. The Army's 155mm ramp was driven by shortages from supporting allies & partners, not DoW's total requirement.
Nobody owns the fix. Authority is spread across A&S offices, a program executive office, service program offices & working groups. The Munitions Acceleration Council, stood up in July 2025, still has no signed charter as of this month.
DoW can't fully trust its own count. Its dept-wide inventory tool pulls from 37 data sources. It hasn't been regularly reconciled w/ the services & has no data management plan.
The plants are old. Many Army ammo plants still run on original WWII-era buildings, equipment & production processes.
The supplier base is narrow. GAO notes 5 large primes dominate the commercial base. @serellium pulled DoW guided missile contracts active since Oct 2025: $182.8B obligated across 33 companies. Lockheed & RTX hold 91%. Less than 4% went thru full & open competition. 83% of the $$ sits on awards that drew 1 offer.
GAO's 4 recommendations cover planning, leadership & data. DoW concurred w/ 3 & partially concurred w/ 1. All needed.
Big question is whether that plan will count new suppliers as part of the answer
@hgandreasen Yes, many. Rayon fiber, laser crystals, radiation detector crystals, certain explosives, ceramic fiber, coal tar pitch to name a few. All flagged in SBIR topics as foreign-dependent or short on U.S. sources
What is so important about this issue is that the govt saw the rayon gap coming for 20+ yrs and said so publicly.
- 2002: space & missile systems center asked industry for U.S. aerospace-grade rayon suppliers
- 2016 MDA SBIR topic states rayon is no longer domestically produced
- 2021 and on: DLA buying imported aerospace-grade rayon for the national defense stockpile
Every one of those was a public demand signal. Market research notices, SBIR topics, stockpile purchases. Any U.S. company with fiber, textile, or carbon materials capability could have read them.
Meanwhile DoD's own 2022 competition report found small businesses in the defense industrial base fell 40% in a decade.
The rayon shortage got the headline but the MUCH BIGGER problem is that the businesses who could answer signals like these never see them.
The Pentagon urgently needs more Patriot missiles. It has a textile problem.
Only one rayon product is approved for use in the most advanced Patriot missiles, and it comes from the Kansas factory of a single supplier, Park Aerospace.
The rayon begins its life as strands of fiber at a factory in the Czech Republic owned by a Thailand-based chemicals conglomerate. That yarn makes its way to France to be woven into fabric. ArianeGroup carbonizes the fabric at its facility near Bordeaux, then sells it to Park Aerospace, its lone North American distributor. Park coats it in Kansas and sends it to L3Harris, which layers it inside the nozzles of solid rocket motors bound for Lockheed's Patriot factory.
Ariane has been clear that supply is limited. When the Pentagon pressed for more, Park's CEO said, "They're a French company. They can't be bullied."
The sole North American producer of aerospace-grade rayon, in Front Royal, Virginia, was shut down in 1989. Its successor in Elizabethton, Tennessee, wound down production in 1998. By the mid-2000s, with no major U.S. options left, the Pentagon had to look abroad.
The aerospace and defense textile supply chain is riddled with chokepoints. Critical components often come from outside the U.S. via complex webs of companies, with few or no American alternatives.
The demand is unlike anything the industry has seen. Park says, "In all our years, we have never seen anything like this." A U.S. factory for the material isn't expected until around 2030, and as Park's CEO put it, "nobody wants to wait 3½ years."
Textiles aren't just uniforms and protective equipment. They're inside the rocket motors of America's most advanced interceptors, and right now the supply depends on one company overseas.
Yes, many. Rayon fiber, laser crystals, radiation detector crystals, certain explosives, ceramic fiber, coal tar pitch to name a few. All flagged in SBIR topics as foreign-dependent or short on U.S. sources
If you build rental listing, property management, matching, or computer vision software, FEMA wants to hear from you.
FEMA’s Office of Direct Housing Operations just released an RFI on modernizing how it finds temporary housing for disaster survivors.
Here’s the problem. When a presidentially declared disaster makes homes uninhabitable, FEMA can lease rental units for eligible survivors for up to 18 months. Today, contracted property finders canvas markets by hand & send FEMA unstructured inventory reports. FEMA staff then manually cross-reference survivor case files against those lists. FEMA says this reliance on manual handoffs & fragmented spreadsheets leads to stale vacancy data & longer placement timelines, especially for survivors who need accessible units.
What FEMA wants to hear about from industry --
1. Real-time rental inventory: pulling listings from MLS, property management software APIs, rental boards & landlord portals, then confirming the units are still vacant
2. AI matching: pairing a household w/ a unit in seconds, factoring in work commutes, school districts, medical providers & household size
3. Accessibility checks: computer vision on floor plans, entry photos & property specs to flag ADA/UFAS features like ramps, ground-floor units, grab bars & wheelchair-accessible kitchens and bathrooms
4. Landlord workflow: self-service portals to submit properties, accept lease terms & coordinate inspections, plus e-signature on govt lease templates
5. Federal-grade security: FISMA Moderate, FedRAMP authorization (or a path to it) & strict protection of survivor PII
Bottom line, FEMA wants to replace spreadsheets w/ a real time matching platform.
Big question is who’s already doing this well on the private side, and whether they’ll bring it into govt.
10-page max responses, due Dec 18
This is the gap @serellium was built to close. We map any capability to real federal demand: which agencies buy it, how they buy it, the $$ moving, and where a new entrant has a realistic path in.
A $4M Patriot interceptor is waiting on a roll of fabric from France. What WSJ reported is alarming but not unusual. They reported that only one carbonized rayon is qualified for the PAC-3 MSE rocket motor nozzle. The fiber is spun in the Czech Republic, woven & carbonized near Bordeaux, coated in Kansas, then built into the motor by L3Harris.
The last U.S. plants making aerospace-grade rayon closed in 1989 &1998. A new U.S. source, if all goes to plan, opens around 2030.
Meanwhile the Army's newest PAC-3 MSE contract w/ Lockheed has $4.76B obligated & a potential value of $58.6B thru 2035. Tens of billions in demand, potentially gated by a fabric with one qualified source overseas.
The govt saw this coming. Space & Missile Systems Center asked industry for U.S. aerospace-grade rayon suppliers in 2002. In a 2016 SBIR topic, MDA stated rayon is no longer domestically produced. DLA has been stockpiling imported rayon since 2021, including a $150M order last December. Per WSJ, that rayon still isn't qualified for PAC-3 MSE.
Building a U.S. rayon plant is one path. Replacing rayon may be another option, and govt has been asking for it for 20+ yrs.
Rayon is just one example.
The defense industrial base is carrying more demand right now than at any point in recent history, on a supplier base that's been thinning since the 1990s. DoD reported in 2022 that the number of small businesses in the DIB fell 40% in a decade. Patriot is not the only program with a chokepoint like this, and most of them will never make the WSJ.
Here's what I keep coming back to.
There's innovation, production capacity, and creative thinking across American businesses, investors, and founders that could help close a gap like this.
Most have never heard of that 2016 MDA topic, and wouldn't know where to start in this market if they had. They know what they make, what they're capable of making. What they can't see is which govt programs depend on it, which agency buys it, how it gets bought, who already holds that work, & how to approach.
The demand is there. The willing businesses are there. The patriotism is there. The connection between them is weak.
That's why we built @serellium. We've built a capability map of the federal market. We take any capability & match it to real federal demand: the agencies & buying offices that purchase it, the $$ moving, how they buy it, and where a new entrant has a realistic path in.
We will not meet today's national security priorities with one qualified supplier per critical material. We need far more of American industry in the fight, pointed at the right buyers faster.
Well that's nuts. It's actually the most open door, maybe ever --
- DoW asked for $1.5T in FY27, largest defense request in history
- More than half of it is for buying munitions, planes, tanks, and ships
Shipbuilding alone:
- Navy asked for $65.8B to build ships in FY27
- Its plan calls for 34 new ships in FY27 and 122 over 5 yrs
- Navy plans to spend $6.7B over 5 yrs to grow the surface shipbuilding supply base
- Only 10% of Navy ship construction happens outside the major shipyards today, Navy wants that to be 50%
- Building it takes welders, ship fitters, pipe fitters, machinists, electricians, painters, electronic techs, + suppliers of high tech castings and forgings
- The submarine program alone needs about 140,000 skilled workers over the next decade, and shipbuilders already can't hire welders, pipefitters, and machinists fast enough
Munitions:
- Senate appropriators' draft put $47.2B into munitions
- $40.6B of that funds new multiyear contracts for 9 munitions, including Patriot, SM-6, THAAD, and tomahawk
$1.9B set aside specifically for new entrant munitions, like low-cost cruise missiles
- DoW wants munitions contracts that run up to 7 yrs, so suppliers can justify building capacity
- Every one of those missiles depends on solid rocket motors, energetics and critical chemicals, metal castings and forgings, and microelectronics
- DoW has flagged rocket motors as significantly concentrated, with production limited to very few U.S. companies
And from 2011 to 2020, the # of small businesses winning DoW contracts fell more than 40% even as DoW spent a record amount with small businesses.
The demand is greater than ever.
Lean into what you know!
Last year a vc told me that @DeptofWar wasn't going to build more ships // munitions.
Separate the signal from the noise. Lean into what you know.
Still more production.
GAO released its space supply chain report yesterday & it's not pretty. In short, the Dept of War plans to spend $33B over the next 5 yrs on proliferated satellite constellations. Some critical parts have exactly 1 qualified supplier, germanium for solar cells comes largely from China, and circuit board lead times just helped push a missile warning launch back 8 months.
Quick roundup of the 5 things the private sector should be watching based on GAO's findings:
1. The bottlenecks are clearly named.
Space Force's integration office ranked its top 5 industrial base risks: encryption devices, satellite-to-satellite laser links, the chips that turn sensor images into digital data, radiation test facilities & germanium wafers space solar cells are built on. Radiation-hardened processors and laser communication terminals also show up as risks across multiple programs.
2. Single points of failure.
There's 1 space-qualified cover glass supplier for solar arrays. Domestic suppliers of certain focal plane arrays (the imaging sensors in missile warning satellites) are down to 2. Aerospace Corp found 12+ assemblies & components w/ only 1 supplier that meets DoW needs.
3. Jan 1, 2027 is a hard date.
EO 1445 directs DoW to stop issuing routine waivers for tungsten, tantalum & rare earth magnets from adversarial nations. Germanium for space solar cells comes largely from China. One program already switched star tracker designs bc the prior supplier didn't comply with U.S> sourcing law.
4. Shortages are already costing schedule.
Resilient Missile Warning & Tracking MEO's first launch slipped from Sept 2026 to May 2027, partly on printed circuit board lead times. The same suppliers are building for the same vendor across multiple platforms. PWSA, Space Development Agency's 300+ satellite missile warning and tracking constellation, keeps hitting delays on laser communication terminal production.
5. Money & visibility are moving below the primes.
PWSA contracted directly with 2nd & 3rd tier suppliers to fund infrastructure. DPA Title III put $18.1M into U.S. germanium refining & $15M into focal plane arrays (part of an $80M effort thru FY29). A DCMA study due Mar 2027 will map production capacity, bottlenecks, and sole source risk for low Earth orbit constellations, then identify needed investments. An Air Force memo directs Space Force programs to map their supply chains down to the component level.
Full report below:
What I love about this deal:
Hadrian just bought one of the biggest bottlenecks in submarine production. It plans to spend $300M upgrading the existing plant & building a new automated factory on the same site.
The reason this is so strategic is because it can scale up naval-grade steel castings for Virginia- and Columbia-class subs, with melting, refining, and casting done all at one location.
Demand for these capabilities in the federal market is absolutely surging.
The Navy’s FY27 shipbuilding plan puts $124.9B into 5 Columbias and 10 Virginias over the next 5 yrs and wants to build 1 Columbia and 2 Virginias every year by FY31.
Today the yards deliver about 1.3 Virginias a year. The Navy’s sub production lead says hitting the target takes roughly 70M labor hours a year, and every one of those boats depends on heavy castings and forgings the supplier base already can’t keep up with.
Dept of War has also put $10B+ into the sub industrial base without knowing how much more it will take.
In March the Navy put $900M behind Hadrian to stand up 3 factories, 2 of them focused on castings, forgings and other critical materials.
Kutztown gives Hadrian a working foundry already inside the Naval Reactors supply chain, so it can start meeting that demand now instead of starting from zero.
Kutztown is already in that supply chain. Its publicly published subcontract awards include 2 in 2026 totaling $198K for a certification package for a forged steel component for Naval Reactors (via Fluor Marine Propulsion) and a NAVSEA-funded industrial base task (via Deloitte).
Buy a legacy supplier & scale it. There are so many opportunities like this across the federal market right now to meet rising demand.
The smartest companies are moving on this.
Kutztown Foundry, pouring metal in Pennsylvania since 1866, is now Hadrian Factory 6.
We are investing $300M to upgrade and expand it, adding the casting capacity America's naval and munitions supply base needs.
Yesterday, Dept of War & Commerce’s Bureau of Industry and Security launched a national survey of roughly 500 additive manufacturing companies. Survey notices will started going out Nov. 16.
The problem they’re trying to solve is that qualifying an additive manufacturing part hits small businesses, new entrants, & lower-tier suppliers hardest, creating significant barriers to entry.
What stands out to me on this one:
The data from these ~500 companies will shape defense procurement policy, common qualification standards across the Military Departments, and where targeted capital investment goes. A small group of respondents is setting the inputs for rules the whole AM base will live with.
If you get a notice, answer it. The survey runs under Section 705 of the Defense Production Act, and past BIS 705 surveys have been mandatory. Read yours closely. Individual responses are confidential and FOIA-exempt, so sharing capex and machine utilization numbers carries less risk.
For investors, qualification cost is the moat right now. Companies that already paid to qualify parts hold a real advantage. If the services move toward common standards, that advantage shrinks, and smaller shops have the most to gain.
My opinion on this is that based on the procurement data we've analyzed, new entrants face a series of challenges on the front end of entering this market. And it all starts with the minimal understanding of their specific federal market opportunity, how to position for the highest demand entry points, and how to win (enough & rapidly) to gain a productive foothold that incentivizes them to keep participating.
The survey covers aerospace, maritime, munitions, space, and ground vehicles. The follow-on DoW and Commerce strategy should tell us which sectors get capital first.
Still missing from the release is a response deadline, a date for the joint strategy, and any $$ figure behind the capital investment.
Can't stop thinking about this one, which is just one of many.
One of the biggest challenges facing the defense industrial base is figuring out how to bring the full capacity of American industry to meet national security needs.
Whether as a prime contractor, subcontractor, a go-to supplier, American businesses should know exactly where their capabilities are needed & how they can participate.
That's exactly the problem we're working to solve at Serellium.
Yesterday, Dept of War awarded a $224M max-value IDIQ for the wool & poly/wool cloth for Air Force & Space Force uniforms (serge, elastique, tropical, heather charcoal, dark blue).
The govt received one bid. Here’s why --
Couple of things at play here.
The Berry Amendment requires DoW to buy military clothing & textiles grown and produced in the U.S. DLA’s notice specified 100% US made, from 100% U.S. materials.
According to DLA’s own Aug 2026 consolidation memo, until recently Burlington Industries was the only domestic source capable of making Berry-compliant dress cloth.
The only other domestic mill named is American Woolen, a small business that makes a small share of Army taupe and Marine Corps scarlet cloth.
The existing vendor base can’t keep up. Since at least 2022, DLA has approved 3 domestic non-availability determinations (DNADs) for dress cloth, which let the govt buy foreign-made cloth when US supply falls short. Right now, 9 of 10 dress cloths DLA buys have one.
Of the 36 DLA awards for serge, elastique, and tropical cloth going back to 2006, Burlington won all 36. Every one awarded since 2014 drew a single offer. The last AF serge contract ($43M, 2023) had one bid. DLA ran a sources sought for these exact cloths in July 2025. Still one bid.
So why aren’t others jumping in?
DLA’s memo says interested companies have made limited progress toward U.S. production. Without a guaranteed baseline of work, the capital cost of standing up compliant capacity is commercially prohibitive.
DLA's trying to fix this. A separate solicitation bundles 10 dress cloths into up to 3 awards (planned for 2 large, 1 small business). It lets new entrants supply DNAD cloth while they build U.S. production.
Meanwhile, this award keeps AF & Space Force cloth with one mill thru 2031.
Yesterday, Dept of War awarded a $224M max-value IDIQ for the wool & poly/wool cloth for Air Force & Space Force uniforms (serge, elastique, tropical, heather charcoal, dark blue).
The govt received one bid. Here’s why --
Couple of things at play here.
The Berry Amendment requires DoW to buy military clothing & textiles grown and produced in the U.S. DLA’s notice specified 100% US made, from 100% U.S. materials.
According to DLA’s own Aug 2026 consolidation memo, until recently Burlington Industries was the only domestic source capable of making Berry-compliant dress cloth.
The only other domestic mill named is American Woolen, a small business that makes a small share of Army taupe and Marine Corps scarlet cloth.
The existing vendor base can’t keep up. Since at least 2022, DLA has approved 3 domestic non-availability determinations (DNADs) for dress cloth, which let the govt buy foreign-made cloth when US supply falls short. Right now, 9 of 10 dress cloths DLA buys have one.
Of the 36 DLA awards for serge, elastique, and tropical cloth going back to 2006, Burlington won all 36. Every one awarded since 2014 drew a single offer. The last AF serge contract ($43M, 2023) had one bid. DLA ran a sources sought for these exact cloths in July 2025. Still one bid.
So why aren’t others jumping in?
DLA’s memo says interested companies have made limited progress toward U.S. production. Without a guaranteed baseline of work, the capital cost of standing up compliant capacity is commercially prohibitive.
DLA's trying to fix this. A separate solicitation bundles 10 dress cloths into up to 3 awards (planned for 2 large, 1 small business). It lets new entrants supply DNAD cloth while they build U.S. production.
Meanwhile, this award keeps AF & Space Force cloth with one mill thru 2031.