> ## Content Index
> Fetch the complete content index at: https://top-margin.ghost.io/llms.txt
> Use this file to discover other available public pages before exploring further.

# The Pentagon Started Buying Factories, Not Satellites
- URL: https://top-margin.ghost.io/the-pentagon-started-buying-factories-not-satellites/
- Published: 2026-09-19T11:40:50.000Z
- Updated: 2026-09-19T11:40:50.000Z
- Author: Marcus Cole

At the Space Symposium in April the head of Space Systems Command told the satellite industry what he wanted from it in a single sentence. If a company is nominally making ten, he said, be prepared to make forty. That is not a request for a better satellite. It is a request for a bigger factory, and the Space Force procurement account behind it has grown from a few billion dollars to something closer to nineteen billion in the current request.  
  
Since then the Defense Innovation Unit has been writing contracts under a program called Adaptive Space Manufacturing and Integration at Scale, awarding companies money specifically to build production throughput rather than to deliver units. A separate award under Defense Production Act Title III went to a maker of space traveling wave tube amplifiers, the power components inside military communications payloads. The dollar values on most of these are not disclosed. What they buy is.

### From the Battlefield to the Balance Sheet

There is a real difference between a government buying a product and a government buying the ability to make one. Unit procurement leaves the capital decision with the supplier, who must guess how long the orders will last before committing to a second production line. Capacity procurement moves that decision, and part of that risk, onto the buyer.  
  
For a manufacturer this changes the arithmetic considerably. The hardest number in any capacity expansion is the one nobody can know, which is how many years of demand will follow. Government money that pays for tooling, test equipment and factory space shortens the payback period on the supplier's own contribution and makes the rest financeable. It is the same instrument that rebuilt munitions lines, applied to spacecraft and their components.  
  
The catch is proportion. Capacity awards of this type are usually small relative to the investment they are meant to trigger, and they are meant to be. They de-risk a decision rather than fund it. So the question for anyone reading these announcements is not what the government put in, which is often undisclosed and modest, but whether the supplier followed with its own capital. A company that takes the tooling money and does not expand has told you what it thinks of the demand.

### **The Dual-Use Reality Check**

Producing satellites at rate is a manufacturing problem that has almost nothing to do with space. It is about fixtures, repeatable test, supply chain qualification and getting variation out of a process that was previously done by hand a few times a year. Those disciplines came from automotive and consumer electronics, and they are moving into spacecraft now because the customer finally wants volume.  
  
The flow back out is real as well. Amplifiers, radiation-tolerant electronics, precision structures and additive manufacturing of propulsion hardware all serve commercial constellations, and the commercial constellation market is the larger buyer by unit count. A line built with defense capacity money and filled with defense orders in its first years is a line that can bid commercial work at a cost the operator could not have reached alone. That is the quiet transfer in this policy, and it is more durable than any single contract.

### **The Capital Signal**

The capital signal is that the Pentagon has started treating manufacturing throughput as something it can purchase directly, and it is doing so through innovation offices and production act authorities rather than through traditional procurement. That widens the field of companies that can win, because the qualification being bought is the ability to build many units rather than a track record of building exquisite ones.  
  
The structural read is to watch the components rather than the spacecraft. A satellite integrator can ramp faster than the suppliers of its amplifiers, radiation-hardened parts, optics and propulsion, and those suppliers are where a rate increase either happens or stalls. The fact that production act money went to an amplifier maker rather than to a satellite builder is the tell. Capacity announcements are cheap to make and slow to verify, so the number that matters is deliveries per month a year after the award, not the award itself. Until that number moves, a four times increase in output remains a request rather than a fact.  
  
*Signal: When a buyer starts paying for tooling instead of units, the constraint has moved from design to throughput, and throughput lives with the component suppliers.*

![](https://storage.ghost.io/c/e3/0e/e30efc82-f19f-427f-b131-8005b5e7416a/content/images/2026/09/the-pentagon-started-buying-factories-cinematic-1.jpg)

Marcus Cole, Top Margin