The $17.5 Billion Program With $398 Million Behind It

On August 11, at the Space and Missile Defense Symposium, General Michael Guetlein put the budget tables into plain language. He told the audience that without reconciliation funding, there is no Golden Dome. The program he directs runs across nine program elements in the FY2027 justification books and totals $17,518.1 million. Of that, $398 million sits inside the discretionary bill that Congress is working on. The remaining $17,120.2 million is a mandatory request that has not been enacted. That means only 2.3 percent of the program is moving through the annual appropriations process. The other 97.7 percent is riding on a reconciliation package that has not moved.

Most large defense programs carry a prior-year discretionary funding level, and that matters for a mechanical reason. Under a continuing resolution, agencies generally keep spending at last year’s rates for ongoing activities, while new programs face restrictions. A program with an established discretionary base can continue spending at roughly its prior pace while Congress argues. Golden Dome does not have that base, because it was funded through reconciliation rather than built up through the annual process. That’s why its discretionary money appears in only two of the nine program elements. The other seven carry nothing. Under a CR, there is nothing to fall back on.

The FY2026 line totals $21,085.1 million, and the FY2027 request comes in at $17,518.1 million, a 16.9 percent decline. The decline itself isn’t unusual. Programs often shrink once they move from initial allocation to actual execution. What stands out is the ratio. Programs at this scale almost always carry a meaningful discretionary floor, built up over successive budget cycles, so that they can survive the annual appropriations fight without going dark. Golden Dome skipped that process entirely.

Invisible to the usual tools

The program is also invisible to the tools that defense analysts normally use. Golden Dome appears in zero rows of the congressional marks dataset, across every fiscal year the data covers. Reconciliation money does not pass through appropriations markup. There is no request for a committee to mark and no line in any committee report. An analyst tracking the four defense committees has visibility into $398 million of the program and none at all into the other $17.1 billion.

The contract record is nearly as opaque, for a different reason. Search federal award descriptions for the program name, and the result is seven actions across five vendors totaling roughly $900,000, all recorded between May 2025 and March 2026. Anduril holds about $500,000 of that. SpaceX holds about $200,000. The figures sound like a study program, not a $17.5 billion line.

That is not evidence the program has failed to spend. At the same August appearance, Guetlein said roughly 90 percent of the approximately $24 billion allocated through last year's reconciliation act has been obligated, with about 95 percent of the $22.5 billion appropriated so far committed to specific work. The spending does not carry the program name. Contracting officers write these descriptions themselves, so a budget line called Golden Dome becomes contract language about interceptors, missile defense engineering, or ground systems integration. The money is moving. It is just moving under different labels.

Follow the work, not the name

Search for the actual work instead of the program name, and the picture changes. Obligations on contracts describing missile defense or interceptor work show a sharp acceleration. To make a fair comparison, the window is restricted to October through March, so each fiscal year compares against the same six months. The first half of FY2023 recorded 229 actions and $320 million obligated. FY2024 ran 197 actions and $550 million. FY2025 came in at 256 actions and $480 million. FY2026 jumped to 470 actions and $1.69 billion. Over the same period, total defense obligations rose 12.2 percent year over year. This category rose 252 percent.

The vendor composition shifted more dramatically than the total. In the FY2025 window, Boeing led at $241.9 million across 16 actions, and Lockheed Martin did not appear in the top eight. In the FY2026 window, Lockheed Martin sits first at $924.1 million across 19 actions, accounting for more than half the category. Boeing dropped to $172.5 million, below where it had been a year earlier. Parsons came in at $99.1 million. Northrop Grumman recorded $90.7 million on 54 actions. Raytheon ran $70.7 million on 47. The entire list of prime contractors for missile defense work reorganized in a single fiscal year.

New names below the primes

Below the primes, 29 vendor registrations appeared in the FY2026 window that were absent from the prior three years combined. IERUS Technologies, a Huntsville engineering firm, recorded $31.9 million. Cummins Power Generation recorded $29.5 million on a single action, a figure that reads as site power infrastructure rather than missile work. Alaska Aerospace Corporation also appeared, the operator of the Kodiak launch complex the Pentagon uses for missile defense testing. Some of those registrations belong to companies already active under a different entity record, so the count of genuinely new firms is lower than 29. Either way, the vendor base is getting broader.

The budget presentation does not separate these two realities. In one set of columns, a program has obligated most of its enacted funding and is pulling new vendors into missile defense work at a pace that the contract record has not shown before. In the adjacent columns of the same table, that same program is requesting $17.1 billion against just $398 million of base. It’s dependent on a package that has not moved, with no prior-year discretionary level to fall back on. The spending and the precariousness sit side by side, in the same row, and the budget book offers the reader no way to weigh one against the other.