Launch Economics

last updated 2026-08-31

Physics / mechanism

Launch economics is the set of cost relationships that determine whether an application can be economically placed in orbit: the price of delivering a unit of mass to a target altitude and inclination, the cadence at which that mass can be delivered, and the fraction of total system cost that launch represents once payload manufacturing, power, thermal management and replacement cycles are included. Because launch price per kilogram scales with vehicle reusability and flight rate, it behaves less like a physical constant than like a manufacturing learning curve, and any orbital business case is a bet on where that curve sits at deployment time.

The 2026 orbital data centre thesis is the clearest current test of the variable. Mach33’s analysis frames the collision between a $600B+/yr AI infrastructure build and terrestrial grid constraints as an arbitrage that orbital capacity can capture, with the market deciding within 18 months. That framing only closes if launch cost per kilogram of compute is small relative to the terrestrial cost of power and interconnection, which makes launch price the pivotal input rather than a line item.

Scale is the other lever. SpaceX’s FCC filing of 30 January 2026 contemplates up to 1,000,000 satellites operating as orbital data centres in 500-2,000 km sun-synchronous orbits, linked to Starlink by optical inter-satellite links. A constellation of that size implies a launch cadence and mass throughput far beyond current commercial practice, so the economics depend on marginal cost at very high flight rates, not on advertised list prices.

Vertical integration changes who pays the launch bill. The all-stock SpaceX acquisition of xAI in February 2026, forming a $1.25T combined entity, was justified by Musk as enabling orbital data centres, and the $20-25B Terafab 2nm facility announced in March 2026 targets 80% of output at orbital AI satellites using a radiation-hardened D3 chip designed for high-temperature vacuum operation. Where launch, payload silicon and the end application sit inside one balance sheet, launch is an internal transfer price and the relevant figure is internal marginal cost, which is not observable to outside buyers.

Competitive landscape

The sources available support only one comparison: orbital versus terrestrial siting of AI compute, where launch cost is the incremental burden orbital carries and grid availability is the constraint terrestrial capacity carries. No competing launch vehicles, price points or providers are compared in the material supplied, so the relative standing of launch options cannot be assessed here.

Evidence base

Frontier (open questions)

Synthesised 2026-08-31 from 1 KB sources by the resynth pipeline; citations are KB source slugs.

Recent mentions

Frontier questions