Jeff Bezos is building his lunar tollbooth on your dime.
It is fair to say, and the press release is at pains to say it, that lunar infrastructure requires heavy lifting — landers, rovers, drones — that the federal government currently does not manufacture, and that handing the procurement to private industry is how the space agency has been forced to operate since the commercial-crew era. It is also fair to say that the Artemis architecture requires a sustained heavy-lift cadence that only two domestic firms can currently supply. The trouble — and here we are obliged to be precise, because the language of “reusable rockets” and “sustainable presence” papers over seven distinct regulatory and labor gaps — is that the political economy governing the Artemis contracts is being written in real time, by the contractors themselves, and the safety frameworks that took terrestrial heavy industry two centuries to build do not exist on the lunar south pole.
The contracts awarded last week to Blue Origin, Astrolab, Lunar Outpost, and Firefly Aerospace represent hundreds of millions of dollars of public appropriations. As the agency continues to outline its multi-phase base plan, the drones will scout; the rovers will haul; the landers will deliver. Before the first Artemis astronauts touch down in 2028, the robotic infrastructure will be in place, and the operational tempo will be set. What will not be in place is the Occupational Safety and Health Act. What will not be in place is the National Labor Relations Board. What will not be in place is the environmental-impact review. The contractor will provide the habitat, and the contractor will provide the airlock schedule.
A lunar lander is not a particularly novel piece of engineering. The Soviet Union landed one in 1966; the Chinese have now done it three times. The rocket equation does not care whether the prime contractor’s headquarters is in Kent, Washington, or in a NASA field centre. What the corporate form does care about is who owns the only ride to the surface once the federal research-and-development money has been spent and the agency has retired its own in-house capability. The commercial-crew program gave us an answer: when a single company holds the certified vehicle, the per-seat price rises to whatever the market will bear, and “the market” means the U.S. Congress, which has no alternative.
Vannevar Bush’s 1945 report Science, The Endless Frontier — the founding text of the postwar compact between the federal government and basic research — argued that public money should build the laboratories and train the scientists, because no private firm could sustain the long-arc investment that nuclear physics or spaceflight demanded. The compact held, in its fashion, for half a century. It worked for penicillin, radar, and the ARPANET. Now the laboratories are being turned into customer-lists. Blue Origin is not being paid a fixed fee to build a lander that NASA will own; it is being paid to build a lander that Blue Origin will own, and NASA will purchase rides on it at a schedule and a price the company will determine. The Astrolab and Lunar Outpost rovers, the Firefly drones — they will all belong to their builders, not to the agency that paid for their development. The pattern will be familiar to anyone who has watched what happened when DARPA funded the core research for Siri, and then Apple bought the spin-out and turned it into a proprietary interface. Public investment; private extraction.
The contracts are procurement vehicles, not treaties; they obligate NASA to pay, and they obligate the companies to perform the specified work, but they do not obligate the companies to grant access to the resulting infrastructure on reasonable, non-discriminatory terms. A future Japanese or European or Canadian mission that wants to land at the south pole will have to negotiate with the company that holds the pad, and the company will name its price. The company will be doing exactly what its shareholders expect. The failure is the agency’s, for writing the contract that way.
Last month the Artemis II crew flew farther into space than any Apollo astronaut. The achievement was genuine, and the crew’s safe return was a credit to the engineers who built the capsule and the heat shield. But the capsule is Orion, a government-owned vehicle. The lander that will carry the first woman and the next man to the surface is not. The distinction matters because history suggests that once the public-sector alternative has been allowed to atrophy, the company that owns the only operating lander will charge whatever it costs to keep the monopoly, plus whatever the shareholders have decided a moon base ought to return. The first astronaut to step off the Blue Moon lander will be standing on a platform whose purchase order was signed by the American taxpayer, whose development was funded by the American taxpayer, and whose operating manual will be held by a Delaware corporation that has no obligation to let anyone else read it.
The political economy of this is not different from the extraction model that subsidized the Alberta oil sands: public capital absorbs the upfront geological risk, private capital captures the downstream monopoly, and the labor force works under the terms the operator sets. On the moon, the operator sets the atmosphere. If the dust-mitigation protocol fails, or if the regolith fines seize a habitat airlock, the contractor who built the seal is also the only entity that can fix it. The independent repair technician does not exist.
The right-to-repair movement has spent the last decade arguing that if you cannot open the machine, you do not own it. The Artemis hardware will be the ultimate test of that principle. A subscription model for terrestrial software is an annoyance; a subscription model for the air scrubbers in a pressurized lunar habitat is a structural dependency. If the lunar-rover battery-thermal management system fouls in the south-pole dark, who opens the casing? The procurement contracts treat the hardware as a service, which is another word for a vendor-locked supply chain that cannot be audited from the ground.
The moon base is not a science project. It is the first tranche of a new terrestrial economy, one in which the physical layer — the landing sites, the power systems, the surface-communications relays — is being partitioned among a handful of U.S. corporations at the same moment the public treasury is paying for the concrete. The same four companies that are building the hardware will, if the pattern of the last decade holds, be the ones that operate it, control it, and charge rent on it for the next thirty years. There is no FCC filing for a lunar toll, and no Competition Bureau review of the south-pole bus franchise. The Outer Space Treaty of 1967 declares the moon belongs to no nation. The contracts say the landing pads belong to whoever poured the sinter. The moon, like the early internet, is being handed to a small group of owners before most of the world has noticed it is being sold.
There is a public-comment period open for NASA’s long-term lunar architecture, routed through the Federal Register. The deadline matters because the only way the public gets written into the operational parameters of the south-pole base is if someone writes them in before the contractor pours the concrete. The records are open. The labor is documented. The extraction is identical.
There is a Polish saying my grandfather used, which translates badly but means, roughly, the work doesn’t care how you feel about it. The work of building the tollbooth is proceeding on schedule. The Sea of Tranquility has no paywall yet. The south pole will, before the decade is out, and the tariff will be written by the same four hands that signed last week’s contracts.