Every frontier settlement in human history required an economic justification. The Spanish came for gold. The British for fur and farmland. The whalers for oil. The Moon has no beaver pelts. The question that will determine whether a lunar settlement survives beyond government funding is simple: what does the Moon sell?

The near-term economic case is propellant. Water ice electrolyzed into liquid hydrogen and liquid oxygen is rocket fuel. A kilogram of propellant manufactured on the lunar surface and delivered to lunar orbit costs an estimated $5,000 to $20,000. The same kilogram launched from Earth costs $50,000 to $100,000. Any cislunar transportation architecture, from communication satellites to Mars-bound spacecraft, benefits from lunar-sourced propellant.

Helium-3 has been called the Moon's gold rush, but it remains speculative. The lunar surface contains helium-3 implanted by the solar wind at concentrations of roughly 10 parts per billion by mass. Extracting meaningful quantities requires processing millions of tonnes of regolith. And helium-3 is only valuable as fusion fuel, for reactors that do not yet exist. We are mining a resource for a technology that is always thirty years away. A beautiful irony.

The more pragmatic mineral case focuses on platinum-group elements and rare earth elements. Lunar regolith contains platinum at concentrations of roughly 8 to 10 parts per billion, comparable to some terrestrial ore bodies. But the economics are currently inverted. At $30,000 per kilogram for platinum and $50,000 per kilogram for Earth-to-Moon transportation, you spend more getting the mining equipment there than the product is worth coming back.

Perhaps we are thinking about this wrong. The lunar economy may not export physical materials at all. It exports something more valuable: knowledge. Every year of operating in the lunar environment produces engineering data, materials science data, biological data, and operational procedures that have enormous value to the broader space industry. The settlement is a laboratory, and its product is information.

The long-term economic model I find most compelling is the lunar settlement as a construction yard. In one-sixth gravity with abundant raw materials and vacuum conditions, the Moon becomes the optimal location for manufacturing large space structures. Solar power satellites. Deep-space vehicles. Orbital habitats. Launching finished products from the lunar surface requires one-twentieth the energy of launching from Earth. The Moon does not sell resources to Earth. It sells infrastructure to space.