Reusable Rockets Changed Space Economics, But Not Everything Is Cheap

The moment a Falcon 9 first stage settled back onto Landing Zone 1 in December 2015, the price of reaching orbit began moving in a direction it had never moved before: down. A single Falcon 9 launch that once cost roughly 60 million dollars dropped, over successive reflights, toward 30 million and below for certain customers. That shift was not cosmetic. It restructured which satellites got built, which orbits became commercially viable, and which space agencies could afford ambitions they had shelved for a decade.

The old model was simple and expensive: build a rocket, burn it, invoice the customer for the whole thing. The upper stage, the fairings, the first stage, all of it fell into the ocean or burned up on reentry. Every launch was a fresh manufacturing run. The economics resembled an airline that incinerated its aircraft after each flight and billed the passenger accordingly.

Reusability broke that logic at the most expensive single component: the first stage, which carries the main engines. Merlin engines account for a substantial fraction of a Falcon 9's total cost. Flying the same booster ten or more times spreads that manufacturing cost across ten missions rather than one. SpaceX has flown individual boosters more than twenty times. At that reflight count, the per-mission cost of the booster itself approaches something close to a consumable maintenance budget rather than a capital write-off.

Fairing recovery added another layer. The two clamshell halves that protect a payload during ascent cost in the range of six million dollars per set. SpaceX began catching them with net-equipped ships. Recovered fairings now routinely fly again. The saving per mission is not enormous in isolation, but across hundreds of launches a year it compounds.

The downstream effect was felt most sharply in the smallsat and constellation markets. When Iridium, OneWeb, and Starlink needed to place hundreds of satellites, the difference between 60 million and 30 million dollars per launch was the difference between a viable business model and a funding crisis. Lower launch costs did not just save money, they made entire categories of mission possible that were previously on paper only.

Propellant is cheap. Liquid oxygen and kerosene together cost a few hundred thousand dollars per Falcon 9 launch, a rounding error against the total. But propellant is not the only recurring cost, and this is where the economics become less dramatic than the headlines suggest.

Refurbishment is real and variable. After a booster lands, engineers inspect every weld, every turbopump, every ablative surface. Early in the Falcon 9 reuse programme, SpaceX was conservative about how much work each returning booster needed. The cost of that inspection and repair cycle was not published, but industry analysts estimated it ran into the low millions per flight for the first several reflights. As the programme matured and confidence in the hardware grew, refurbishment scope shrank. But it never reached zero, and for a new vehicle, Starship, New Glenn, Neutron, the refurbishment learning curve starts over.

The upper stage is still expendable on Falcon 9. The second stage, which does the final push to orbit, is not recovered. It burns up or remains in orbit as debris. That means every Falcon 9 mission still manufactures and destroys one upper stage. The economics of reusability apply to roughly half the vehicle by mass, not the whole stack.

Payload capacity takes a hit on reuse missions. A Falcon 9 flying in full expendable mode can deliver around 22,800 kilograms to low Earth orbit. A reuse mission, which must reserve propellant for the boostback burn and the landing burn, delivers less, roughly 15,600 kilograms to the same orbit in a standard return-to-launch-site profile. Customers with heavy payloads sometimes choose expendable flight for this reason, which means they pay more and the booster is lost. The tradeoff between payload mass and booster recovery is a genuine engineering constraint, not a marketing footnote.

SpaceX's Starship is designed to be fully reusable, both the Super Heavy booster and the upper stage ship. The ambition is to make the economics look less like aviation and more like trucking: the vehicle is an asset that depreciates over thousands of flights, and the per-flight cost approaches the cost of propellant, crew time, and maintenance. Starship uses liquid methane and liquid oxygen, both of which can in principle be synthesised on Mars from local resources, a design choice that reflects ambitions well beyond Earth orbit.

The Mechazilla catch system, which uses mechanical arms on the launch tower to grab the returning Super Heavy booster rather than landing it on legs, is an attempt to eliminate the time and cost of moving a booster from a landing pad back to the launch mount. If it works reliably at scale, it compresses the turnaround. In late 2024, SpaceX demonstrated booster catch on the first attempt. Whether that translates to a 24-hour or 48-hour turnaround in routine operations remains to be proven at the cadence the company projects.

New Glenn, built by Blue Origin, is also designed for first-stage reuse. Its first launch in early 2025 reached orbit but did not recover the booster. Rocket Lab is developing Neutron with reuse in mind. The competitive pressure SpaceX created has made reusability the baseline assumption for any new heavy-lift vehicle, not a premium feature.

The infrastructure costs of catching and processing returning rockets are not trivial. The drone ships, the catch towers, the processing facilities, the range safety systems, these are capital investments that only make sense at high launch cadence. A company flying four times a year cannot amortise a Mechazilla tower. The economics of reusability are inseparable from the economics of volume.

ISRO has watched the reusability shift carefully. The Reusable Launch Vehicle Technology Demonstrator, RLV-TD, has conducted a series of landing experiments at Chitradurga in Karnataka, testing the autonomous landing logic that a recovering winged vehicle needs. Gaganyaan, India's crewed mission launching from Sriharikota, is not a reusable vehicle in the SpaceX sense, but the broader programme has pushed ISRO to think about next-generation launch costs in a market where Falcon 9 has reset the global benchmark.

ISRO's LVM3, which carried OneWeb satellites to orbit in 2023, competes in a market that SpaceX prices. The cost pressure is structural. India's advantage has historically been low labour cost and frugal engineering, Mangalyaan reached Mars for a budget that astonished the world. But frugal expendable rockets face a different competitive landscape than they did before December 2015. The next chapter of Indian launch economics will depend on whether ISRO or a private Indian launch company, Agnikul, Skyroot, or others, can bring reusability into a vehicle that reaches commercial scale.

The revolution in launch costs is real, but it is partial. What changed is the ceiling, the worst-case price of access to orbit has dropped sharply, and the market has expanded to fill the space that opened up. What has not changed is the physics of propulsion, the metallurgy of engines running at extreme temperatures, and the basic fact that getting mass off Earth requires enormous energy. Reusability spreads the fixed costs of hardware across more flights. It does not eliminate those costs, and it does not make space cheap, it makes space less ruinously expensive, which turns out to be enough to change almost everything.