At what launch price and hardware cost could electricity beamed from orbit compete with firm power on Earth, and what should decision-makers take from the evidence in 2026?
The short answer. not soon, and only if two costs fall by large factors at once. Space-based solar power (SBSP) collects sunlight in orbit, where it shines almost all the time, and sends the energy to the ground as microwaves or light. NASA's 2024 assessment put the cost of electricity from two reference designs at about $0.61 and $1.59 per kilowatt-hour, ten to thirty times terrestrial renewables, with launch the largest cost. Setcoin's model reaches about $357 per megawatt-hour for a design delivering power at 10 kg per kilowatt, launched at $1,000 per kilogram to low orbit, with $2 per watt of hardware; it falls to about $95 at $200 per kilogram and about $50 if mass and hardware costs halve as well. But space hardware today costs hundreds of dollars per watt, a hundred times more than the design assumes. The field has moved from paper to hardware: demonstrations beamed power in orbit (Caltech, 2023), over 8.6 km on the ground (DARPA, 2025) and are now flying (Cowboy Space's Reason-1, October 2026), and Meta has signed for up to 1 GW from Overview Energy from 2030. The first markets are likely to be military bases, remote sites and power for other satellites, not national grids.
Ten findings
| # | Finding | Chapter |
|---|---|---|
| 1 | Sunlight in high orbit is available about 99% of the time, against roughly 10–25% capacity factors for ground solar | Chapter 9 |
| 2 | Beaming physics sets the size of the receiver: a 1 km transmitter at 5.8 GHz from geostationary orbit lights a spot about 4.5 km across | Chapter 25 |
| 3 | Only about 18% of the sunlight reaching a microwave system's arrays becomes grid power; about 6% for laser systems | Chapter 27, Chapter 29 |
| 4 | Launch dominates cost: about 92% of capital in Setcoin's reference case | Chapter 53 |
| 5 | Hardware cost is the hidden second gap: space arrays cost €200–350/W, a design needs about $1–2/W | Chapter 47 |
| 6 | Published cost estimates span $48 to $1,590 per MWh because assumptions differ by orders of magnitude | Chapter 55 |
| 7 | To match new nuclear ($141–251/MWh) at $2/W hardware, launch must fall to roughly $350–650 per kilogram to low orbit | Chapter 57, Chapter 59 |
| 8 | A 2 GW plant at 10 kg/kW weighs about 20,000 tonnes: some 200 Starship-class launches | Chapter 43 |
| 9 | Defence and remote power will pay first; grid power comes last | Chapter 69 |
| 10 | Beam safety, spectrum, astronomy and public acceptance are real but manageable constraints | Chapter 35, Chapter 77 |
Key numbers
| Measure | Value | Source |
|---|---|---|
| NASA reference designs, cost of electricity | ~$610 and ~$1,590 per MWh | NASA OTPS, 2024 |
| Launch share of NASA design costs | 71% and 77% | NASA OTPS, 2024 |
| Setcoin reference LCOE (10 kg/kW, $1,000/kg, $2/W) | ~$357/MWh | Chapter 53 |
| Same at $200/kg launch | ~$95/MWh | Chapter 53 |
| New nuclear, unsubsidised (US) | $141–251/MWh | Lazard, 2025 |
| Longest ground power-beaming record | >800 W over 8.6 km | DARPA, 2025 |
| Largest announced offtake | Up to 1 GW (Meta–Overview Energy) | Companies, 2026 |
| Mass of a 2 GW plant at 10 kg/kW | ~20,000 t | Setcoin calculation |
The five interactive models
| Model | Question it answers |
|---|---|
| Cost of electricity (LCOE) | What does a megawatt-hour from orbit cost for given mass, launch price, hardware cost and finance? (Chapter 53) |
| Efficiency chain | How much sunlight survives each step to the grid, for microwave and laser systems? (Chapter 27) |
| Beam and receiver size | How large must the transmitter and receiver be, and how intense is the beam? (Chapter 25) |
| Launch campaign | How many launches and years would a plant take? (Chapter 43) |
| Break-even map | Which combinations of launch price and hardware cost beat a target power price? (Chapter 59) |
Each model can be downloaded as a spreadsheet with live formulas.
How the study is organised
| Part | Chapters | Content |
|---|---|---|
| I. Foundations | Chapter 2 to Chapter 8 | Definitions, history, why now, sunlight in orbit, method |
| II. Concepts and programmes | Chapter 10 to Chapter 23 | Designs; Europe, UK, US, Japan, China; companies; the Meta deal; demonstrations |
| III. Physics | Chapter 25 to Chapter 39 | Beam size, efficiency, lasers, mirrors, receivers, safety, spectrum, weather |
| IV. Engineering and cost | Chapter 41 to Chapter 51 | Mass, launch, assembly, hardware cost, orbit choice, operations |
| V. Economics | Chapter 53 to Chapter 67 | LCOE, published estimates, firm power, break-even, sensitivities, value, finance, carbon |
| VI. Markets and regulation | Chapter 69 to Chapter 80 | Defence, data centres, islands, power for satellites, rules, law, acceptance |
| VII. Outlook and tools | Chapter 82 to Chapter 98 | Scenarios, roadmap, gates, case for and against, lessons, checklist, implications, risks, signals, questions, method |
Appendices. A, programme and demonstration reference table; B, glossary; C, Setcoin reference data.
Reading paths
| Reader | Suggested path |
|---|---|
| Utility or energy buyer | Chapters 1, 9, 53, 57, 63, 71, 82 |
| Space company | Chapters 10, 25, 41, 43, 47, 59, 94 |
| Government and defence | Chapters 12, 15, 19, 69, 77, 78, 84 |
| Capital provider | Chapters 1, 21, 55, 61, 65, 88, 96 |
| Newcomer | Chapters 2, 4, 9, 25, 100 |
How this study relates to the series
This is the eleventh study in Setcoin's space research programme. Study S3 (space-based datacentres) supplies the cost of space solar arrays; Study S5 (launch economics) supplies launch prices; Study S7 (supply chain) covers solar cells and minerals. Facts are as of October 2026; Setcoin calculations and illustrations are labelled.