Why did solar become the cheapest electricity in history?

Because solar PV is a technology, not a fuel. Technologies follow Swanson's Law: every doubling of cumulative module shipments cuts the price by roughly 20%. Solar cost about USD 100 per watt in the early 1970s and has since fallen by more than 99%. By 2021 utility-scale solar reached USD 24–41/MWh, below coal at USD 42–152/MWh and nuclear at USD 131–204/MWh (Lazard LCOE Analysis v16).

The decisive distinction is technology versus fuel. A fuel's price is set by geology, markets and war, and does not systematically fall because you use more of it; a technology's price falls along a learning curve. When Bell Labs demonstrated the first practical silicon PV cell in 1954, it was 6% efficient and cost the equivalent of USD 1,785 per watt today — its first market was not on Earth but in space, powering the Vanguard 1 satellite in 1958. By the early 1970s the cost was around USD 100 per watt. Between 1977 and 2025 module prices fell by 99.9%. For contrast, the steam engine was refined for two centuries, moving from 1% to 40% efficiency with costs falling linearly; solar's decline has been exponential, with levelised cost down roughly 90% from the 1970s to 2021.

The cost structure compounds the effect. Fuel accounts for 60%–80% of the lifetime cost of a gas-fired plant — a variable cost that swings with global markets and conflicts. At a solar farm, 90% of the cost is incurred on the day it is built, and for the next 30 years the fuel is free. That makes solar a low-risk asset, and once the modules are paid off the marginal cost of the next unit of electricity is essentially zero. This zero-marginal-cost reality is sweeping through power markets: on sunny afternoons wholesale prices often fall to zero or negative, forcing utilities to rethink their business models entirely.

The third factor is manufacturing scale. China now controls roughly 80% of the entire solar supply chain, from polysilicon refining to module assembly, a position built on large integrated industrial parks and accumulated manufacturing experience. In response, the United States and others introduced measures such as the Inflation Reduction Act (IRA), offering billions in subsidies to bring solar manufacturing back onshore. The result shows up in deployment: global cumulative solar capacity passed 2,392 GW by the end of 2025 (IRENA, 2026), with 510 GW added in 2025 alone — about 75% of all renewable capacity added worldwide that year — making solar the largest single renewable technology by installed capacity.

The 'cheapest' claim needs one qualification: it describes utility-scale solar as a new-build generation source. Rooftop solar costs far more per watt because it lacks the economies of scale of large projects. Rooftop solar thrives not because it is the cheapest way to generate power but because it competes with the retail tariff: if the utility charges 30 US cents per kWh and a household can produce its own for 12 cents, the fact that a desert solar farm can do it for 3 cents is irrelevant to that household. This mismatch is where the utility death spiral debate begins — customers leave the system, costs rise for those remaining, and they leave in turn.

In most of the world, wind and solar are now the cheapest new-build generation humanity has ever had — not the cheapest renewable, not the cheapest 'clean' option, but the cheapest, unconditionally.

This question is covered in depth in The Full Spectrum Every Energy Source Explained — A Singapore Perspective，第七章 7.5