Cheap future energy
In my 2016 book 'Britain and the world in 2050,' I suggested that energy would be extremely cheap by 2050, with widespread effects. Although energy costs have been rising recently, there is a strong case that my prediction for 2050 will come about.
The key is to separate two things that the recent headlines blur, the cost of making energy and the price people pay for it.
The recent rises are not a failure of the technology. In Britain they come mainly from gas and from what is stacked on top of it. Because gas plants are so often the marginal source, gas has an outsized effect on UK electricity prices. In 2021 gas supplied 40% of generation but set the wholesale price in an estimated 97% of half-hour periods.
Then wholesale gas prices doubled in early March 2026 when the Middle East conflict began. Networks add more; for businesses, non-commodity costs now make up around 64% of the electricity bill, and network charges rose by roughly 60% from April 2026. Neither is a statement about what a kilowatt-hour costs to generate from new plant.
Generation keeps getting cheaper, and here the trend is relentless. Utility-scale solar fell 90% between 2010 and 2024, to $0.043/kWh, 41% below fossil fuels. My memo's "roughly 90 per cent in fifteen years" is spot on. Onshore wind was cheaper still, at $0.034/kWh.
Storage, the old objection to intermittency, is collapsing in cost too. Stationary storage packs dropped to $70/kWh in 2025, 45% lower than a year earlier. The lowest observed pack prices were $50/kWh, and similar lows last year suggest these are no longer outliers. Notably, battery prices held down even when cobalt and lithium prices rose, which suggests a maturing industry rather than a lucky run.
Firm, round-the-clock power is following. Enhanced geothermal has moved from theory to contracts. Fervo's Cape Station in Utah expects first power in 2026, reaching about 100 MW by early 2027, and Google has signed a 396 MW deal with an option for nearly 1 GW. Drilling time fell 70% compared with its first horizontal well in 2022. That is the learning curve of shale, applied to heat.
Technologies with near-zero fuel cost and falling capital cost tend towards very cheap output. What can stop consumers seeing it is planning delay, grid bottlenecks and market design? IRENA, the International Renewable Energy Agency, itself warns that higher costs are likely to persist in Europe and North America because of permitting delays, limited grid capacity and higher balance-of-system expenses.
So my prediction holds globally with some confidence. Whether Britain shares in it depends on whether it lets people build: generators, pylons and storage facilities alike. Places that build freely will reach 2050 with energy almost too cheap to meter. Places that don't do that will still be metering it.
There is even some early evidence the transition is already biting: ECIU analysis suggests renewables cut UK wholesale electricity prices by a third in 2025, by pushing the costliest gas plants out of the merit order. My 2016 forecast looks less bold now than it did then, and the main thing standing between Britain and cheap energy is Britain. Britain could have virtually limitless cheap energy if it wanted to. But it has to want to.
Madsen Pirie