Have emissions from electricity peaked?

 From EMBER.

Global [electricity] demand rose by 369 TWh (+2.6%) in H1-2025, compared with 731 TWh (+5.3%) in the same period last year. The smaller increase was due to a few factors, including a more measured pace of industrial growth in China and India, but also fewer heatwaves in May and June in India.

In China, demand grew by 198 TWh (+4.2%), compared with a much stronger increase of 326 TWh (+7.5%) in the same period last year.

Despite the smaller rise, the global demand increase in H1-2025 was close to the 10-year annual average of 2.7% for 2015-2024.

In India, demand growth was particularly low at 12 TWh (+1.3%), compared with +75 TWh (+9%) last year when heatwaves drove higher demand. Ember estimates that if weather during May and June 2025 had matched last year’s heatwave conditions, demand growth in India would have been closer to +3%.

The world’s four largest polluters accounted for 81% of the global demand rise in the first half of 2025: China 54% (198 TWh), the US 21% (76 TWh), India 3.3% (12 TWh) and the EU 2.4% (9 TWh).

Solar generation grew by 306 TWh (+31%) in the first half of 2025, its fastest absolute growth on record. If this pace continues, solar is on track to remain the fastest-growing source of electricity for the 21st consecutive year and to outpace wind growth in absolute terms for the fourth year in a row. 

Solar’s global share was 8.8% in the first half of 2025, more than doubling in the last four years, from 3.8% in 2021. In many countries, solar now makes up a considerably higher share of the electricity mix.Several economies set new records. Among the top 20 largest solar generators in absolute terms, seven countries — Hungary, Greece, the Netherlands, Pakistan, Spain, Australia and Germany — generated 20% or more of their electricity from solar in the first six months of 2025.

Hungary led with nearly 30% share of solar generation, ahead of Greece and the Netherlands, which both surpassed 25%, up from just over 10% only four years ago (in the first half of 2021). Meanwhile, based on Ember’s estimate, Pakistan saw the largest increase in share, from 4.4% in H1-2021 to 21.9% in H1-2025. The increase was driven by the rapid adoption of rooftop solar by households and businesses in response to high electricity prices, as reported previously by Ember.

Based on available monthly data, at least 29 countries generated over 10% of their electricity from solar from January to June 2025, up from 22 countries in the same period of 2024 and only 11 countries in the first half of 2021.

China remained the leader in absolute growth terms for the third consecutive year, accounting for 55% (168 TWh) of the global increase in solar in the first half of 2025. The US accounted for 14% (44 TWh), the EU 12% (37 TWh) and India 6% (17  TWh). In contrast, solar generation fell marginally in Japan by 1.4 TWh (-0.4%), partly due to record-high curtailment. Solar also declined slightly in Vietnam (-0.5 TWh, -1.7%).

Solar capacity additions also grew at a record pace, reaching a new high of 380 GW in the first six months of 2025 – 64% more than the 232 GW added in the same period last year. A record surge in May, driven by accelerated installations in China ahead of new pricing rules on 1 June 2025, was a key contributor. Overall, China accounted for 67% of total solar capacity additions in H1-2025.

Renewables overtook coal in the electricity mix for the first time on record, rising by 363 TWh (+7.7%) to 5,072 TWh in the first half of 2025. Their share increased to 34.3%, up from 32.7% in the same period last year. Coal fell by 31 TWh (-0.6%) to 4,896 TWh, with its share dropping to 33.1%, down from 34.2%.

My view on these figures:

The output of solar is rising by 30% per annum, and because solar panels and battery storage keep on falling in cost, that growth rate will continue, and may even accelerate.  This half year, solar output rose 306 TWh over the year, so in H1 2026, it could easily rise by 400 TWh, H1 2027 by 517 TWh, and in H1 2028, by 673 TWh.  Assume there is no acceleration in wind output, i.e., the annual increase in output remains at 100 TWh.   

World growth is likely to pick up.  The switch to EVs is accelerating, and datacentres are gobbling electricity.  Demand growth for electricity will be higher over the next year, and higher still in 2027. But, if world electricity demand rises by, say, 500 TWh in the year to H1 2026 (compared with 369 TWh in H1 2025), this will be satisfied by the increase in renewables supply.  Similarly, in H1 2027, demand could rise by more, say 550 or 600 TWh, and still be met by the increase in output from renewables.  By H1 2028, demand would have to rise by a record 770 TWh for renewables not to match this increase.  And thereafter, renewables will eat more and more deeply each year into generation from fossil fuels.

In other words, emissions from electricity generation have probably peaked.  

Declines will be small initially, but the stronger rise in electricity demand is (partly) to supply electricity for EVs. The decline in emissions from land transport will be accelerating.  

We are at, or are close to, the point where not just emissions from electricity generation, but also total global emissions, are peaking.  This is not a cyclical peak, caused by a recession or Covid.  It is a secular peak.  We have at last started on the long road to zero emissions.  Does that mean temperatures have stopped rising? No, unfortunately.  For that, we would have to cut emissions by at least 90%.  However, we're getting there.   Emissions could halve by 2045, and halve again by 2060. Definitely too long and too slow.  But we are at last on the right road.  Now we must try and bend the curve to get there faster.