Year in electricity 2025 -- EMBER report

[woodgeek]

A new state of global electricity report dropped from the (Euro) think-tank EMBER

Global Electricity Review 2025 | Ember

Record renewables growth led by solar helped push clean power past 40% of global electricity in 2024, but heatwave-related demand spikes led to a small increase in fossil generation.

ember-energy.org

Always good to remember that the world is getting better.

Lots of stats about the crazy growth of solar. A few numbers;

-- Solar production is now 7% of global electricity (similar to the US fraction, we are not leading)
-- Solar production in 2024 was 2134 TWh, the same as 240 GW continuous generation (or 30W continuous per human).
-- Solar production is continuing to double roughly every two years.
-- At current growth rates Solar will pass global nuclear and global hydro in the 2026 time frame.

Here is a chart:

It shows that in the 2020-2024 the world ex-China and India decreased its total fossil use for electricity production, despite increasing demand.
China's fossil usage for electricity is expected to peak in the next couple years. India will take longer to peak, but the global figure is expected to fall in the next 4 year cycle.

Think about that... total global fossil use for electricity will soon be falling, even as global demand rises.
A good start.

 

[EbS-P]

At current growth rates Solar will pass global nuclear and global hydro in the 2026 time frame.

That is really amazing!

Solar production is now 7% of global electricity (similar to the US fraction, we are not leading)

So we were average. With the changing incentives that is sure to drop.

 

[woodgeek]

So we were average. With the changing incentives that is sure to drop.

In context, the US is a petro-state. Currently the largest oil producer. Historically, the original petro-state, and the largest oil producer for a very long time. We freaking invented oil.

That industry will not go quietly into the night.

Given that context, average on solar might be seen as pretty good?

 

[dougstove]

Ummm, actually Abraham Geisner invented kerosene outside Moncton, NB, Canada

 

[begreen]

Ummm, actually Abraham Geisner invented kerosene outside Moncton, NB, Canada

And James Young in Scotland 1847. According to sources in Wikipedia, the Russians were on it at this time around Azerbaijan. But long before that, the Chinese were using refined petroleum around 4 BC and had gas wells with bamboo piping by 340 AD.

 

[DBoon]

With the changing incentives that is sure to drop.

I try to find a reasonable middle ground between over-optimism and alarmism.

I don't like the elimination of the incentives for solar PV and wind (in my mind, they equalize the playing field for all of the externalities imposed on us by fossil-fuel pollution, CO2 being just one part of it, or the costs of storing and ultimately "disposing" of nuclear waste).

However, if the incentives were removed 10 years ago, that might have really hit the brakes on solar PV and wind deployment. I think they were critical to get these technologies to scale and be pervasive. 10 years ago, there was basically no solar PV on the ground in NY state and people would try to tell me that these technologies didn't work. Now I routinely get people in my rural area pulling into my driveway to talk to me about my solar PV array and ask how much money I save. So people know the technologies work and they see it everywhere. Now they are trying to figure out who is telling the truth - those who say it costs too much or those who have it and describe how much it actually costs and how much they save. To me, that is a big win over a 10 year period. And we've gone from solar PV being <1% of total US electricity to 7%, and increasing by 1% more every year. Granted, 7% but doesn't sound like a lot, but it is increasing quickly and will continue to increase quickly, despite what some politicians want. Just look at what is happening in Texas, despite all of the rhetoric from the politicians.

Will solar PV installations slow in the US in 2026 and 2027? For sure - I bet we see a big 2025 year as projects get fast-tracked to take advantage of incentives, and this will spill into 2026 for large installations. 2026 may be flat compared to 2025 (but still a good year) and 2027 probably dips by 30% in terms of installed kW. Then I would expect resumed 30% growth as the market adjusts. Why? It's still the cheapest form of energy, and it's still the fastest to be able to be deployed. Trump can piss and moan all he wants about wanting more gas generation, but that is a 5-year lead time and nobody has 5 years to wait. Meanwhile, solar PV can be distributed, plugged in anywhere and deployed in 1-2 years. New York state has been running this playbook for the last 5 years and is adding (on average) 1 GW of solar PV capacity per year (or more) and NY state isn't even what I would consider a leader in solar PV deployment. I don't think that is going to slow down. And that (now) almost 7 GW of capacity in NY state will be 10-12 GW of capacity by 2030 and more after that. We have already reached a California-style "duck curve" in my NYISO region during periods of peak solar PV production, and I can see where the entire NYISO hits this in 5 years or so. Battery deployments will quickly reduce the afternoon peaks. All this is going to happen regardless of these incentive reductions, it will just happen a little more slowly - but it will happen.

The big concern is how much new AI-driven data centers will suck up this new capacity. Again, here, I think that is overblown on a 5+ year time scale, and the power utilities now it, which is why they are scrambling to lock-in high-cost gas generation and new transmission so they can earn their 9% guaranteed returns on these investments. Past experience with non-AI data centers was that the original servers in the data centers were energy hogs. That became a big problem as the data centers scaled up. Then, surprise! Companies started making more energy-efficient chips and designing more power-efficient servers and lo and behold, data center demand didn't break the grid. The same thing will happen with the chips that power the AI data centers - it has too happen, and it always does happen.

 

[begreen]

Meanwhile the subsidies and tax breaks for fossil fuels and corn based ethanol keep rolling on without challenge.