'Expectations v. reality'

Publication of the International Energy Agency’s World Energy Outlook is a keenly awaited event. The document (officially not a forecast) is a bible for the energy industry, though it is often accused of leaning too heavily on its origins in the oil crisis of 1974.

One particular complaint is that its projections for growth of solar capacity have undercooked actual outcomes by inexplicably large margins. But the 2025 version published last week is a strange document even by historical standards.

The first of the Trump 2.0 era, it brings changes to the modelled scenarios that downgrade the priority given to emissions cuts aligned with Paris Agreement pledges, as well as major changes to the structure of the report.

The net effect is to front load the scenarios that align least with Paris Agreement emissions cuts (with the oil and gas sector continuing unimpeded), and reduce the prominence of Net Zero Emissions (NZE) — the scenario that best aligns with Paris and would show much larger cuts to oil, gas and coal output and consumption.

WEO 2024 included a stated policies scenario (STEPS) based on implemented or announced policies, to give a “direction of travel". STEPS was consistent with global temperatures rising no more than 2.4°C above the average of the industrial age. It also included an announced pledges scenario (APS) that embraced longer term climate targets and led to a maximum temperature rise of 1.7°C. Finally, a net zero emissions scenario was modelled that limited temperature increases to 1.5°C.

WEO 2025 has downgraded the ambition, and upgraded the pain and distress envisaged for the world. Announced pledges is out, replaced by a current policies scenario (CPS) under which the world continues on its merry way with no further climate action, oil and gas demand booms, greenhouse gas emissions increase further and temperatures rise by nearly 3°C with disastrous consequences for humanity and Earth’s environment. NZE now envisages temperature increases exceeding 1.6°C for much of the rest of this Century, topping out at 1.65°C. STEPS envisages a 2.5°C increase.

'Missed it by that much'

Best for humanity

CPS is given equal top billing with STEPS in WEO 2025, and NZE feels like a bit of an afterthought. The report’s structure might as well have been designed to placate the Trump administration.

Most media coverage to date has focussed on CPS and STEPS, so The Energy is here focussing on NZE. We are fortified by a post from Our World in Data’s Hannah Ritchie rubbishing CPS scenario assumptions that electric vehicle sales stall everywhere but China and Europe (“bewildering”), and that global deployment of solar stalls at 2024 levels (“overly conservative and pessimistic”).

Some might think this unrealistic; we think the scenario where humanity does the right thing (after exhausting all the alternatives in the case of the US) should get an airing.

An even better reason to focus on NZE is that this is the best outcome for humanity. Access to electricity and clean cooking improves by far the most rapidly under NZE and ACCESS (“accelerating clean cooking and electricity services scenario”), and hardly at all under CPS and STEPS.

Household air pollution and ambient air pollution from the use of fossil fuels are one of the world’s biggest killers, accounting for approximately 2.5 million and 4.5 million premature deaths respectively in 2024.

But providing more access to clean cooking and clean electricity more quickly can sharply reduce these unnecessary deaths, which are concentrated almost entirely in low-income countries.

Not surprisingly, they fall most sharply under NZE, with combined deaths from household air pollution and ambient air pollution plunging by 4 million by 2035.

Under STEPS and CPS, deaths from household air pollution fall much less sharply, and deaths from ambient air pollution jump instead of falling. In aggregate, the legacy of these do little or nothing scenarios is 130,000 to 620,000 more premature deaths.

Cheaper

NZE is not only better for human health, it’s better for bank balances. In rich economies, households spend about US$650 a year more on electric vehicles, building retrofits and heat pumps, about US$100 more on electricity, and US$1300 less on oil products by 2035, for a net gain of about $US550 each year compared to 2024 bills.

In emerging and developing economies, households spend a couple of hundred dollars more on energy bills, EVs and other equipment than in 2024, but they are getting more energy services in exchange (and their governments may redirect fossil fuel subsidies into clean energy subsidies).

Under STEPS, annual spending on energy services and goods in advanced economies falls by a net US$400 — less than under NZE — while under CPS the savings are smaller still. The same is broadly true in emerging and developing economies.

...and more efficient

Solar squib

Where the rubber hits the road is in the roll out of clean energy. Under NZE solar and wind explode to dominate generation, as would be expected, hydro and nuclear make modest contributions, “unabated gas” shrinks to a very small role, hydrogen, ammonia and “fossil fuels with CCUS” are just visible on the chart and coal disappears rapidly.

(Under CPS and STEPS, gas and coal retain a much larger role; growth in wind and solar is still substantial but much less than under NZE.)

Given the IEA’s record of underestimating solar, it’s worth asking if they’re doing it again.

Ember, a UK-based think tank with a pro-energy transition bias, last week published a chart showing the accelerating growth of solar generation over the past four years. The compound annual growth rate (CAGR) from 2022 to 2025 (2025 is projected to 31 December for 2025) averaged 28%, accelerating from 24.5% in 2023 to 31%.

What does WEO 2025 say about this? It slams on the brakes. Under NZE CAGR for solar falls to 19% by 2035 and 11% by 2050. (Under STEPS the comparable rates are 14% and 8.5%, and under CPS 13% and 7.8%).

That could be justified by “reversion to mean” modelling assumptions. Then again, under relentless pressure from China’s massive capacity expansion, solar panel prices continue to slide, which doesn’t sound conducive to decelerating growth.

To again quote Hannah Ritchie: “I still think it’s unlikely that we’re at peak solar PV deployment globally. Some might disagree.”

Energy mix

​

With thanks to OnlyFacts​

NEM Renewables Breakdown
​Last week (11 Nov - 17 Nov) vs. same week in 2024:

• Renewables: 51.6% (+6.2%)
• Fossils: 48.4%

SWIS Renewables Breakdown
​Last week (10 Nov - 16 Nov) vs. same week in 2024:

• Renewables: 61.3% (+8.1%)
• Fossils: 38.7%

Emissions Intensity (NEM & SWIS) This month so far vs. Nov 2024​
• NEM: 473.9 kg CO₂e/MWh so far this month (-8.3%)
• SWIS: 297.0 kg CO₂e/MWh so far this month (-19.0%)