Key takeaways
- Global nameplate battery capacity passed 4 TWh at the end of 2025, up about 30% on 2024, with China holding over 80% of it — a concentration figure, not a growth story.
- EVs displaced about 1.7 mb/d of oil in 2025 and the IEA projects roughly 5 mb/d by 2030: an increment of about 0.7 mb/d a year.
- The IEA recorded a 4.8 mb/d year-on-year contraction in oil demand in 2Q26 — seven times the annual EV increment, and the context for North Sea Dated moving from US$68 to US$77/bbl in weeks.
- Public charging points passed 7 million at end-2025, but China went from 3.4 million to 4.7 million and took over 75% of global growth.
- The IEA does not say wind and solar cover over 90% of global demand growth; it says demand growth to 2030 will be met by renewables, natural gas and nuclear together.
Four transition metrics, and only one of them reaches a price
The energy transition produces a steady supply of large, quotable numbers. Few reach a security's cash flows. The useful test for each is what mechanism would carry it into a price — and whether that mechanism exists.
Battery capacity above 4 TWh is a glut signal, not a growth signal
The IEA's Global EV Outlook 2026 puts global nameplate lithium-ion manufacturing capacity at more than 4 TWh at the end of 2025, up roughly 30% on 2024. Actual EV battery deployment in the same year was 1.2 TWh. Nameplate capacity is therefore running at more than three times the volume it serves.
That gap has a direction. Average battery prices fell 8% in 2025, and LFP packs were more than 40% cheaper per kWh than NMC alternatives. Capacity growth of 30% against a utilisation rate near one-third is a margin story for cell makers, and the sign is negative. Quoting "4 TWh" as evidence of transition momentum inverts what the number says about pricing.
The same figure carries a jurisdiction problem. China accounts for over 80% of that capacity and over 80% of cell production. A sleeve benchmarked to battery manufacturing carries more than 80% single-country exposure in the underlying physical asset base, whatever its holdings' listing venue says.
EV oil displacement is real, slow, and swamped by the cycle
The IEA reports that the global EV fleet avoided about 1.7 mb/d of oil consumption in 2025, and projects that displacement triples to around 5 mb/d by 2030. Across those five years the average annual increment is roughly 0.7 mb/d.
The IEA's July 2026 Oil Market Report has global demand contracting 4.8 mb/d year-on-year in 2Q26, easing to -1.7 mb/d in 3Q26 and returning to +1.2 mb/d in 4Q26, for a full-year 2026 decline of about 1 mb/d and forecast growth of 2 mb/d in 2027. North Sea Dated fell US$31/bbl over a month to US$68/bbl in early July before trading back to US$77/bbl.
The chart puts these on one axis as annual changes in mb/d: a structural EV increment of 0.7 against cyclical swings of -4.8 and +2.0. EV displacement is roughly one-seventh the size of the shock the market was pricing, and it arrives on a five-year clock rather than a five-week one. It is a genuine metric, but not a marginal price-setter, and a portfolio positioned for oil demand destruction on the strength of EV adoption in 2026 was positioned against a 13% price rally that had nothing to do with cars.
Charging infrastructure is a capex line in the wrong currency
Public charging points passed 7 million globally at the end of 2025, with 1.8 million added in the year — growth of more than 33%. China went from nearly 3.4 million points to over 4.7 million, taking over 75% of global growth. Europe grew about 20%. The United States finished the year with roughly 70,000 fast and ultra-fast points.
The aggregate is a headline; the distribution is the investable fact: most of the world's charging build-out is Chinese, state-directed capital expenditure that does not pass through a Western listed operator's income statement.
The generation number is the one that transmits — and it is misquoted
Solar PV posted the largest year-on-year increase of any generation source in 2025, at 620 TWh. Global electricity demand grew 3% in 2025 and the IEA forecasts 3.6% a year through 2030, with solar and wind rising from 17% of generation today to 27% by 2030. Displacing incumbent generation on price, at that scale, is a mechanism that reaches a utility's realised power price and a fuel supplier's volumes.
What does not survive contact with the source is the claim, repeated across four previous posts on this site, that wind and solar cover over 90% of demand growth. Electricity 2026 says demand growth through 2030 "will be met by renewables, natural gas and nuclear." The 100% framing exists only regionally: the IEA expects renewables to meet all EU demand growth, and all of China's additional demand to come from low-emissions sources. Globally, gas and nuclear are in the sentence — so a thesis that gas volumes are structurally capped by renewable build-out has no support in the IEA's own forecast.
Where this framework breaks
Nameplate capacity is not utilisation, and the IEA does not publish plant-level utilisation, so the "three times oversupplied" reading is an inference from two numbers rather than a measurement. Charge-point counts say nothing about throughput; idle and busy chargers count the same here. And the 0.7 mb/d EV increment is a straight-line average of a projection, not a forecast path — if displacement front-loads, the comparison with cyclical swings narrows faster than this piece assumes. The falsifying observation would be a year in which the change in EV oil displacement exceeds the change in cyclical demand. 2026 was not that year, and on the IEA's own numbers 2030 is not either.
The metric worth a watch-list slot is none of the four: battery price per kWh — the only one of these series that is simultaneously falling, measurable, and directly attached to a margin. A second 8% decline in 2026 would say more about which transition equities survive than any capacity headline, and it would land in the same reports that this year's energy-driven inflation shock is still working through — a shock that has already reshaped the front end of the rates curve.
This content is for informational purposes only and does not constitute financial advice. Always do your own research or consult a qualified financial adviser before making investment decisions.