A Polemicist's Valid Concerns, Buried Under a Lifecycle Fallacy

Overall Score: 3/10

Final Verdict: Misleading

• Factual accuracy: 4/10 — Several individual claims are accurate (fossil fuel share of the grid, DRC cobalt dominance, child labour, higher manufacturing emissions for EVs), but the headline claim — that EVs are worse for the planet than petrol — is contradicted by the weight of lifecycle evidence, and the 'nearly double' manufacturing-emissions figure is an extreme end of the range that applies only to coal-intensive production.
• Evidence quality: 2/10 — No specific studies, datasets, or named sources are cited anywhere in the piece; assertions are presented as self-evident facts, making independent verification of the author's specific figures impossible.
• Logical coherence: 3/10 — The argument consistently compares EV manufacturing emissions against petrol operational emissions rather than like-for-like lifecycle totals, a structurally invalid comparison that undermines the core conclusion.
• Completeness: 2/10 — The piece omits the most consequential counterargument — that EVs recover their manufacturing carbon debt within tens of thousands of miles of driving and thereafter emit far less than equivalent petrol vehicles over a full lifetime — as well as the improving trajectory of grid decarbonisation.
• Source independence: 5/10 — No author identity or affiliations are disclosed; the piece is labelled only as a 'motoring blog', so a specific undisclosed interest cannot be established, though the framing is consistently and one-sidedly favourable to the internal combustion engine.
• Precision of claims: 3/10 — Claims such as 'nearly double' manufacturing emissions and 'a small fraction' of batteries recycled are vague, unsourced, and selectively drawn from the worst-case end of a wide empirical range, lending them the appearance of precision without the substance.
• Currency: 5/10 — The 60% fossil fuel electricity figure is broadly current for 2024, and the cobalt and recycling concerns are live issues; however, the piece does not acknowledge the rapid direction of travel — grids decarbonising, battery chemistry shifting away from cobalt, recycling capacity growing sharply — all of which materially affect the argument.

Summary

The piece's fatal flaw is structural rather than merely factual: it repeatedly isolates the manufacturing phase of the EV lifecycle and compares it against the operational phase of a petrol car's lifecycle, which is not a valid comparison. When the full lifecycle is assessed — as the IEA, Transport & Environment, Argonne National Laboratory, and ICCT consistently do — EVs produce substantially lower total greenhouse gas emissions than equivalent petrol vehicles in virtually every grid scenario, including grids with a significant fossil fuel component. The claim that an EV is 'worse for the planet' is not a heterodox view courageously spoken; it is the conclusion that emerges only if one stops counting once the car leaves the factory gate.

Several of the supporting claims are, taken in isolation, legitimate. The DRC cobalt and child labour problem is real and well-documented, with over 70% of global cobalt mined there and credible estimates of 40,000 children involved in artisanal extraction. The 60% global fossil fuel share of electricity generation is accurate per IEA 2024 data. EV manufacturing does generate higher upfront emissions than petrol car manufacturing, and battery recycling rates remain low relative to the scale of production. These are genuine issues that the industry and policymakers should address.

However, the piece consistently omits the evidence that cuts against its thesis: that EVs break even with petrol cars after roughly 15,000–30,000 miles and thereafter diverge increasingly in EVs' favour; that grids are decarbonising, compounding EVs' lifetime advantage over time; and that cobalt use per battery is declining as chemistry evolves. No sources are cited anywhere, making the specific figures — such as 'nearly double' manufacturing emissions — unverifiable and, as the evidence shows, drawn from the worst-case tail of the distribution rather than the central estimate.