Every so often a piece of research lands with a title that could have been written for this network. On 1 September 2026 the Technical University of Munich (TUM) presented a white paper called «Defossilisation, Not Decarbonisation» — an argument for a new, life-cycle-based standard for the CO₂ regulation of passenger cars. It was produced by CirculaTUM, the university’s interdisciplinary research network, under the lead of professors Johannes Fottner, Magnus Fröhling and Tim Büthe, and commissioned at the start of the year by TUM President Thomas F. Hofmann ahead of this autumn’s political decisions on the future of the European car industry.
The paper is not a position piece. It is a synthesis of 19 studies covering 47 modelled scenarios, and by the authors’ account currently the most comprehensive assessment of the climate balance of cars across their full life — production, use and end-of-life recovery. Its central finding is uncomfortable for everyone who assumed the current rules were roughly right.
The number that does the damage
Across the studies reviewed, battery-electric vehicles reduce life-cycle CO₂ emissions compared with combustion vehicles by 41% on average, and deliver a reduction in around 92% of the cases modelled. The spread is wide: from an 89% reduction at the good end to a 21% increase at the bad end, depending largely on how the car and its battery were produced and what electricity it runs on.
Under Regulation (EU) 2019/631, the same vehicles count as 100% climate-neutral. The reason is simple: the regulation measures CO₂ at one point only, the exhaust pipe. Fossil carbon released in manufacturing, in the electricity used to charge the car, or lost because materials are not recovered at the end of life, does not appear in the metric at all.
Why a measurement problem becomes an investment problem
This is the part that matters for Power-to-X. If a manufacturer invests in near-CO₂-free steel, if a supplier defossilises battery production, if a recycler keeps critical raw materials in circulation — none of it moves the regulatory number by a single gram. A regulation that cannot see a contribution cannot reward it. And what is not rewarded does not get financed at industrial scale.
Fottner’s point from the press briefing is that the logistics picture is better than the regulation suggests: traceable supply chains for green steel and material recovery at the end of a vehicle’s life are already making real progress that the current rules simply do not capture. Industry, on this reading, is willing — it is the accounting that is missing.
The same blind spot applies to renewable synthetic fuels. A car running on e-fuel made from renewable electricity and recycled CO₂ emits fossil carbon at the tailpipe on paper, and is treated accordingly. The tonne of fossil carbon that was never extracted upstream is invisible. That asymmetry is not a detail of e-fuel policy; it is the same structural flaw seen from a different angle.
The 2030 gap
The paper puts a figure on the consequence. Drawing on one of the reviewed studies (Tang et al., 2023), it finds that at the EU’s intended pace of electrification the passenger car sector would deliver roughly 80 million tonnes of CO₂ reduction by 2030, against the roughly 188 million tonnes needed to meet the EU’s own target — a shortfall of more than 100 million tonnes. Combustion vehicles are expected to make up around 78% of the fleet in 2030 regardless.
The reform of Regulation (EU) 2019/631 is part of the Automotive Package the European Commission presented in December 2025. The plenary vote in the European Parliament is currently scheduled for 23 November 2026. According to the TUM analysis, the credits for steel and fuels contained in the package as it stands would cover at most 10% of a manufacturer’s required CO₂ reduction — helpful, but not enough to close the gap. Hence the call for a life-cycle standard in addition.
What the researchers actually propose
The proposal is a mandatory, verified life-cycle standard: performance-based crediting that treats every tonne of fossil carbon genuinely saved the same way, regardless of where in the life cycle the saving occurs — and that is operable in practice, not only in theory. Fröhling’s argument is that this is not regulatory uncharted territory: the Ecodesign Regulation and the Carbon Border Adjustment Mechanism already show that Europe can regulate on a life-cycle basis. Steel, with the most reliable data available today, is proposed as the sensible starting point.
Büthe adds a political reading worth keeping in mind: the dispute between groups in the European Parliament is, at its core, not about whether the life cycle of a vehicle should be captured in regulation, but about how binding that should be. That distinction tends to disappear in the headlines, and it is precisely the distinction on which a pragmatic compromise depends over the coming months.
The Swiss angle: we already know how to do this
For SPIN this is a welcome piece of evidence, and a familiar argument arriving from an unfamiliar direction. Switzerland’s CO₂ Act has long been built around measurable CO₂ reductions rather than the monitoring of individual vehicles — a logic much closer to what CirculaTUM now proposes for the EU than to the tailpipe metric itself.
But before anyone in Bern nods along too comfortably, there is an inconsistency in our own house. Switzerland demands life-cycle assessment almost everywhere in environmental policy. Ecobalances are the working currency in construction materials, packaging, agriculture, public procurement and product declarations. Whenever an environmental claim has to be substantiated here, the answer is an LCA.
Everywhere, that is, except in the one place where it would matter most for Power-to-X and for the reduction of CO₂ emissions caused by road traffic. When the CO₂ reduction achieved by e-fuels and other renewable synthetic fuels is calculated, the accounting reverts to a pure tailpipe view. The renewable carbon bound into the fuel, the fossil carbon that was never extracted to produce it, the upstream emissions avoided along the way — none of it enters the calculation.
Where the Swiss error comes from
The origin of this is worth naming precisely, because it changes what needs to be done about it. The tailpipe view is not something Swiss legislators chose. It enters through the implementing ordinance: in drafting and applying the CO₂ Ordinance (CO2-Verordnung), the administration orients itself closely on EU legislation — and imports the EU’s exhaust-pipe metric together with it.
The superordinate CO₂ Act (CO2-Gesetz), however, departs markedly from that approach. It is built on technology neutrality, and it provides for renewable synthetic fuels to be credited towards emission reductions. In other words, the ordinance is narrower than the law it is meant to implement. Swiss legislators deliberately chose a different logic from Brussels, and the implementing practice has quietly narrowed it back down again by following the EU template.
That is what makes this a solvable problem rather than a political battle. Correcting it does not require reopening the CO₂ Act or picking a fight over drivetrains. It requires bringing the ordinance back into line with the law that governs it.
The consequence of leaving it as it stands is twofold, and both parts are avoidable. Scientifically, the reduction actually delivered by renewable synthetic fuels is systematically understated. Economically, the compensation obligation becomes more expensive than the physical carbon balance warrants, because more offsetting has to be purchased than the fuels’ real climate effect requires. A rule that is stricter than the science is not caution; it is a self-inflicted cost on defossilisation.
SPIN is therefore calling on the federal authorities to amend the CO₂ Ordinance accordingly and apply to fuels the same life-cycle logic that Switzerland requires of nearly every other product and process. This is not a request for privileged treatment of e-fuels. It is a request for consistency: one method of measurement, applied evenly, so that every verified tonne of fossil carbon avoided counts once — wherever in the chain it was avoided. And it is a request that the ordinance reflect the technology neutrality the Swiss Parliament wrote into the Act.
The wider point is one this network keeps returning to. Defossilisation is the goal; a specific drivetrain is not. A regulation that counts fossil carbon wherever it is avoided is technology-neutral by construction: it rewards green steel, circular materials, renewable electricity and renewable synthetic fuels on the same terms, and lets engineering and economics decide the mix. A regulation that counts only what leaves the exhaust pipe forecloses that question before it has been asked — and, on TUM’s numbers, misses its own target by more than half.
The white paper is open access and worth reading in full before the November vote.
Source: Technische Universität München (ed.): «Defossilisation, Not Decarbonisation. Ein Weg zu einem neuen, lebenszyklusbasierten Standard für die CO₂-Regulierung von Pkw.» White Paper, Munich, September 2026, DOI 10.14459/2026md1861116; and TUM press release of 1 September 2026 (tum.de).
🤖 AI transparency: This text and its visual were created using AI, reviewed and approved by a human.
Power-to-X Congress Switzerland 2026 — «Reality Check with Net Zero». Join us on 22 September 2026, 10–18h, at the Kursaal Bern. Co-organised by energie-cluster.ch and SPIN, with partner Réseau H2 Suisse Romande. Learn more and register →