Electrification could cut global energy demand 40%, but Europe still faces a costly decade

Electrifying transport, heating and industry would cut global final energy demand from 416 to 247 exajoules, about 40%, according to an analysis published on 29 September by Hannah Ritchie's By the Numbers newsletter. The same day, CBC News reported that Newfoundland and Labrador's energy minister said the province's door is "open for business" to AI data centres.

The two stories point in opposite directions. One says a fully electrified world needs far less energy than today's fossil system. The other shows how much electricity new computing demand wants to pull out of the grid before that transition is anywhere near finished.

What the 40% actually measures

Ritchie's piece works through numbers from Nick Eyre, a professor at Oxford. It compares global final energy demand today with a post-transition system where suitable sectors are electrified and the rest run on hydrogen. Electricity demand rises from 110 to 189 exajoules, but total final energy demand drops from 416 to 247 exajoules. The model assumes no efficiency gains beyond electrification and the switch to hydrogen, so Ritchie calls it "a fairly simplistic model" that likely underestimates the reduction.

The mechanics are not mysterious. An electric car converts around 80% of its energy into motion. A petrol car manages about 20%. Buildings shift from gas boilers to heat pumps. The model puts 90% of space heating on heat pumps and 10% on hydrogen, because full substitution is uneconomic during winter demand peaks. Cooking is fully electrified. Buses are 80% electrified, short-haul aviation is electrified while medium and long haul runs on hydrogen, and only 10% of marine transport, short trips such as ferries, is electrified.

Industry is where the model runs into limits. High-temperature processes cannot be electrified with current technology, so they move to hydrogen. Steel gets a partial shift, from around 25% electric arc furnaces today to 50%, roughly the current OECD mix. Low-temperature steam switches to electric boilers, which the piece says are about 20% more efficient. Drying and separation below 120C can use heat pumps.

The result is a 40% cut in final energy demand, not primary. That distinction matters for anyone reading the number as a forecast of what Europe will actually consume.

Ritchie's framing is careful. She notes the comparison does not account for energy growth as countries develop, and that other analysts would push some sub-sectors toward biofuels or other alternatives. Those choices, she argues, do not change the overall conclusion.

Europe's decade problem

The wider European picture is messier than the model. Recent headlines, which are context rather than facts this article relies on, describe high energy prices pressing on EU economies, diesel prices surging with Ukraine only part of the story, and a warning from Reuters columnist Ron Bousso that Europe's electrification dream will not spare a painful decade. Those headlines are not part of the dossier, but they frame the gap between a clean model and a continent mid-transition.

Oil gives a sense of how sharp the swings can be. Writing in The Conversation on 29 September, two energy economists note that the price at the primary US crude hub was $66 a barrel in late February 2026, before the US and Israel attacked Iran, and $101 a barrel on 13 April. The Strait of Hormuz closure is the mechanism they cite. When a significant source of oil is blocked without much notice, prices rise sharply, and the money flows mostly to producers.

Where it lands depends on ownership. Saudi Arabia's government controls nearly all production, so high prices benefit state finances. In the US Permian Basin, the windfall goes largely to shareholders through dividends, debt reduction and buybacks. In the UK, private shareholders benefit but an additional tax on oil and gas profits means the government collects a significant share. Norway channels revenue into its Government Pension Fund Global, valued at over $2 trillion. Russian crude is subject to price sanctions: Western shipping, insurance and financing can be used only if the price is below $60 per barrel.

That is the fossil system Europe is trying to leave, and it is still setting the price of the transition.

New demand arrives before the old system leaves

Data centres are the clearest example of new load arriving early. CBC News reported on 29 September that Newfoundland and Labrador's energy and mines minister, Lloyd Parrott, told the House of Assembly during the special session on the new Churchill Falls agreement in mid-September that the province could consider selling excess Churchill Falls power to a data centre company. "We said our door is open for business and we're looking forward to doing business with the world," Parrott said at the time. Premier Tony Wakeham and lead negotiator Barry Perry also floated data centres as a potential use for that power.

A department spokesperson, Brodie Thomas, confirmed by email that the provincial government has been approached by companies about potential data centre development, but said commercially sensitive discussions prevent comment on specific proponents. TechNL CEO Andrea King told CBC that Labrador appears to meet what data centre proponents want: low-carbon hydro electricity, cold weather and a lot of land. She added a caveat: "It's not just 'does this make sense?' by itself. You have to look at economic costs and opportunity costs of what else you could do with that electricity."

Tech journalist Paris Marx, author of Hyperscale, made a similar point to CBC, asking whether giving electricity to data centres makes sense when there might be better uses. Marx acknowledged some companies mitigate impacts, such as through closed-loop cooling that recirculates water.

The scale is not small. CBC reported that Meta's planned data centre in Sturgeon County, Alberta, is expected to cost $13 billion and come online in two to three years, and that Meta and Bell have announced billions for data centres in Alberta, Ontario and Saskatchewan. Nova Scotia Premier Tim Houston has set five starting conditions for any proponent, and EverWind Fuels has expressed interest in building in that province.

At the chip level, the pitch is efficiency. Efficient Computer said on 29 September that it raised a $97 million Series B led by TQ Ventures, taking its total raise to $173 million at a $650 million valuation. CEO Brandon Lucia wrote that the company claims 10 to 100 times better energy efficiency than traditional CPU architectures, and cited Amdahl's Law: if 90% of a computation is accelerated to zero cost, the maximum benefit is still 10x, because the remaining 10% sets the ceiling. The company says it is launching its Electron E1 processor at volume for embedded physical AI systems such as robots, drones and wearables.

Those two claims sit awkwardly together. A 40% cut in final energy demand depends on electrifying most things and making the rest efficient. A data centre boom adds load first and efficiency later, if at all.

What the historical record says

The US Energy Information Administration's photovoltaic timeline, updated in the dossier, traces the cost curve that makes the electrification case possible. The first PV cells converted less than 1% of radiant energy into electricity; some cells today reach a 40% conversion rate. Bell Labs researchers Daryl Chapin, Calvin Fuller and Gerald Pearson built a 6% silicon cell capable of running everyday electrical equipment. The Energy Tax Act of 1978 created a 10% investment tax credit for photovoltaic applications, and the Solar Photovoltaic Energy, Research, Development and Demonstration Act of 1978 committed $1.2 billion over 10 years.

Progress was slow and subsidised. Boeing and Kodak fabricated the first thin-film cells above 10% efficiency. A University of South Florida thin-film cell hit 15.89%, breaking the 15% barrier. By 1993, record efficiencies approached 16% for polycrystalline thin film and 30% for single-crystal devices. None of that happened on market signals alone.

David MacKay's 2008 book Sustainable Energy Without the Hot Air made the arithmetic case that dominated later debate: "Numbers, not adjectives." Its endorsement list included Tony Juniper, then executive director of Friends of the Earth, and Robert Sansom of EDF Energy. The book's site was last modified in August 2015. Its central demand, that energy plans be judged on quantities rather than rhetoric, is the same demand Ritchie's 40% figure is making now.

Two caveats hang over the number. The model assumes hydrogen covers everything that cannot be electrified, which is a large assumption about infrastructure that does not exist at scale. And it is a global average: Europe's starting mix, its industrial base and its winter peaks mean its own path will look different from the aggregate. Ritchie's own note says the estimate likely understates savings, but understating savings and delivering them are separate problems.

The near-term question for Europe is not whether electrification lowers energy demand. The model says it does. It is whether the continent can add data centre load, replace Russian pipeline gas, and keep industrial production inside its borders at the same time, without the decade costing more than the transition saves.