The most quotable moment in today's lead story is about magnets, not cars. Michael Dunne's argument is that China's position in the car industry was assembled decades before anyone was choosing an EV, in materials deals and state credit that looked boring at the time. If he is right, the useful question stops being how everyone else catches up on manufacturing, and becomes what they do with the ground they still hold. Today's other four stories read, without meaning to, like an inventory of that ground.
Two of them answer the same problem from opposite directions. BMW's answer is hardware brute force: a 15 minute charge that bought 238 km of motorway driving, off the back of a very large pack and a 400 kW peak. XPENG's answer, as its e-powertrain vice president told Forbes, is that pushing peak charging speeds higher now brings diminishing returns. Its bet is the software layer, which is what the Munich self-driving demonstration is really arguing. Both are expensive to build. Only one of them is expensive to copy.
The Turlock Irrigation District is the outlier, and quietly the most encouraging story today. Putting solar canopies over irrigation canals is not a supply chain play at all. It is the reuse of infrastructure a public utility already owns, on land nobody has to buy, next to the pumps that need the power. That kind of asset cannot be relocated, undercut on price, or bought out from under a state, and by the New York Times' account California has roughly 4,000 miles of major open canals.
Then there is the Twizy, the one that keeps nagging. Two people spent a day and around a thousand dollars discovering their own car was already the fast version, then nearly halved its 0 to 28 mph time with a phone app. None of it is strategically significant. But enthusiasm for cars as objects rather than appliances is one thing scale does not manufacture, and it is where every aftermarket and club series starts.
Two dates matter over the next six months. UC Merced's forthcoming economics report will decide whether solar canals spread past demonstration status or stay a very well documented experiment. European approval for XPENG's urban self-driving, which Forbes suggests is unlikely before 2027, will decide whether the software argument gets tested in front of real customers or stays a video shot for journalists.
Bottom line: Dunne is almost certainly right that nobody out-manufactures China this decade. The winners outside it will be whoever stops trying and builds the things that do not fit in a container: regulated software, infrastructure already in the ground, and people who actually want to own the cars.
Original editorial from nexusEVnews. It links to our commentary on third-party videos, where figures are as presented in the source and have not been independently verified. Spotted an error? Tell us and we will correct it.