This week, writing in the Daily Mail, Matt Ridley produced a devastating takedown of the government’s 2030 ban on the sale of new conventionally powered cars. He plans to pre-empt the ban himself by buying a brand-new petrol car in 2029.

I thought he was right about almost everything, except perhaps that final prediction. He’s right to be sceptical about the environmental benefits of electric cars – especially in countries such as China (and, to a lesser extent, Germany) where electricity is largely generated from the filthier forms of coal. Indeed Germans, if they truly want to benefit the environment, may be better off buying electric cars not for themselves but for random French people, since France invested sensibly in nuclear power.

Above all Matt is right to be highly sceptical about arbitrary simultaneous deadlines imposed by governments. We can tell these deadlines are arbitrary, because the years proposed for any ban are all suspiciously a multiple of five. The original date for the ban was to be 2040. That became 2035 and then 2030. The date for the deferred ban on sales of new hybrid cars will be, you’ve guessed it, 2035. As a friend remarked, if you want to maintain at least the pretence of rational calculation, it might have made sense to randomise these dates a bit.

In any case, I don’t see why you would ban plug-in hybrids. For people who mostly drive very short distances (which is most people in Britain), it may make no sense to cart around half a ton of battery on every school run, in case you want to take a driving holiday in France in 2032. That battery might be better deployed elsewhere.

It’s not as if we haven’t seen the same mistake before. In 2007 a ban was imposed on selling new incandescent light bulbs. This was criminally stupid, since it forced everyone to buy utterly useless fluorescent bulbs only a couple of years before better, more efficient LED bulbs became available. In a similar vein, if Toyota fulfils its promise to have created workable solid-state batteries by the end of the decade, there is the risk that every car buyer is forced to buy an inferior battery when a better alternative is only a few years away. Innovation happens gradually and delivers its benefits unevenly – therefore it is stupid to impose it indiscriminately on everyone all at once. It is a classic case of the recurrent problem James C. Scott identifies in his masterwork Seeing like a State, where the need by governments to view the world through the lens of averages leads them to impose uniform prescriptions that prevent individuals from exercising ingenuity or discrimination based on specific knowledge of their circumstances.

But where I think Matt may be too damning about electric cars is – surprisingly, for a renowned evolutionary biologist – in not acknowledging that, for all its virtues, the internal combustion engine is a bit of an evolutionary dead-end. If not by 2029, at least in the next 20 years, I anticipate a Cambrian explosion in transport innovation which would not have been possible with engines powered by dinosaur juice. The electric motor is gloriously simple, smooth, silent, energy-agnostic, efficient, powerful and scaleable, and ‘talks to’ software in a way no conventional engine can. Innovative ideas like the Heathrow Pod (a kind of personal tram) or even the cargo bike are only possible if we back Faraday, not Daimler or diesel. With technology such as in-wheel motors and V2G (vehicle-to-grid) charging, personal transport may see the emergence of a diversity of new forms – and manufacturers – lacking from conventional car-making for more than 50 years.

So when Matt walks into his car dealership in Cramlington in 2029, I’m willing to bet him 0.1 bitcoin he drives off in something electric.