I am puzzled about the response of the highways to the (currently being tested in public) driverless horseless carriages. Those who claim to be "in the know" have been telling us years that this innovation is coming literally down the pike in a few years. We do not really know what sort of highways these things are likely to require. Yet highways are still being proposed and constructed despite the risk these will be "stranded assets" in light of driverless cars. And yes, highway schemes over run on cost at least as much as those on railways.
Yet when the equally unknown quantity of bi-modes emerges, this is takes as reason to cancel electrification. Bi-modes are at least known to benefit from electrification.
Perhaps a consistent logic is required across modes?
A road is a road. New build roads are generally of much higher standard than the average existing road. Driverless cars will have to be designed almost entirely for the roads which already exist - we are not going to rebuild the entire road network for them, unless you want to implement the Serpell Report to fund it.
'We shouldn't build roads because driverless cars are coming' is nonsense. They are still cars and are not in themselves likely to radically change the need for roads to exist.
'We shouldn't electrify railways because of new technologies' makes rather more sense. Bi-modes, batteries, whatever else would make wires redundant to a greater or lesser extent, and more importantly deliver better value for money, the government hopes, reasoning nothing could be worse than electrification. In the nowadays rare cases where a new road is built it is deemed a better investment than the alternatives.