Locos are now around £3 million. IEP cars are around £2.8 million. Hauled vehicles are around £1 million if you need new ones, and around £300k for a thorough refurbishment of an old vehicle such as a mark 3, which will run for another 20 years at least. The differentials do not stop with the initial costs. Several power cars are inevitably going to cost more to maintain than one locomotive.
If the country end of a route is much less busy than the London end,
then either the train is overcrowded at the London and or it is
under-utilised at the country end. If the line is not electrified
throughout, either there is diesel running under the wires - undesirable
but not a catastrophe, or a change of traction. Either there is dead diesel traction on the electrified route, or dead electric traction on the non-electrified route.
The neat solution to the problem is to electrify out from London to the
end of the busy section, split the train and push/pull enough of the
vehicles to satisfy the need at the country end of the route. The
undesirable alternative is to make people change trains, in which case
an ordinary DMU can be used.
In the face of criticism, it was claimed that another great benefit of bi-mode is the ability to "self-rescue" if the power is down. This is fools' gold. There are indeed situations where this might be a benefit, but even then the train is not going to get further than the train at the front of the queue that does not have this "self-rescue" ability. And the best way to prevent loss of power is not to skimp on the specification of the overhead line and to maintain it, and the equipment on the trains, to a reliable standard.
IEP cleverly manages to get the worst of all worlds at maximum expense
as a result of a civil service specification which asked for a train
that could do everything. The same thing has been going on in defence
procurement for decades. It is what happens when projects are left to