The internal combustion engine is fantastically cheap for what it entails. It's as complicated as a bloody rocket engine, but an entire vehicle can be bought for a few months California rent. It took literally trillions to engineer it to what we have today. Companies don't want to abandon that investment.
And yet... That's what they'll have to do. The electric car will be so much cheaper to make (and is so much cheaper to operate and maintain) due to its fundamental simplicity that as soon as battery costs get lower, nearly everyone will be forced to transition to electric.
Tesla was recently replacing $10,000 electric motors on their Model S because they were wearing out under normal usage - some customers got 3 or 4 of them, if I remember correctly. http://www.greencarreports.com/news/1101153_two-thirds-of-ea...
And your point about how good internal combustion engines are for the price is exactly what I said earlier. It's pretty amazing that we HAVE been able to make ICEs so cheap and (relatively) reliable, considering how complex they are with so many moving parts.
The fact that Tesla had teething issues with their early electric motors does not change the fact that fundamentally, brushless electric motors are simpler, more wear resistant than internal combustion engines in principle by a huge margin.
Tesla isn't the only one who will be able to take advantage of the simplicity of a pure electric drivetrain.
Advances would be more likely to come in the form of reduced weight (better magnets for PM motors, e.g.) and durability.
After all, it's not like we haven't been using electric motors for roughly the same amount of time as we have IC engines. Heck, most of us own way more electric motors than IC -- kitchen appliances, HVAC, computer drive motors, laundry equipment. Even the car usually has a dozen or two electric motors in it compared to one IC engine. :-)
Also it is not before the next generation of cars (i.e. 10 years) that the bulk of their sales could really be mainly electric. (the world is not like California. A lot of people interested like me simply can't get an electric car - not even because of lifestyle, just because of infrastructure) In the meanwhile they will need to provide hybrid or simpler hybrid.
And traditional constructor are experienced at designing roomy interiors. Designing a single interior across a range is probably more a win than a loss.
BTW it seems that nowadays every constructor has some model ready to commercialise either in 2018 or 2019. Will be interesting to see if they really can hit their target and what Tesla competition is really looking like. Interesting times.
The major German marques often seem to be investing in shared platforms for next generation models that support combustion and electric drive options, but it strikes me that a platform for a combustion based car has to support a ton of things the electric one often doesn't - combustion engines have typically needed much larger/complex gear boxes and drive train in general, plus all the big firewalls needed to keep engine fires out of the cabin and so on. A purely electric car company like Tesla can presumably build out its platform technology without having to worry about a lot of these things. I'd guess you can even build more cabin space for a vehicle of the same size and so on thanks to eliminating many of them. I wonder if shared platforms can leave in otherwise artificial constraints on the electric version for compatibility reasons?