Look at it the other way round: they could reduce losses from 6% to 4%, which lead further savings in needing less cooling tech.

Note that the impressing 9x% efficiency rates that we hear about are peak efficiency rates. There is something to gain in widening the operating ranges of this efficiency also. See for instance https://www.researchgate.net/figure/Efficiency-maps-of-sever...

what is the cost of the motor vs the cost of the input materials? what % of the cost of a vehicle is the cost of the electric motor? My understanding (however limited here) is that car-worthy motors are still priced at a bit of a premium to their input materials, but it doesnt really matter because overall they are still not that expensive compared to batteries.

Yeah, some motors have rare-earth permanent magnets which are expensive. Induction motors don't, they just have a weird squirrel-cage structure made of (I guess) copper. Probably the manufacturing inputs for an induction motor are cheap, but the motors are a bit inefficient.

I'm currently doing a conversion that uses a Netgain Hyper9 [1]. They cost about four and a half thousand dollars (including controller) and they're pretty bulky and heavy. They're really efficient though. An OEM manufacturer I'm sure could have a motor made much more cheaply and design it to run at a much higher voltage and produce correspondingly more power. I think the Hyper9 just uses magnetized iron or something like that for the permanent magnets. No rare-earths, so it might be pretty cheap to make something like that in volume. They serve the conversion market though, so it's kind of niche product. I have no idea what Nissan spends to make something like a Leaf motor.

[1] https://www.go-ev.com/motors-warp.html#HyPer_9HV