I expect local governments to enact per mile taxes, replacing gas taxes, which at some point makes this unaffordable to constantly have your vehicle on the move for you (but still affordable if its part of a rideshare fleet generating revenue).
So in a hypothetical wherein Robotaxis have acheived full penetration, maybe we'll need 1/3 of the vehicles we use today.
There would likely be a lot less steel in those vehicles as well, since they'd largely be optimized for single commuters, and would need less of the redundancies built into current cars for safety.
http://data.bls.gov/timeseries/LNS11300000
https://i.imgur.com/5YFhBpx.png
As well as the labor force declining in general:
http://www.factcheck.org/2015/03/declining-labor-participati...
"Still, Priebus’ comment, tying the entirety of the drop in the labor force participation rate to “the Obama economy,” ignores some of the demographic and structural forces that have been driving the participation rate down for more than a decade, and that are expected to continue to drive the rate down for decades to come."
This is why I chuckle when everyone says "Look how few Tesla vehicles are produced!". They don't have to replace every vehicle purchased, just the vehicle miles driven. Cars sit idle 90-95% of the time, which means if you increase utilization with the Tesla Network, they need only 20-30% of total worldwide manufacturing capacity to replace all cars.
Total cars sold in the US this year? Just shy of 18 million. How many does Tesla need to build to replace that capacity if they're self driving? 4-6 million cars/year to start, decreasing rapidly depending on rideshare uptake. Easily doable with their automated manufacturing efforts.
Gonna be a bitch when Tesla manufacturing curve hits vehicle miles demand, and existing automaker sales plummet.