It seems like self driving cars would reduce the importance of the network effect and make the barrier to entry a capital one (cost of the fleet). I'd like to

It seems like self driving cars would reduce the importance of the network effect and make the barrier to entry a capital one (cost of the fleet). I'd like to see how much utilization the drivers in a typical market are getting. In order for the self-driving fleet-ownership model of future-uber to work, I think the vehicle utilization would need to be relatively high to make it economically competitive with car owning. It also occurs to me that ride-hailing may not be a good substitute for those who use a car to commute. $70B starts to look awfully high if most people still need to own a car to get to work....