A lot of economists fantasies involve "peak pricing" and "surge pricing" and making sure prices adjust instantaneously in response to changing demand in things like electricity markets, rush hour traffic, and now taxis. Not that these are *always* bad ideas, but analysis often overlooks the fact that people *really really really* hate price uncertainty. It's actually a big reason why people hate highway congestion so much. It isn't the slowness or the bumper to bumper traffic - though it is that, too! - it's the not having any idea how long it's going to take you to get home on any particular day. If people knew that their commute would be 50 minutes and cost 5 bucks in tolls every single day they probably wouldn't be thrilled but they'd accept it. Trying to smooth out traffic flows by adding varying tolls just shifts the annoyance from varying driving times to varying prices. The focus is always on "how much would people be willing to pay to reduce their travel time?" But the real question is "do people really prefer price uncertainty to time uncertainty" and I don't think there's a clear answer about that. These cunning plans just shift from one to the other.
And the same with Uber. Everybody puts on their very smart Econ 101 hat and says, while stroking their chin, that this is efficient because it matches supply to demand, but that's based on a particular model of preferences which doesn't factor in aversion to price uncertainty. If I take my Uber to work every morning, I really don't want to wake up every day wondering if it's going to cost me 6 bucks or 40. One problem - I don't know how long I am going to wait for my Uber - is solved, but replaced with another - don't know how much it's going to cost. The models focus on the waiting time/stuck in traffic time as the issue, but the bigger issue is price/time uncertainty. It's why people pay a lot to lock in long term contracts.