argumate:

argumate:

if you’re running a taxi service you can’t just charge for driving someone from A to B, you have to charge for driving from A to B to C, where you pick up the next available passenger.

if more people take taxis then there are more passengers to pick up so the distance from B to C will decrease, allowing you to charge less for the drive from A to B, increasing the number of people willing to take taxis in a virtuous cycle.

however, trips are not random and at any given time there is a net flow of passengers in a particular direction: into the city, out of the city, from north to south, or whatever.

this is bad news as in the worst case it doubles the trip length: if every passenger gets picked up in the city and dropped off in the suburbs 10km away, then the fare must reflect the total round trip for that passenger of 20km.

Uber’s subsidies do not solve this problem as they cannot manufacture demand for trips that no one wants to take.

one can postulate some kind of nanotech self-driving car that simply dissolves into its constituent atoms at the end of the journey instead of making a round trip but absent some technological break through the fundamental problem still remains weighing the convenience of being able to travel somewhere no one else is going against the cost of travelling somewhere no one else is going.

someone has to pay for the return trip and it is either the passenger in the form of higher fares, or the driver in the form of lower pay, or the Uber investors in the form of subsidies.

there is an upper limit on the fare that passengers are willing to pay for a trip before they consider alternative arrangements, and a lower limit on the pay that drivers are willing to accept before they consider alternative employment, so the only unknown quantity is exactly how much capital can be ploughed into subsidies before the investors baulk.

– Crazy Taxi (1999)