A lot of it is in the weakness of the gig model. If prices rise, then demand falls. But the lower demand is, the less likely people are to try and drive, so the

A lot of it is in the weakness of the gig model. If prices rise, then demand falls. But the lower demand is, the less likely people are to try and drive, so the more expensive it becomes to do rides. A significant chunk of their "overhead" isn't engineers, but marketing to get new drivers into their system. It's not stable. Turnover is very high. Because they don't have a stable workforce, their problems compound more quickly than in other business models. Surge pricing is what happens when there's not enough drivers. If you raise the base price, and now have lower demand and fewer drivers, you'll likely need to surge on top of that increased base. You can imagine a taxi company with 20 cabs on the road has a fairly predictable supply. A gig model is always playing the odds. When there's too few drivers, Uber both misses customers and has to charge the customers extra to pay the drivers more. When there's too many drivers, Uber can probably charge less and meet all demand, but the drivers have a worse experience. This is not so bad when the overall level of activity is high, but is not good when the overall level of activity is low. That is, it's much worse to have 5 drivers and 10 riders than it is to have 55 drivers and 60 riders. tl;dr if you're charging high prices, you have a small, upper class target market. The gig model is suited to a mass market model.