> If you have costs that scale sublinearly with volume, such as software development or idle driver time, then margins really do improve with volume

And if you're subsidising rides below cost because you're trying to grow your market and you have piles of VC cash so you don't care that you're burning through money, then eventually you're going to run out of that money and will be in trouble if you can't raise any more.

Time will tell whether Uber was building a sustainable business or just burning through VC cash.

Uber was just burning through VC cash based on the idea that 'if you're the biggest jerk on the block in every possible sense, capital will decide you're going to be the winner because you're meaner than everybody else'.

Winners don't quit, so Uber is dead to capital now: it was always based on maximum evilness and all the stuff about disrupting and ridesharing was mere window dressing.

Note to capital, wherever it is: this is what you get when you go by personality rather than studying the fundamentals of a business. You can't simply pick the most toxic individual or company, claim they're going to kill everybody else, and then prop them up with valuation. The valuation didn't fail but your pet Dr. Evil did, and that was your proxy for maintaining the 'killer' behavior. Unless or until capital can be personified as evil AIs that cannot die, this was never really an optimal strategy for capital.

Considering that Lyft is viewed as the "nicer" company, do you think its fundamentals are more sound? Will it reach profitability?

Here is what Uber has built, save your car EMI's.

When you're #1 and losing money on every transaction, that didn't work.

Only if you are in #1 in a pie that isn't getting bigger. Ridesharing is most definitely not a developed market globally. There is a lot of room to increase the size of the pie.

Ridesharing is definitely not a developed market globally because in most of the rest of the world the margins on organising transport around cities are already measured in cents rather than dollars. These markets are not going to be more profitable for Uber than the ones they started in.

Particularly not if you can't set up an operation in an Western city with expensive taxi services without losing money on every ride you operate even before centralised overheads are taken into account.

Is there evidence that Uber is has negative gross margins in developed Western cities? (Not a rhetorical question - I'm open-minded and very curious.)

From their 2015 financials, it looks like revenues have been consistently higher than cost of sales.

Source: http://www.nakedcapitalism.com/2016/11/can-uber-ever-deliver...

Is there evidence that Uber loses money on every transaction?

According to financials from 2015, they have higher revenues than cost of sales:

www.nakedcapitalism.com/2016/11/can-uber-ever-deliver-part-one-understanding-ubers-bleak-operating-economics.html

This suggests to me that marginal rides are profitable, even if they aren't profitable enough to cover fixed costs of software development or investments into new geographic markets.

When you're giving drivers more cash than the customer is paying you, there's no way to 'scale' out of that.

Is there evidence that this is the case? And if so, could you share it?

From their 2015 financials, it looks like revenues have been consistently higher than cost of sales.

Source: http://www.nakedcapitalism.com/2016/11/can-uber-ever-deliver...

Granted, it was only promos in certain cities to drive growth of both drivers and riders.

Quick Forbes link I found https://www.forbes.com/sites/ellenhuet/2014/07/02/ubers-newe...

Thanks for the source and the reply. It's good to know that in 2014 they had promos in developed city markets that dropped their gross margins into negative territory.

Nonetheless, my impression is that this is more the exception than the rule, based on the financials I linked to earlier.

But not Uber.