In, say, 2006, $884M was spent on shipping and $567M in shipping revenue[1], so there was a subsidy of $317M. This was in total revenue of $10.71B[2]. That's a 3% subsidy, and is within a profit margin.

Uber is a bit harder to pin down from a quick google [3]. I'm seeing $750M loss in the recent quarter, which are "largely" attributed to driver subsidies by third parties, but Uber isn't talking. Let's say $500 as a rough estimate, which is 10% of the $5B in bookings that quarter.

So I guess Uber's subsidies aren't as high as I thought, but are still significantly higher than Amazon's. They both operate in cut throat industries with slim margins. Whereas Amazon invested in warehouses and their infrastructure. Uber may have something up their sleeve, and I hope they do, but it's not clear for an outside observer how they are anywhere as likely as Amazon was to have profits to justify their valuation.

[1] https://www.statista.com/statistics/236503/amazons-annual-sh...

[2] http://www.wikinvest.com/stock/Amazon.com_(AMZN)/Data/Revenu...

[3] https://www.bloomberg.com/news/articles/2016-08-25/uber-lose...