The proposal suggests to me that they think their customers are not extremely price sensitive but they're also not loyal to a specific supplier. In other words, they'll take almost as many rides even if the price goes up but they'll still go with whoever is cheapest.
It makes sense, if something is cheaper, you use more of it. If its more expensive, you start using substitutes, as the benefits of rideshare gets dwarfed by its extra cost. It’s literal econ 101
If they can establish a regulatory framework that favors them (Uber), they can raise prices as competition is further restricted, to more than offset the $21 / hour. Their ideal has to be to establish that favorable competition environment, enabling considerable price increases that get them closer to profitability (ie exceeds the wage cost increase by a large margin). In the current situation, Uber is facing a bankruptcy scenario with their low sales growth, extreme burn rate and with a competitive market where they can't freely increase prices. That tells you what their goal has to be.
If they were certain their customers were not price sensitive, Uber would have already spiked their pricing far higher to push toward profitability.
Many/most Uber customers would use a materially cheaper equivalent service if available because the switching costs are low.
However, that's not the same as price elasticity (of demand) which is what the term usually refers to. Price elasticity is about how much less they'd use any price-competitive service as the price increases.
I'm honestly not sure what that curve looks like. Within reason, pricing wouldn't affect my usage much at all. But I'm a light user of these services and use them almost entirely for business. Certainly I'm a very different profile from a young urban professional who doesn't own a car.
ADDED: Between surge pricing and other pricing experiments I'm sure they conduct, I would imagine that Uber has a pretty good idea of what demand and driver supply curves look like.