"cartel (n): an association of manufacturers or suppliers with the purpose of maintaining prices at a high level and restricting competition."
Having two competitors doesn't necessarily create a cartel. Are you trying to argue that Uber and Lyft are working together to maintain prices at a higher level?
I'm saying that only two participants in a space does not typically yield the benefits of true competition.
The stronger may use competitive pricing in the short term to weaken the other, driving it to withdraw from the market or be acquired, but this hardly provides long-term benefits to the consumer like true competition would.
For example, note the Office Depot / Officemax / Staples relationship. Though competitors, they never really did compete on price on everyday products (and in fact often charged the exact same price for common items -- a price much higher than, say, mail- and internet-order sources such as Amazon), hence the ability to advertise local price-matching without genuine risk to profit.
The result? Three became two, and two are becoming one, resulting in no competition between them at all. The Uber / Sidecar / Lyft relationship is tracking very similarly.
Can you name any retail space with only two players of significance that has not ended up having the same effect on consumers as a cartel? I honestly can't think of one.
The Cellular One (now AT&T) / GTE Mobilnet (now Verizon) relationship is another example of this behavior. Prices of entry and of usage stayed high until additional, independent competing networks (e.g. T-Mobile, Sprint/WiMax) entered the space.
Barriers to entry are high for a mobile network, but probably pretty low for a ridesharing app. If Uber and Lyft raised their fares massively, then they're leaving a wide open opportunity for someone else to come and undercut them, even if just for a short time. There might be some 3rd app waiting in the wings to pick up price-sensitive riders and drivers whenever the "cartel" puts prices too high.
Anti-competitive behavior is easier when you have fewer players to coordinate, or—put another way—few opportunities for a defector.
I wouldn't venture to guess that Lyft/Uber is a good long-term state of affairs for the ride-scheduling market (or whatever you want to call it), but it's better than just Uber.
Also, consider that Lyft and Uber are spending big money to change the rules in every market—something that is only going to lower barrier to entry for new competitors. Based on real-world experience, I'm convinced that Uber is very worried about the follow-on effect and looking for ways to lock in customers, drivers, etc.
Anyway, this whole ride-scheduling thing is only going to last until it's possible to hop on the never-ending train of self-driving cabs.