TLDR: Ridesharing companies’ massive valuations are based off of a driverless taxi future. However, it seems that in this future almost all the profits will go to the companies that have solved the hard tasks of manufacturing the cars and designing the self driving AI software, not the relatively easy to replicate task of building the app that customers use to order the taxis. What am I missing here?
In light of the Uber IPO next year I’ve been thinking about the multi billion valuations for ridesharing companies like Uber/Lyft/DiDi/Grab.
As far as I can see, their perceived value comes from driverless cars. The idea is that in the future many people won’t even own a car, they will just call up a driverless taxi when they want to go somewhere (the combination of no drivers’ salaries and the fact that cars being able to drive 24/7 will reduce the number of cars that need to be bought will make the taxi services cheaper than owning your own car). Ridesharing companies, as the tools with which customers call their robo-taxis, will hence get a big slice of this multi-trillion dollar market.
However, I can’t understand why these ridesharing companies get a place in this driverless future at all. As far as I can see, we can break down the driverless taxi service into three components:
- Manufacturing the actual physical cars. This is a very difficult engineering challenge; it requires both great industry knowledge and a large number of factories. This “moat” makes it hard for someone to just come in and start their own car manufacturing company, which in turn ensures that the company can have a decent markup/profit on their cars.
- Designing the AI software that drives the cars. Again, this is a very hard engineering challenge which ensures that there’s a “moat” that stops competitors and hence safeguards profits.
- Building the app that people use to call up the taxi. I’m not saying this task is entirely trivial, but (speaking as someone who works in tech/AI) it’s not that hard for a reasonably large tech company to quickly make a reasonable looking app that lets users order a taxi, and it’s definitely nowhere near as hard as the first two. As far as I can see, there’s very little “moat” here, anyone can easily move into this area, which means that there’s very little profit available in just running the app that people use to contact the cars.
As far as I can see, if driverless cars become feasible/reliable, then it’s very likely that the companies with the driverless AI software will just buy cars (or partner with car manufacturing companies) and launch their own taxi service because there’s no way that they’re going to let Uber/Lyft/DiDi/Grab act as a middleman and keep a big slice of the profits. I suppose that the existing ridesharing companies could stay in business by cutting prices until they’re only marginally higher than the cost of cars+driverless software, but then they’re not gonna make much profit and hence don’t justify their current massive vaulations.
I’ve seen a couple of counterarguments to this, but they don’t make sense to me:
- The ridesharing companies can build/are building their own driverless car technology. Firstly, driverless car technology is extremely difficult, and from what I’ve read, companies like Uber are well behind the industry leaders like Google’s Waymo and GM Cruise. Furthermore, I don’t see why starting work on driverless cars (or claiming you can make them in the future) justifies the huge valuations that ridesharing companies currently have; surely it makes more sense to give a large valuation to the startups that have made the most progress in driverless cars (the technology that’s hard to replicate) rather than the startups which have so far only proved that they can build a taxi app (the technology that’s very easy to replicate).
- The network effect/brand loyalty. From what I can tell, this just doesn’t really apply much to taxi services. All the available evidence has shown that consumers pick the cheapest service that can keep waiting times below 3 minutes (i.e. there’s no extra network effect once your company has a certain number of cars in a given city). If I recall correctly the driver’s salary is about 85% the cost of a taxi ride; while ridesharing companies can cut this cost a bit, the reality is that a company with good driverless car tech could easily undercut ridesharing companies by a ridiculous margin while still remaining profitable. In terms of reaching the number of cars needed to keep waiting times below 3 minutes, a large company like Google can just buy the cars needed to achieve this in a given city, take over the market in that city, and then move onto the next one (this is a bit different to something like Facebook where there’s a global network that is harder to disrupt unless you can convince a large number of people all over the world to move to your service at the same time).
Of course, there’s a reasonable chance that driverless cars will prove to be infeasible with current technology. But then the main “nobody will own their own car any more” value proposition for these companies is gone. And furthermore, due to the lack of proper network effects that I mentioned earlier, it’s unlikely that these companies will be able to make much profit at all (since someone else will come in the moment that they try to jack up prices). Hence again I can’t see why these companies have such high valuations.
What am I missing here? Why are the valuations for these companies so high?
Disclaimer: Consult your financial professional before making any investment decision.