An Economist's Dream
Steve Levitt captured my sentiment precisely in his recent Freakonomics episode "Why Uber is an Economist's Dream" when he commented about the demand curve, "I've been dreaming of the day I could answer this question, and it probably says a lot about me."
A little background: Levitt, along with Peter Cohen, Robert Hahn, Jonathan Hall and Robert Metcalfe, used consumer reaction to Uber's surge pricing data to estimate a demand curve and the resulting consumer surplus. Uber's data makes this calculation easy to conduct since a consumer's willingness to pay can be tested for almost the exact same product (getting a ride) at many different price levels. They estimated that for every dollar that someone spends, he would have happily spent another $1.60, that $1.60 being the consumer surplus. The economists contend that "back-of-the-envelope calculations suggest that overall consumer surplus generated by the UberX service in the United States in 2015 was $6.8 billion"! Wow, people are willing to pay a LOT to get a ride!
Another dream come true would be figuring out at what price the drivers would have driven at compared with what they actually got paid, or what is known in economic theory as the producer surplus. In order to know the producer surplus, the supply curve itself would have to be estimated by figuring out how many drivers would be willing to drive at every given price. The upward sloping supply curve transpires because the number of drivers willing to drive increases at each price level.
Ride-sharing apps in reality do espouse many virtues of perfect economics experiments. They are cross-sided or two-sided marketplaces where each side requires different functionality from the platform: the driver requires a rider and a rider requires a driver with a vehicle. Ride-sharing also has merits that other two-sided marketplaces lack: near perfect competition. Although Uber clearly dominates the ride-sharing apps today, I think there is a necessity for multiple apps due to the unique dynamics of this two-sided marketplace and the propinquity of the industry to a perfectly competitive market.
Two-Sided Marketplaces: No study of two-sided marketplaces is complete without revisiting the war between VHS and Betamax. This famous incident has the makings of everything needed for a "winner take all" landscape: people weren't going to buy multiple players or videotapes in both formats before one became the clear standard, and then, manufacturers began choosing sides. In the end, JVC won with its VHS design. Will there be a single winner in the ride-sharing market the way there was in the videotape market? I don't think so, and here's why.
In the article called "Strategies for Two-Sided Markets", Eisenmann, Parker and Van Alstyne proffer that a networked market is likely to be served by a single platform when the following three conditions apply:
- Multi-homing costs are high for at least one-user side. "Homing" costs comprise of all expenses network users incur-including adoption, operation, and the opportunity cost of time, in order to establish and maintain platforms affiliation.
- Network effects are positive and strong, at least for the users on the side with high multi-homing costs.
- Neither side's users have strong preference for special features.
Ride-sharing apps fail test #1. Multi-homing costs are the cost that consumers had of having two different players for videotapes, and that just doesn't exist with ride-sharing. It's easy for riders to have both Lyft and Uber on their phones and to check both for time-until-ride and price. There's no cost of having both apps on the phone and very little opportunity cost of time spent checking both apps. Multi-homing costs aren't even an issue for drivers, as many drivers drive for both companies.
Network effects are strong and positive for both sides and neither side really have a preference for special features, but given that it's really easy to download and compare trips from both apps or to drive for both companies, I don't think ride-sharing is inherently a winner-take-all industry.
Near Perfect Competition: Another reason that ride-sharing apps are such a great case study is that the service of ridesharing is almost (but not quite) a perfectly competitive market. In economic theory, a perfectly competitive market has the following characteristics:
- No barriers to entry: drivers can decide to enter and exit the market as he pleases without too much trouble. Because of this, there are too many drivers in the market to measure.
- No single firm (in this case, driver) can influence the market price or condition. Each driver is a price taker, and price is determined based on supply and demand.
- Homogenous, almost identical output: one ride may be better than another, but main output is to get from point A to point B. (The slight difference in each experience actually does matter, as discussed below).
- There is perfect knowledge with no information failure or time lags. The ride-sharing app platforms make the transmission of information much easier than in the non-virtual world.
Ride-sharing ought to be a perfectly competitive market if all information is available. That is, absent regulation, no one entity ought to be able to set a price, given that there are lots of drivers who provide a seemingly indistinguishable service. One would think prices would naturally be set dynamically where supply and demand meet.
In practice, figuring out the demand curve is difficult, as discussed above in reference to the paper that Levitt co-authored. You may accept a ride at $5.00 but not at $5.50, but it's difficult for Uber to know that without asking you about both of those prices. And this is the reason the market will support at least one more competitor to Uber, to ensure that there is a price checking mechanism for a commodity service. Without Lyft or other competitors, would Uber have created the consumer surplus that it did in 2015?
Another aspect that contributes to the likelihood of multiple ride-sharing apps existing is that, unlike in a theoretical perfect competition, each ride is a little bit different from the one before. The variance in each individual experience results in much more uncertainty about customer satisfaction about the output than if the output were a consistent product, like Campbell's soup. It only takes one bad ride for a rider to desire another option for an app. Uber tries to solve for this with the drivers' ratings system to ensure a consistently pleasant experience. That's not fool-proof (the driver can just have a bad day), however, and it doesn't solve for the negative experience being exogenous to the driver.
The price and treatment checking will happen for drivers too. Anecdotal data suggests that more drivers prefer Lyft to Uber for various reasons, and the normal ebb-and-flow of business policies will cause either one of the apps to be the darling at any given time. Regardless, I think the people on both sides of the ride-sharing equation like having the choice, and that choice serves as a bit of a price-check on the market.
What's interesting is that obviously Uber the platform itself is in nothing close to a perfectly competitive market: barriers to entry are enormous, which is why the company is so successful. But it is facilitating transactions that, if done correctly, would buoy the perfect competition of the ridesharing world. It could eschew the need for price-checking by competitors if it could figure out how to let the market set the price dynamically based on supply and demand; that is, if somehow, each ride could be priced by the rider instead of the company itself. This still doesn't solve for riders just wanting the second choice in the case of disenchantment with one of the platforms.
Ben Thompson Disagrees: Since I get so much inspiration from Stratechery, I'd be remiss if I didn't mention that Ben Thompson kind of disagrees with me. Thompson argues that since the number of riders is far greater than the number of drivers, and as of now, Lyft has fewer number of riders (less demand), drivers will be too busy serving Uber customers, which will lead to a winner-take-all dynamic.
He explains further, "It doesn’t matter that drivers may work for both Uber and Lyft. If the majority of the ride requests are coming from Uber, they are going to be taking a significantly greater percentage of driver time, and every minute a driver spends on a rider job is a minute that driver is unavailable to the other service. Moreover, this monopolization of driver time accelerates as one platform becomes ever more popular with riders. Unless there is a massive supply of drivers, it is very difficult for the 2nd-place car service to ever get its liquidity to the same level as the market leader (much less the 3rd or 4th entrants in a market)."
Thompson also claims that people build allegiances to a brand and persist with that brand, unless they are given a reason to change; it's simply not worth the time and effort to constantly compare services at the moment of purchase, and that Uber and Lyft would ensure that their prices are pretty similar anyway.
To his first point about drivers, I think people act irrationally, and that explains why drivers drive for Lyft and other apps when liquidity might be the best at Uber. Just check out the reasons people give for enjoying Lyft; there are people who will drive with the less-busy app because they like it that way, so I don't think the liquidity of drivers, at least in large cities, will be an issue.
To the second point about riders having allegiances, again, I think people are fickle. One bad experience, which is easy to have in ride-sharing, and they will open up the "other" app for the next ride. Because unlike a can of soup by Campbell, each ride by Uber actually isn't exactly the same.
Also, people do gravitate towards their niches, especially in dense areas, so ride-sharing apps with a twist could also do well. A pet-friendly car service? Paw-fect! A neighborhood service for hauling kids by a certified parent? It takes a village to drive a child, right? And I'm sure there's already lots of cannabis-friendly ride-sharing schemes being conjured up in a few of the states.
A Note on Financing: According to Crunchbase, Lyft has raised $2B to date while Uber has raised $8.7B. The thesis of having two successful ride-sharing apps is predicated upon having two apps with pretty good brand awareness and customer acquisition ability. And to do those two, you need financing. Lyft, the #2 player has been able to pay for its customer acquisition costs thus far, but if the financing market dries up and the company has challenges raising its next round, it will have trouble keeping up with its customer acquisition goals. A stalled financing market is bad for both companies, of course, but especially bad for Lyft, since it's trying to play catch-up. A prolonged tightness in the fundraising market, therefore, would end up aiding Uber in retaining and expanding its market dominance.
Update: We're seeing financing woes play out for Ola, India's Uber rival, as reports indicate that the Indian ride-sharing app may settle for equity financing at a 40% lower valuation. Still, the company would be valued at $3 billion and is planning on raising approximately $600MM, which would still give it enough ammo to acquire customers. Until funding dries up entirely, I don't think we'll see any of Uber's competitors shutting down.
