In recent days, since their PR troubles, there has been much discussion as to why Uber seems to be so aggressive. Reasons ranged from being inept, to the challenges of fighting politics against taxi regulations to a claim that Uber’s market has a ‘winner take all’ nature. It is this last one that is of particular interest because it suggests that Uber has to fight hard against competitors like Lyft or it will lose. It also suggests that Uber’s $20 billion odd valuation is based on beliefs that it will win, and win big.
I am not sure that this is really the case. Despite the name ‘Uber’ connoting, ‘one Uber to rule them all,’ the theory underlying the notion of winner take all is rather special and is far from being proven in cases like this.
So let’s review the ‘winner take all’ theory. Here is the logic of Ben Thompson who put forward this argument: suppose that Uber has the majority of riders that means that more drivers will serve Uber customers, those more drivers mean that Uber’s service level will improve (that is, there are more cars available, more quickly), that better service drives Uber to have even more riders and the cycle continue. However, what is a virtuous cycle for Uber, Thompson claims, is a vicious cycle for Lyft.
Said that way, the effects seem to make sense. But the simple story ignores price (although Thompson does discuss it). He argues that perhaps consumers will be ‘sticky’ in the sense that they will adopt one service and they will choose the service that offers the best trade-off for ease of getting a ride and the price of that ride. If the larger service is trying to grab the entire rider market, then it can match price with the smaller one and still win (because its ‘product’ is of higher quality). But there will come a point where that extra share isn’t worth it relative to their ability to price their product higher without losing ‘sticky’ customers who derive value from their greater driver availability. This suggests an end to that dynamic and probably an end well before Uber has 100 percent of the rider market.
Even accepting this, price is a double edge sword. The lower Uber’s price and the more other drivers are on Uber, the less incentive an additional driver as to join them rather than Lyft. To a driver, fewer competitors on the same platform means more profit. You’ll find a ride quicker and so there will be less downtime. The point here is that lots of drivers being available means lots of drivers being without rides and that is bad news for drivers.
So while there may be forces that could pull riders to coordinate on one platform, the very same forces must push drivers to move towards the smaller platform.
This assumes, of course, that drivers are sticky or exclusive to one platform. For taxis and limos, Uber is it for the moment. So what we are talking about is UberX versus Lyft. But, in the absence of either contractual exclusivity or incentive payments that give you exclusivity by proxy, a given driver has an incentive to be available on both networks and to accept the first ride that comes along. In this situation, whether a driver is available on UberX or Lyft is irrelevant in the aggregate. Both systems will have precisely the same availability. And in this case, size will not matter for consumers in terms of their choice of platform … only price. And on that score, neither platform can profit from having a lower price than the other as this will, in effect, reduce their pool of available drivers.
Thus, when you sort through the equilibrium effects of all this (as economists are want to do) we have the conclusion that Uber and Lyft aren’t fighting for drivers at all but are competing for customers. That is, of course, unless there are moves towards exclusivity or, for that matter, other things that make drivers more likely to favour one platform over the other. I suspect that is still to come.
I should also add that I am not at all sure Thompson’s customer ‘stickiness’ assumption is right. The one thing taxi apps have done is make signing up and using them easy. For instance, it is far easier for an Uber user to check Lyft if a ride isn’t available than it is for a Starbucks drinker to pop down to Dunkin Donuts when the queue is too long. Given that, as I argue above, availability may look the same on both platforms in the long-run, then the consumer’s decision will all come down to price.
My point here is that there is a ton of innovation going on. But fundamentally, the market is about the riders (demand) and drivers (supply) with the platforms are intermediaries. And this may well be a market where the edges win and the intermediaries can capture value by competing in the market and having better service quality and not by competing for the market and a long-term monopoly bottleneck that they hope to get with it.
10 Replies to “Is Uber really in a fight to the death?”
The post (and the referenced Thompson post) looks at a couple pieces of Uber’ operating/selling equation but I think misses the critical bottom-line: will a relatively mature/post-equilibrium Uber have significant competitive advantage over all competing taxi/livery/limo services? The question isn’t whether Uber can find profitable niches (i.e. wealthy/tech savvy customers in select large/dense cities), or whether it can continue to grow (with the generous pricing/commissions any company might use during startup). Why does anyone think the Uber+contractor business model will be sustainably better than the existing taxi/livery/limo services? On the cost side, the current model has clearcut advantages in terms of vehicle ownership, maintenance and insurance purchasing, and the costs of financing these things. Uber will have higher labor cost for at least the medium-term, and cannot grow unless drivers are convinced they will make significantly more with Uber, after accounting for all their capital costs/risks and direct operating costs. Above a minimum scale existing companies would seem to have a strong advantage tailoring capacity to local market demand, and ensuring very high service reliability. If Uber provides a premium product (bigger/nicer cars, better trained drivers) this would come at a premium cost. Nothing in the Uber business model (i.e. smartphone apps) could possibly overcome these cost/service issues and create sustainable advantage. Where exactly is the “ton of innovation”? If it can be readily copied, and doesn’t clearly drive a major cost or service quality advantage, it is not important to any discussion of longer term competition in the car service market. For a more detailed discussion of these issues see http://talkingpointsmemo.com/edblog/understanding-the-economics-of-uber
If the Uber+contractor model had the clear potential to produce much higher quality service at much lower price, the company would have no need to be investing huge PR sums into vilifying and intimidating every competitor, regulator and journalist in sight. If Uber had a compelling quality/efficiency story it would be utterly irrational to hide that story in order to better communicate that the company is led by arrogant jerks who enjoy turning potential supporters into committed enemies. Again, the question of whether Uber or its drivers are making money in San Francisco today is utterly irrelevant. All of the $30 billion (or whatever) valuation discussions depend on Uber capturing a massive share of the revenue existing companies earn, and driving much of their capacity out of the market. But there’s no evidence that future Big Uber will be able to produce car service more efficiently than current producers do. And if that’s true, several logical hypotheses would explain Uber’s hyper-aggressive hostility. The “appearance of inevitable success” certainly helps all of their financing activities (and those have the most direct bearing on the owner’s hope of achieving spectacular wealth). Like many other startups, Uber might be happy to forgo profits for years in the hope of achieving a position where it could then exploit pricing power and more leverage over its drivers. The ability to intimidate local regulators and journalists would certainly help near-term cash flow.
Everything boils down to the question of whether the future Big Uber model has clear potential to overwhelm existing competitors because of sustainably lower costs and superior service. If they do, then Uber is not in a fight to the death, because it will “win” naturally, as passengers and drivers willingly choose them over higher cost/lower quality alternatives, who have no ability to match Uber’s superior efficiency. If Uber has clearly superior productivity, then one must also accept that much of its recent actions have been incredibly wasteful and self-destructive. If Uber has no competitive advantage, then yes, they must pursue scorched-earth, no-hold-barred warfare against anyone and everyone who might stand in the way of the megagrowth that a $30 billion valuation requires. If Uber has no competitive advantage, then the huge sums it plans to spend to drive existing players into bankruptcy will destroy massive economic value. I have looked high and wide for independent evidence of sustainable Uber competitive advantage. All I can find are vague PR claims or expressions of tribal/ideological affinity with the views of Uber managers. Have you found any such evidence? Can you share it?
“But there will come a point where that extra share isn’t worth it relative to their ability to price their product higher without losing ‘sticky’ customers who derive value from their greater driver availability. This suggests an end to that dynamic and probably an end well before Uber has 100 percent of the rider market.”
I think the question is whether that’s actually true. First, part of the reason why monopolies can pursue price matching is because they may lose profit on the short term, but they’re large enough to absorb that to ensure market dominance, which then allows them to control price over the long term. Second, it’s entirely possible that as you get extra shares your availability continues to go up. After all, what drives the scale effect here is that more riders then leads to more drivers, which then leads to more riders. After all, when we say “level of service goes up”, that level of service is availability, which is directly driven by the addressable market of riders, which is an endogenous variable, and not some other exogenous “service” variable.
“Even accepting this, price is a double edge sword. The lower Uber’s price and the more other drivers are on Uber, the less incentive an additional driver as to join them rather than Lyft. To a driver, fewer competitors on the same platform means more profit. You’ll find a ride quicker and so there will be less downtime. The point here is that lots of drivers being available means lots of drivers being without rides and that is bad news for drivers.”
The above paragraph misses that as Uber’s price lowers, the number of riders should also increase, so that you don’t have excess capacity (though that also depends on elasticity). Furthermore, because the market isn’t Uber or Lyft specific, both Uber and Lyft may experience excess capacity if demand is tapped out, which then favours the bigger player. Also, fewer drivers on the same platform may mean fewer competitors, but because the market is fundamentally driven by ridership demand, and because fewer drivers also impacts service and costs (costs in terms of greater distance to pick up drivers, and service mainly in terms of availability), there is a countervailing force that prevents this from being an advantage and actually turns into a disadvantage.
The market is driven by both drivers and riders. Slash prices and you’ll get loads of riders but no drivers. Hike prices and you’ll see the reverse.
My thought is that this cuts to the heart of the article and why it is true. This is also the reason why Amazon doesn’t make a profit.
Just as any fool can drive a car, anyone can sell books/gramophone records/eBooks online. If Amazon or Uber or Lyft try to make significant profits by raising prices they’ll find someone will undercut them. If they do it by squeezing suppliers too hard they’ll find suppliers will go elsewhere. In the eEconomy there are no super profits to be made retailing stuff that can be bought elsewhere.
That’s how it normally works, but not in a situation where there’s a network effect.
And Amazon doesn’t make a profit because it deliberately reinvests any profit into business development.
And not in a market that hasn’t tapped out demand. It’s an elasticity problem where if the increase in demand from a price cut creates new customers supply actually goes up to meet it.
To put it in econ 101 terms this is a situation where the effect of price isn’t a movement along the supply and demand curve but a shift in the supply and demand curves.
Josh, I think you misunderstand the way cutting price works, though I could be missing something. You can cut price and still keep the driver revenue flat. Say uber takes 20% of each ride cost normally. Then say to drive out lyft they want to cut prices. They can cut their take to zero (or to negative) and the driver still makes the same money and is happy. But the customer shifts to the 20% lower price. Do this until Lyft is bankrupt. You have a monopoly. Because they have a tremendous cash/fundraising/valuation advantage they can afford to do this until Lyft bleeds dry trying to match them on price.
Then, once Uber has a monopoly, seems easy to keep. You can wield this tool, and many others, until no one funds new entrants.