Get email delivery of the Cadence blog featured here
In the last couple of years, one of the biggest changes in China has been the arrival of bike sharing everywhere. At one level, you can think of it as Uber for bicycles: there's a smartphone app, you get a bike, you use it, you get charged through the app.
Let me first explain how it works, and how the companies doing this work. Then we can look and see whether there are any lessons we can learn about how self-driving car sharing might work. There's a real sense in which bike sharing is more like self-driving car sharing than Uber today, since it doesn't require a driver (well, of course, on the bike, you are the driver but you don't need anyone to show up or to be getting paid).
To have the full experience, your Breakfast Bytes blogger rented a Chinese bike a few weeks ago (proof on left).
It is worth understanding how bike sharing works to have an idea if and how it will work in the US. All over the cities there are bikes everywhere, primarily from two companies called Ofo (yellow) and Mobike (orange and grey). Ofo is backed by Alibaba, and Mobike by Tencent. Ofo is not just in China, it is in 20 other countries (along with over 250 cities in China). In total, the two companies have raised over a billion dollars in funding.
Each bike has GPS (so it knows where it is) and it uses narrowband IoT to communicate back to base. The batteries for this don't need replacing since they are charged when someone rides the bike. The bikes are "dockless" in that you pick them up where they are, and can leave them anywhere. You don't have to go to a docking station to get a bike (and there aren't anyway), nor to return it to one.
To rent a bike, you can look at the app and see where bikes are in your area, but I spent one day when I was out and about and I never managed to find a moment when I couldn't see a bike within 50 yards. When you get the app, you put some money into it which is partially a deposit and is used to pay the rental fees. I think that there are other mechanisms that link it to electronic payment systems. When you want to actually use a bike, you simply point your camera at the barcode on the back of the bike, the app recognizes it, and sends you a code. You type the code on the keypad (to the left of the barcode in the picture to the right) and a moment later there is a beep and the lock opens. You ride off. When you are finished, you slide the lock knob over, which pushes a metal rod through the back wheel, immobilizing the bike, and also, since it is spring loaded, setting it up ready to spring open. Of course it also signals back to the network that your ride is over. The cost seems to vary (since there are introductory rates) but around 1¥ per 30 minutes, so even more roughly 25¢ (US) per hour.
In total there are supposedly 12M bikes and 50M rides per day. I'm not sure if that is just China or the other countries, which are on a smaller scale anyway, so there may not be that big a difference.
One big difference from Uber, apart from the obvious one that bikes are not cars, is that the companies own all the bikes. It is not a system for people who already have bikes to rent them out. This makes the business fairly capital intensive. Obviously, the bikes have to be purchased before they can be rented out. Even with the scale of millions of bikes, the rental fees of 2¥ per hour takes a long time to pay off. You certainly see lots of people riding the bikes, and bike shops are complaining that they are going out of business because nobody wants to own their own bike anymore. But, as the pictures through this post make clear, there are large numbers of available bikes everywhere. I took all the pictures in Guangzhou (fka Canton, and still with CAN as its airport code).
One red flag about the business, or rather a blue one, is that there used to be a third company called Bluegogo, but it went bankrupt. There seems to have been something of a Ponzi scheme about it since there are widespread complaints of people not getting their deposits back. In the initial fast growth phase, the paid-in deposits can be used as working capital, but that only works while new users are joining.
Another challenge is that there are no network effects and so the barriers to entry are low. The only advantage of using Ofo over a new entrant is that you don't need to download another app and set up a new deposit. Plus, it's not like you care if you use the same color bikes as your friends, so there is no Facebook-type effect.
Don't just take my word for it. Here's an optimistic take from last year Bike-sharing boom in China pedals to new heights. Or for a different take Wait, Chinese Bike-Sharing Doesn't Make Any Sense.
I'm just going to look at what's relevant for comparing bike businesses in the two countries. For example, they speak Chinese in China, but that doesn't seem to be significant here. However, two things I think are very significant:
Obviously, bikes are not cars, but it is worth listing the relevant differences for this business:
The current bike business is very different from today's Uber and Lyft since the current business is built around drivers, who also provide and maintain their own vehicles. In the future, Uber and Lyft (and Waymo and Ford...) will probably own their own vehicles. This makes the business a lot less capital efficient than the current model. It is a big change though, making the business one-sided instead of two-sided. The companies just need to find passengers, and don't face the issue of having to recruit both drivers (by having lots of passengers) and passengers (by having lots of drivers). That sort of two-sided market makes it hard for new entrants (and even old entrants: the original company in the space, Sidecar, was driven from the market and now just holds a lot of patents that might be very valuable).
But a one-sided market also means that the barriers to entry are low, so there are likely to be more suppliers. There is no particular reason not to install apps for all of them and pick the one that can come the fastest, or offers the cheapest price. One prediction: somebody will make an app that compares all the prices and summons the best car for you, and the Ubers and Lyfts will find a way to shut it down. Another prediction: at least some of the suppliers will go to a subscription model (so much a month for all your trips, or pay $200 and get $300 of trips).
There has been a lot of docked bike sharing in the US. The business model has been to find a city like San Francisco, get them to pay you to put the docking stations in place and rent the bikes out. It is more like building a light rail line, where the construction companies find a city who wants one built for political reasons (customers are never demanding them), makes all their money up front, and doesn't care much whether anyone uses it afterwards since that's not their business. OK, I'm a bit cynical. But the most topsy-turvy situation is the silver line in Washington DC which runs alongside the freeway from Dulles Airport. The silver line is largely paid for by tolls on the freeway, meaning that each person driving on the freeway (and thus not using the train) is paying more per person for the train than the people who are actually using it.
There seem to be dockless systems being set up in Dallas and LA. Here's a positive a16z podcast with council-members and a VP from LimeBike (a16z's entrant in the space, and green bikes I'm guessing). And if a16z like them, that's enough to get full coverage in Wired: The Bike Share Wars Heat Up with the Latest Funding.
Or if you want the contrarian view, read Dallas Bike Sharing Companies Are Doomed which wonders who all LimeBike's customers are. Dallas is not a major tourist destination, doesn't have a subway, and everyone already has a car (and a bike).
I'm very dubious about the business economics. Of course, the prices in the US can be higher than in China. If someone will rent a bike for 25¢/hour they will probably do it for $1/hour. The bigger challenge is to find a customer who wants a bike, and doesn't have one of their own already. The economics seem dubious in China, although it is hard to tell. In the early phase of this type of market, profit is a waste, it is better to plough it into even faster expansion to lock up all the markets before someone else does. The economics of building out the cable network in the US was like this, where it was more important to get another city cabled (forever) than make some profit now.
However, there aren't really any economies of scale. For every million more users, you need a certain number more bikes, and that number doesn't go down. The cost of the bikes may go down a little, but when you are already buying bikes by the million, it doesn't seem like there is likely to be a lot more fat to cut out. Furthermore, as you get outside the very dense central parts of cities, you need people to go and fetch the bikes from places where they are likely to languish, or where they are piling up in ways that are causing obstructions, or are unlikely to be rented since there is too much supply in one place.
Sign up for Sunday Brunch, the weekly Breakfast Bytes email.