So comfortable and regulated are we in the formal world that we have fallen into the foolish belief that all networks must be the result of organized and conscious creation, when in fact the genius of markets lies in their ability to use the dynamics of cellular automata – hundreds, thousands, or even millions of isolated individual decisions, each one optimizing based on local phenomena – to create and maintain a robust, ever-changing, and efficient system, such as that of the Nairobi matatus, whose network was made visible by a project referenced in The Atlantic, Cities (February 3, 2014):
This Is What Informal Transit Looks Like When You Actually Map It
Even without captions, you can find the city center
Even without captions or context, the map is revealing.
To begin with, it’s trivial to find the city center – where all the bus routes cluster, and where there are huge transfer junctions (the while ovoid rectangles in the center). And indeed, in such spots the matatus do gather as if they were ants in a hive.
As transit systems go, the “matatus” in Nairobi exist somewhere between underground gypsy cabs and MTA bus service. The minibuses themselves aren’t owned by any government agency.
Why should they be?
Not owned by the government, are they?
The fares aren’t regulated by the city.
Why should they be? If people don’t want to use the buses, they don’t have to – and if the bus drivers can’t stay in business, they won’t. Regulated fares might be necessary if one operator held a monopoly (unlikely in the riot of commerce that is an emerging-world city), or if the system had embedded public subsidies so that market fares would result in a government-financed windfall for the lucky license holders
The economics work because the rides are full
The routes are vaguely based on a bus network that existed in Nairobi some 30 years ago –
In other words, the government once maintained a bus network, then canceled the service as unaffordable, and private markets moved in to the void. That, gentle reader, is pro-poor market action.
– but they’ve since shifted and multiplied and expanded at the region’s edges.
Rather than asking what’s wrong with that, let us observer what’s right with this – a network is being filled in, operated, and maintained, by individual entrepreneurs who’ve made the capital investment to buy a matatu and who’ve shown the wherewithal to succeed in a pure Adam-Smith marketplace.
A man with invisible hands
This sounds like controlled chaos, although it more or less describes how transit works in much of the world outside of North America and Europe. But amid the 130 or so unregulated matatu lines in metro Nairobi, there’s an admirable logic. In the absence of a formal public transit system in Kenya’s capital, people have created a comprehensive – if imperfect – one on their own.
It even offers modern services like free wifi on board:
The back window panel advertises VumaOnline, enjoy free internet on board
A matatu driver on Route 45 in the northeast part of Nairobi may know next to nothing about the lines that service the other half of town.
Why should he? He needs to know the routes that intersect with his, and he needs to know the major transit hubs (nearly all of them in the downtown, where the jobs cluster), but beyond that, how does it profit him to know the rest of the network?
Distinctive decorations can help you spot a familiar bus
Even the selection of unique route numbers is a natural market-response of species identification. It’s in every bus’s interest that its route be distinguishable from other bus’s routes, and it’s in the interest of the buses as a swarm that their customers can distinguish them from one another, because each bus in the network does better when the network as a whole is thriving.
Not surprisingly, many passengers on board know little about them, either. Riders who navigate the matatu system rely on it in parts, using only the lines they know and the unofficial stops they’re sure actually exist.
In short, the matatu network is a living economic example of a cellular automaton – a game between riders and drivers where the individual goals of self-maximization lead, remarkably, to a meta-level optimization across the whole metropolitan area of Nairobi.
And now we know that it looks like this.
The network is irregular – some areas have higher concentrations than others. This is partly because cities are inherently messy, and partly because a network is never ‘perfect,’ it’s always evolving as the city which hosts it is likewise evolving.
The University of Nairobi, where our definition of sculpture is similarly evolving
Researchers and students at the University of Nairobi, the Center for Sustainable Urban Development at Columbia University, and the Civic Data Design Lab at MIT produced that map – and the underlying data behind it – after carrying cell phones and GPS devices along every route in the network.
Like so many maps, this one is a visual delight by itself, so I urge you to download it.
“We recognized that if there was going to be any kind of improvement of this system in Nairobi, then people would need to be able to see it and visualize it and speak about it as a system,” says Jacqueline Klopp, an associate research scholar at Center for Sustainable Urban Development.
Late last week, that map was unveiled to the public, the press, and the matatu drivers themselves in Nairobi, prompting headlines and revelations large and small from riders who’ve long used the system.
One student at the University of Nairobi who worked on the project took a look at the finished map and realized for the first time that he could travel across town by a better route.
Neat – and a demonstration of the value of open-source information.
Anyone else who eyeballs this map is bound to notice a larger pattern: the heavily centralized system no doubt complicates congestion in downtown Nairobi –
Kibera from space: probably the densest human habitation on earth
Kibera owes its existence to its great location, close to Nairobi’s downtown.
Not on formal maps, but on Google’s image
In fact, Kibera is not identified on the matatu map, it can be located by triangulation. It lies west and a little south of downtown Nairobi, between the Ngong Road and the Langata Road, just north of the “Nairobi dam” (which doesn’t contain any water) and east of the Ngong Road Forest.
Just northwest of the blue ovoid representing the Nairobi Dam
Zooming in on the map, we can find it:
The white space the network routed around
If you lived in eastern Kibera, you would walk out of the slum to the Langata Road, then take a 15 or 34 matatu (brown lines) to the Blue Sky/ T-Mall and connect to the rest of the city.
Kibera from the air: matatus cannot enter (there is no space)
If you lived in west Kibera, you’d want the Ayany/ Kibera matatu routes, 8 and 32 (coral lines) and thence into downtown.
– suggesting that there might be some ways to improve it.
Perhaps – or perhaps people go through downtown Nairobi because that’s where the jobs are, and the highways are.
Cities meet in their downtowns
Observe the linkage between needing to change the routes and needing to know where the routes could or should be changed — supply must always connect to demand.
As we saw with the proposed trans-Egyptian highway imagined by Professor Farouk el-Baz, one cannot simply pave a highway between two points and expect a city to grow around it. Nor can one expect commuters to change their journeys without buses to take them there. One might expect bus operators to try new routes, but they’ll do this cautiously, and will not know how to market, unless they have the evidence in front of them.
Another map, produced by Kenya Buzz
So the open-source information adds value and enables the system to change itself.
The project was also intended for the benefit of government officials, who distribute matatu licenses for specific routes, but who had long since given up on trying to plan the system. Now the researchers hope that will change.
“Look, these people have planned your system from below!’” Klopp says. “It is not as chaotic as people think it is. They have routes, they have numbers. There’s very, very regular stops that the city didn’t plan. I think it really helps people to see that there is this system that you can then improve on, that it’s not just a chaotic mess.”
Klopp believes in “seeing the system” from below.
Exactly! It’s a dynamic mess, and one responsive to the needs of perhaps a hundred thousand people a day.
The project also amassed a developer-friendly open dataset akin to the GTFS data produced by U.S. transit agencies and used by Google Maps and route-planning apps. Open data about informal transit networks is inherently different from its MTA cousin. Matatus abide by no set schedule. And the service itself is more likely to respond to the needs to riders, slightly diverting off-route or depositing passengers outside of designated stops.
Within reason, why not?
These are traits that transportation advocates might not want to eliminate from systems like the one in Nairobi, even as local governments try to leverage these informal networks into something that’s more, well, formal.
Route flexibility in response to changing dynamics could be quite striking.
In this case, for instance, the Nairobi government wants to adopt the above diagram as the city’s official matatu map.
A good idea – but of course the matatu network will change constantly.
A mat-eat-man society?
Perhaps the map should be updated periodically via an app For instance, by locating a GPS transponder in every matatu and feeding that information into a monthly refresh?
This project, though, illustrates that it’s possible to wrap your head around what looks like a scattered, unplanned transportation system.
And it speaks to the reality that transit is so essential to urban life that people will find a way to develop it even without government help.
(It says something about the Atlantic author’s experience that she is agog at a network emerging without the government to organize it.)
“When the government does not step in, these informal economies are developed to meet a certain need that the government should be taking care of,” says Sarah Williams, the director of the Civic Data Design Lab. “That’s exactly what’s happened here. And it’s fascinating to see, because it’s totally driven by need.”
Driven to find things driven by need: Sarah Williams