This story is over 5 years old.

Why Elon Musk’s Tunnel System Can’t Solve Congestion in LA

You can’t solve traffic congestion by putting more cars on the road.
Image: OnlineInnovation/Flickr

As a former urban studies grad student, there is something I find hard to accept about Elon Musk's worldview.

Of course, the man has a formidable mind and powerful business acumen, so it's not from any doubt that he knows about cars, or rockets, or tunnels. It's that something in his particular brand of technological solutionism—which shares much overlap with key elements with urban design, like transit systems and energy use—is so intent on disrupting the received wisdom on any subject that it ignores many of the plain truths we already know.


His latest venture, The Boring Company, is a case in point. In an origin story ripe for future dramatization, Musk was trapped in one of LA's notorious traffic jams and tweeted that he would build a tunnel boring machine and just start digging (perhaps the scene could be played by Michael Douglas?).

Just six months later—a huge credit to his ability to turn words into action—and Musk sits in a plush grey chair at the TED conference in Vancouver, detailing his early-stage experiments with tunnelling machines and unveiling a slick concept video for the tech press to fawn over. And here's where I dust off my town planning textbooks and scratch my head: Because it's not that the tunnel scheme itself is impossible (as Motherboard already wrote); it's that for a range of reasons, as planners understand, you can't just ease traffic congestion by building more roads.

The first and more esoteric of these is stated by Braess' paradox, named after mathemetician Dietrich Braess. Working in Germany in the late 1960's, what Braess observed was that in an already congested road network, construction of an additional road can actually increase average journey time for all drivers, and conversely, closing an existing road can reduce journey time overall. Though this outcome doesn't always occur, further mathematical modelling by two American academics concluded that as a general rule, the addition of a new road to a network had roughly 50/50 odds of increasing travel time.


The second reason is that the level of demand for a road network cannot be considered in isolation to the supply (i.e. the capacity of the roads). In traffic flow just as in economics, supply and demand tend to seek an equilibrium: After a certain point, traffic jams and slow roads disincentivize the use of cars for unnecessary journeys, and the number of vehicles in the system remains roughly constant. But this also means that there exists a "latent demand," people who would be more inclined to use their cars if the roads were better; and as long as this latent demand exists, building more roads will tempt more people to drive, making it difficult to ease congestion.

So the best way to improve traffic flow, very simply, is not to build more roads, but to have fewer vehicles on the roads. And the way to have fewer vehicles on the road is to encourage higher vehicle occupancy and invest money in public transport. Jarrett Walker, a consultant in public transit policy, made this point well in a blog post titled Does Elon Musk Understand Urban Geometry?, written in 2016 before The Boring Company was even conceived.

"Musk assumes that transit is an engineering problem, about vehicle design and technology," Walker writes. "In fact, providing cost-effective and liberating transportation in cities requires solving a geometry problem, and he's not even seeing it."

It's a geometry problem because for improving traffic flow, the key measure isn't one of speed, but space. It's a metric called Vehicle Miles Travelled—the amount of distance a vehicle must cover per passenger in order to get people to their destination—which is something that can't be reduced by putting more small vehicles onto the road. ("Technology never changes geometry," Walker concludes.)


Juan Matute, Associate Director of the Institute of Transportation Studies at UCLA, told Motherboard that he judged Musk's plan to be impractical for different reasons. In a phone call, Matute highlighted the difficulty of obtaining access rights to subsurface property: Since a property owner has rights to the land beneath their house, tunneling below private property requires what's known as an easement, the permission for a third party to operate in the underground portion of the land. If a private property owner refuses to grant an easement, a government actor can compel them to do so through the power of eminent domain, but a private company has no such right, meaning Musk would be forced to divert the tunnel route around any landowners who refused to cooperate.

"All of these practical considerations which happen as a matter of construction make it very difficult for a private entity to engage in this type of infrastructure building," Matute said. "And if it were to happen, the price would likely be very expensive, and it would be primarily for rich users of the transportation system."

As a more effective alternative, Matute suggested that adding congestion pricing to some parts of the freeway system—of the kind that has reduced both traffic and accidents in London—would be a way to tackle congestion at a much lower overall cost.

A house built by Brad Pitt's Make It Right Project in New Orleans. Image: Geoff Livingston/Flickr

But the problem with wealthy philanthropic geniuses like Musk is that they're usually unwilling to take a dull, practical, unsexy approach to solving a problem if there's a more glamorous way to attempt it, however ill-advised. It's why Brad Pitt and his architect friends descended on New Orleans to build a handful of cutting edge $350,000 homes for displaced residents instead of building cheaply for as many people as they could, or why Mark Zuckerberg's foray into international development was a plan to bring the internet to some of the poorest people in the world with a fleet of giant drones. And it's why Musk is more inclined to undertake the vastly complex and extremely expensive job of drilling a subterranean tunnel network than think seriously about how to get more people riding buses or less people using the freeway in the first place.

"I'm not trying to be anyone's saviour," says Musk in an eminently tweetable soundbite. "I'm just trying to think about the future and not be sad." It's an admirable sentiment, because it's all too easy to be filled with anxieties about the future given some of the pressing problems we face. But personally, what makes me sad isn't that we don't have the technology we need; it's that the direction of technological development is shaped more by the whims of billionaires than the wants of ordinary people.

As one example of this, I'm concerned that a man with the brains and the resources to consider drilling tunnels under LA would do so to make a fancy way of slinging cars around instead of trying to extend the subway system. And I'm concerned with a model that looks to improve the urban transportation system by focusing on seamless individual journeys (a ride from point A to point B without ever leaving your vehicle) rather than developing better fixed route networks for the collective good.

I'm concerned that, though he's a man of many talents, Elon Musk does not understand urban design.

Subscribe to Science Solved It, Motherboard's new show about the greatest mysteries that were solved by science.