Does transportation serve people or do people serve transportation?

No big city has solved either of two problems: making car traffic flow smoothly or making parking simple and cheap. The issue isn’t that governments in every city are bad, though sometimes they are, or that traffic engineers didn’t anticipate the future, though sometimes they didn’t, or that drivers drive too aggressively, though sometimes they do. The issue is that these problems are impossible to solve due to basic geometry. You can fit more people close together than you can fit cars close together. In a small town, you can fit all the people and the cars together without much difficulty. But as a city grows larger, it’s easier to accommodate people, which are relatively small, compared to cars, which are relatively large.

There are a million ways to think about this that all come to the same conclusion. Caleb Pritchard has pointed out that there is more space downtown devoted to parking cars in downtown Austin than there is space in all of downtown (some of the parking spaces, of course, are stacked in multi-level garages, so the math works):

New office buildings in Austin devote as much space to storing cars as they do to storing people.

This is ugly and wasteful and terrible for the environment and extraordinarily expensive. It requires people to spend a long time and sometimes a lot of money circling up floors or circling city streets to find parking spaces and people hate the process with a fiery passion. People’s (un)willingness to circle up floors to park is the binding factor limiting the size and creativity of downtown office buildings in Austin. Car parking is the reason our office buildings are shorter than our residential towers — illustrated nicely by how the Sullivan’s Tower got shorter when it was converted from apartments to office. Car parking is also the reason our office towers are shorter than office buildings in less car-oriented cities like, say, New York or Houston. But despite all that, it kind of works. Every work day, 100,000 people find a place to store 2,000 lbs of metal within a square mile of land, so that they can be close enough together to hold a meeting in the conference room or walk to the courthouse or the state house or the post office or city hall or each other’s offices or any other places they need to go as part of their work day.

As bad as the parking situation can get, it’s good news compared to figuring out the logistics of moving them around the city. We can make many more multilevel parking garages in downtown Austin before we run out of space, but stacking cars on top of one another when they’re driving (by building tunnels or elevated streets) is prohibitively expensive for more than a few major highways. Even then, the best you can hope for is two or maybe three levels that people hate to drive on, look at, or live near. You can widen streets to handle more cars, but the more space you use for streets, the less space you have for the places people are using those streets to go.

Another option is to spread buildings out. Instead of a huge chunk of people coming together to work downtown, make many different employment nodes around the city. This can work, to some degree. But the reasons people needed to be in close proximity to one another don’t go away, so the cure is often worse than the problem. Instead of finding a way to get people who need to be in proximity with one another into a small walkable area, you require them to drive half an hour to get to any meeting they have, creating even more traffic and making things even worse!

There’s simply no solution that allows people to get wherever they want in a big city in a car without facing traffic along the way. Implicitly, when we develop new buildings, our development rules recognize this fact. New buildings are required to perform a “Traffic Impact Analysis” which uses arcane, inaccurate, and context-free rules to model how many times per day somebody will arrive at or leave a building, depending on what kind of things people do inside the building. Rules which require a TIA often limit the number of so-called “trips generated” for a particular development.

The basic idea behind trip generation limits is the understanding that big cities and cars are fundamentally incompatible. Rather than limiting the use of cars, limits on “trips generated” are a vain attempt to limit a city from growing too large to outgrow the usefulness of cars. The reason why this attempt is in vain should be apparent: trips aren’t generated by buildings, but by needs. Sprawling trips out over a bigger area merely forces people to drive even further, creating even more traffic.

Of course, just because cars and big cities have fundamental conflicts doesn’t mean cities can’t continue to outgrow the constraints of car-oriented development. There are countless solutions: buses, trains, bikes, scooters, sidewalks, dollar vans, etc. The details of each solution vary but the overarching idea is very similar: 1) make trips shorter so people don’t need to travel as far to get to the places they want and need to go, and 2) fit people closer together while they’re making trips to make more efficient use of each street.

There have been some positive ideas on reforming TIAs to use these solutions, California being the best example in the US. But I’d like to propose a different way of thinking what a TIA could be. Instead of assuming cities are tied to cars only and then limiting developments to the constraints that cars have, we could start with the development we want and then require that this developments accommodate whatever transportation system is necessary to complement it. Instead of trying and failing to scale cars to a level that they simply aren’t capable of handling, we find the transportation system that is actually capable of handling the development.

For example, if a developer proposes a development whose size or density tips past the point where most trips can be comfortably done with cars, she would be responsible for providing street designs that include different ways of getting around: bike lanes and bike corrals, or designated places to store dockless scooters, or proximity to a transit line.

There is a sense in which we do this already. Austin’s land development code, like many others, requires infrastructure for cars (mostly parking) but allows developers to provide less of it if they provide alternative transportation infrastructure (e.g. parking spaces designated for shared cars like Car2Go, showers for bicyclists). But these efforts are piecemeal and complicated. All too often, TIAs are wielded as weapons to prevent homes and offices, when instead, we could make them a tool to make any development work.