On reaching the natural endpoint of suburban development

Paul Krugman wrote a good post on Substack yesterday about how Atlanta is approaching the limits of sprawl. What I really want to nail down is how this is really a limitation imposed by road capacity more than anything else.  If you build your infrastructure and your cities with only drivers in mind, you reach capacity real fast.

The gold standard for how many people you can move in any given direction is PPHPD, which stands for "People Per Hour Per Direction," for one specific lane or track.  

These numbers are general figures, because the details matter.  (For example, Vancouver's SkyTrain maxes out around 18,000 PPHPD, using 250'/76m long trains, while the old Independent Subway in NYC was designed for 60,000 PPHPD, using 600'/185m trains.)

The math sounds ridiculous at first glance: how on earth can a single bus lane have more capacity than a four-lane freeway? But it really is so - it's just simple geometry.  A picture is worth a thousand words.

Now, think of Atlanta for a second, which is pretty much endless suburbs as far as the eye can see, and compare it to, say, Barcelona.  (The map here is from 35 years ago, where traffic was a huge problem already!)  And you can't densify those suburbs because of the zoning laws.

Pre-demic, over 20% of Atlanta's jobs were located in 0.7% of the land area, and it doesn't take a genius to figure out that everyone having to drive to the concentrations of jobs is a recipe for gridlock. If you rely only on the private car, there is no way short of massive urban demolition, to add more roadway capacity if you want a non-lousy commute. 

("Unaffordable" areas are orange, and the jobs-rich areas are in black.)

I suspect it's not just Atlanta that's going to have this issue going forward, because Atlanta's land use and transport patterns are so, well, normal for the United States, especially cities that came of age after the Second World War.  Something has to change if we want to fix this.