This post, which points out that transit can’t be judged on the low ridership of services where ridership isn’t the goal, drew this question from Rob:
I have a question about cities with weak urban cores. I don’t know much about Seattle, but the story that the numbers tell is that the city’s population is currently near its historical high. But what do you do in cities that are losing population, like many in the rust belt? In my hometown, Cleveland, the population is lower than it’s been since 1900. Many urban neighborhoods are no longer the densest areas (there are 3 inner-ring suburbs more dense than the city-proper). What do you think?
As I explained, if Seattle’s King County Metro were pursuing a pure ridership objective, it would cut almost all service in the low-density suburbs and put all those buses in Seattle as higher frequencies on dense corridors. The principle is the same in any network:
If your goal is ridership, follow patterns of dense development with intense service.
This principle explains why San Francisco has more than half of the Bay Area’s transit ridership despite having barely a tenth of its population. It explains why, if you rank any transit agency’s services by productivity (passengers per unit of service cost), the high-frequency lines in the densest part of the city always come out on top.
The answer to Rob is that different cities may have different patterns of density, so in your city the highest-ridership area may not be the historic core. On the other hand, even substantially neglected cores trigger strong ridership because even if density is lower than it could be, it’s often still higher than in the suburbs.
When I started working at Portland’s Tri-Met in 1983, the most productive bus routes were already all in the inner city, in areas that would look pretty neglected to you if you visited them via time machine from the gentrified Portland of today. Even when flight-to-the-suburbs was at its maximum, there were lots of people left in the inner city, and their relative poverty was only part of why they were still on the bus system. It was also because the bus system worked well for them, and worked well in the kind of geography they lived in.
(If anyone can find productivity-by-line statistics for some “neglected-core” cities like Detroit or Cleveland, I’d love to see them. But I bet you’ll find that even in such places, there are a lot of transit riders left in areas that look abandoned to you, and that inner-city service is still performing pretty well.)
This is also an example where “design” comes in. If you press on the “follow patterns of dense development” rule, what’s under it is the principle of a radius of demand. A transit stop’s market is the area that’s within a fixed walk distance radius. Since it’s a fixed radius, it’s a fixed area, so the number of people and jobs and activities there is determined by density. However, some people may be within the fixed radius but not able to walk to the stop because of barriers in the street network; such barriers are much more common in newer suburbs than in old core cities. The fully connected grid street networks of old urban cores generally minimize walking distances and thus maximize the real radius of demand.