New York City is the most urban city in America, with the largest subway network by far, and yet, even here, ridership levels have yet to recover to their pre-pandemic levels. Recent data shows subway ridership hovering between 70% and 80% of 2019 levels, and the MTA anticipates that it will remain "at about that level through 2029."

The obvious explanation is that office workers continue to work from home on occasion, and that's certainly a significant part of the story here. But it doesn't appear to be the entire story.

For example, looking at station ridership recovery across the city, there visually appears to be a geographic correlation with areas in Upper Manhattan, the Bronx, and the outer boroughs in general not recovering to the same extent as Manhattan.

In the early days of the pandemic, ridership levels were mostly correlated with median household incomes. Ridership remained higher in the outer boroughs, while residents in wealthier neighbourhoods simply worked from home. Since then, that correlation has weakened and the geography has inverted.

This suggests to me that in addition to WFH, there has also been a structural mobility shift for many households. We know that car registrations in NYC spiked during the pandemic, and presumably that means some new mobility habits were formed.

Cover photo by Igor Wang on Unsplash

Chart from Subway Recovery Tracker

These charts, from the School of Cities at the University of Toronto, are an excellent use case for a ternary chart. A ternary chart is a triangular plot with three different variables that all sum to a constant. In this case, the charts compare driving, walking/biking (active transport), and taking transit, with the constant being 100%. The data is from the 2022-23 Transportation Tomorrow Survey.

What is clear from the first image above is that Old Toronto (also known as the former City of Toronto before amalgamation) is unlike any other municipality in Ontario. Its tight, pre-car urban fabric and transit coverage mean that nearly 65% of people walk, bike, or take transit! This is compared to about 43% for the entirety of the city today; meaning, Old Toronto is the only place where driving isn't the majority mode.

The second chart above shows mode share by age. What is interesting to see, though not surprising, is that younger people are more likely to walk, bike, and take transit. Here, the figures peak between the ages of 15-19 with over 66% of trips falling into these categories, which is just slightly above the Old Toronto figure. Then, as people get older and have greater incomes (which is another one of the charts), they move toward driving.

Of course, this doesn't mean that all older people must drive. There are lots of older people who live in Old Toronto where doing something other than driving makes up the majority of trips. Urban form, density, and access to transit play the most important roles in determining what modes of transport people will choose and what they find most convenient.

Cover photo by Mitch Hodiono on Unsplash

Charts from the School of Cities

One generalized truism is that European cities are walkable and transit-supportive, and North American cities are not. This is not universally true, but it's often thought to be directionally true. However, a recent paper called "Car Dependency in Urban Accessibility" reveals that this may not be as true as we think.

The study introduces something called a Car Dependency Index (or CDI). What it effectively does is compare accessibility to jobs and services within a city by car versus public transit. They did this for 18 European and North American cities, and here's what they found:

A positive score (red on the map) means that a car can access more opportunities than public transportation, and a negative score (blue on the map) means the opposite. What's not surprising is how car-dependent the outskirts of most cities are, including European cities. Car dependency was high in over 70% of the urban territories that they analyzed.

What is more surprising to me is that most cities don't have much, if any, blue. The best-case scenario seems to be a lot of white (which represents accessibility parity between cars and public transit). Hmm. Does Manhattan really not have any blue? The glaring exception is Paris and, to a lesser extent, Zurich, though keep in mind these are only city proper boundaries.

Another finding is that car dependency remains a primary driver of car ownership, even when accounting for income. What this means is that if you took two people with the exact same income, one living in transit-rich Paris and the other living in the suburbs of Rome, the person in Rome is much more likely to own a car.

Once again, this supports the obvious fact that if we design cities so that they're inconvenient to navigate without a car, well, then more people will get cars. It's not easy to build a transit network that can compete. Individual lines won't do it. The key word is "network." And you need the right land-use policies to support it.

Cover photo by Alessio Ferretti on Unsplash

Charts from "Car Dependency in Urban Accessibility."