Nationwide across the US, transit ridership is only at about 70% of where it was in 2019 before the pandemic. But this is not the case in all cities around the world. According to this recent Bloomberg article, Madrid, Hong Kong, and Paris are all above their 2019 ridership levels. Seoul and Shanghai are also close at just over 90%, and London is at 85%.
So this problem of fewer people riding transit seems to be a North and South American phenomenon. Rio de Janeiro is at 73%, Mexico City is at 70%, and San Francisco is somewhere near or at the bottom at 44%. The obvious explanations for this are that Europe and Asia are generally denser and less car-oriented, their return-to-office patterns have been much stronger (less WFH), and their governments probably care more about transit (and spend more money on it).
Broadly speaking, I think this is all true, but I'd love to know more precisely what's driving these differences. Because it's not exactly obvious. Consider, for example, Paris and London. Paris is at 103% of its 2019 levels, whereas London is only at 85%. Why is that? Both cities share a lot of similarities. They have a river that weaves through the middle, they're dense, they have lots of trains, and both are alpha global cities.
So why the delta? What exactly is Paris doing that is encouraging more transit usage?
Charts via Bloomberg
I spent three years living in Philadelphia for grad school and one of the things that I appreciated the most was its walkability. I walked and took transit everywhere. Much of this has to do with the grid system that was laid out for the city in the 17th century. But there are also lots of more recent developments that help to reinforce this fabric.
CityLab, for example, just published this article on Penn's Landing Square, which is a housing complex in Philadelphia's Society Hill neighborhood. Built in 1970 and designed by Canadian-American architect Louis Sauer, the modernist complex occupies an entire 2.37-acre block and contains an assortment of 118 low-rise homes, many of which are connected through small interior laneways.
In addition to its handsome architecture, what is noteworthy about Penn's Landing Square is that its site plan makes it quite a dense low-rise development. At 118 homes, this translates into just under 50 units per acre. CityLab estimates that this means the development holds about 174 people per acre (~412 people total), which would make it more dense than Stuyvesant Town in New York (~158 persons per acre).
However, this is based on the assumption that there are almost 3.5 people living in each of these homes. While generally large, I don't know if this is the case. It would be higher than the average US household size. But regardless, from a unit per acre standpoint, it remains a great example of dense, family-oriented, and grade-related housing.
For fun, let's compare this to a more intense form of infill development. Our Junction House project, for instance, contains 151 homes and sits on a 0.48-acre piece of land. This translates into about 315 units per acre. I don't know off hand the average number of occupants per household, but I reckon that, given our larger average suite size, we should be on the higher end compared to most mid-rise condominiums. So I would say that we are probably 400+ people per acre.
It's unfair to compare a single development to an entire neighborhood, such as Stuyvesant Town. Circulation and other open spaces will necessarily pull down your average density. But these individual development examples do speak for themselves. There are many parts of North America where you might find 1 home or a handful of homes per acre of land. At Penn's Landing Square, this number is 50 units per acre. And at Junction House, it's 315 units per acre.
Here's some recent data, via CityLab, suggesting that Americans are walking less and driving less, but killing more people when they do drive around. (The report is based on data from 2019 to 2022.)
My first reaction to these high-level findings is that they seem to make sense. This time period was the pandemic. And people were locked away at home (though I used to take some seriously long walks around downtown during this dark time).
So I don't know, I'm not sure we can conclude that walking less is truly a structural phenomenon. Similarly, I'm not sure that we can immediately conclude that cars are becoming increasingly more dangerous.
According to Wikipedia, deaths per capita, deaths per billion vehicle miles traveled, and total deaths, have all been generally declining in the US since the 1960s.
However, I do wonder if there's some sort of correlation between people walking less and car-related fatalities. The most dangerous streets, in my mind, are often the ones that don't have a lot of pedestrians.
That's why, broadly speaking, it feels safer walking around Manhattan than it does Los Angeles. So maybe less people walking is enough to trigger an increase in pedestrian fatalities.