Great article! I think we urbanists shoot ourselves in the foot when we try to argue that cars are inferior in terms of mobility. It's just not true. But it is true that the costs are not worth the mobility benefits.
One complaint: I read a decent amount of history and Marchetti's Constant is one of the worst theories I have ever read. It's a great construct for understanding the modern world, but it's completely misleading for any time before the nineteenth century. Commuting itself is a relatively new practice. Maybe you can find some odd cases of it as Marchetti did, but it was not a norm in pre-industrial society.
That's interesting, thanks! I haven't studied the history of Marchetti and how he came up with the formulation in detail, but my understanding is that cities throughout history have had a very strong tendency to concentrate within whatever distance is "one hour across," which is perhaps coincidentally a 30-minute commuting radius from some central point.
In my mind "30 minute radius" is pretty easy to explain, and it does map relatively well to modern American development patterns. When cities today reach that point they start putting new hubs of employment on the edge and then becoming polycentric.
But to me the more important core idea is that there's roughly a one-hour diameter for economic agglomeration that defines the "gravity well" of a city, and polycentric places don't get the full effect.
Case in point, having lived and worked in both San Francisco and Houston, I found that in both metros there were very big differences between living and working at the center versus one of the edge nodes. Eg. most people who live and work in SF don't regularly mingle with people who live and work in San Jose. It's one of those things that people often lament -- "I go down for a conference now and then, but it's just too far to go there regularly."
IRCC, Marchetti claims that the time constant is neurologically wired as a balancing of two neurophysiological needs: opportunity (expressed as range) and security (expressed as time-distance from home). He also supposes that this applies to all settlements for all eras. Marchetti cites only one pre-modern example.
I’ve thought a lot about this, though not particularly systematically.
I think many metros would benefit by eliminating cars from accessing the downtown metro (much of Manhattan would be improved by eliminating passenger vehicles and limiting trucks and delivery to specified corridors).
The improvement of battery technology has made personal mobility options much more attractive.
My family lives in Seattle, in a dense by Seattle neighborhood but not downtown. We have a folding electric scooter that cost 400 dollars that makes the bus system here much more usable. We have one car, but that scooter has meant we haven’t needed a second car. Public transit systems could probably be optimized around such solutions, giving people much wider range of what amenities are served by a transit corridor (and might be a way out of that transit trap between 5k and 10k people.
The problem is primarily political. Seattle is on that bubble of density, and a lot of people oppose new development because of parking, but the transit system doesn’t really let you live a car free existence either, and plans to expand the transit system are glacially slow.
Thanks for the reference! I think this is a pretty obvious observation once you stop and think about it, I'm sure there are many other sources that mention this concept as well.
Nice piece! Our region is bustling with very little transit presence, and a decent micro presence (protected bike infrastructure). I just finished a write up on our region’s challenges and would like to try to steer our politics away from wasteful road widening, and consider new ways to enhance car intensive mobility before we get too far down the road mimicking most western US metropolis’.
Drive to the City of London on a Sunday or Tel Aviv centre on a Friday and the driving is easy. Now these are extremely dense areas, and all the people who live there still go shopping and on other errands on the weekend so clearly the problem isn't out roads handling the 20k people per square mile per se.
Instead the problem is almost all super dense areas are also places people commute to. Now a commute has the following problems:
1. It's long distance. Walking isn't an option and cars spend a long time on the road.
2. Everybody's going in the same direction. Too many cars squeeze on to the same under capacity roads so everything gridlocks.
3. The density temporarily shoots up way past 20K per square mile.
So using cars as a way to carry out local errands is fine. Using them for commutes is problematic. We should focus policy on cutting out commutes. That can range from various forms of congestion charges and tolls, to encouraging public transport alternatives, and in practice a mix of both.
That's true. Similarly, I lived for years in San Francisco, in one of the densest residential neighborhoods in the US, and I still had a car and found it handy to use for certain kinds of trips. As you point out, I wasn't downtown, so I didn't have a lot of contention with commuters to deal with.
But also London, Tel Aviv, and San Francisco all existed before cars, so they were never built upon a car-only design pattern.
The tricky thing in the US is that at the point the majority of inhabited land area *was* developed post-car and most of that was developed for cars only. It's not a coincidence that our most car-oriented city, Los Angeles, is also one of the most intensely NIMBY places in the country.
One way to think of this is that walkability soaks up a huge number of neighborhood trips. But the counter is also true -- car-only design quickly leads to traffic and parking problems when density reaches even modest levels.
The most expensive thing about cars is in buying the car or driving the car, it’s the cost of housing when we have exclusionary zoning in place to simply make room for the car homes are about 30% more expensive because we banned multifamily homes in most of our towns just so we can have parking and we don’t have to deal with too much traffic as we drive a car around. And the side effect of cars is that 40% of Americans can’t even save for retirement.
I would add one more huge advantage that cars have: they are mobile storage lockers. Most daily drivers aren't used to carrying things, whereas if you're on foot or on a bike you have to.
One thing you might be underestimating on the strength of car culture is that cars taking up lots of space is valuable to the owners. It's space they have a permanent option of using. I wrote an article recently "Dare I praise the earth gods" which gets in the head of the full benefit of automobile ownership and the reasons I am pessimistic about the current paradigm of "urbanism hopes" being more than hopium.
Fantastic piece. I'm curious to what extent you think certain new technologies - namely, electric scooters and ride-sharing - can solve for the missing middle by filling in the last mile of commutes.
Great article! I think we urbanists shoot ourselves in the foot when we try to argue that cars are inferior in terms of mobility. It's just not true. But it is true that the costs are not worth the mobility benefits.
One complaint: I read a decent amount of history and Marchetti's Constant is one of the worst theories I have ever read. It's a great construct for understanding the modern world, but it's completely misleading for any time before the nineteenth century. Commuting itself is a relatively new practice. Maybe you can find some odd cases of it as Marchetti did, but it was not a norm in pre-industrial society.
That's interesting, thanks! I haven't studied the history of Marchetti and how he came up with the formulation in detail, but my understanding is that cities throughout history have had a very strong tendency to concentrate within whatever distance is "one hour across," which is perhaps coincidentally a 30-minute commuting radius from some central point.
In my mind "30 minute radius" is pretty easy to explain, and it does map relatively well to modern American development patterns. When cities today reach that point they start putting new hubs of employment on the edge and then becoming polycentric.
But to me the more important core idea is that there's roughly a one-hour diameter for economic agglomeration that defines the "gravity well" of a city, and polycentric places don't get the full effect.
Case in point, having lived and worked in both San Francisco and Houston, I found that in both metros there were very big differences between living and working at the center versus one of the edge nodes. Eg. most people who live and work in SF don't regularly mingle with people who live and work in San Jose. It's one of those things that people often lament -- "I go down for a conference now and then, but it's just too far to go there regularly."
Here is a link to Marchetti's original paper. Unfortunately Elsevier is an expensive subscription that I no longer have access to:
https://www.sciencedirect.com/science/article/abs/pii/0040162594900418?via%3Dihub
IRCC, Marchetti claims that the time constant is neurologically wired as a balancing of two neurophysiological needs: opportunity (expressed as range) and security (expressed as time-distance from home). He also supposes that this applies to all settlements for all eras. Marchetti cites only one pre-modern example.
Thank you. Wonderful piece.
I’ve thought a lot about this, though not particularly systematically.
I think many metros would benefit by eliminating cars from accessing the downtown metro (much of Manhattan would be improved by eliminating passenger vehicles and limiting trucks and delivery to specified corridors).
The improvement of battery technology has made personal mobility options much more attractive.
My family lives in Seattle, in a dense by Seattle neighborhood but not downtown. We have a folding electric scooter that cost 400 dollars that makes the bus system here much more usable. We have one car, but that scooter has meant we haven’t needed a second car. Public transit systems could probably be optimized around such solutions, giving people much wider range of what amenities are served by a transit corridor (and might be a way out of that transit trap between 5k and 10k people.
The problem is primarily political. Seattle is on that bubble of density, and a lot of people oppose new development because of parking, but the transit system doesn’t really let you live a car free existence either, and plans to expand the transit system are glacially slow.
Thanks, now I do not have to write about this!
The geometry problem was described by Jarrett Walker a while ago.
https://humantransit.org/2016/07/elon-musk-doesnt-understand-geometry.html
Thanks for the reference! I think this is a pretty obvious observation once you stop and think about it, I'm sure there are many other sources that mention this concept as well.
A good explanation of the transportation transect and it interacts with the land use transect.
Nice piece! Our region is bustling with very little transit presence, and a decent micro presence (protected bike infrastructure). I just finished a write up on our region’s challenges and would like to try to steer our politics away from wasteful road widening, and consider new ways to enhance car intensive mobility before we get too far down the road mimicking most western US metropolis’.
Great piece!
Drive to the City of London on a Sunday or Tel Aviv centre on a Friday and the driving is easy. Now these are extremely dense areas, and all the people who live there still go shopping and on other errands on the weekend so clearly the problem isn't out roads handling the 20k people per square mile per se.
Instead the problem is almost all super dense areas are also places people commute to. Now a commute has the following problems:
1. It's long distance. Walking isn't an option and cars spend a long time on the road.
2. Everybody's going in the same direction. Too many cars squeeze on to the same under capacity roads so everything gridlocks.
3. The density temporarily shoots up way past 20K per square mile.
So using cars as a way to carry out local errands is fine. Using them for commutes is problematic. We should focus policy on cutting out commutes. That can range from various forms of congestion charges and tolls, to encouraging public transport alternatives, and in practice a mix of both.
That's true. Similarly, I lived for years in San Francisco, in one of the densest residential neighborhoods in the US, and I still had a car and found it handy to use for certain kinds of trips. As you point out, I wasn't downtown, so I didn't have a lot of contention with commuters to deal with.
But also London, Tel Aviv, and San Francisco all existed before cars, so they were never built upon a car-only design pattern.
The tricky thing in the US is that at the point the majority of inhabited land area *was* developed post-car and most of that was developed for cars only. It's not a coincidence that our most car-oriented city, Los Angeles, is also one of the most intensely NIMBY places in the country.
One way to think of this is that walkability soaks up a huge number of neighborhood trips. But the counter is also true -- car-only design quickly leads to traffic and parking problems when density reaches even modest levels.
The most expensive thing about cars is in buying the car or driving the car, it’s the cost of housing when we have exclusionary zoning in place to simply make room for the car homes are about 30% more expensive because we banned multifamily homes in most of our towns just so we can have parking and we don’t have to deal with too much traffic as we drive a car around. And the side effect of cars is that 40% of Americans can’t even save for retirement.
I would add one more huge advantage that cars have: they are mobile storage lockers. Most daily drivers aren't used to carrying things, whereas if you're on foot or on a bike you have to.
One thing you might be underestimating on the strength of car culture is that cars taking up lots of space is valuable to the owners. It's space they have a permanent option of using. I wrote an article recently "Dare I praise the earth gods" which gets in the head of the full benefit of automobile ownership and the reasons I am pessimistic about the current paradigm of "urbanism hopes" being more than hopium.
This is one of the best urbanist essays I have ever read. I love how clear and unjudgmental it is.
I enjoy imagining you narrating the things that you write
Fantastic piece. I'm curious to what extent you think certain new technologies - namely, electric scooters and ride-sharing - can solve for the missing middle by filling in the last mile of commutes.
I'm really bullish on them! Two previous posts:
- https://postsuburban.substack.com/p/the-missing-middle-of-transportation
- https://postsuburban.substack.com/p/the-future-of-transportation