A Transit City is a Centralized City
In New York, a large fraction of employment clusters in a rectangle bounded roughly by 59th Street, 2nd Avenue, 42nd Street, and 9th Avenue. Although it’s a commonplace that New York employment is centralized around Manhattan, in reality most of Manhattan is residential, and employment is concentrated in a few square kilometers in the heart of Midtown. This is where the subway lines converge from all directions – elsewhere there simply isn’t enough capacity. Of course it wasn’t always like this: Manhattan’s population in the 1890s was the same as it is today, and it was clustered toward the southern third of the island, but employment was relatively evenly distributed in the downtown area. What has happened since then is that New York became a transit city.
There’s a strong correlation between the form of a city and the mix of transportation options people use. This extends well beyond density, but the principle is the same. Transit is at its best at high intensity, because this is what supports high-frequency service. Cars are the opposite: even on a normal urban street, a car alone will beat any rapid transit line, but every additional car will slow down the road dramatically, so that at even the moderate intensity of an edge city gridlock ensues.
Although usually this principle is stated in terms of density, it’s equally true for work centralization. The pedestrian city and the bus city will be dense all over, and feature high job density scattered across neighborhoods: walking is too slow for the transit city pattern to emerge, and buses have too little capacity. But dedicated rapid transit wants to serve an area right next to the stations, and once a network is built, a CBD grows around the central area. This CBD is typically small, just a few square kilometers. Even vaguely CBD-ish locations, such as Penn Station, are too far, as one commonly quoted figure about work locations demonstrates. The CBD isn’t even large enough to encompass all of the 34h-59th Street strip that the tourist guidebooks define as Midtown. The subway lines only form a tight mesh in a subset of that general area.
The job density of such a CBD is measured in hundreds of thousands per square kilometers, requiring many high-rise towers, several of which are supertall. In contrast, most of New York’s residences are mid-rise, and Tokyo’s are low- and mid-rise; their residential densities in the low tens of thousands per square kilometer are high enough that they are considered the epitome of density, but their CBDs are an order of magnitude denser.
Of the major transit cities of the world, Paris is the only one that’s resisted this trend with its height limit, but instead a transit-like CBD started out in La Défense, and the same pattern that comes from the subway in New York or Tokyo or the L in Chicago emerges with the RER. Of course, Paris maintains very high residential density, but its job distribution is more in line with that of a bus city – employment is dense all over, and the Downtown Paris employment density peak is less pronounced than in comparable transit city downtowns.
This does not mean a transit city needs to have empty trains going in the reverse-peak direction, as Cap’n Transit, Jarrett Walker, and others charge. A transit city will have job destinations outside the CBD, growing around rapid transit junctions: for example, Tokyo has Shinjuku, Shibuya, and Ikebukuro, all of which are so replete with high-rises it’s hard easy to forget they’re secondary job centers. While there is still a pronounced peak direction, people rely on transit so much that they take it for regular errands, supporting very high off-peak frequency by the standards of trains with drivers.
New York has something similar in Downtown Brooklyn, Jamaica, and Long Island City, but the modal split of those job destinations is much less favorable to transit – 50% in Downtown Brooklyn and Long Island City and 30% in Jamaica, according to a study of New York’s secondary job centers that I can no longer find. This is a general feature of many old American cities: the core looks like a transit city, but beyond it is a car-centric city, filled with edge cities and edgeless cities. Because the layout beyond the core is car-centric, the off-peak and reverse-peak traffic that supports high all-day bidirectional frequency on the Tokyo rail network, or for that matter on most New York City Subway lines, does not exist. The preference of American commuter rail agencies for peak-only service comes partly from an operating model that makes it impossible to run frequent off- and reverse-peak service, but also from a job distribution that makes the market for such runs small even under the best industry practice.
A corollary of this fact is that the multipolarity of other cities, for example Los Angeles, is not an asset. It would be an asset if those job centers were intense and could be easily served by transit; in reality, they have moderate intensity, nothing like that of the secondary centers of Tokyo or even New York, and serving many of them requires digging new subway lines. Burbank, on the legacy Metrolink network, could make a reasonable site for a transit-oriented secondary center, if commuter rail operations were modernized and local transit lines were extended to it; the Westside and Santa Monica do not, and the hope is that the investment in the Subway to the Sea could enable them to grow to reasonable size.
The key here is that the reason Shinjuku, Ikebukuro, and Shibuya are as transit-oriented as Central Tokyo is that they historically arose as connection points between the Yamanote Line and the private railroads. In particular, they already had rapid transit fanning out from multiple directions when they became major job centers. But Tokyo’s transit development history is peculiar; most other cities did not have large electrified rapid transit systems terminating at the edge of the urban core prior to building local subway lines.
A second corollary then is a strategy that sought to make New York a more transit-oriented city would treat centralization differently. It should turn the secondary centers into transit nodes in their own right, with tails extending as far out as reasonably possible. Jamaica already has some of the infrastructure, but it’s used poorly because of antiquated LIRR practices; the same can’t be said of Flushing, so a priority should be to build reasonable-quality transit from multiple directions, connecting Flushing with College Point and Jamaica and modernizing the LIRR so that it could connect it with Bayside.
A point that many people writing about this neglect (with pleasant exceptions like Cap’n Transit, the Streetsblog crowd, and Paul Barter) is that this requires both the carrot of more transit and the stick of less parking. In any case it’s hard to create high job densities when much of the land is used for parking. But on top of that parking mandates make it difficult for transit to be competitive when it’s expected to include railyards and depots in its budget and roads are not.
But what a transit city doesn’t need is job dispersal. The importance of creating secondary centers is strictly as alternatives to auto-oriented edge cities and edgeless cities, since whatever happens, not all jobs will be in the CBD. A large city with rapid transit connecting to all major neighborhoods will automatically have high transportation capacity. Rapid transit is good at transporting tens of thousands of people in one direction in the peak hour; let it do what it’s good at.
Pedestrian Observations 
<blocThe preference of American commuter rail agencies for peak-only service comes partly from an operating model that makes it impossible to run frequent off- and reverse-peak service, but also from a job distribution that makes the market for such runs small even under the best industry practice.
These two things form a positive feedback system, jobs are not there because of bad service, and service does not have to be improved because there are not enough jobs to justify it. This leaves service bad … and so on. This kind of Catch 22 can be broken by either fact changing, for example if Metrolink is modernized so it allows faster and more frequent service than some threshold, the location attractiveness for bussinesses can multiply quite quickly and this in turn can invoke further interest for better transit service and so on.
A Jamaica-Flushing rapid transit line should continue to LaGuardia and Manhattan, not to College Point. The ridership would be much higher.
Don’t necessarily think of heavy rail. Surface-running streetcars in dedicated lanes might work well in this environment. The distances aren’t that long. Jamaica to Flushing is less than four miles. Parsons and Kissena Blvds are both at least 48 ft. wide and for most of their lengths devote their curb lanes to parking. A Jamaica-Flushing line along that alignment would be relatively cheap. Eliminating street parking might be considered a feature rather than a bug. Add a similar line along Main St. to the E-F station at Van Wyck and another along Northern Blvd into Woodside and you’re starting to get enough transit infrastructure into Flushing and at much lower than Second Ave. Subway costs.
The ends are problematic for a surface streetcar. The merge of Kissena Blvd and Main St is one of the worst congested Intersections in NYC. Sections of Parsons Blvd south of Grand Central Parkway are too narrow for dedicated lanes, unless you want sidewalk eliminating street widening and eminent domain of front yards and buildings. Otherwise you’ll have conflicts with school buses, fire trucks, large box vans/trucks, and anything 8′ – 8.5′ wide. Main St option may be the way to go, and might be better if extended down Queens Blvd to Jamaica Ave and LIRR Jamaica Station.
Alon, something else that occurs to me is the way this intersects with the way we speak about urbanisms. The so-called Traditional Urbanists (like Nathan Lewis, say) are in practice advocating a pedestrian-oriented city, whereas suburbia is historically auto-oriented. In practice, true transit cities are so rare, usually present only in the largest city of a Western country, e.g. London, Paris, New York, Milan, that they don’t even have a typology associated with them.
Part of this is the onset of land use management–and particularly its onset in terms of preventing densification. Zoning, itself, has proven the most powerful tool ever made available to NIMBYs, in large part because it was developed as a legal tool for them. (In American jurisprudence, it is an outgrowth of nuisance law.) Charlie Gardner gives the example of Vancouver.
My own thinking on this matter, is that cities tend naturally to fall into equilibria based on their dominant transportation system, and that urban advocates for any particular side want the system changed to support their equilibrium. Yet the most likely equilibrium a city in the developed world will fall into would be the transit city–where the transit network is good enough to offer it.
Funnily enough, the other major angle of development, the bike city, would create neosuburbs at essentially the same spatial scale but with a continuing destructive disconnect between transit and urban amenities.
I’d be interested to see an employment spatial analysis of Vancouver. While they invented large scale, high density TOD, I wonder if their non-CBD employment centers are concentrated in these districts. I suspect suburban job sprawl is more rampant than we think, even in metro areas with progressive land use and transportation policies.
What you seem to be argueing is that a city with a high transit mode share must have centralised employment. (Correct me if I’m wrong).
So, does that it is impossible to make a non-centralised city have a high transit mode share? (Without centralising things)
It depends on what “centralized” means. There’s centralized like Paris, and centralized like New York or Tokyo. Paris is centralized in the sense that a large fraction of regional employment is in the city proper, but it has no analog of Midtown Manhattan or Central Tokyo, which cluster hundreds of thousands of jobs in just a few square kilometers. In other words, Paris centralization involves a broad employment peak in a diffuse center, whereas Tokyo centralization is spiky around transit nodes. But Paris is moving in the Tokyo direction somewhat: the RER has wide stop spacing, which encourages spiky development, while the Arc Express orbital may be marketed as a circumferential line, but it’s for the most part just a north-south line through La Défense that’s being run together with other lines to potential secondary centers.
I’m going to completely punt on whether this is required for a high transit mode share in a smaller city. But I’ll note that in Karlsruhe, poster child of successful small-city transit, the center is dominant, and the reason tram-trains work is that they only need to be in streetcar mode from the train station to the city center.
@Tom West
I think he’s just saying that in a transit-oriented city, employment and density are quite nodal, like the transportation network. This follows pretty naturally if you have a good transit network, but of course typically the two patterns evolve in concert. Tokyo is actually quite decentralized compared to most big American cities.
Interesting post. A few people have been writing about centralization in cities (I wrote a response to Stephen Smith’s piece at Forbes a few weeks ago here) and I think yours is a unique view.
I think the natural tendency of many transit cities to be highly centralized is based on the fact that transit will develop in more mature cities, which have a greater tendency to already be centralized. Pedestrian oriented cities came first, so when mass transit was developed afterwards cities already had a pedestrian-density core to commute to. Newer cities (or even newer areas of older cities) that never really had a pedestrian oriented phase are more decentralized because they were built with the car in mind.
I’d argue that we’ve never been in a situation where existing car-centric cities are facing pressure to develop transit. It’s interesting that you say buses are a more decentralized form of transit: I’d say this is a historical quirk as well, and that buses were developed as the main mass transit fix in car-centric cities.
I don’t think there’s anything conceptually wrong with a grid-like evenly distributed transit system. If anything, it’s more efficient because you have less issues with peak hours and empty trains going in the opposite direction. I think someplace like Los Angeles should develop with this in mind, and develop different nodes connected by a grid rather than focusing all of its development attention on Downtown LA.
For someplace like New York, the peripheries should be treated like LA. You could make it easier to take a train to Manhattan, but that means people will simply return to their car dependent lifestyle once they get back home. If you start building more transit connections to places other than the center, you’d give people more car-free options and I’d say you’d get more people out of their cars than if you just focused on more center oriented transit development.
The treatment I have of bus cities vs. transit cities isn’t really applicable to North America or Europe. It’s a distinction I learned from Paul Barter’s thesis, which discusses East Asian cities; it applies to cities that industrialized late enough (for example, after WW2) that they had buses before they had trains. For those cities, the cheap transit as they grow too big to be pedestrian cities consists of buses, including jitneys and minivans. For example, most Chinese cities today are bus cities, but some, especially Shanghai, are transitioning to the transit city model by building enormous rail networks. The transition to a car city involves a natural rise in car ownership causing congestion and making buses slow (while also promoting a shift away from the congested center), and also a fair amount of government rules that regulate private jitneys out of existence.
For a large city, and this includes Los Angeles, the advantage of a strong center is that it can be served in all directions. Of course any place can be served on a grid, but the problem is that rapid transit is expensive, so you can’t build it on a grid except in the center, where demand is very high. Even Paris has a radial transit system outside the M2-M6 ring; the Métro isn’t very radial, but close to half of rail ridership in Paris is not on the Métro but on the RER. Los Angeles’s geography is such that several of its secondary centers, such as Long Beach and Santa Monica, don’t have too many directions to go in and this can simplify building out a polycentric system, but if the choice comes to building more in Downtown LA or building more in Pasadena or Anaheim, go with Downtown.
What you mention at the end, about the car-dependent lifestyle at home, is the same chicken-egg problem dejv mentions in his comment. In the US, all commuter lines and most new light rail lines, are built with park-and-rides, and due to a combination of NIMBYism and bad planning there’s no attempt to have any commercial or high-intensity residential development near stations. This isn’t how things look if you go to the subway in Manhattan, or other places where pedestrian-centric rapid transit coexists with a dominant center: parking at stations is hard to find just like parking in the center, and there is a mixture of uses near stations, allowing people to use transit for non-work trips. For an American example of how to get it right, look to the Washington Metro in Arlington and Alexandria, which preferred upzoning to park-and-rides; as a result, it extended the walkable center of DC outward, whereas most other American transit lines instead extend the drivable suburbs downtown.
I was thinking of a grid like system with a bit higher density… There will always be a periphery. You mentioned in your reply to Tom it depends what you mean by centralized. When I think about grid like transit, I’m thinking more about something like the Triboro RX rather than something connecting the ends of metro north lines.
This also means is there’s a big difference between talking about places like Downtown Brooklyn and LIC and places like White Plains and Newark. While Long Beach will probably be an edge city that would be best for more periphery focused TOD, there are a few centers you could focus on west of Downtown. Santa Monica would be the western edge (and Burbank possibly the northern edge), but there are quite a few employment clusters in this area. I visit financial clients in the area and actually spend most of my time in Century City, and even have one client in a former industrial area Google maps calls “alsace” which is a bit south of Culver City and seems to have a lot of newer offices in former lofts.
I definitely agree with the chicken and egg problem you and Dejv mention, but this is why I think developing like this would make sense in LA more than anyplace else. Since LA has a more spread out geography, you don’t have to artificially induce external job centers to build up demand, they’re already there. You could either lean development towards the center and keep other areas at a similar lower density they’re at now, or build up a more grid like transit system and (with proper planning) induce increases in density everywhere. I would think you would get more people out of their cars by doing the latter.
Also thanks for the clarification on bus cities, I totally misunderstood what you meant by that!
The New York City subway is radial if you consider the center to be Fulton Street. All the more reason to drag NJTransit and the LIRR to Fulton.
Good post. Not sure why you appeared to differ with me, as I think we agree completely.
hard to forget they’re secondary job centers
…I think you meant “easy to forget.”
I did, thanks. But it’s mild by the standards of writing a long comment excoriating thinktanks and putting in the sentence “if a thinktank disagrees with a peer-reviewed study, the thinktank is always right.”
“Inter city” travel in much of the world is “medium distance suburban commuter” travel (or even short-distance) in sprawl-land.
Concentration (of employment, housing, infrastructure, etc) is indeed vastly more efficient, and is personally attractive to me, but can also be so on less than a NYC scale.
In central Europe, cities with their own Central Business Districts and with their own S-Bahns, etc, can be separated by less the length of some US commuter bus runs. The Rhein-Ruhr is just the most familiar poster child.
Multi-polar or uni-polar? It depends on the scale used.
This is where the subway lines converge from all directions – elsewhere there simply isn’t enough capacity.
It’s nearly uniformly the same between 53rd and Houston. South of Delancy, if anything, it gets higher since the lines start to converge as the island gets narrower. South of Chambers it seems like there’s a subway entrance on every corner. … which made sense 100 years ago because the employment center was Wall Street and to a lesser extent Downtown Brooklyn, for people who wanted to live in suburban Harlem or Flatbush.
Even vaguely CBD-ish locations, such as Penn Station, are too far, as one commonly quoted figure about work locations demonstrates.
Midtown work locations. How do they define “Midtown”? If the definition is “north of the center of 34th Street”… that means jobs in Penn Station – 5 Penn Plaza? – or the Vornado extravaganza that will replace the Hotel Pennsylvania? – aren’t in Midtown. Jobs in the Empire State Building aren’t in Midtown. Most the jobs in the Garment District aren’t in Midtown. Macy’s is a Midtown department store, Gimbel’s isn’t…. if there still was a Gimbel’s.
Most cities would give their eye teeth to have their CBD to look like Herald Square. Not that there are many places in the world that could fill the Empire State Building. Or generate enough shoppers to support a department store the size of Macy’s Herald Square. Or fill Madison Square Garden.
….most metro areas in the US could comfortably fit all of their residents in Manhattan. Only two, Los Angeles and Chicago wouldn’t fit into New York City. DC wouldn’t but only if you include Baltimore. Brooklyn, if it was it’s own Metropolitian Statistical Area would be 17th or 18th on the list. Set Brooklyn and Queens adrift in the Dakotas and it’s the biggest metro area northwest of Chicago….
Well, I’m talking relative to the size of the metro area. Besides, the transit city with the multiple radial lines converging on the CBD from all directions is a large city – smaller cities don’t have the population to support so much rapid transit. So while 34th Street is a huge employment center by the standards of the Dakotas, or for that matter the standards of Queens, it pales in comparison to what’s happening 20 blocks up. It’s more comparable to Union Square, which, important as it is, nobody is going to confuse with the center of the city. The best way to see this is to look at transfer patterns at QBP: in the am peak, more people transfer from the 7 to the N/Q than the reverse, and in the pm peak, it’s the reverse. In other words, 59th and West 49th are bigger draws for commuters than Grand Central and Bryant Park. So if 42nd is on the margin, think how out of it 34th is.
There aren’t all that many jobs south of 34th. There’s the Empire State Building and the towers above Penn, but the other Midtown skyscrapers in the 200+ meter range are farther north, starting around 42nd and peaking in the 50s. And south of 34th, soon the buildings drop in heights and the neighborhoods start having real names like Koreatown.
You’re right that a hundred years ago the employment peak was at the southern tip of Manhattan, but the employment patterns have evolved. First, Midtown has eclipsed Lower Manhattan, and Downtown Brooklyn became a decidedly secondary center. And second, within Lower Manhattan, the same shift I’m describing happened, creating an intense concentration of employment around Fulton and Wall Streets. If you plot the locations of the skyscrapers in Lower Manhattan, with few exceptions they’re all south of Chambers; a Park Row terminal today would be almost as inappropriately located as a train station on 34th Street. North of Chambers again you get into neighborhoods with names, and to the east of City Hall you get approach ramps and a Moses-built auto-oriented form, hardly conducive to a transit city.
In other words, 59th and West 49th are bigger draws for commuters than Grand Central and Bryant Park. So if 42nd is on the margin, think how out of it 34th is.
Or they are changing at Queens Plaza to get to Union Square. Much easier to cross the platform than wade through the crowds at Grand Central to get on an overcrowded Lex. Ave train. Times Square and Grand Central are the busiest subway stations, Third busiest is 34th and 6th, fifth busiest is 34th and 7th, sixth busiest is 34th and 8th. They aren’t all there to go shopping at Macy’s or going to an event at Madison Square Garden.
If I did the arithmetic correctly 34th street has an average weekday ridership that’s 90% of 42nd. I wasn’t in the mood to try out figure out Wall Street… is the City Hall station “Wall Street”, How about Whitehall? Or try to figure out the 50s. Flushing Main Street is very busy, lots of that is Malba to Rockefeller Center… some of it’s to Union Square and some is to Wall Street and some of it to Corona or Jackson Heights.
The best way to see this is to look at transfer patterns at QBP: in the am peak, more people transfer from the 7 to the N/Q than the reverse,
Partly because the line to Astoria is shorter and more suburban than the line to Flushing. 6 stops to the end of the line in Astoria and 14 to Flushing. … assuming every thing was perfectly symmetrical 30 percent of the riders changing trains would be from the BMT to to the IRT and 70 would be from the IRT to the BMT. ( 6/20ths versus 14/20ths )
Partly because the BMT has more stops in Manhattan…. if I’m starting out at 40th street in Sunnyside and I want to get to Union Square walking across the platform might be more interesting then fighting the crowds at Grand Central…. might even be faster….and the N doesn’t screech when it pulls into Union Square….
North of Chambers again you get into neighborhoods with names,
That’s partly geology. The schist and the gneiss take a decided dip in the 30s and don’t come up again until Chambers…. there’s a reason why the water tunnels dip deep under Chelsea, Greenwich Village, SoHo and Tribeca…. and SoHo and Tribeca are relatively recent neighborhood names… and Greenwich Village was literally a village that suburbanites commuted to, by ferry, when it was developed.
A huge amount of the ridership on 34th is people connecting from Penn Station onward. On 42nd this is less true – first because there’s higher non-train station ridership, second because Grand Central’s subway ridership is a little less than the combined total of the two Penn Station stops, and third because Grand Central has only about half as many commuter rail riders and many more of them can walk to work from the station.
Higher non train ridership like the people getting off the bus in the Port Authority Bus Terminal? So they can get to their job down on Wall Street?
My sense is that job centers are labile around Manhattan south of the Park. A few years ago, AP moved from 50 Rock to the Ziggurat at 33rd and 10th, for example. Since the transport infrastructure doesn’t move, any statistics about how many jobs are reachable from which transport node will be out of date fairly quickly. Part of the point of the New York subway is that most any point in Manhattan south of the Park (at least between 2nd and 9th Aves) is within walking distance of some subway stop, so if jobs move within that space, they’ll still be subway accessible.
Alon, interesting two posts (including the newer one) on an interesting topic! I mostly disagree (and, possibly, also agree, a bit), but don’t have the time right now to give the topic the justice it deserves — so, this is a “place saver,” so to speak, for a possible future comment.
– – – – – – – – – – –
RIchard Mlynarik wrote:
“Inter city” travel in much of the world is “medium distance suburban commuter” travel (or even short-distance) in sprawl-land.
In central Europe, cities with their own Central Business Districts and with their own S-Bahns, etc, can be separated by less the length of some US commuter bus runs. The Rhein-Ruhr is just the most familiar poster child.
Multi-polar or uni-polar? It depends on the scale used.
Benjamin Hemric writes:
Interesting information, and a good point! This is similar to one of the points I hope to make.
– – – – –
Charley Ferrari wrote:
I’m thinking more about something like the Triboro RX rather than something connecting the ends of metro north lines.
Benjamin Hemric writes:
Hadn’t heard of the Triboro RX. Don’t know if I think it’s a good idea or not – but it’s good to see that people are at least exploring such possibilities.
– – – – –
Jim wrote:
Surface-running streetcars in dedicated lanes might work well in this environment. The distances aren’t that long. Jamaica to Flushing is less than four miles. Parsons and Kissena Blvds are both at least 48 ft. wide and for most of their lengths devote their curb lanes to parking. A Jamaica-Flushing line along that alignment would be relatively cheap.
Al wrote:
Sections of Parsons Blvd south of Grand Central Parkway are too narrow for dedicated lanes, unless you want sidewalk eliminating street widening and eminent domain of front yards and buildings. Otherwise you’ll have conflicts with school buses, fire trucks, large box vans/trucks, and anything 8′ – 8.5′ wide. Main St option may be the way to go, and might be better if extended down Queens Blvd to Jamaica Ave and LIRR Jamaica Station.
Benjamin writes:
As someone who grew up in Jamaica Hills, I’ve been thinking about such a surface line for years. However, I agree with Al that sections of Parsons Blvd. south of Grand Central are at least problematic because of their width.
By the way, I’m guessing that the Parsons Blvd. route is also problematic because of its grade – as there actually used to be a streetcar line that connected Jamaica with Flushing and it crossed over Parsons Blvd. at 87th Ave. because, and this is just a guess on my part, Parsons Blvd. in that area was too steep. So, instead of continuing along Parsons, which would seem to have been the natural route, the trolley went along a private right of way between Parsons and 164th Street (Glenn Ave. on Google maps), which I believe is at a lower grade, and from there it continued onward to Flushing.
For those who might be interested in this trolley line, which was part of the New York and Queens Railway, it is discussed on pages 9 and 10 of “Lost Trolleys of Queens and Long Island.” The book says it was mostly a private right of way line, so the idea that the trolley used the private right of way between Parsons and 164th because it was at a lesser grade is only a guess on my part. The line was discontinued in 1937. There are also some nice photos (about 20 which pertain to this particular line) and route maps (inlcuding one of this line). The books is by Stephen L. Meyers and is part of the “Images of Rail” series of Arcadia Press.
Benjamin Hemric
Sat., November 26, 2011, 3:55 p.m.
Your focus on subways and intra-Manhattan commuting seems off. I think the density of work areas in midtown is driven by the locations of Penn Station, Port Authority, and Grand Central, which all bring people from the suburbs. Or am I wrong?
I don’t have exact numbers for Midtown, but Manhattan-wide, a large majority of workers live in the city, and a plurality (30%) live in Manhattan.
In Taiwan you have all three of these situations: every city used to be a bus city; recently Taipei has been moving towards becoming a transit city while the rest of the island becomes car cities. Taipei is a bit different from Tokyo or NYC though: there’s high transit ridership and a clear central CBD, however you don;’t get the crush loads you find elsewhere and the CBD is far from dominant, perhaps because there is a newer greenfield CBD (with poor transit and very wide roads) in the city’s east which takes some demand for office space, and also because the area where the high-capacity lines merge is pretty hostile to pedestrians and very poorly used (bus depots, empty plazas, surface parking lots). Overall though employment seems to be relatively evenly spread, and a lot of people drive, getting around (some of) the density issues cars present by driving scooters and parking on sidewalks. So perhaps it’s an example of a half-formed transit city.
Taiwan is weird because of the high motorcycle ownership; I don’t really know how a motorcycle city will look like. Taipei is also unique is that it maintained high transit ridership despite building rapid transit quite late in its development; one study I’ve read on the matter claims it’s because it built its MRT system very rapidly, unlike US cities (or Bangkok and Kuala Lumpur for that matter).
The unwalkable-CBD problem is not unique to Taipei. Tel Aviv has the same issue, since the emerging CBD is built out near the highway, while the old city center, which is fairly pedestrian-oriented, is not getting much new development. Is Taipei’s CBD a relatively new development, or is it the city’s old center, as is the case in Tokyo and New York?
Taipei’s CBD is its old city center. Inside the CBD itself it’s actually not so bad, it’s just the train station area next to it, which used to be occupied by at-grade rail tracks, that is truly horrible. On second thought though calling it a CBD is a almost an exaggeration- it’s definitely one of the largest conglomerations of jobs in Taipei and has the most expensive office space, but as far as I can tell it’s overshadowed by the greenfield CBD in the east, which supposedly sucked a lot of the life out of the old center after it was built. In that sense I find it hard to fit Taipei into your model- it’s very dense (about 20,000-35,000 people/ sq km throughout the city) and does have high transit ridership by US standards (around 15% share for MRT and 18% for buses for all of greater Taipei) and low auto use (just over 50%), but its job distribution only partially follows a typical transit city pattern.
I’d guess geography has something to do with the success of Taipei’s MRT. Taipei is surrounded by mountains, so there is no room for low-density expansion, which may have hobbled the vicious cycle that paper describes.
Pretty much every Taiwanese city outside greater Taipei is a motorcycle city, with motorcycles having a 62% transit share in Kaohsiung. I’m not very familiar with these places, but they seem to have dense (10k-20k ppl/ km2) but mostly low-rise old centers, surrounded by sprawl consisting mostly of attached single family houses, and with employment spread evenly, in minor centers, or in suburban industrial parks.
I was thinking about your concept of the bus city vs the transit city, and it occurred to me that this doesn’t really explain the situation of all but the largest US cities 1890-1920 – post development of street car networks but pre-automobile (and pre-truck)
The transportation mode in these cities most closely resembles the bus city, but with mixed traffic streetcars taking the places of buses, and with similar limitations. Nonetheless, the typical US city of the period was a very spiky place, with a strongly centralized business district surrounded by ever more spread out residential areas with little local employment. Despite the absence of anything more than surface transit, places like Denver (pop 130,000 in 1900) had more transit trips per capita than large European cities (per Fogelson).
How big were those business districts relative to city size? The point of transit city spikiness is that a metro area of 20 million can have a CBD of about 4 square kilometers. Of course a city of 130,000 would have a small CBD surrounded by lower density – at that scale it’s hard to distinguish bus cities from transit cities. The distinction starts making sense at somewhat higher population levels.
They were generally fairly compact business districts in cities of all sizes, but I don’t have any numbers off hand. But the Loop in Chicago was already a huge business district before the Union Loop was constructed giving rapid transit access in the 1890s relying entirely on street cars and steam trains before then. Before the Loop Chicago had already seen built the world’s tallest load bearing brick building, the Monadnock, demonstrating the very peaky central business district’s demand for space. This was noted by observes at the time to be in contrast to the dense-all-over, work-all-over Parisian type.
Perhaps there are some technologies that play into city form here too, like the telephone and cheap drayage by truck that contribute to the modern patterns that that the bus city has that the streetcar city of 1900 did not
One aspect of Southeast Asian bus cities is their ethnic diversity, with no assimilation whatsoever. Thus there were separate commercial sectors for the local Malay or Filipino population, the Chinese community, the white colonial community, and (in some cities) the Indian community. These sectors were all adjacent, but separate, creating a geographically large low-rise central business district, surrounded by suburbs.
Of note, Nairobi today, entirely a bus city, seems to be intermediate in typology between a bus city and a transit city. Like a transit city, it has a small CBD, and a handful of off-CBD nodes. But as in a bus city, people still tend to work in the neighborhood where they live or in the CBD, without the specialization that modern transit cities have with secondary districts.
Of course, in 1900, the major US cities had unassimilated ethnic minorities like the Jews, the Italians, the Poles, and (occasionally) the blacks. Did they have commercial sectors of their own? I know that in New York at the time, Brooklyn was still not integrated into the Manhattan-centric urban layout, so people could live their entire lives without leaving the borough, and workers were more likely to commute to Downtown Brooklyn than to Downtown Manhattan. Development was also fairly uniform along streetcar lines, which is still the case in New York today with the subway, but isn’t really the case in Tokyo, which has redeveloped a lot more.
To note about Nairobi, is that when there was a project to map the routes of the private Matatu buses of Nairobi, they were found to be strongly radial (https://www.wired.com/2015/08/nairobi-got-ad-hoc-bus-system-google-maps/#slide-3)
With respect to ethnic neighborhoods, Chinatown in San Francisco and Vancouver (and wee little Salinas) are adjacent to the CBD (and in the case of Vancouver, Chinatown is only a couple blocks from the original CBD locus in what’s now the Downtown East Side. At least Fogelson argues that in most US cities, that substantial ‘neighborhood’ business didn’t really emerge until the 1920s
Hmmm. The stereotype of people who live their entire lives in Brooklyn and never visit Manhattan is older – more late 19th century than early- or mid-20th century.
Chinatown is a good example of an unassimilated ethnic neighborhood, yes. New York has internal private Chinatown buses, connecting Chinatown with Flushing and Sunset Park; it also has an MTA-branded bus connecting South Williamsburg with Borough Park, for the ultra-Orthodox population. Singapore, too, still has Chinatown and Little India. However, these are commercial centers more than job centers. Singaporeans of all ethnicities are likely to work in the CBD, at Orchard, in Marina Bay, etc.; New Yorkers of all ethnicities are likely to work in Midtown or Lower Manhattan. Ethnic centers like Chinatown (in both New York and Singapore) and Harlem just aren’t major job centers today.
In Vancouver the situation is more complex, I think – Richmond blew away the traditional Chinatown. Immigrants to Canada also assimilate very rapidly; Surrey and Richmond look like just two suburbs to me, rather than like cultural centers for stable ethno-linguistic groups. (And Vancouver itself is a mix of bus, transit, and auto city.)
I’ve seen the matatu map! It is fascinating. I may or may not have drawn whimsical crayon for a Nairobi subway based on it. I will note that the matatu map is more like traditional US commuter rail – exclusively radial, poor connections between different lines – than like radial rapid transit.