I've noticed similar in some European cities in terms of an apparent higher average density. I've always attributed it to cities that have existed longer, and that were already significantly more established by the time technology (bicycles, trains, cars) helped people commute further.
In the absence of this technology, I've always assumed people tended to stay more local, which meant that cities ended up a cluster of local communities, rather than one large central district with a large number of commuters.
That said, this seems to conflict with the example of Tokyo, which I understand was significantly rebuilt after the second world war.
>That said, this seems to conflict with the example of Tokyo, which I understand was significantly rebuilt after the second world war.
Roughly half of Tokyo was destroyed by US bombing, I'd imagine the practical and psychological result of this plays a large role in this phenomenon. And many European cities faced intense bombing in the war too, so there may be some link to what you saw.
There seem to have been two choices for the rebuilding: rebuild what was there, maybe with better trams and subways; or allow Modernist planners to take over and build Brutalist suburban car-centered systems.
I'm amazed that the author didn't include european cities in the comparison.
There are many zoning laws to ensure that cities provide good quality of life by approximating the "urban village".
This is done by keeping the density uniform and avoiding overly specialized areas like shopping malls, company campuses, sprawls and so on.
It allows people to walk or use public transport for a short time for most of your daily needs.
Of course there are economical ramifications: there are laws requiring a minimum building size, in order to provide housing at reasonable prices, instead of setting a maximum to make prices go up.
European 'zoning' regulations don't necessarily forbid mixing shops, houses, light industry, bars, clubs, etc. This means that you can often find housing, work, entertainment and food within walking distance of each other. The US rules seem to encourage segregation; not just racial segregation but also segregation of the types of use to which buildings are put.
There's also the fact that, not only was Tokyo rebuilt after the war, but that it is continually being rebuilt in a way that U.S. cities aren't. My understanding is that there's a strong preference in Japan for living in newly constructed residences, which means that older housing stock is being replaced at a much higher rate than elsewhere.
There is a strong preference for new houses, add in their superstition where if someone dies in a house it might as well be considered unmarketable and you get a market where you have to always keep building.
> I've noticed similar in some European cities in terms of an apparent higher average density. I've always attributed it to cities that have existed longer, and that were already significantly more established by the time technology (bicycles, trains, cars) helped people commute further.
I don't think this is necessarily correct. Dublin is a fairly non-dense city by European standards, and always has been, but it is currently densifying, and the number of commuters using cars for their work commute is falling.
The pattern in the previous bubble was increased sprawl, here, but densification definitely seems to be the way it's going now.
There are all sorts of possible reasons for this; traffic has gotten worse, public transport has somewhat improved, and zoning rules require minimum densities in some places.
It's mainly zoning since the 1970s that has caused prices to shoot up. If London was allowed to go up and out more prices wouldn't be do high. It's similar for central Paris.
There are definitely lot of policy, infrastructure, and cultural differences in play as well, but this is a huge factor the author missed.
Japan didn't allow any tall buildings until 1963, so the development of tall urban centers came very late, and as others have mentioned, followed patterns of existing rail infrastructure.
This blogpost lays out a pretty convincing case that difference between Japanese and (say) American zoning laws are the main reason for this type of "continuous urbanization" in Japan.
It's not unique to Japan: many other Asian cities exhibit dynamic, multi-polar cities and extraordinary continuous density.
> 1- Zoning is a national law, not a municipal by-law
This is pretty much it.The fact that strong central governments can ensure good density and growth means that Asian cities are growing as needed to serve the national economic agenda. In the West, particularly Europe and the US, cities are ruled by local municipal powers who all have strong incentives to stop growth and increase prices/rents. There's really no mystery here.
* As noted, the zoning laws and general policies surrounding real estate have made it so that residential development keeps with demand. In every developed area, the buildings consistently climb to 8 stories high. I can't speak for Europe but this is something the U.S. simply cannot do within its existing policy regime, now that the automobile suburbs are built out. It's stuck with an escalating price bubble for the foreseeable future.
* The shinkansen itself cannot be underestimated either. Other cities tend to be like "suburbs of Tokyo" in the Japanese model, taking pressure off the city to go supertall.
* Although you can find neighborhoods with large office parks and highways(e.g. Akasaka), the zoning laws permit using your home to do business, which reduces average commutes and requirements for dedicated workspace, and goes a long way to explain the high degree of similarity between any two random urban streets in the country.
* The rail lines are also real estate companies, which creates a virtuous cycle around every transit stop to pile on amenities and maximize the attractiveness of each neighborhood. You have to go quite far out to find "sleepy" train stations.
* The keiretsu business model that built contemporary Japan is likewise intertwined with the transit and development pattern. The economics of the conglomerate dictate that business units should complement and support each other which allows for a certain degree of robustness, further encouraging neighborhoods that have many similar patterns. Even in Sapporo, with it's 19th-century gridded blocks designed by an American, the overall development is distinctly Japanese. U.S. style patterns tend to favor superlative extremes of optimization - the biggest, the fastest, the most profitable - and not average case outcomes. Manhattan may be exceptional simply because we've assumed every city should be exceptional.
* Tokyo is in a very active earthquake zone. Building earthquake-proof supertalls is even more expensive than usual, and not practical in the large portions of the city built on reclaimed land.
* Sunshine laws make it very difficult to build tall buildings even when zoning regulations otherwise allow it. This is why you often see weirdly shaped (triangular etc) buildings in improbable places.
Shinkansen commuting is also a fairly marginal phenomenon, the cost is prohibitive and I recall reading that the number of active Shinkansen commuter passes is measured in the thousands (for a megalopolis of 30 million plus).
I agree. Tokyo is my favourite city, for many reasons. Livability and comfort being two of the top. I've already lived mid-term in Tokyo, and I see myself returning in the future.
right. it looks a bit ugly at first if you're not into it, but living there after a while it all really makes sense. great people too generally - friendly and welcoming like in a big village, yet leaving you alone when you need it like in a big city. perfect for my tastes.
And perhaps the most important difference from the American economy:
Housing is viewed as a disposable commodity, with an approximately 30 year shelf-life and depreciating value. Land in Japan does hold its value, but (at least post-bubble) at nowhere near the level of insane, speculative fluctuations driven by US-style “financialization.” Land values mostly track with inflation, like they did in the US before housing became an investment vehicle for Global Capital.
yes, it's unfortunate that we lose sight of the main purpose of a home: to be a safe place to live and commune with others, not a cold financial instrument.
when you think about how housing prices should change, you'd expect it to correlate with a macroeconomic growth measure like gdp or inflation. if it doesn't, then you'd probably first expect something obvious like
(1) many people are looking for homes (high demand)
(2) people are fighting fiercely for prized urban properties (focused demand/limited supply)
(3) we've run out of land/homes (short supply)
but rather, it's excess capital chasing excess returns, to accumulate more excess capital, with no thought to the underlying asset.
> this is something the U.S. simply cannot do within its existing policy regime
The only policy regime preventing this is a cultural one. Americans think of their houses as piggy banks, not places for them (and their neighbours) to live.
The development permissiveness in Tokyo is amazing. In the US, space under overpasses is wasted, and often becomes desolate and quite scary for pedestrians to traverse. In Tokyo, you’ll see retail built under the overpass.
After thinking a bit more, I also remembered Pershing Square Cafe underneath Park Avenue across from Grand Central Station. Kinda the same thing on a smaller scale.
Berlin has the S-Bahn, which is an elevated train network, and so there are a lot of overpasses, and elevators, stairs, and so on. And big stations are full of shops. Pretty great.
> I’m still at a loss as to what caused this dynamic. My best guess is that the city developed alongside a mature, well-supported rail system that could whisk people from one part of the city to another, so the advantage of being in the middle of it all was lower. Manhattan’s subway system could be described in much the same way though, so this answer is not satisfying.
I think the author is correct in saying that it's due to the rail system. The difference between Tokyo and New York is that the Yamanote line runs in a circle, whereas Manhattan's subway lines don't.
The Yamanote line not only connects most of Tokyo's major districts (Shinjuku, Shibuya, Ginza, Ikebukuro, etc.), but it's also the line that connects other train lines together. It's kind of like the beltway/orbital/ring road of a typical American city, just that it's a railway not a freeway. My understanding is that after WW2 when the city was being rebuilt, it was a conscious decision not to define a central neighborhood, and instead have all the lines connect to any station on the Yamanote line instead.
Note that this is not a Japanese thing - Osaka's tall buildings are clustered around Umeda, and Nagoya's are in the area around Shin-Nagoya station. Those cities also don't have a central circle line like the one Tokyo has.
Ah you're right, I don't know how I missed that! I guess it just doesn't follow the city's main neighborhoods like Tokyo's one does. There goes my theory...
It's pretty depressing that Tokyo is kinda the high point of urbanism and even it has so many issues. The bar is set really low for nice cities
Tokyo has many city centers, but I think it fails pretty badly when it comes to office space distribution. There's still concentration of large companies in a couple central areas, meaning everyone is going into a couple spots every morning, and leaving every evening.
There's a big status thing involved where companies want to be in the good neighborhoods so their address tells people that the company's successful. It means that huge swaths of the city are empty during the day. This compounds into lack of as good public services in certain areas, basically creating dreadful suburb communities where there's nothing but housing and convenience stores for miles, and local businesses can't survive on the paltry foot traffic
Meanwhile everyone and their dog is in the same 5 neighborhoods
Tokyo is lucky in that there's permissive zoning, so theoretically this can be corrected. But there's almost no incentive to do so because there's still a lot of prime construction to be had in other parts of the city
The city is going to end up with a bunch of holes in its urban planning because theres not much glory in developing the less popular places
The dark side of the transit network being in place is that there's not as much incentive to ensure local growth. So everything is a 40 minute train ride away.
(To a greater extent the Shinkansen did this for neighboring cities. You can now get to Tokyo in an hour, so there is less pressure to have stuff in your town. People end up getting used to it and moving to Tokyo)
My understanding was that the actual ground in NYC roughly determined building height. Manhattan was originally a swampy island, so you need to find something hard underneath (bedrock) to build on if you want 50 stories.
Depth to Bedrock and the Formation of the Manhattan Skyline, 1890–1915
New York City historiography holds that Manhattan developed two business centers—downtown and midtown—because the bedrock is close to the surface at these locations, with a bedrock “valley” in between. This article is the first effort to measure the effect of depth to bedrock on construction costs and the location of skyscrapers. We find that while depth to bedrock had a modest effect on costs (up to 7 percent), it had relatively little influence on the location of skyscrapers.
London is similar to Tokyo. After the Second World War, Blitzed-out land was bought cheaply by local councils and high-rise, high-density (but generally good quality) housing was built on those sites, hence an uneven spread of high-rise buildings throughout the city. Historically though the tallest buildings had to be church spires and the first true "skyscraper" in London was probably the Post Office Tower (now BT Tower) in Fitzrovia in the 1960s. It now stands pretty isolated as most of the tallest buildings in London are concentrated around the City and Canary Wharf, in the east and east docklands areas.
Looking out over Tokyo from one of the city centres (Tokyo is really a prefecture, and a collection of smaller cities), you can see that the other cities look quite built up, and the areas between are the smaller buildings the author is talking about.
I don't think the image provided does it justice, as it just showed a small subsection of Tokyo, and not what the city actually looks like.
One factor not mentioned is that Tokyo was almost completely destroyed by bombs and especially by firestorms during World War 2. This required rebuilding much of the city after the war. Cities like New York have never been destroyed. San Francisco was destroyed by earthquake much further back (in 1906 before modern skyscrapers were common). So some of what we see here may be rethinking following destruction as well as other factors.
On the other hand, the cities in the US that were built out postwar tend to be sprawly and not very dense. San Francisco arguably got lucky to be (re-)built before detached single family homes were in vogue.
Even the US cities that have grown post war generally had their city centers long before the war, and their tallest buildings tend to be there. This is about tall building density, not population density.
Not only was Tokyo destroyed in the war, the previous business and financial class was largely destroyed in the occupation. So the ones doing the rebuilding had ties to new areas of the city. Plus, having just witnessed their city being destroyed, the fear of it happening again would inspire them to limit clustering.
How can the author claim the lack of density in a city such as Tokyo when it has districts like Shibuya or Akiba which you really can't go any more dense without adding stories to the existing building.
51 comments
[ 3.0 ms ] story [ 100 ms ] threadIn the absence of this technology, I've always assumed people tended to stay more local, which meant that cities ended up a cluster of local communities, rather than one large central district with a large number of commuters.
That said, this seems to conflict with the example of Tokyo, which I understand was significantly rebuilt after the second world war.
Roughly half of Tokyo was destroyed by US bombing, I'd imagine the practical and psychological result of this plays a large role in this phenomenon. And many European cities faced intense bombing in the war too, so there may be some link to what you saw.
There seem to have been two choices for the rebuilding: rebuild what was there, maybe with better trams and subways; or allow Modernist planners to take over and build Brutalist suburban car-centered systems.
e.g. Coventry: https://www.theguardian.com/cities/2017/mar/14/brutal-inspir...
vs Dresden: https://www.gettyimages.co.uk/event/aerial-views-of-historic...
There are many zoning laws to ensure that cities provide good quality of life by approximating the "urban village".
This is done by keeping the density uniform and avoiding overly specialized areas like shopping malls, company campuses, sprawls and so on.
It allows people to walk or use public transport for a short time for most of your daily needs.
Of course there are economical ramifications: there are laws requiring a minimum building size, in order to provide housing at reasonable prices, instead of setting a maximum to make prices go up.
I don't think this is necessarily correct. Dublin is a fairly non-dense city by European standards, and always has been, but it is currently densifying, and the number of commuters using cars for their work commute is falling.
The pattern in the previous bubble was increased sprawl, here, but densification definitely seems to be the way it's going now.
There are all sorts of possible reasons for this; traffic has gotten worse, public transport has somewhat improved, and zoning rules require minimum densities in some places.
Paris and London were rebuilt to the maximum height that was practical then.
The rennovation - or destruction of old Paris really - was dramatic:
https://en.wikipedia.org/wiki/Haussmann%27s_renovation_of_Pa...
It's mainly zoning since the 1970s that has caused prices to shoot up. If London was allowed to go up and out more prices wouldn't be do high. It's similar for central Paris.
I always thought the urbanization of Japan was mostly limited by natural conditions.
Indeed, there is no point in making tall buildings if they are more likely to fall and cost more in repairs and anti-seismic technology
Japan didn't allow any tall buildings until 1963, so the development of tall urban centers came very late, and as others have mentioned, followed patterns of existing rail infrastructure.
http://urbankchoze.blogspot.com/2014/04/japanese-zoning.html
> 1- Zoning is a national law, not a municipal by-law
This is pretty much it.The fact that strong central governments can ensure good density and growth means that Asian cities are growing as needed to serve the national economic agenda. In the West, particularly Europe and the US, cities are ruled by local municipal powers who all have strong incentives to stop growth and increase prices/rents. There's really no mystery here.
https://www.economist.com/china/2018/06/23/china-is-trying-t...
* As noted, the zoning laws and general policies surrounding real estate have made it so that residential development keeps with demand. In every developed area, the buildings consistently climb to 8 stories high. I can't speak for Europe but this is something the U.S. simply cannot do within its existing policy regime, now that the automobile suburbs are built out. It's stuck with an escalating price bubble for the foreseeable future.
* The shinkansen itself cannot be underestimated either. Other cities tend to be like "suburbs of Tokyo" in the Japanese model, taking pressure off the city to go supertall.
* Although you can find neighborhoods with large office parks and highways(e.g. Akasaka), the zoning laws permit using your home to do business, which reduces average commutes and requirements for dedicated workspace, and goes a long way to explain the high degree of similarity between any two random urban streets in the country.
* The rail lines are also real estate companies, which creates a virtuous cycle around every transit stop to pile on amenities and maximize the attractiveness of each neighborhood. You have to go quite far out to find "sleepy" train stations.
* The keiretsu business model that built contemporary Japan is likewise intertwined with the transit and development pattern. The economics of the conglomerate dictate that business units should complement and support each other which allows for a certain degree of robustness, further encouraging neighborhoods that have many similar patterns. Even in Sapporo, with it's 19th-century gridded blocks designed by an American, the overall development is distinctly Japanese. U.S. style patterns tend to favor superlative extremes of optimization - the biggest, the fastest, the most profitable - and not average case outcomes. Manhattan may be exceptional simply because we've assumed every city should be exceptional.
* Tokyo is in a very active earthquake zone. Building earthquake-proof supertalls is even more expensive than usual, and not practical in the large portions of the city built on reclaimed land.
* Sunshine laws make it very difficult to build tall buildings even when zoning regulations otherwise allow it. This is why you often see weirdly shaped (triangular etc) buildings in improbable places.
Shinkansen commuting is also a fairly marginal phenomenon, the cost is prohibitive and I recall reading that the number of active Shinkansen commuter passes is measured in the thousands (for a megalopolis of 30 million plus).
The Shibuya Cast building opened last year. 30 stories?
The Stream building , 38 stories hasn't opened yet but it's close. Google Tokyo will move there in late 2019.
The new station building is 35-40 stories
Across the tracks from. The Stream building a new 40 story building will start soon in the area with all the music instrument stores.
South exit, the new Tokyu Plaza will be 19 stories.
Up Tamagawa Dori across from The Cerulean is a new 2O story building .
Behind 109 a 29 story building is going in.
1/2 a block from Tokyu Hands a new 20 story building is up.
The new Parco will be 19 stories.
The Shibuya City Hall next to NHK is under construction 2 buildings . one 20 stories. the other 39
Right across from them another 20+ story building just finished.
Behind Tower Records an 20 story hotel is going in.
In a Omotesando in the neighborhood behind the Apple Store a giant 30+ story complex is going in.
Only one has not started construction the businesses in the area to be reconstructed are closing.
Oh and not Shibuya but there's a 41 story apartment building going in at Musashi Koyama station.
Housing is viewed as a disposable commodity, with an approximately 30 year shelf-life and depreciating value. Land in Japan does hold its value, but (at least post-bubble) at nowhere near the level of insane, speculative fluctuations driven by US-style “financialization.” Land values mostly track with inflation, like they did in the US before housing became an investment vehicle for Global Capital.
when you think about how housing prices should change, you'd expect it to correlate with a macroeconomic growth measure like gdp or inflation. if it doesn't, then you'd probably first expect something obvious like
(1) many people are looking for homes (high demand)
(2) people are fighting fiercely for prized urban properties (focused demand/limited supply)
(3) we've run out of land/homes (short supply)
but rather, it's excess capital chasing excess returns, to accumulate more excess capital, with no thought to the underlying asset.
The only policy regime preventing this is a cultural one. Americans think of their houses as piggy banks, not places for them (and their neighbours) to live.
http://bedfordandbowery.com/2016/12/your-next-favorite-book-...
http://gothamist.com/2014/11/18/dim_sum_dance_party_you_guys...
https://www.google.com/maps/@40.7521215,-73.9777846,3a,75y,1...
I think the author is correct in saying that it's due to the rail system. The difference between Tokyo and New York is that the Yamanote line runs in a circle, whereas Manhattan's subway lines don't.
The Yamanote line not only connects most of Tokyo's major districts (Shinjuku, Shibuya, Ginza, Ikebukuro, etc.), but it's also the line that connects other train lines together. It's kind of like the beltway/orbital/ring road of a typical American city, just that it's a railway not a freeway. My understanding is that after WW2 when the city was being rebuilt, it was a conscious decision not to define a central neighborhood, and instead have all the lines connect to any station on the Yamanote line instead.
Note that this is not a Japanese thing - Osaka's tall buildings are clustered around Umeda, and Nagoya's are in the area around Shin-Nagoya station. Those cities also don't have a central circle line like the one Tokyo has.
The main stations along this line are probably Kyobashi and Tennoji, the latter of which has at least the tallest building in Japan.
I would mainly divide Osaka in two centers, with Umeda being the CBD and Namba/Shinsaibashi area to be the center for people going out.
Tokyo has many city centers, but I think it fails pretty badly when it comes to office space distribution. There's still concentration of large companies in a couple central areas, meaning everyone is going into a couple spots every morning, and leaving every evening.
There's a big status thing involved where companies want to be in the good neighborhoods so their address tells people that the company's successful. It means that huge swaths of the city are empty during the day. This compounds into lack of as good public services in certain areas, basically creating dreadful suburb communities where there's nothing but housing and convenience stores for miles, and local businesses can't survive on the paltry foot traffic
Meanwhile everyone and their dog is in the same 5 neighborhoods
Tokyo is lucky in that there's permissive zoning, so theoretically this can be corrected. But there's almost no incentive to do so because there's still a lot of prime construction to be had in other parts of the city
The city is going to end up with a bunch of holes in its urban planning because theres not much glory in developing the less popular places
The dark side of the transit network being in place is that there's not as much incentive to ensure local growth. So everything is a 40 minute train ride away.
(To a greater extent the Shinkansen did this for neighboring cities. You can now get to Tokyo in an hour, so there is less pressure to have stuff in your town. People end up getting used to it and moving to Tokyo)
For height limits you have the architectural stutter https://www.architectural-review.com/Pictures/web/p/y/l/spee...
San Francisco vs Japan seems very much like monocentric vs polycentric development https://architecturehereandthere.files.wordpress.com/2017/01...
-- and organic expansion through duplication https://3.bp.blogspot.com/-5JgoOuUmeDg/WMhJTUWeNWI/AAAAAAAAr...
In this talk here he talks about the classical vs vernacular layout styles https://youtu.be/kFiYL8AvvnY?t=1150
This is actually something of a myth [1].
[1] https://www.google.com/amp/observer.com/2012/01/uncanny-vall...
New York City historiography holds that Manhattan developed two business centers—downtown and midtown—because the bedrock is close to the surface at these locations, with a bedrock “valley” in between. This article is the first effort to measure the effect of depth to bedrock on construction costs and the location of skyscrapers. We find that while depth to bedrock had a modest effect on costs (up to 7 percent), it had relatively little influence on the location of skyscrapers.
https://www.cambridge.org/core/journals/journal-of-economic-...
I don't think the image provided does it justice, as it just showed a small subsection of Tokyo, and not what the city actually looks like.