GDP per capita of the European and North American metro areas

[SHiRO]

Originally Posted by Crawford

London does not have 20 million people. That's absurd by any definition.

Probably around 12-15 million people using the most expansive U.S. CSA definition.

How can you claim such an utterly moronic thing when just a couple of posts above there is a whole PDF which states that metro London in 2007 had 13-14 million people (depending on the "building blocks" you use) AS A COMPARABLE METRO AREA TO A MSA?
And we know that Greater London alone grew by 1.4 million (number used in the PDF for GL is 7.1 million and GL currently sits at 8.5 million).
So currently an equivalent for a MSA has to be at least 14.5-16 million (exactely what I claimed).
Any CSA equivalent has to be >16 million. To claim 12-15 million at its most expansive is nothing short of ridiculous and denying the facts (as usual).

[muppet]

ok folks, this is how it works.

The US CSA counts are VERY generous, and getting increasingly so. It's basically a case of different metro definitions.

They count the urbanised sprawl, but then started counting commuting rates from immediate satellite towns. This could be as low as 10%. We'll call this Metro 1 stage. from back in the early noughties.

Then they widened it to include the satellite towns that commuted to the next town along (as opposed to the centre city), then enlargened it again to 'might aswell-add-on' their respective counties. By this stage the catchment area became huge, with thousands of km encompassing large swathes of farmland and forest, or in the case of LA, mountain and desert too. NYC even included Pike/ Pitt? County on the grounds that they were receiving the NYC TV signal (though I don't know why they wouldn't just use the commuting criteria as that would have made it). This we'll call Metro 2 and 3 stages.

Then they added again even more towns along, effectively doubling even that ginormous area. In short less than 1-3% commute into the actual old city limits, and the area verging on the size of mainland Scotland or Serbia by this stage. Metro 4 stage


European counts do not take commuting into account - but if they did, well the Blue Banana is the densest network of urbanity in the world, and bearing in mind youre only counting the commuting between small towns, it would stretch into Italy, including just as 'short' hops across you know, a sea or mountain chain.

The same area of NYC CSA over the London-SE England- E England (both 20,000 sq km) catchment that the Greater London Authority uses in it's London Plan is about the same. 21 million by now for the London 'mega-region'

http://www.america2050.org/Healdsbur...e_pp_59-67.pdf


The density of 'commuter-hoods' (new developments in light pink) created by the Green Belt - note the absolute density of them and the utter failure of the Green Belt. This would be the London equivalent to Metro 1 stage - immediate satellite towns and dormitory developments. London forces it's sprawl into a peppering of high density developments, whereas across the pond it's blanket suburbia. In this regard NYC is undeniably the higher count for a contiguous city - 16.7 million as opposed to 12 million, but that depends if you want to discount the commuter peppering as 'suburbs' or not:


http://geology.com/world-cities/lond...ed-kingdom.jpg


This is the metro 2-3 stage - including the proper (read: larger regional) satellite towns not just the immediate ones in the vicinity. This is the most used 'metro' when we talk about the London metro area, but not the one for NYC. NYC would at this stage be at 18 million and London at 15 million.





This is the metro 4 stage - including every town that commutes into the next one, and their respective counties thrown in for luck. In London they refer this as the mega-region, in NYC it's just NYC metro. To reiterate both are at the 20,000 sq km stage by now and equally dense, and the numbers have now 'levelled off' with 21 million for both cities:

[SHiRO]

Originally Posted by 10023

ShiRo,

The 11.9m figure was the number given by the source I found that also offered GDP data (Eurostat). It was close enough to the 13m definition for the per capita GDP to be relevant. That's it.

No it's not close enough. 14-16 million is close enough.

The most common definition of metro London, and the one used by the Greater London Authority (see my link), has a population of 13 million people. I have already acknowledged that the population of this region has likely increased by 5-10% (650k to 1.3m people) since that data was collected, so the population of metro London is probably pushing 14 million these days.

No, like Crawford, you don't get to arbitrarily decide what the metro population for London is either.
And you don't get to arbitrarily decide the growth in the region either. We KNOW that GL grew by 1.4 million and the region by close to 2 million. These are easily verifiable facts.
So going back to the real numbers as opposed to the ones made up by you, this gets us a London "MSA" of 14.5-16 million.

If you include "cross-commuting" between towns from which people don't actually commute to London itself, then practically the whole of England from Brighton to Leeds is a single metropolitan area. It's a densely populated country with a good transportation network. That doesn't make these places a single metropolitan entity, and including most of Hertfordshire, let alone areas outside of the Home Counties, is ludicrous. You'd have to include Oxford, Swindon, Bournemouth, Northampton. These areas commute to cities like Birmingham and Bristol as well, so why don't we just toss those into London as well.

Nope, wrong again. The outer perimeter of the London commuter zone/CSA is roughly as follows: Ipswich-Cambridge-Peterborough-Milton Keynes-Oxford-Basingstoke-Brighton-Ashford. This has been established by people with actual interest and knowledge about the area over a decade ago. All these places are within a 100 km or an hour train ride radius. Nothing special or hyperbolic about it, especially considering the far flung areas supposedly part of American CSAs.

American metro area definitions are more expansive because American metros ARE bigger. People commute longer distances. When I tell people that I used to drive 3-3.5 hours each way every weekend between Manhattan and Montauk, they look at me like I'm crazy. Here it's 4 hours by train to Edinburgh, but no one (at least not a statistically relevant number of people) commutes to Birmingham, let alone the North of England or Scotland. It's partly cultural and it's partly due to sparse population between urban centers in the US (away from the Northeast US at least).

Ah the old "the USA is special" card... Except noone is talking about 4 hour commutes or Leeds being part of London metro. It's just the usual smoke and mirrors. I've proven to you USING YOUR OWN LINK that London MSA is 14-16 million and not the 11.9 million you claimed. And if you would stop and think instead of being a contrarian know it all, it isn't at all out of the ordinary to include satellite cities less than an hour away by train in a London CSA equivalent.

You have always claimed some sort of superior knowledge of all things European in arguments against American forumers, but as a half-Brit and London resident, I am telling you that you are full of shit. Also, say more things like "you probably don't know how a CSA works" and I'm just going to block you, and you can have fun arguing with yourself.

Please do and while you are at it block threads like these where you offer nothing of value besides the usual denial of facts and know-it-all behavior.

[SHiRO]

Originally Posted by New Brisavoine

Lille is also less than an hour by train from Paris.

First of all Lille is a large city in a metro area of over 1 million of its own. It's doubtfull that it sends 15% of its working population to the Paris area which isn't even adjacent to it. It's only an hour away on the TGV, normal trains take close to 2 hours (?). The distance between Lille and Paris is 200 km.
Peterborough is a city of 190,000 that is 53 minutes away by normal train. Dozens of trains depart from there each morning to King's Cross. It is also directly adjacent to the London metro area so all we really need is enough cross commuting between the two.

What you are doing is just your usual smoke and mirrors, bringing absurd comparisons. 10023 did it too with Wilmington and Providence. It only weakens your argument.

As often said, CSA is a concept that works in North America, where densities are low, and wouldn't work in Europe, especially in the dense parts of Europe, like England. So it makes sense to use CSA for the US metro areas (it makes perfect sense for SF and LA for instance, since their MSAs don't even include all of their urban areas), but it makes no sense to try to carve "CSA" territories in Europe.

Absolute rubbish. You can just as easily define CSA like areas in Europe, just like there are CSAs in the US Northeast no problem.

[muppet]

in short:

1. city contiguous: NYC > London

However, city + suburbs: NYC>London still, but London catching up considerably (the term 'suburbs' if you are fine with taking 'Green Belt' peppering as an urban contiguity. If not just do 'city contiguous' above).

city metro 1 : NYC > London. city + immediate satellite towns This is the stage London uses when it talks about its' 'metro'.

city metro 2: NYC > London city+ all 'satellite' towns

city metro 3: NYC = London, city+ all 'satellite' towns + counties. Both cities about 18 million

city metro 4 within 20,000 sq km: NYC = London city+ all 'satellite' towns + counties + 'sattelite' towns to the counties of the 'sattelites' - and their counties. Both cities now about 21 million. This is the stage NYC uses when it talks about its 'metro'. To reiterate only 1-3% of the enlargened catchments actually commute into the city by this stage.


however...

city-metro-3-counted-the-same-way-as-NYC-does-on-commuting-between-towns: London > NYC as that would mean most of England (near 47 million in area the size of Maine, with very dense commuting).

[New Brisavoine]

Guys, the larger an area you take for London, the lower its GDP per capita is going to be, and so the lower London is going to rank in the list. Therefore I don't understand why some people here insist on taking an area as large as possible for London.

In the list I've used Greater London + the 6 home counties, which is the area that corresponds the best to a metro area of London (Greater London alone would be too small, and more than the 6 home counties would be quite large). Again there are no GDP data at a sub-county level in England, so either you include all 6 home counties, or you include none. Can't include just the parts of them closest to Greater London.

This territory (Greater London + 6 home counties) has a land area of 15,110 km² (5,834 mi²) and a population of 15,290,600 (in mid-2011). This compares to a land area of 12,012 km² (4,638 mi²) for the Paris Region, and 10,820 km² (4,178 mi²) for the Rhine-Ruhr (as per the definition I've used in this list).

As for the London metro area proper, the only official definition we have is that of the London LUZ provided by Eurostat with assistance from British statisticians. The London LUZ has a land area of 8,922 km² (3,445 mi²), and a population of 13,011,200 (in mid-2011). There are no GDP figures available for this territory. Its GDP per capita would be a bit higher than the one I've given in the list, but not by very much (perhaps, say, $5,000 more at the max).

[SHiRO]

I don't care about which is bigger or has the higher GDP/capita nor do I care about "carving out" an as large as possible area for whatever city.

I care about morons maintaining that NYC has 23 million and in the same breath claiming London has 12 million "at its most expansive".



And what you are doing Brisavoine and what makes your "lists" less than usefull is using things that are not metro areas as metro area proxies and then use the excuse that there is no data for the entities that really make up metro areas. In case of London this time it kinda works out, but Barcelona province is not Barcelona metro area (for instance) and things like that skew the results big time.

And Brisavoine, monsieur "you can't have CSAs in Europe", then why do you insist on comparing US CSAs to European provinces/arrondisements/LUZ (anything but metro areas it seems). One would think the first thing you would start with is to use US MSAs no?

[kool maudit]

i don't think commuting patterns in western europe are as centralized, as the smaller towns and cities here GENERALLY function in a more self-contained degree than do those of north america. hence one can be fairly close to copenhagen or hamburg, but you aren't "copenhagen" in the same way that a similarly outlying area in southern ontario is "toronto."

the trouble is that people use commuter-shed figures to demonstrate city-size rankings, and so you get absurdities in which boston is seven times the size of amsterdam or whatever, when on the ground they are very similar in scale and in your sense of how much activity surrounds you. north americans move around more, both for cultural reasons and because many north american small towns and cities are rather degraded.

europe's big cities do not absorb their surroundings to the same degree, and this is due in part to the fact that these surroundings are often as old and as identity-possessed (helsingør has a meaning apart from copenhagen; laval has no meaning apart from montreal) as the central cities.

[New Brisavoine]

Originally Posted by SHiRO

I care about morons maintaining that NYC has 23 million and in the same breath claiming London has 12 million "at its most expansive".

Personnally I would say New York 20 million, London 13 million. That's about the ratio.

Originally Posted by SHiRO

but Barcelona province is not Barcelona metro area (for instance) and things like that skew the results big time.

Barcelona province had 5,511,900 inhabitants in mid-2011. The Barcelona metro area is usually listed as having 5 million inhabitants. So it's very close. The province is perhaps 10% too large in terms of population compared to the metro area. Let's recall that most of the Barcelona province in uninhabited, and 90% of its population live in the metro area of Barcelona. If we could have a GDP per capita figure for the metro area proper, it wouldn't be very different from the one for the province that is in my list.

Originally Posted by SHiRO

why do you insist on comparing US CSAs to European provinces/arrondisements/LUZ (anything but metro areas it seems). One would think the first thing you would start with is to use US MSAs no?

I use CSAs essentially because of LA and SF Bay Area. So I use CSAs for the other US metro areas as well for the sake of comparability. Usually the MSA figure is not very different from the CSA figure anyway. When the MSA figure is quite different from the CSA figure I have indicated it (Seattle, Washington, Boston).

[10023]

Originally Posted by New Brisavoine

Personnally I would say New York 20 million, London 13 million. That's about the ratio.

This is correct.

I also agree with New Bisavoine that the real reason for looking at CSAs and North America is when there are two MSAs which clearly constitute a single urban region, but are split up at the MSA level because of two or more roughly equal-sized "hubs". The SF Bay Area is the perfect example of this. In Europe, the Ruhr or the Rotterdam-Delft-Hague metroplex are good examples. Extending London all the way to Peterborough is not. That's like extending Chicago's metro to Rockford.

London to Peterborough:
https://goo.gl/maps/KYgCy

Chicago to Rockford, IL:
https://goo.gl/maps/C0VEL


You say "it's one hour by train" as if, in a densely populated country with fast trains, that was the only determinant of what is or isn't a single urban area. Fucking Paris is 2 hours by train...

ShiRo you are as ignorant about North American cities as you claim Americans are about European ones. It's clear that for you, everything is a competition between Europe and the US, but please stop being so obnoxious. New York and London are both great global cities, I know both quite well, but New York is quite simply a lot bigger than London.


For what it's worth, I do think the US CSA definitions can often be too expansive. For instance it's bullshit to include DeKalb in Chicago... the metro really ends at Aurora, Geneva, etc. But because the US is so sparsely populated outside of its major urban centers, extending the CSA into DeKalb County doesn't distort the numbers too much (it only adds 105k people). In England or other densely populated parts of Europe, using this sort of spurious definition can add 50% to the city's population, and is clearly a load of crap.

Location of DeKalb:
https://goo.gl/maps/RZyxR

About:
http://en.wikipedia.org/wiki/DeKalb_County,_Illinois

[SHiRO]

Originally Posted by kool maudit

i don't think commuting patterns in western europe are as centralized, as the smaller towns and cities here GENERALLY function in a more self-contained degree than do those of north america. hence one can be fairly close to copenhagen or hamburg, but you aren't "copenhagen" in the same way that a similarly outlying area in southern ontario is "toronto."

the trouble is that people use commuter-shed figures to demonstrate city-size rankings, and so you get absurdities in which boston is seven times the size of amsterdam or whatever, when on the ground they are very similar in scale and in your sense of how much activity surrounds you. north americans move around more, both for cultural reasons and because many north american small towns and cities are rather degraded.

europe's big cities do not absorb their surroundings to the same degree, and this is due in part to the fact that these surroundings are often as old and as identity-possessed (helsingør has a meaning apart from copenhagen; laval has no meaning apart from montreal) as the central cities.

I couldn't agree less. And another perfect example of a post talking about "thinking" rather than "knowing".

Europeans do commute. Europeans commute long distances (on average not as long as Americans but still). There are bedroom communities, satellite cities and purposely build new developments in abundance.

And first and foremost, there are places on the internet that analyse and discuss these things just like American cities and suburbia are analysed and discussed at great lenght here. It is beyond appearant that noone here visits these places...

[SHiRO]

Originally Posted by New Brisavoine

Personnally I would say New York 20 million, London 13 million. That's about the ratio.

Absolutely ridiculous and it is quite appearant that you have an agenda downplaying London. Never change Brisavoine...

How anyone can and dares to maintain such a ridiculous claim WITH THAT PDF ON THE PREVIOUS PAGE is beyond me...

London
City: 8.5 million
Urban: 10 million (due to Green Belt)
Metro area: 14-16 million
Combined metro area: 20 million

New York
City: 8.4 million
Urban: 18 million (due to connectivity by low density sprawl)
MSA: 19.9 million
CSA: 23.5 million

This is the ratio. See the PDF by the GLA on the previous page as a reference for the London figures and the US Census for the New York figures.

What you are claiming is incorrect both by official figures and by common sense.

This is Paris btw:

City: 2.2 million
Urban: 10.5 million
Metro area: 12.3 million (an MSA equivalent would be slightly higher 13-14 million maybe)
Combined metro area: 14-16 million

[SHiRO]

Originally Posted by 10023

the Rotterdam-Delft-Hague metroplex

(What?)

Extending London all the way to Peterborough is not. That's like extending Chicago's metro to Rockford.

I thought it was like extending NYC to Wilmington or Providence?

You say "it's one hour by train" as if, in a densely populated country with fast trains, that was the only determinant of what is or isn't a single urban area. Fucking Paris is 2 hours by train...

More false equivalencies. LDN to Paris is 2 hours 15 minutes on the Eurostar and a ticket costs €223,50...

And we're talking about combined metro areas, not "single urban areas", you can't even keep your definitions straight. Are you really that confused?
Dozens of commuter trains leave Peterborough for London each morning, between 6 and 7 AM there are 9 alone!

ShiRo you are as ignorant about North American cities as you claim Americans are about European ones. It's clear that for you, everything is a competition between Europe and the US, but please stop being so obnoxious. New York and London are both great global cities, I know both quite well, but New York is quite simply a lot bigger than London.

The reality off course is that I have been studying cities in general all my life and American cities specifically on this forum for 15 years. I've been drawing maps, crunching numbers, looking over atlases since I was a kid. I even had a map of the US hanging over my bed. I can name every state capital and every city over 100,000, point them out on a map and have a rough idea of their population. I can draw a map of NYC from memory... I don't want to make assumptions about you, but I'm pretty sure I've put in the hours especially when it comes to cities in Europe/cities outside the US. (plus I didn't even make any claims about US cities other than citing their official US Census defined populations. Contrary to others who keep insisting making up their own numbers for London).

New York is not a lot bigger than London. City proper wise LDN has recently overtaken NYC but the urban cores are pretty much equal (maybe a slight edge to NY). NY however is surrounded by a lot of lower density sprawl which according to the US Census definition counts as urban area, whereas London's urban area gets cut short by the Green Belt (sort of). That doesn't mean that London as a functional city ends at the Green Belt, in fact what it means is that people commute from even further distances (leapfrogging, helped by the excellent transport connections).

But what baffles me the most is that YOU were the one who posted that PDF from the GLA. That PDF clearly states that metro London as an equivalent to a MSA had 13-14 million in 2007. Greater London grew by 1.4 million and the region by almost 2 million so the conclusion must be that in 2014 the London "MSA" is in the 14.5-16 million range.
So you are contradicting yourself by maintaining that it is 13 million, after you contradicted yourself earlier by previously claiming it was 11.9 million. So which is it 10023? Are you now also disagreeing with the PDF you yourself posted?

[lio45]

Originally Posted by muppet

This is the metro 4 stage - including every town that commutes into the next one...

FWIW, this (ridiculous) metric of two commuting steps would obviously have all of Philly within NYC, as there's commuting to both from places in central NJ.

This "Greater Princeton" is quite a bit bigger than London...

[Nouvellecosse]

Originally Posted by New Brisavoine

A fraction of the population, but a significant part of its economy.

Alberta, Yukon, Northwest Territories, Nunavut, and Newfoundland produce 20% of Canada's economy. Add in Saskatchewan (also a very high GDP per capita), and you've got 23.5% of Canada's economy.

Without these 6 provinces/territories, the GDP per capita of Canada in 2011 wouldn't have been US$51,791, but it would have been US$47,148, which is only slightly higher than the GDP per capita of Detroit CSA (US$46,802).

Those areas contain about 12% of the population, so 20% of the economy is certainly a higher than average proportion but the suggestion that Canada's productivity is created by resources rather than people in large cities is clearly false. Especially considering that the largest chunk of that population is in Alberta (about 3.6 milion)which contains two of Canada's ten largest cities and would have a good chunk of the GDP with or without the oil and gas industry. That sector is currently important to the economy there but it's far from the only industry.

[SignalHillHiker]

Originally Posted by muppet

ok folks, this is how it works.

The US CSA counts are VERY generous, and getting increasingly so. It's basically a case of different metro definitions.

They count the urbanised sprawl, but then started counting commuting rates from immediate satellite towns. This could be as low as 10%. We'll call this Metro 1 stage. from back in the early noughties.

Then they widened it to include the satellite towns that commuted to the next town along (as opposed to the city), and their respective counties, then did it again. At this stage the catchment area was huge, thousands of km encompassing of course large swathes of farmland and forest, or in the case of LA, mountain and desert too. In the case for NYc it even included Pike/ Pitt? County on the grounds that they were receiving the NYC TV signal (though I don't know why they wouldn't just use the commuting criteria as that would have made it). This we'll call Metro 2 and 3 stages.

Then they added again even more towns along, effectively doubling even that ginormous area. In short less than 1-3% commute into the actual old city limits, and the area verging on the size of mainland Scotland or Serbia by this stage. Metro 4 stage


European counts do not take commuting into account - but if they did, well the Blue Banana is the densest network of urbanity in the world, and bearing in mind youre only counting the commuting between small towns, it would stretch into Italy, including just as 'short' hops across you know, a sea or mountain chain.

The same area of NYC CSA over the London-SE England- E England (both 20,000 sq km) catchment that the Greater London Authority uses in it's London Plan is about the same. 21 million by now for the London 'mega-region'

http://www.america2050.org/Healdsbur...e_pp_59-67.pdf


The density of 'commuter-hoods' (new developments in light pink) created by the Green Belt - note the absolute density of them and the utter failure of the Green Belt. This would be the London equivalent to Metro 1 stage - immediate satellite towns and dormitory developments. London forces it's sprawl into a peppering of high density developments, whereas across the pond it's blanket suburbia. In this regard NYC is undeniably the higher count for a contiguous city - 16.7 million as opposed to 12 million, but that depends if you want to discount the commuter peppering as 'suburbs' or not:


http://geology.com/world-cities/lond...ed-kingdom.jpg


This is the metro 2-3 stage - including the proper (read: larger regional) satellite towns not just the immediate ones in the vicinity. This is the most used 'metro' when we talk about the London metro area, but not the one for NYC. NYC would at this stage be at 18 million and London at 15 million.





This is the metro 4 stage - including every town that commutes into the next one, and their respective counties thrown in for luck. In London they refer this as the mega-region, in NYC it's just NYC metro. To reiterate both are at the 20,000 sq km stage by now and equally dense, and the numbers have now 'levelled off' with 21 million for both cities:

Awesome post. Thanks for the info!

[SHiRO]

Indeed. I especially like the last two maps, both of which I hadn't seen before. That first satellite pic has become kind of a muppet staple I'm sure it has been posted about a hundred times on this forum by now (it's all good though).

The second one really shows that a London metro of 15 million is nothing outrageous and that in fact it can be justified on morfology alone (the US Census would probably treat most of it as a single urban area). Hard to imagen there are people still claiming that Reading (the biggest blob to the west of London) or Luton (the biggest blob to the north-northwest containing one of LDN's airports) are somehow not part of metro London.

The last map is also great and illustrates perfectly what the meaning of "combined metro area" is. Even if you omit the outlying satellite cities in this, you'll still end up with 18-20 million people in what is clearly an interconnected economic area centered on London. Also hard to imagen that people will deny this while accepting US CSAs which often times have far more flimsy connections and occupy relatively larger areas (put a map of the Atlanta CSA to scale next to it if you want to have a laugh at the hypocricy).

[Nouvellecosse]

That's interesting. I didn't realise London is actually just as large as NYC but of course it can be hard to find real apples to apples comparisons.

[SignalHillHiker]

I didn't realize London was so much larger than Paris. London certainly felt bigger, but not to the degree that it actually is.

Now, I've said that to friends on Facebook - most of whom have been to both - and they were all incredulous.

"London is a combination of places, so it can feel smaller than it is in any given place, but it is WAY bigger than Paris!"

[Boris2k7]

Originally Posted by Nouvellecosse

Alberta (about 3.6 milion)which contains two of Canada's ten largest cities and would have a good chunk of the GDP with or without the oil and gas industry.

Alberta's population is 4.1-4.2 million actually. The 3.6 million figure is from 2011.