Waning for decades, England’s big cities are growing fast again

[SHiRO]

Just 12 million as a metro area for London is completely ridiculous and not fair at all when comparing to North American ones! See what I mean with downplaying areas you're not familiar with? And are you really going to use 2001 figures after nito just showed the region grew about 2 million people over the last decade? The GLA definition is the one that gave London 18 million on 27,000 km2 in 2001. It has slightly different boundaries than the green on your map, this is because it includes some satellite cities that have very good connections to London and take you there in about one hour (Peterborough, Cambridge, Oxford). A lot of countryside is omitted in that definition. The New York CSA is 30,000 km2 so I don't see what's so outrageous about the London one.

Like I said; 14-16 million for a MSA like metro area and 18-20 for a CSA like one.

What are your thoughts on the Golden Horseshoe and the 33,000 km2 it takes up? Is it the GTA that grew by 972,000 or the Golden Horseshoe?

And Greater London is not a "large semi-metropolitan entity" it's exactely the same thing as the City of Toronto!

[Minato Ku]

Metropolitan areas are based on commute, not on size.
What is important is not the number of people living in a xxxx km2 area around London but the ratio of people commuting to London from a Y area.

There are also some fact, more a city is in a densely populated region more its metro area is compact in size.
Why ? Because there are less kilometers separating the important employment center in dense region than in empty one.

In France, Lille and Toulouse have a similar metro population but Lille metro is less than 1,000 km² while Toulouse métro is over 4,000 km².
Lille is in Blue Banana megalopolis, if you take 4,000 km² around Lille there is more inhabitants than in Toulouse but much of these inhabitants don't need to commute to Lille as they have other medium cities close to them.
In the 4,000 km² around Toulouse, there are nothing, people living here don't have much choise that working in Toulouse.

Toulouse

Lille


Don't forget a metropolitan area is based on commute, not on a geographic area while many people commute to London from the whole SE England, it is not enouth to make a metro area.

[SHiRO]

We've been over this before. If you apply US Census (commuter based!) formulas on London you got 18 million in 2001. nito is correct to point out that it's closer to 20 million probably now a decade later.

I'm really struggeling to see why for some people it's so hard to grasp that London is a HUGE commuter magnet. 12 million common! Paris is 12 million with the very strict INSEE formula (40% commuters instead of the 25%/15% of the US Census formulas) and Paris is a much more compact city in a much less densely populated area that draws commuters from places that are much closer by. It's all so damn obvious!

[nito]

Originally Posted by MonkeyRonin

Because Greater London accounts for the large majority of the hypothetical metropolitan London. Going by Nito's numbers of 9.2 million or 11.9 million, Greater London would account for 69-89% of the metro area, whereas those other municipalities range from 30-45% of the metro population - and which are fully built out. London still has plenty of untouched land for potential development (though I imagine most of it is protected).

As mentioned previously, the ONS and Eurostat have produced their own figures, but these can't be really compared to US figures. For instance the 9.2mn ONS figure is based on 75% working in the London area, while the figure used for US metro's is 25%. In essence, metro area comparisons are even less legitimate than city area comparisons.

Originally Posted by someone123

You say LA is an exception and grew as much as London, but LA is dead last in the graph, with less than 1/8 the population growth of London. Similarly I also pointed out that most of Toronto's population growth didn't make it onto that graph, and that's probably true of all four that are not London. I'm not sure that it's true of London.

The position of cities such as Los Angeles and Toronto is a reflection of low core growth and high periphery growth; presumably suburban sprawl, although that would require further research to confirm.

Originally Posted by MonkeyRonin

This isn't supposed to be a contest to see who's grown the most. The problem is the comparison of a large semi-metropolitan entity to smaller single-tier municipalities. But that comparison right there is at least a little more comparable than the one in the graph. Though for the record, Greater Toronto grew by 972,365 between 2001 and 2011.

I believe the confusion relates to the impact that the Green Belt had on sprawl in London, relative to North American cities where it got out of control. Instead intense development occurred in and around entirely new towns and former rural-focused towns & villages dotted across the South East and East of England. My home town for instance was previously a market town that sent meat and vegetable produce down to London. The food has now been replaced by bankers, lawyers and media-types that utilise the excellent rail links into Central London, and en-masse; the town has become a dormitory settlement.

Naturally the potential US-equivalent metro population of London is no way equivalent to the entire population of the South East & East of England, but a significant amount would most likely fall under such an area.

The following map I created earlier in the year is a tad better at explaining the situation; it shows a snapshot of the number of train services heading into Central London that arrive between 0800 and 0859 in the average morning. These services simply wouldn't be viable if there was a condensed metro area commuting population.




Finally, while Toronto as a city is smaller than London (difference of 942 sq km), the contrast with London and the Greater Toronto Area is even more stark (5,553 sq km). An alternative way to look at the situation would be to look at the individual cities population growth on a per sq km basis:

London - 542
Toronto - 212
New York - 138
Houston - 90
Los Angeles - 75

[Minato Ku]

Originally Posted by SHiRO

I'm really struggeling to see why for some people it's so hard to grasp that London is a HUGE commuter magnet. 12 million common! Paris is 12 million with the very strict INSEE formula (40% commuters instead of the 25%/15% of the US Census formulas) and Paris is a much more compact city in a much less densely populated area that draws commuters from places that are much closer by. It's all so damn obvious!

Who care of the compactness of the city, what is important is the proximity to other medium cities (large enouth to have their own commuter belt).
If you read well my previous message, you would see that if Lille has a much smaller commuter belt area than Toulouse, it is not because the city is more compact but because the city is closer to others medium cities, big enouth to have their own commuter belt.

More a city is in an empty area more its comuter belt is large in area.
Not in population but in area.
A lower density also means that road are less congested and train faster (because of less stop) making a long distance commute faster.

By exemple in Paris.
The closest metro areas with over 150,000 inhabitants of Paris are at over 100 km of Central Paris. Rouen, Orleans, Amiens.
The area around Paris is quite empty, if you take the metropolitan area and excluded Paris (population and area) we have a density of 121/km².

In London, Reading is at only 60 km, Luton at 50 km, Milton Keynes at 70km of Central London.
Right, there are a signifant flow of people who work in London (between 1,000 and 10,000), it is a lot of people but it is not enouth to be part of London metropolitan area by using a 25% commuter rate (even in mixing inbound and outbound commute).
To be part of London metro, these cities would need to be 2 to 5 times smaller in population.
(Maybe using CSA, Luton would be part of London metro but CSA tend to give exagerated number in density populated region).

In 2008, there were about 800,000 people commuting to London everyday, mostly comming to the closest suburbs (outside the GL) or exuburbs.
A lot of people come from futher but not enouth to make a metropolitan area.

[SHiRO]

Bollocks, Luton and MK are the very definition of commuter towns for London. MK was even specifically build for it. Reading is a satellite city if there ever was one. There's no question all these places are part of metro London!

[M II A II R II K]

Originally Posted by MonkeyRonin

This isn't supposed to be a contest to see who's grown the most.

It's supposed to be about British cities.

[Minato Ku]

Originally Posted by SHiRO

Bollocks, Luton and MK are the very definition of commuter towns for London. MK was even specifically build for it. Reading is a satellite city if there ever was one. There's no question all these places are part of metro London!

But none give over 25% of their working force to London, the minimum to be part of a metro area.

[M II A II R II K]

So defining factors are proximity, accessibility, and how economically tied it is to the main city. Particularly bedroom communities.

[SHiRO]

Originally Posted by Minato Ku

But none give over 25% of their working force to London, the minimum to be part of a metro area.

No that's just one way to define a metropolitan area, the way the US Census does it for what they call a MSA. And it's not as simple as "giving 25% of the workforce to London". First of all, what London? Greater London? The urban area? The counties containing the urban area?

http://en.wikipedia.org/wiki/Metropo...atistical_Area

In any case you haven't provided a source about the % of the workforce from Luton, MK or Reading working in "London" (again, what is London?) so please do so, because anyone can look at a map and see that Luton is part of the London metropolitan area based on proximity alone.

You can argue that places like Swindon, Southampton, Brighton, Ipswich, Cambridge, Peterborough, Northampton and Oxford are on the outer edge of London's commuter belt but Luton contains one of London's airports and MK was specifically build to be a satellite city to London.

Noone (to my knowledge) has ever attempted to calculate a London metro based on the MSA definition so I can't say for sure it would include Reading and MK. But the attempt has been made for a CSA and Reading and MK are definately included in that. Be my guest if you want to take it upon yourself to calculate a MSA for London.

[Minato Ku]

London as Greater London area.
A metropolitan area is not based on proximity but on commute pattern. As I earlier wrote a metro area located in a densely populated region spreads less than a metro area in an empty one.
If no attemp has been made to calculate a London MSA, so it is not possible to calculate a London CSA because CSA are based on MSA.
You cannot combine metropolitan area without calculate those.

Milton Keynes by exemple, 83% of the working force worked in Milton Keynes in 2008. The rest work mostly in close neighboring areas, while several thousand work in London, it is not enouth to make it part of London metro.
http://www.milton-keynes.gov.uk/coun...10.02.2011.pdf Page 12.

We all agree that London is a great and big city, that it had a high growth this decade but do we really need to inflate the data.

[SHiRO]

Originally Posted by Minato Ku

London as Greater London area.

But Greater London is not the London urban area. A entirely one-on-one comparison is not posible because of the different nature of the administrative entities. Which brings me to the next point:

If no attemp has been made to calculate a London MSA, so it is not possible to calculate a London CSA because CSA are based on MSA.
You cannot combine metropolitan area without calculate those.

You have a point here, but I guess they took Greater London or some other combination of entities as stand in for the "MSA". I don't remember as this was more than 10 years ago on SSC. I do remember on forumer having done a lot of work calculating all commuters from each individual council and it adding up to 14 million. Maybe they took that as the "MSA"?

A metropolitan area is not based on proximity but on commute pattern. As I earlier wrote a metro area located in a densely populated region spreads less than a metro area in an empty one.

You're contradicting yourself here. If a metro area (for which different definitions can apply, also ones without factoring in commuters) is solely based on commuting patterns where does density come into play? Of course those commuting patterns are also based on something, hint: proximity! And more density means more people in the proximity so there are all different kinds of correlations. But it's not correct to first state that metro areas are always and solely based on commuter patterns and then contradict yourself by bringing density into play.

Milton Keynes by exemple, 83% of the working force worked in Milton Keynes in 2008. The rest work mostly in close neighboring areas, while several thousand work in London, it is not enouth to make it part of London metro.
http://www.milton-keynes.gov.uk/coun...10.02.2011.pdf Page 12.

OK, so you're right about Milton Keynes. But instead it should be part of the "CSA".

Reading you're not right about, as 51% of the total workforce commutes into Reading and 57% of the resident workforce commutes out (these are 2001 figures). It's not specified where to, but you can be sure that a high enough % goes into metro London.
http://www.reading.gov.uk/council/ke...rough-profile/

Luton I couldn't find any statistics about, but I did founf about a billion references to it being a "London commuter town". Maybe nito can help us out?

We all agree that London is a great and big city, that it had a high growth this decade but do we really need to inflate the data.

You don't need to downplay it either and that's exactely what you're always doing.

[Minato Ku]

Originally Posted by SHiRO

But Greater London is not the London urban area. A entirely one-on-one comparison is not posible because of the different nature of the administrative entities. Which brings me to the next point:

Well, the urban area of London outside Greater London is not really big, it will not change much the result.
To make a signifiant change, these area should be major employment center.
It is not like New York where New York City is only 50% of the urban area population.

Greater London is not at all comparable with New York City.
Greater London included a lot of suburbs while New York City is almost only made of dense city district.

Originally Posted by SHiRO

You have a point here, but I guess they took Greater London or some other combination of entities as stand in for the "MSA". I don't remember as this was more than 10 years ago on SSC. I do remember on forumer having done a lot of work calculating all commuters from each individual council and it adding up to 14 million. Maybe they took that as the "MSA"?

I know but his method has huge flaw and infact misunderstood the way of calculating MSA and CSA.
I know him very well, as he was a french forumer of SCC, I had numerous debate with him.

I don't know how many inhabitants would have London with the MSA, about 13 million inhabitants, today, I guess.

Originally Posted by SHiRO

You're contradicting yourself here. If a metro area (for which different definitions can apply, also ones without factoring in commuters) is solely based on commuting patterns where does density come into play? Of course those commuting patterns are also based on something, hint: proximity! And more density means more people in the proximity so there are all different kinds of correlations. But it's not correct to first state that metro areas are always and solely based on commuter patterns and then contradict yourself by bringing density into play.

Actually no, you don't understand the point.
I said that more a city is located in densely populated area, more its metropolitan area tends to be small in area size because in a densely populated region, important cities are closer to each other.
This is why a metro areas in dense region are smaller than metro areas in empty region.
The density is not a factor for calculating a metro area, only the commute pattern are important but it does not change this observation when we compare several metro area calculated in the same way.

Don't forget Lille and Toulouse.
If your idea was true, why Toulouse metro area is 4 time bigger than Lille, while Lille is a much more densely populated region ?
We can find the same pattern for other case.

Originally Posted by SHiRO

OK, so you're right about Milton Keynes. But instead it should be part of the "CSA".

Not even sure, while we need the commute pattern of whole MK metro area, for MK only it would be quite difficult.

Originally Posted by SHiRO

Reading you're not right about, as 51% of the total workforce commutes into Reading and 57% of the resident workforce commutes out (these are 2001 figures). It's not specified where to, but you can be sure that a high enough % goes into metro London.
http://www.reading.gov.uk/council/ke...rough-profile/

In 2001, there were between 0 and 5,000 people from Reading commuting to London.
In the highest case, this means 10% of the working force would work in London.

I don't believe that reading would be part of London MSA, but I quite sure that it would be part of a London CSA.

Originally Posted by SHiRO

Luton I couldn't find any statistics about, but I did founf about a billion references to it being a "London commuter town". Maybe nito can help us out?

What does it proves, I can find dozen of reference about wrong thing.
Thirdly, it depends what criteria we give to the word commuter town.

According a PDF that I read, I can say that already 85% of Luton employed population don't work in London and its suburbs.
60% of Luton empoloyed population work in Luton, it is quite a town as its own when I read these stat.
www.semlep.com/modules/downloads/download.php?file_name=6

Originally Posted by SHiRO

You don't need to downplay it either and that's exactely what you're always doing.

London is the most populated metro in EU, but it doesn't have the size of New York unlike what many people believe.

[SHiRO]

Originally Posted by Minato Ku

Greater London is not at all comparable with New York City.
Greater London included a lot of suburbs while New York City is almost only made of dense city district.

You mean like Staten Island?

I know but his method has huge flaw and infact misunderstood the way of calculating MSA and CSA.

Common man, you weren't even on SSC 10 years ago. Plus you're the one misunderstanding the definition for MSA (see below)

I know him very well, as he was a french forumer of SCC, I had numerous debate with him.

This was not Manuel, this was a British forumer.

I don't know how many inhabitants would have London with the MSA, about 13 million inhabitants, today, I guess.

I'd say 14-16 so we're not too far off in our estimations.

Actually no, you don't understand the point.
I said that more a city is located in densely populated area, more its metropolitan area tends to be small in area size because in a densely populated region, important cities are closer to each other.
This is why a metro areas in dense region are smaller than metro areas in empty region.
The density is not a factor for calculating a metro area, only the commute pattern are important but it does not change this observation when we compare several metro area calculated in the same way.

There are no cities in the vicinity of London that are even 1/100th the importance of London. London is a certified primate city (like Paris) and a huge magnet not only for commuters, but for everything. The fact that the region is densely populated supports my position (of a higher population of the metro area) not yours. How can you not see that?

Don't forget Lille and Toulouse.
If your idea was true, why Toulouse metro area is 4 time bigger than Lille, while Lille is a much more densely populated region ?
We can find the same pattern for other case.

Doesn't say anything about London. Even if it was true for every other city on earth (it isn't), that wouldn't prove it would be true for London as well.

In 2001, there were between 0 and 5,000 people from Reading commuting to London.
In the highest case, this means 10% of the working force would work in London.

And here we have the critical flaw in your thinking. Commuters don't have to go into Greater London (because that is just the "city proper"), that's not how MSAs are calculated in the US either. They have to go into the CBSA. You are severly underestimating how loose of a definition MSA really is. Technically, Reading is already part of the MSA just because it's in Berkshire in which part of the London urban area overspills. I'm willing to acknowledge that in a European context it's not suitable to add entire counties just because part of the urban area overspills into it, but you have to keep being fair and do include commuters to that part of the urban area at least! Seriously, where do you think that 57% of the Reading workforce goes? It has a huge magnet (London urban area) just to the right, but you expect us to believe there are job centers that are bigger pulls that are further away and are not London or Reading itself?

According a PDF that I read, I can say that already 85% of Luton employed population don't work in London and its suburbs.
60% of Luton empoloyed population work in Luton, it is quite a town as its own when I read these stat.
www.semlep.com/modules/downloads/download.php?file_name=6

I wonder who or what the 22 trains between 0800-0859 are carrying then. Somehow these numbers must be off...

EDIT- Do you not see the 15% commuters going to London from Luton and the 10% to Hemel Hempstead?! You got your inclusion in the MSA right there even without the inbound commuter percentage!

So you have proven that MK is not in the London MSA and that Luton is definately in it! Thanks, clears up a lot. I underestimated MK as a jobcenter, guess they made it not just a bedroom community but also a relatively large economic center.

London is the most populated metro in EU, but it doesn't have the size of New York unlike what many people believe.

It's the other way around. Most people don't believe/know how big London is in an apples to apples comparison with New York. It's not as big, but if New York is 22 million, London is 18-20 million.

[Minato Ku]

Originally Posted by SHiRO

EDIT- Do you not see the 15% commuters going to London from Luton and the 10% to Hemel Hempstead?! You got your inclusion in the MSA right there even without the inbound commuter percentage!

Read better, it say that 15% of the outward traveller commuter, not the total number of worker.
That's means that 15% of the 40% worker who work outside Luton work in London (15% of 40%, it is 6%).
If we add the other 10%, (10% of 40%, it is 4%).
It is around 10% of the employed working force, not enouth to be part of London metro area.

[SHiRO]

Oh brother...

I know it says that but the numbers don't add up. It only adds up to 43% for 2010. A bit above that it says that in 2008 60% works in Luton, 8% Central Bedfordshire, 4-5% in Milton Keynes, 1-2% in Stevenage and 1-2% in Bedford. Added up that's 74-77%. So where does the other 23-26% work? Answer: 4-5% in Dacorum (Hemel Hempstead) = London, 1-2% St Albans = London and 1-2% Watford = London. That still leaves 17-20% unaccounted for so you tell me where they work? I say London (off course!!!) but according to you there probably is some other big regional jobcenter nearby, anything but London right? Maybe they commute to Paris?

If you really think that just 6 or 10% or whatever work in London you clearly don't know the area and shouldn't be commenting in the first place. I think you are trying a little too hard to prove London is not a huge commuter magnet as it is. Anyone from there or who has family living there (I do) knows Luton is a London commuter town. I mean why wouldn't it be at just 26 minutes by train from Central London and 20+ trains going there hourly and the M1 leading straight into it. Please use a little common sense...

[Derek2k32]

Originally Posted by SHiRO

Why not? In case of LA and Houston it's totally fair. Don't see that huge a problem with comparing to New York City or the (amalgated) city of Toronto either.
Great graphs nito!

Land area:
New York City: 790 km2
Greater London: 1,572 km2

Metro New York grows slowly because it's expensive and strict land use policies that encourage low density sprawl.

[SHiRO]

Originally Posted by Derek2k32

Land area:
New York City: 790 km2
Greater London: 1,572 km2

We already covered this...

[MonkeyRonin]

Originally Posted by SHiRO

Just 12 million as a metro area for London is completely ridiculous and not fair at all when comparing to North American ones! See what I mean with downplaying areas you're not familiar with? And are you really going to use 2001 figures after nito just showed the region grew about 2 million people over the last decade? The GLA definition is the one that gave London 18 million on 27,000 km2 in 2001. It has slightly different boundaries than the green on your map, this is because it includes some satellite cities that have very good connections to London and take you there in about one hour (Peterborough, Cambridge, Oxford). A lot of countryside is omitted in that definition. The New York CSA is 30,000 km2 so I don't see what's so outrageous about the London one.

Like I said; 14-16 million for a MSA like metro area and 18-20 for a CSA like one.

This all coming right after I said how absurd the CSA thing was...



Anyway, the thing with populations is...determining the exact size of a "city" is impossible, and they're never going to be precisely comparable, even within a single country. Comparing between countries makes it even harder.

But we can certainly find an approximate "range", which for London, definitely seems to be somewhere in the 11-14 million range (evidenced by much Minato's data). While say, New York is in the 16-19 million range, Los Angeles in the 10-15 million range, Houston and Toronto in the 5-6 million range.

And so, it all goes back to my original point that Greater London is comparatively quite a bit larger than the municipalities it was compared to. Its growth is still larger than any of the others, but to say it was to the extent which was conveyed in the graph is misleading.

What are your thoughts on the Golden Horseshoe and the 33,000 km2 it takes up? Is it the GTA that grew by 972,000 or the Golden Horseshoe?

Calling the Golden Horseshoe a "metro" would be completely absurd, just as it would be to call the southeast of England or an American CSA a metro. They're densely populated regions composed of multiple cities in close proximity, not metropolitan areas.

And the 972,000 figure I gave was for the GTA, as I said when I posted it.

And Greater London is not a "large semi-metropolitan entity" it's exactely the same thing as the City of Toronto!

Wrong. Each borough of London has its own city council, with larger regional government. Its not a unitary political body like Toronto. Its more like the old Metro Toronto government, which was pretty much the definition of "semi-metropolitan entity".

Originally Posted by SHiRO

You mean like Staten Island?

Staten Island is about 400,000 out of 8.4 million people. The vast majority of New York is highly urban in nature, beyond that and some areas on the edge of Queens it otherwise doesn't have significant suburban areas within its borders.

Originally Posted by nito

The position of cities such as Los Angeles and Toronto is a reflection of low core growth and high periphery growth; presumably suburban sprawl, although that would require further research to confirm.

Is the bulk of the growth in London really core growth though, or is it growth in the inner suburbs? (this would seem to suggest a bit of both) I'm not sure about Los Angeles, but in Toronto, the growth is divided between the fast-growing inner-city and outer-suburbs (probably split evenly between infill and greenfield development) while the inner-suburbs (part of the municipality) stagnate.

[Minato Ku]

The GLA never wrote that London metro area is 18 million inhabitants.
It wrote "Metropolitan region", a thing that can say anything.
Don't compare the area size, it is a big error. (Lille and Toulouse exemple)

New York has a much larger urban area than London (17 million vs 9 million), it quite usual that the NYC has much more populated metro.
True London has a Green Belt but it cannot explain a such high difference appart the fact that New York is bigger than London.
True London urban area is bigger than Greater London but it is 90% of the urban population compared with less than 50% in NYC.

In London, only 800,000 out of the 4.8 million employment are not living in London. That's 18% of the job, a lower number than people imagine.

Other thing, SE England is a well developed region with many employment centers, people don't need to go in London to have a job.
SE England is much more than a region made of commuter town of London.

People are so much into the overstatement for London that they don't realise anymore when a number is exagerated.
We had the same case for the 400,000 french people in London.
What is even more wrong is that it is the people who critize the overstatement that need to prove it self when it should be the oposite.