Where are largest cities in the world? 2015 edition

The world's biggest cities from above. Image: Getty.

The number of people who live in a city, as we may have mentioned before, is surprisingly hard to work out. That's partly because populations shift and change, unnoticed by demographers.


Mostly, though, it's because you quickly run into the fact that different authorities define their cities in entirely incompatible ways, rendering any attempt to compare them a bit like comparing apples with trucks. What you really need is a single source that follows a single set of rules: the results may not be perfect, but at least they'll be consistent.

That’s where Demographia comes in. The St. Louis-based consultancy was founded by the urban planner Wendell Cox (who, unusually for someone in that line of work, likes to write articles for conservative publications about the benefits of private car ownership). It also has a website that is, with the best will in the world, not massively easy on the eye.

But every year Demographia publishes something rather good: the World Urban Areas Report, a sort of bumper book of city population stats. You can check out the 11th edition here.

Here’s how it defines a city:

An urban area ("built-up urban area," urbanized area or urban agglomeration) is a continuously built up land mass of urban development that is within a labor market (metropolitan area or metropolitan region). An urban area contains no rural land... [It] is best thought of as the “urban footprint” – the lighted area that can be observed from an airplane (or satellite) on a clear night.

This year's edition of the World Urban Areas Report covers more than a thousand of the largest cities in the world, from Tokyo (pop. 37,000,000), right down to tiddlers like 850th-ranked Leicester (Pop. 534,000): all 1,009 cities, in fact, with a population of half a million of more.

Here are the top 10. See if you can spot any pattern.

 

Seeing it yet? Need a clue?

Okay, here's the same chart again. This time, we’ve coloured the cities by continent.

 

Yep.

Asia's dominance continues as you move down the league tables, too. Here's the top 20:

 

The world's largest continent, in fact, is home to 33 of the world's 50 largest cities, and 50 of the top 100. After that we got bored and stopped counting.

This isn't quite as extreme a result as it may appear: Asia, after all, is home to just shy of 60 per cent of the world's population. So the way in which the 2.1bn people who live in cities of half a million souls or more are distributed around the globe is largely a function of where people actually live. Look:

Comparing those two graphs, you can see that the Americas and Australia are unusually urbanised, Africa unusually under-urbanised, and Europe and Asia are pretty close to what you'd expect.

There are two other striking things about these league tables.

China rules

The first is the dominance of China, which takes 12 places in the top 50, and 22 of the top 100.

Again, this shouldn't be surprising – it's the world's most populous country, home to nearly a fifth of humanity – but nonetheless, it's striking how many of these megacities you almost certainly haven't heard of.

Where power lies

The other striking thing here is how far down the league tables many of the world's most powerful cities fall. Here's where some of the places that commonly make the “world city” league tables fall:

 

Size, it seems, isn't everything.

The data

Here, for completism's sake, is a chart showing the figures for all the city population charts above. You can find the full data set on Demographia's website, here.

Rank City Population
1 Tokyo-Yokohama 37,843,000
2 Jakarta 30,539,000
3 Delhi 24,998,000
4 Maniilla 24,123,000
5 Seoul-Incheon 23,480,000
6 Shanghai 23,416,000
7 Karachi 22,123,000
8 Beijing 21,009,000
9 New York 20,630,000
10 Guangzhou-Foshan 20,597,000
11 Sao Paulo 20,365,000
12 Mexico City 20,063,000
13 Mumbai 17,712,000
14 Osaka-Kobe-Kyoto 17,444,000
15 Moscow 16,170,000
16 Dhaka 15,669,000
17 Cairo 15,600,000
18 Los Angeles 15,058,000
19 Bangkok 14,998,000
20 Kolkata 14,667,000
25 Shenzhen 12,084,000
28 Tianjin 10,920,000
29 Paris 10,858,000
31 Chengdu 10,376,000
32 London 10,236,000
37 Chicago 9,156,000
41 Dongguan 8,442,000
43 Wuhan 7,509,000
45 Hangzhou 7,275,000
46 Hong Kong 7,246,000
47 Chongqing 7,217,000
50 Quanzhou 6,710,000
57 Nanjing 6,155,000
58 Shenyang 6,078,000
59 Xi'an 5,977,000
62 Qingdao 5,816,000
68 Singapore 5,624,000
72 Suzhou 5,246,000
76 Zhengzhou 4,942,000
77 Washington DC 4,889,000
79 Harbin 4,815,000
89 Boston 4,478,000
90 Xiamen 4,420,000
94 Dalian 4,183,000
97 Berlin 4,069,000
99 Fuzhou 3,962,000
101 Dubai 3,933,000
299 Amsterdam 1,624,000
832 Geneva 599,000
 
 
 
 

What other British cities can learn from the Tyne & Wear Metro

A Metro train at Monument. Image: Callum Cape/Wikipedia.

Ask any person on the street what they know about Newcastle, and they’ll list a few things. They’ll mention the accent; they’ll mention the football; they’ll mention brown ale and Sting and Greggs. They might even mention coal or shipbuilding, and then the conversation will inevitably turn political, and you’ll wish you hadn’t stopped to ask someone about Newcastle at all.

They won’t, however, mention the Tyne and Wear Metro, because they haven’t probably heard of it – which is a shame, because the Metro is one of the best things the north-east has to offer.

Two main issues plague suburban trains. One is frequency. Suburban rail networks often run on poor frequency; to take Birmingham for an example, most of its trains operate at 30-minute intervals.

The other is simplicity. Using Birmingham again, the entire system is built around New Street, leading to a very simple network. Actually, that’s not quite true: if you’re coming from Leamington Spa, Warwick, Stourbridge, Solihull or a host of other major minor (minor major?) towns, you don’t actually connect to New Street – no, you don’t even connect to the ENTIRE SYSTEM BUILT AROUND NEW STREET except at Smethwick Galton Bridge, miles away in the western suburbs, where the physical tracks don’t even connect – they pass over each other. Plus, what on earth is the blue line to Walsall doing?

An ageing map of the West Midlands rail network: click any of the images in this article to expand them. Image: Transport for the West Midlands/Centro.

But Newcastle has long been a hub of railway activity. Tragically, the north-east has fewer active railway lines than any other region of the UK. Less tragically, this is because Tyne and Wear has the Metro.


The Metro was formed in 1980 from a somewhat eccentric collection of railways, including freight-only lines, part of the old Tyneside Electrics route, underground tunnelling through the city centre, track-sharing on the National Rail route to Sunderland, and lines closed after the Beeching axe fell in the early 1960s.

From this random group of railway lines, the Metro has managed to produce a very simple network of two lines. Both take a somewhat circuitous route, the Yellow line especially, because it’s literally a circle for much of its route; but they get to most of the major population centres. And frequency is excellent – a basic 5 trains an hour, with 10 tph on the inner core, increasing at peak times (my local station sees 17 tph each way in the morning peak).

Fares are simple, too: there are only three zones, and they’re generally good value, whilst the Metro has been a national leader in pay-as-you-go technology (PAYG), with a tap-in, tap-out system. The Metro also shares many characteristics of European light rail systems – for example, it uses the metric system (although this will doubtless revert to miles and chains post-Brexit, whilst fares will be paid in shillings).

 

The Metro network. Image: Nexus.

Perhaps most importantly, the Metro has been the British pioneer for the Karlsruhe model, in which light rail trains share tracks with mainline services. This began in 2002 with the extension to Sunderland, and, with new bi-mode trains coming in the next ten years, the Metro could expand further around the northeast. The Sheffield Supertram also recently adopted this model with its expansion to Rotherham; other cities, like Manchester, are considering similar moves.

However, these cities aren’t considering what the Metro has done best – amalgamated local lines to allow people to get around a city easily. Most cities’ rail services are focused on those commuters who travel in from outside, instead of allowing travel within a city; there’s no coherent system of corridors allowing residents to travel within the limits of a city.

The Metro doesn’t only offer lessons to big cities. Oxford, for example, currently has dire public transport, focused on busy buses which share the same congested roads as private vehicles; the city currently has only two rail stations near the centre (red dots).

Image: Google.

But it doesn’t need to be this way. For a start, Oxford is a fairly lateral city, featuring lots of north-south movements, along broadly the same route the railway line follows. So, using some existing infrastructure and reinstating other parts, Oxford’s public transport could be drastically improved. With limited engineering work, new stations could be built on the current track (blue dots on the map below; with more extensive work, the Cowley branch could be reinstated, too (orange dots). Electrify this new six-station route and, hey presto, Oxford has a functioning metro system; the short length of the route also means that few trains would be necessary for a fequent service.

Image: Google.

Next up: Leeds. West Yorkshire is a densely populated area with a large number of railway lines. Perfect! I hear you cry. Imperfect! I cry in return. Waaaaaah! Cry the people of Leeds, who, after two cancelled rapid transit schemes, have had enough of imaginative public transport projects.

Here’s a map of West Yorkshire:

Image: Google.

Here’s a map of West Yorkshire’s railway network:

 ​

Image: West Yorkshire Metro.

The problem is that all of the lines go to major towns, places like Dewsbury, Halifax or Castleford, which need a mainline connection due to their size. Options for a metro service are limited.

But that’s not to say they’re non-existent. For example, the Leeds-Bradford Interchange line passes through densely populated areas; and anyway, Bradford Interchange is a terminus, so it’s poorly suited to service as a through station, as it’s currently being used.

Image: Google.

With several extra stops, this line could be converted to a higher frequency light rail operation. It would then enter an underground section just before Holbeck; trains from Halifax could now reach Leeds via the Dewsbury line. The underground section would pass underneath Leeds station, therefore freeing up capacity at the mainline station, potentially simplifying the track layout as well.

 

Image: Google.

Then you have the lines from Dewsbury and Wakefield, which nearly touch here:

Image: Google.

By building a chord, services from Morley northwards could run into Leeds via the Wakefield line, leaving the Dewsbury line north of Morley open for light rail operation, probably with an interchange at the aforementioned station.

Image: Google.

The Leeds-Micklefield section of the Leeds-York line could also be put into metro service, by building a chord west of Woodlesford over the River Aire and connecting at Neville Hill Depot (this would involve running services from York and Selby via Castleford instead):

The path of the proposed chord, in white. Image: Google.

With a section of underground track in Leeds city centre, and an underground line into the north-east of Leeds – an area completely unserved by rail transport at present – the overall map could look like this, with the pink and yellow dots representing different lines:

Et voila! Image: Google.

Leeds would then have a light-rail based public transport system, with potential for expansion using the Karlsruhe model. It wouldn’t even be too expensive, as it mainly uses existing infrastructure. (Okay, the northeastern tunnel would be pricey, but would deliver huge benefits for the area.)

Why aren’t more cities doing this? Local council leaders often talk about introducing “metro-style services” – but they avoid committing to real metro projects because they’re more expensive than piecemeal improvements to the local rail system, and they’re often more complex to deliver (with the lack of space in modern-day city centres, real metro systems need tunnels).

But metro systems can provide huge benefits to cities, with more stops, a joined-up network, and simpler fares. More cities should follow the example of the Tyne and Wear Metro.