The latest instalment of our weekly series, in which we use the Centre for Cities’ data tools to crunch some of the numbers on Britain’s cities. 

Last week in this slot, we looked at how the populations of Britain’s largest cities had changed between 1981 and 2014. What we found is

  • that London’s population boom hadn’t been replicated in the other core cities;
  • that most cities had in fact shrunk, as a share of the population of the UK; and
  • that those which had shrunk the most (Liverpool, Glasgow) were those which had built up around ports.

Here’s the obligatory chart.

These, though, are only the largest British cities. Do the same trends hold if you look at every significant British city? What do you get if you plot the way populations have changed in all 62 British cities followed by the Centre for Cities, relative to their 1981 population?

A mess, that’s what.

(Full disclosure at this point: I got halfway through colour coding them by their size, realised the result would still be a mess, albeit a mess with slightly fewer colours in it, then decided to try something else. So, there you go.)

That chart is clearly illegible. If we’re going to learn anything at all we need to break things down a bit. So, since that’s where we’ve started, let’s do it by size.

I’ve broken our cities into four groups of 15-17 cities each, based on how big they were back in 1981.

Here’s the largest group: cities which, 35 years ago, had a population of 400,000 or more. (The smallest back in 1981 was Leicester, though it’s since been undertaking by Middlesbrough.) So that you can see how they’ve fared relative to the UK as a whole, I’ve included the trend in national population as a thick black line.

This is basically last week’s graph with a few extra cities thrown in so, once again, the most notable trend is that most cities have grown more slowly than the UK as a whole – have, as a share of the national population, shrunk.

Portsmouth looks like an exception to our theory that shipbuilding cities are in decline. Perhaps it’s protected by fact it’s in south east of England and commutable to London; or perhaps the theory was never up to much. More on this below.

The other thing to note is that Bradford and, especially, Leicester have actually grown pretty quickly. One plausible explanation for this is that both have very ethnically diverse populations: diverse cities tend to be younger than mostly white ones, and thus produce more kids.

Anyway. On to the next group down: cities which, in 1981, had populations of between 275,000 and 400,000. Back then, the largest was Huddersfield (377,000) and the smallest Cardiff (287,000). By 2014, the largest was Bournemouth (478,000), and the smallest Sunderland (277,000).

Once again, the majority of cities have grown slower than UK average. Two of those that have shrunk the most are, on some definitions, suburbs of the larger cities that shrunk most in group one (Birkenhead is sort of Liverpool; Sunderland is sort of Newcastle).

But unlike on the previous chart, in this group there are also a few cities that have grown fairly steadily over the past 30 years (Cardiff, Preston). Several of these – Brighton, Southampton, Bournemouth – are on the south coast of England. Perhaps that’s the trend that allowed Portsmouth to break the curse of the dockyards.

The other city worth noting here is Coventry which, after a long period of decline, seems to have grown unusually quickly since around the time the credit crunch hit. No idea what that’s about.

On to Division Three: cities which, in 1981, had populations of between 165,000 and 275,000. Back then, the largest was Hull (274,000) and the smallest was York (165,000); by 2014, those rankings were held by Reading (320,000) and Dundee (148,000).

Which means that York is now bigger than Dundee. Go figure.

Anyway:

Dundee and Hull seem to fit our narrative that having booming shipyards in the early 20th century is generally a great way of seeing your population collapse by the 21st century. Chatham is another partial exception to that – but, like Portsmouth, was probably saved by its proximity to London.

Aberdeen is another town that’s seen a radical reverse in its fortunes, albeit more explicably: that’s almost certainly the oil boom at work.

And, yet again, the vast majority of cities shrunk compared to the UK as a whole. This is becoming weird now.

Last and also, frankly, least are the tiddlers: those which in 1981 were home to fewer than 165,000 people. The largest then was Luton (just shy of 165,000); the smallest was Crawley (just over 82,000). By 2014, the smallest was Worthing (107,000), the largest was… well, look:

Yep. In 1981, Milton Keynes had a population of 126,000. By 2014 it was 259,000. It’s the only city anywhere in these rankings to have doubled in population. It’s grown so much, in fact, that it sort of breaks the graph, so here’s a version without it:

In this category, for the first time, all the cities have grown, and most have grown faster than UK as a whole. The smaller the city, the more likely it is to have grown faster than the UK average. Here’s a table:

There are number of possible explanations for this that I can see. One is that the absolute figures are smaller: it’s probably easier to grow by 10 per cent if you’re the size of Exeter (and so need to build about 5,000 extra houses) than if you’re the size of Manchester (and so need to build around 100,000).

It probably also reflects land use policies. Most big UK cities are ringed by green belts, intended to stop them growing. As a matter of deliberate government policy, growth was funnelled to the New Towns – which include the likes of Milton Keynes, Swindon, Telford, Crawley, Peterborough, Basildon… Spotting any patterns here?

A prominent theory in the world of urban economics concerns the benefits of aggregation: larger cities, it’s argued, are generally more productive than smaller ones. The UK is a freak in that it doesn’t fit this model at all.

I can’t help but wonder if, just maybe, that’s the result of deliberate government policy to direct growth away from major cities. Perhaps, if we want to solve productivity puzzle, we’ll need to change that.

Jonn Elledge is the editor of CityMetric. He is on Twitter, far too much, as @jonnelledge.

Want more of this stuff? Follow CityMetric on Twitter or Facebook.