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.

Below, excitingly, you’ll find a chart of how the populations of Britain’s ten largest cities changed between 1981 and 2014, according to Centre for Cities (CfC) data*. The chart doesn’t show absolute numbers, because the sheer size of London will render any changes in the population of the other nine all but invisible; instead, we’ve indexed them against their 1981 population (100, on the chart). The black line represents the change in population of the UK as a whole.

We’ve also removed the labels. Just, y’know, to make things more fun.

The 10 cities are: London, Birmingham, Manchester, Glasgow, Leeds, Liverpool, Sheffield, Newcastle, Nottingham and Bristol. See if you can guess which is which.

 

The first thing to note is that almost all these cities are in relative decline. The population of the UK as a whole – the black line – has increased almost throughout this period (it shrunk very slightly in the early 1980s, but then has increased every year since). Yet two of these ten cities have seen their populations increase faster than the UK as a whole; three have actually shrunk since the early 1980s.

It’s possible to over-state the link between population growth and prosperity; and, as ever, where you draw city boundaries is a factor, too. But nonetheless, successful and economically vibrant cities will generally attract more people to live in them. The fact that eight of Britain’s ten largest cities have grown less quickly than Britain as a whole is probably not a good thing.

That’s the general trend – but which city do you think is which? Looks to me like we have a five different groups here:

  • Two boom towns, which have grown faster than the UK population;
  • Two that basically flat-lined until around 2001, then began to grow quite slowly;
  • Three tightly bunched cities (seriously; two of those lines basically cover each other), that declined throughout the late 20th century, but began to recover around the millennium;
  • A fourth, that looked like it was going to be in that group but arrived late at a weaker recovery;
  • Lastly, two that declined much more steeply, and have barely recovered since; they’re still over 10 per cent smaller than they were in 1981. This, given the UK as a whole is 14 per cent bigger, is a really quite significant decline.

You can probably guess that the boom towns are the only two southern English cities on the map. Since 1981, the population of London, on CfC definitions, has grown by nearly a quarter (24 per cent) and Bristol by nealy a fifth (18 per cent).

The others take more explanation so here’s a labelled version of that chart:

 

Why Nottingham and Leeds should have sustained their populations when most similarly sized British cities didn’t is quite frankly a mystery to me. Leeds, one can speculate, was helped out by having one of the north’s more diversified (and richer) economies; the same can’t be said of Nottingham, though. If anyone has a theory, do write in.

The exceptions at the bottom of the graph, though, do have something in common. Newcastle, Liverpool and Glasgow are all built on major rivers (Tyne, Mersey and Clyde), all of which once held shipyards. It’s possible that what we’re looking at here is the decline of British sea power.

That said, London and Bristol, of course, were port cities too, in their day. As ever, the problem is not the decline of old industry, but the lack of new ones.

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.

*By population, incidentally, we don’t mean official population. Rather we’re looking atprimary urban areas: a collection of local authorities which, when you take urban build-up and commuting patterns into account, are as good a way as any to define the functional geography of a city.

PUAs are slightly arbitrary, but so is almost every other option. At least these are comparable. And while the PUA boundaries have changed, the figures we’re using here cover consistent areas right back to 1981, so there.