A short post about cities and statistics:

Cities are remarkably hard to define. You would think otherwise, given the amount of data we have at our disposal. There are few cities on earth that have not been comprehensively mapped and photographed. But certain problems of definition remain: even in a single context, let’s say the United Kingdom, the definition of a city’s limits might vary from one case to another; we often compare apples with oranges. I was reminded this in recent coverage of a survey which measured the relative ‘vibrancy’ of different UK cities. I didn’t think much of the survey (it was a ‘fucking bag of bollocks’ according to the architectural journalist Ian Martin). One reason I didn’t was that its definitions varied according to city. So ‘London’ was, strictly speaking, the Greater London metropolitan region, with a population of 7.5 million or so. Its ‘Manchester’ was the political city, which is to say, a constituent part of a much larger urban region. Its 510,000 population makes it more like a big London borough. In this particular case, smallness played to Manchester’s advantage, and it came it top. For this exiled Mancunian, of course that’s good news, but I still recognise it as a flawed comparison.

I explored the problem a little more in a class today. I showed the group a list of population figures, all of which define ‘Manchester’ correctly:

‘Old City’ (1800 boundary)                                          25,000
Political boundary city 2013                                       510,000
Urban core, including Salford and Trafford            1,200,000
Greater Manchester, political boundary, 2013       2,600,000
Postcode city including all GM codes                     3,000,000 (estimate)
Travel-to-work city                                                  3,500,000 (estimate)
MUFC supporters worldwide, 2012 estimate      659,000,000

Manchester City fans will want to discount the MUFC stat immediately, which is fair enough. It does however give an indication of the potential importance of a non-resident population, all of who, arguably have an interest in what goes on in the city. Anyway, put that figure aside. We still have a remarkable discrepancy between the core city population of 25,000 (up from a historic low of 250 in the 1990), and what is arguably the ‘real’ city, the travel-to-work city of 3.5 million. It’s a differential factor of 140. Even if we were to take the ‘low’ figure as the 2013 political city, 510,000, we’re still looking at a factor of 7.

Manchester’s far from alone in its discrepant representations. Buenos Aires seems to suffer a similar problem. Is it a European-scale large city of 3 million? Or an emerging-market megalopolis of 16 million? Well, both, and neither. It just depends who is speaking and what they are trying to prove. That is the lesson for all of us, and all cities.

NOTES:

Vibrancy survey by Experian, reported in FT (9 October 2013)  http://www.ft.com/cms/s/0/2bc0e682-2f50-11e3-ae87-00144feab7de.html#axzz2iSUwJSb1