The World City System

by Kieran Healy on February 16, 2004

The “latest issue”: of the “American Journal of Sociology”: [subscription required] has a number of interesting articles, but given the, ahem, cosmopolitan nature of the crew here at CT, a paper by Alderson and Beckfield on “Power and Position in the City World System”: [also “pdf”:] caught my eye. They examine power relations between three and a half thousand cities in a network analysis, operationalizing ties with a measure of HQ and branch locations of the world’s 500 largest corporations. The authors develop a “blockmodel”: to identify clusters of “regularly equivalent”: cities. Roughly, members of regular equivalence sets have similar relations to members of other equivalent sets, so equivalent cities stand in the same relations to other groups of cities.

As you might expect, the core of the city world system is the block made up of London, New York, Paris and Tokyo, and these four cities are much more powerful than any of the others. But outside this core group, the analysis suggests some patterns that aren’t visible from less formal approaches. Outside the “L-N-P-T” block, there are six other “Primary” blocks:[1] Amsterdam, Basel, Atlanta, Caracas, Cologne and … “Bristol”: “Chris”: will be delighted.

fn1. These are “cliques whose members are involved in high levels of relations with outsiders. More specifically, they are blocks with greater than expected ingroup preference (their cliquishness), but also greater than expected outdegree and indegree.”



praktike 02.16.04 at 2:17 pm

Ah, yes, but where is by own beloved Pittsburgh?


Matthew 02.16.04 at 3:24 pm

I don’t understand (it’s hard when you can’t read it). If LPNT form a ‘block’ then do the other cities you list form a ‘block’ or are they one city each in other blocs? For I can see that the cities listed are similar in many ways (well except Caracas, of which I know nothing) but I don’t quite see how they could be the only other ‘block’?


Matthew 02.16.04 at 3:27 pm

Incidentally (i guess) this is a similar analysis


Matt McIrvin 02.16.04 at 4:17 pm

I’d have guessed Hong Kong as a fifth.

This reminds me of the Harvard US dialect survey question about what specific city is “the City”. Every part of the country had a local city that was referred to as “the City”, but the only city that was commonly “the City” everywhere in the US, with a distribution resembling the overall sample, was New York. (Chicago had a respectable showing in the Midwest; LA came up much less often than I’d expect.)


Stentor 02.16.04 at 4:26 pm

I’ve seen world city network analyses before, but they never put Paris in the top tier. I wonder if that’s due to different methodology, or whether Paris has been gaining influence in the past few years.


Josh 02.16.04 at 6:27 pm

LA came up much less often than I’d expect.

Not really surprising, when you consider just how large the physical city of Los Angeles is, and just how little it’s defined by its downtown.


yami 02.17.04 at 3:04 am

They didn’t list which cities belong to which cliques! How am I supposed to write snarky remarks on urban stereotypes as they tangentially relate to globalization without tabulated statistical fodder?

Comments on this entry are closed.