GaWC Research Bulletin 39

GaWC logo
  Gateways into GaWC

This Research Bulletin has been published in Environment and Planning A, 33 (6), (2001), 949-954.


Please refer to the published version when quoting the paper.


Being Economical with the Geography

P.J. Taylor

In a previous editorial comment1, I argued that the embedded statism in social science has meant that there is an evidential crisis in the study of world cities. With published statistics being largely about attributes of and in states, trans-state relations between cities are conspicuous only in their absence. Of course, the latter is precisely what makes a world city a world city. The result is a classic case of a literature where the empirics lag far behind the theoretical.

This situation is a result of particular tension in our thinking brought about by the rise of contemporary globalization. From a geographical perspective, this is a tension between different types of space. Put crudely, the plight of world cities research is a consequence of what we might call 'mosaic social science'. As a product of 'state-istics', the geographical bedrock of social science is territorial; it describes the world as a space of places. But what is required to understand world cities is a 'network social science', a research framework that describes the world as a space of flows. These alternative 'world-spaces' are not, of course, purely inventions of social science. They reflect, first, the nationalisation of the humanity over the last two hundred years, and second, contemporary globalization's challenge to this prioritisation of places over flows.

In this editorial comment I attempt to show the tension between places and flows in one important organ of the world's business press, The Economist. This UK publication is a truly global enterprise with 20 editorial offices across the world in what is a role call of leading world cities: London, New York, Tokyo, Paris, Hong Kong, Frankfurt, Los Angeles, Brussels, Beijing, Washington, Sao Paulo, Berlin, San Francisco, Bangkok, Beirut, Delhi, Johannesburg, Mexico City, Moscow and Edinburgh. The latter city provides the only hint of a British bias (especially given the omission of more important cities, such as Chicago, Singapore and Milan, from the list). Nevertheless this is a magazine seemingly 'plugged in' to the world city network, located to report on our new 'globalising' world. And yet the magazine remains dominantly territorial in its view of the world, it provides its readers with reports on regions and countries. Its text describes an international economy as a space of places: I refer to it as The Economist World I. However, an alternative picture can be found in the magazine between the pages of text; the advertisements describe a network world. They engage with a global economy as a space of flows: I refer to this as The Economist World II.


The attraction of The Economist is to be found in the mixture of its reporting and analysis of current politics and business. It is not surprising, therefore, that a large chunk of its weekly contents is organised territorially. The world is divided into six regions: Britain, Europe, United States, The Americas, Asia, and 'International' (equals what is left over, i.e. Africa and the Middle East; it is by no means obvious why it is not so named). Of course, this is a very particular 'Anglo-American' view of the world where the UK is not in Europe and the USA is not part of the Americas. No matter, the point here is the territorial frame. Remaining sections reporting on business and finance/economics also have many place-based stories such as 'Germany's struggling retailers' and 'Japan's coalescing insurers' (both from 2-9-00 issue). However, the 'Indicators' section at the end of each issue best illustrates the mosaic nature of the magazine's view of the world.

Every week three pages of The Economist are allocated to statistics. There is a page each on 'Economic Indicators' and 'Financial Indicators', both of which feature all the main European economies (11), plus Australia, Canada, Japan and the USA. The third page, entitled 'Emerging-market Indicators', provides information on the 25 largest economies in what used to be called the Second and Third Worlds.2 Attributes of these forty countries are described every week thus creating a continuously up-dated world of containers, of national economies that together constitute an international economy. This is a world space of places where information on flows is largely missing.3 It is portrayed as The Economist World I in Figure 1. And this is what we would expect given that the magazine relies on the same 'state-istics' as mosaic social science.

Going beyond the magazine itself, the state-centric nature of the world of The Economist is confirmed by The Economist Intelligent Unit's 'Country Forecasts' which provide up-to-date surveys of 60 national economies supplemented by six 'Regional Overviews' and a 'Global Outlook'. This nested mosaic of data on forecasting economic and financial attributes for places describes an international economy using a spatial framework commonplace to the twentieth century.


Advertisers in The Economist are far less concerned with an international world. On their pages we are 'en route to a borderless world' (according to NTT DoCoMo) where there are 'communications without boundaries' (Avaya). These provide a 'global reach' (Global One) where there is a 'global market call for global players' (Deutsche Post) who 'play to win on a global scale' (Cartesis). This is a 'new new economy' (SAP), with 'a new opportunity, a new paradigm' (J P Morgan), a 'new world, new thinking' (IBM), which is 'unrelenting thinking' (Goldman Sachs) in a world where you 'never stop thinking' (Infineon). If you are not sure you are 'really up to speed for global change' (Commerzbank) you may wish to enrol in 'the only truly integrated, truly global Executive MBA' (Trium) or 'learn, globally, study locally' (Fletcher School). But on the other hand there are 'answers to questions you don't even have yet' (quidnunc) and 'solutions to tomorrow's questions' (Dimension Data) readily available. This will enable us to 'change the world' (BT Openworld, PriceWaterhouseCoopers) and 'bring order to chaos' (PriceWaterhouseCoopers) so that the 'the possibilities are infinite' (Fijitsu) and 'unlimited' (Concert).

Behind all this optimism there is one basic message encapsulated by Morgan Stanley Dean Witter's bold slogan Network the World. The promise is of a 'global network with seamless connectivity' (Concert) or 'seamless global network' (infonet), 'a network of 13,000 open minds' (UBS Warburg), 'networks of confidence' (Bull) and 'a network of excellence' (Goldman Sachs) and, with 'the new internet' (Nortel Networks), 'next generation networks' (telltabs) with 'global crossing' of access and information (Global Crossing) based upon 'more superhighway, less road rage' (Agilent Technologies). Among this invocation of virtual networks, there are less obvious but equally noteworthy references to more concrete networks such as 'the airline network for Earth' (Star Alliance). And behind all infrastructure networks there have to be people that make the system work. These people work in offices and several companies proclaim their prowess in this respect: at one level 'the flexibility of 325 offices' (Maersk Sealand) and at another with 'over 5,000 offices world-wide' (HSBC). Such offices in their high tower blocks provide the familiar silhouettes of world cities and, more important, can be used to define the world city network.4

Many advertisements mention cities. Not just their headquarter location, but as the particular network of cities in which the advertiser operates. I have focussed upon just those advertisements that mention five or more cities for the nine months May 2000 to January 2001. There were 46 such advertisements5, which had a total 681 mentions of cities between them. In all 154 networked cities were identified with London, not surprisingly, having the most mentions (appearing in 39 of the 46 advertisements). In Figure 2 just the cities that appear in five or more networks are shown. These define a skeletal network of a global economy, the nodes of a world space of flows. This is The Economist World II, the picture readers get from the magazine's advertisements.


The notion of 'one world' has become a very worn cliché. Here I mean it literally. The two 'worlds' uncovered above are not autonomous and separate experiences - you cannot choose to leave one and enter the other - rather they are different perspectives, alternative representations of our contemporary economic world. As socially constituted, the spaces of places and flows complement one another, each being necessary for the other's reproduction. Thus it is not a matter of 'choosing' between them, the question is their relationship and what this tells us about the distribution of power in the contemporary world. For many, globalization has undermined the power of states in a world dominated by the space of flows, for others states maintain their power in a continuation of the space of places. The interesting thing about The Economist is that both interpretations can be gleaned from their pages.

The contrast between Worlds I and II is, at one level, unremarkable. Obviously advertisements will aim at a narrow set of readers - particular groups of practitioners - while the general text of the magazine has to appeal to a wider audience. Nevertheless the contrast between the two parts of the publication does provide much food for thought. My take on this is that we are living in a period of massive transition in which socially constituted spaces are central. In this interpretation globalization represents a metageographical moment, a time when the taken-for-granted way in which, collectively, we organise our knowledge of the world as spatial structures is being eroded.6 Globalization challenges the mosaic metageography of states with a new putative network metageography of connections. The pages of The Economist illustrate this spatial tension. The magazine continues to base its information and knowledge on the old mosaic metageography while many of its advertisers are promoting a new network metageography. To live in an era of competing metageographies is unusual; it is certainly not a time to be economical with the geography.


1. Taylor, P. J. (1999) "'So-called world cities': the evidential structure within a literature", Environment and Planning A, 31, 1901-04.

2. Other countries are sometimes featured in additional tables that appear on all three pages, for instance a list poor countries showing % of population undernourished (28-10-00 issue).

3. Where there are data on flows it is still incomplete in the sense that it does not have both an origin and a destination (e.g. on foreign direct investment and trade balances). Occasionally there are tables on corporations and cities but in both cases it is attributes that are described not flows or connections e.g. ranking telephone firms by total amount of traffic (4-11-00) and ranking financial advisers on mergers and acquisitions by total value of deals (13-1-01); and lists of cities showing property prices (2-9-00) and cost of living (20-1-01).

4. Taylor, P. J. (2001) "Specification of the world city network" Geographical Analysis 33 (available in pre-publication form at

5. Many of these advertisements were published several times, they are each counted just once in the data.

6. I take the concepts of metageography from Lewis, M. W. and Wigen, K. E. (1997) The Myth of Continents (University of California Press, Berkeley, CA), p. ix. The argument about metageographical moments is developed in Taylor, P. J. (forthcoming) "Metageographical moments: a geohistorical interpretation of embedded statism and globalization" in Odysseys Ed. B. Denemark and M. A. Tetreault (Routledge, London) (available in pre-publication form at


Figure 1: The Economist World I: the international economy as mosaic 


Figure 2: The Economist World II: cities as nodes in a global economy 


Edited and posted on the web on 31st January 2001

Note: This Research Bulletin has been published in Environment and Planning A, 33 (6), (2001), 949-954.