GaWC Research Bulletin 390

GaWC logo
  Gateways into GaWC

This Research Bulletin has been published in International Journal of Urban and Regional Research, 38 (1), (2014), 98-115 under the title 'Beyond the Global City Concept and the Myth of "Command and Control"'.


Please refer to the published version when quoting the paper


Beyond the Global City Concept

R.G. Smith*


This paper prepares the ground to move the theoretical and empirical agenda of global urban studies beyond its dominant neo-Marxist tradition towards a broadly conceived poststructuralist approach to include research in actor-network theory, cultural-economy, social studies of finance, and post-continental philosophy. First, a previously unacknowledged account of the neo-Marxist back-story to the global city concept is explicated to foreground how Sassen’s concept presides over a common epistemological tradition in urban studies that envisions capitalism as a totalizing structure to assume it is commanded and controlled through strategic urban loci. Second, Taylor’s interlocking network model is interrogated to reveal how it not only contradicts Sassen, but is a structuralism bedevilled by a sorites paradox problem where cities are indistinguishable from, and therefore totally reducible to, structures. Finally, the paper moves beyond the global city through recourse to a broadly poststructuralist literature that is construed to foreground an attentiveness to the limits of capitalist control – viz., the global city – and the necessity for a conceptualization of cities as multiple, eventful, socio-material assemblages, that are never quite in control because as multiplicities they are alive with that which is not yet fully actual.


Why the global city concept caught on and met with wide support as a seemingly valid way to understand contemporary globalization is an interesting and fundamentally important question for those engaged in the study of cities in advanced economies. The idea of certain cities controlling, co-ordinating and dominating the world economy emerged as a lexeme for neo-Marxist urban studies in the context of the political-economy of the 1970s and 80s. Then in Britain and the US, Thatcher and Reagan forced a replacement of the Keynesian post-war democratic settlement with a neo-liberal formula(tion) founded on the ‘free-market fundamentalism’ of Milton Friedman and an abject obeisance to global capitalism – ‘legitimized’ through such venal ideologies as ‘wealth creation’, hyper-individualism, privatization, financialization, computerization, liberalization, denationalization, deregulation, free trade and the ‘invisible hand’ of market forces – that skewed their national economies both sectorally towards finance and servicing globalization, and geographically with those sectors becoming ever-more concentrated in the urban centres/city-regions of London and New York. In other words, neo-liberalism came to town and the global city concept chimed with the need of neo-Marxists for the world’s financial centres – exemplified by New York, London and Tokyo – to be understood as centres of power, as the world’s latest command posts in control of an economic and financial globalization that represents just the latest stage of capitalist development.

In this paper I make the argument that this way of thinking about cities in globalization, about the role and function of some cities in the global economy as the nec plus ultra locations of power, control and command, is a mistake. To make that argument the paper proceeds in three parts. Firstly, the origins of the global city concept are thoroughly explored for the first time – from Hymer (1972), Heenan (1977) and Cohen (1981) to Friedmann (1986) and Sassen (1991) – to specify how the nature of the global city concept – its empirical particularism and abstract universalism – is rooted in a neo-Marxist envisioning of the world economy as a structured totality that needs a few strategic cities to be in ‘command and control’ of it. Secondly, the flawed logic and sorites paradox at the root of Peter Taylor’s construction of a ‘world city network’ via an interlocking network model (INM) is exposed to discuss how whilst purporting to be an extension of Sassen’s (1991, 1994, 2001, 2002) work Taylor’s model not only fundamentally contradicts the global city concept in its technical minutiae (Smith and Doel (2011) explain the unresolvable contradiction – independent (Sassen, see Figure 1) ≠ interdependent (Taylor, see Figure 2) – leaving the neo-Marxist urban field in a kind of theoretical limbo), but also contains no explanation as to why centres of authority are needed by the network. Taylor’s INM represents the third millennium end-game for neo-Marxist work on cities in globalization because its emphasis on relations to the detriment of terms is nothing more than the apogee of structuralism. When the function and identification of terms, attributes, cities, is entirely determined by the number and nature of the relations, connections, networks, then the city vanishes into structure to be merely concerned with riding the tiger – with just documenting the everyday reproduction of capitalist networks (e.g. see Hanssens et al., 2011; Taylor, 2011). Finally, the paper moves beyond the global city concept, and the rubrics of neo-Marxism and structuralism through an interpretation of empirical results from the SSF literature which is understood as a kind of evidential foreground for understanding the multiplicity and eventfulness of leading financial centres as socio-technical assemblages that service a globalization which is never quite under control because in an inter-connected world that which cannot be accounted for has the force of sub-traction (see Smith & Doel, 2011).

Figure 1

fig 1

Figure 2

fig 2

The global city concept de novo

‘We all worry about the circumstances of our death; the circumstances of our birth, however, are less worrisome to us’ (Michel Houellebecq, 2005)

How did the global city concept of a triptych of cities ‘commanding and controlling’ the world economy come to be a great tradition within neo-Marxism? Well the idea that certain cities now act as centres of command and strategic control for the globalization of distanciated economic activity has a long foreground, first emerging in the 1970s, long before the global city concept was popularized by Sassen1, and a decade before experts currently assume: ‘The starting point of this contemporary debate is generally regarded to be the presentation of a research agenda by Friedmann and Wolff (1982)’ write Newman and Thornley (2011: 32) in their truncated discussion on the development of the world city hypothesis (e.g. also see Friedmann  (1995), Taylor (2001), and Surborg (2011) for the same post hoc starting point). An understanding of the long background to the global city concept is essential for understanding it. Let me explain in the necessary detail.

The idea of a world economy articulated through certain cities – a shortlist of unofficial capitals of the world including Rome, Constantinople, Venice, Amsterdam, London and New York – is longstanding (Braudel, 1984; Hall, 1998). In 1966 Peter Hall followed Sir Patrick Geddes (1915), who coined the term ‘world city’, to define them as ‘certain great cities, in which a quite disproportionate part of the world’s most important business is conducted’ (1966: 7) to predict that ‘The economic life of the world will be concentrated into a few major information centres’ (1966: 240), namely seven such centres: London, Paris, Randstad Holland, Rhine-Ruhr, Moscow, New York, Tokyo (Hall’s book was influential running into three editions, 1966, 1977, 1984). However, the specific idea of certain cities having a strategic role in co-ordinating and controlling the world’s economy is relatively new and can be traced back to a handful of authors writing in the 1970s and 1980s. Stephen Hymer (1972), David Heenan (1977), Robert Cohen (1981), and John Friedmann (1986; with Goetz Wolff, 1982) stand out in this regard. These authors all focused on multinational corporations (MNCs) in cities as the locus of command and control for the world economy. However, it was Cohen who most directly foreshadowed Sassen’s and Taylor’s theses because he not only considered cities as centres of corporate headquarters, but also of international banking and strategic corporate services: ‘Only a place with a wide range of international business institutions can be truly called a world city’ (Cohen, 1981: 302). For his part, Hymer (1972) suggested a correspondence principle whereby a centralization of control within MNCs relates to a centralization of control within the international economy. Hymer argued that the emergence of MNCs produced a tendency for high-level corporate decision-making to be centralized in a few key cities, so that ‘[T]he world’s major cities … will be the major centers of high-level strategic planning’ (1972: 124). Consequently, Hymer envisioned an emerging urban hierarchy with a centralization of ‘high-level decision making occupations in a few key cities in the advanced countries’, and that these would be ‘surrounded by a number of regional sub-capitals’, with the rest of the world confined ‘to lower levels of activity and income’ (1972: 114). Overall, the basic pattern would be one of ‘superior and subordinate, head office and branch plant’ (1972: 114). In short, it was Hymer who first outlined a ‘global’ hierarchy of cities (cf. Taylor, 2001: 181) and he allocated control of the international economy to some cities rather than others (1972: 124, provides a list of such cities), and consequently his paper stands as a kind of avant la letter ur-text for all subsequent neo-Marxist work on cities in globalization.

Heenan (1977) was the first urban researcher to coin the term global city according to its current usage, explicitly stating that his ‘article focuses on the emerging phenomenon of the global city’ (1977: 81). He linked the emergence of global cities to the world economy several years before Cohen made the same point: ‘For the most part, global cities are evolving in response to fundamental changes in the world’s industrial system’ (1977: 82). What is even more remarkable about Heenan’s article is his argument that the globalization and regional organization of MNCs is ‘creating a need for global cities’ (1977: 82, my emphasis). Heenan regards the emergence of global cities as being necessitated by those MNCs that are organizing themselves on a regional basis: global cities are a system requirement. It is this structural change in the organization of MNC-led economic globalization that is the basis for the thrust of Heenan’s overall argument. With globalization, cities need to be ‘knowledge-oriented command posts’ (1977: 82). They need to demonstrate global and regional leadership2.

The dominance of the neo-Marxist tradition in shaping the world city concept meant that Heenan’s ground-breaking article was dismissed by Friedmann and Wolff as ‘superficial’. They criticized him for ‘failing to take into account the dialectical relationship between the world system and the “global city”’ (1982: 332). Following Friedmann and Wolff, to my knowledge Heenan is never cited again in the global and world city literature. Failing the neo-Marxist litmus test he is absent from both general overviews of the research field (such as those by Alger, 1990; Brenner and Keil, 2006; Knox and Taylor, 1995; Smith, 2000; Yeoh, 1999) and specific studies. Rather, it is Cohen’s article3, published some four years later in 1981, that is heralded as having developed significantly the idea that certain cities command and control the world economy.

For Cohen (1981: 288), ‘[G]lobal cities act as centers of corporate control and coordination for the new international system’. Importantly, Cohen extends Hymer’s focus on MNCs in cities to consider corporate demand for advanced producer services, and he ‘links these new demands to … the emergence of a series of global cities’ (1981: 288) that ‘serve as international centers for business decision-making and corporate strategy formulation’ (1981: 300). Thus, Cohen lays the path for Sassen’s subsequent emphasis on advanced producer services, rather than MNCs, in her global city thesis because he identifies the agglomeration of corporate services and other key international functions in a few cities. It was Cohen (1981: 302) who first argued that to measure the importance of a city as an international business centre, ‘One has to know if the city is a strong center of international banking and strategic corporate services.’ However, just as important to the formation of the global-cities research paradigm was the fact that Cohen – like Heenan before him – also broadened his argument. Crucially, Cohen argued that ‘[Global cities] have emerged as cities for the coordination and control of the NIDL [new international division of labour]’ (1981: 300). In other words, Cohen established the strategic control principle that comes to be the very essence of Sassen’s global city thesis, and which forms the central tenet of all of the subsequent literature.

The last significant author on global control before Sassen is Friedmann (Friedmann and Wolff, 1982; Friedmann, 1986). Whilst Cohen was inspired by Palloix (1975) to consider a world hierarchy of cities – Cohen reduced Hall’s (1966) list of seven world cities to argue for the predominance of New York, Tokyo, and London –, he primarily concentrated on delimiting the new structure of urban hierarchy in the United States. Cohen argued that in ‘drawing decision-making activities away from national or regional centers’ (1981: 311), the NIDL had led to an increased centralization of ‘international corporate decision-making and corporate services’ (1981: 300) in some US cities (New York and San Francisco are the only US cities Cohen classifies as global cities). Consequently, it was left to Friedmann to expand on the idea that the relative economic power of some cities could be understood in a hierarchical fashion by outlining a world-wide urban hierarchy of global control.

Friedmann and Wolff (1982) explicitly set out to establish an agenda for research and action on the premise that the world’s dominant cities (i.e. those few cities ‘in which most of the world's active capital is concentrated’ (page 309)) are increasingly integral to the formation of a world-wide economic system. Friedmann and Wolff contended that capital assigns spatial dominance to the city, ‘World cities are the control centres of the global economy’ (Friedmann and Wolff, 1982: 319), that trans-national capital allocates control to just a select few of those cities, and that without these ‘world’ cities ‘the world-spanning system of economic relations would be unthinkable’ (1982: 312). This axiom of the world city ‘approach’ allowed Friedmann and Wolff to speculate as to the possibility of a hierarchy of world cities, of global control, in the capitalist world. Indeed, four years later, Friedmann (1986) went on to outline his famous ‘world city hierarchy’ that – following an idea from ‘world-systems analysis’4 – classified thirty world cities as either primary or secondary in either the core or the semi-periphery of the world economy: of those, only six core cities, and two semi-peripheral cities, are classified as primary.

It was by working within the established neo-Marxist tradition and epistemological framework of the world city – where ‘The most inherent feature of the world city is its global-control function’ (King, 1991: 25) –, that Sassen could propose, in 1991, her global city concept and a ‘new strategic role for major cities’ (1991: 3). In her neo-Marxian magnum opus she envisioned a poly-nodal world economic system with New York, London, and Tokyo cemented as the ‘primary loci’, the points de capiton (anchoring points), of a globalization (not urbanization) with a triumvirate lay-out across North America, Western Europe and Asia. Leading urban centres, termed global cities, were those that now functioned ‘as centers of finance and as centers for global servicing and management’ (Sassen, 1991: 324) having both specialised attributes – ‘as key locations for finance and specialized service firms, [and] as sites of production, including the production of innovations, in these leading industries; and … as markets for the products and innovations produced’ (Sassen, 1991: 3) – and specific internal geographies of inter-service firm relations – ‘the new dynamic of agglomeration’ (Sassen, 2001: xx) whereby services from competing firms are packaged together for transnational clients in central business districts. Thus, just a few global cities are afforded a powerful centrality with regard to the structure and function of the world economy through their capacity for the service, management, command, control, and domination of globalization.

An understanding of the neo-Marxist theoretical heritage of the global city concept (vide ante) exposes its inherently familiar limitations as a form of economism and structural-functionalism. In addition, however, it must also be noted that the empirical basis of the global city concept is an assumption, not a proven fact. The internal geography of the global city concept – its assumption of service firm interactive agglomeration – has always lacked a convincing evidential basis: ‘The dominance of London, New York and Tokyo, for example, is more often asserted than demonstrated’ (Short et al., 1996: 698); or, ‘a configuration of global cities is still conceptually and empirically suspect’ (Gottdiener & Budd, 2005: 41), are just two quotes that echo the acknowledged evidential gap (also see Abu-Lughod, 1995; Storper, 1997; Smith, 2001). The carapace of the global city concept has not been broken open to experimental test; no proof-of-concept has been advanced, let alone proven. The validity of the concept is dependent on two assumptions, and therefore may well be premised on a double fiction. On the one hand, the fiction of inter-service firm interaction within individual cities ‘packaging’ their services for TNCs, and on the other hand, the fiction that such activity amounts to a strategic control of the global economy rather than just one aspect of its geographical management. Tout court, it is commonplace nowadays for economists (e.g. Glaeser, 2011) and economic geographers (e.g. see Storper & Venables, 2004) to look beyond capital, labour, and land to argue that clustering, concentration and density in cities and their wider metro-regions are the élan vital for economic growth. However, the specificity of Sassen’s global city concept is such that her argument is quite different and consequently accepting it may well be akin to believing in phlogiston or vitiated airs: do firms really co-operate or do they just sell services to one another? And does that really amount to being in control of the global economy? It all seems highly unlikely, especially after the subprime financial crisis (2008–pres.) which led to the portrayal of both London and New York as out-of-control ‘gambling dens’ or ‘casinos’ (see Zaloom, 2010a).

Lastly, we can turn to the latest phase of neo-Marxist scholarship that has been undertaken by Taylor (and numerous co-authors). Taylor has proposed the existence and importance of what he calls a ‘world city network’. However, Taylor’s aim to build on the neo-Marxist epistemological heritage in urban studies (vide ante), and specifically on Sassen’s assertion as to the importance of advanced producer service firms for strategically controlling the world economy through some leading cities, was not to address the lack of empirical evidence detailing connections and collaborations between advanced producer service firms in global cities, but rather to aim to describe the external relations of world cities to demonstrate those cities in command and control of the global economy. It is to this very latest work in the neo-Marxist tradition that the paper now turns, not only to strengthen the case for the dismissal of the neo-Marxist approach in toto as an inadequate envisioning of the global economy, but also to carefully debunk the ‘world city network’ as a structuralism bedevilled by a sorites paradox, as a non-sequitur because a ‘world city network’ of global command does not logically follow from its underpinning INM.

The global city concept interregnum (a.k.a. the interlocking world city network impasse)

‘What is there under your wallpaper?’ (Georges Perec, 2008)

A number of critiques of Taylor’s interlocking world city network have been made, but they are not particularly telling because they often just repeat what Taylor himself says about his world city network and INM. First, critics point out that the agents in Taylor’s interlocking world city network are firms. But as Taylor himself points out (2004b) he does this to avoid reifying cities as agents, and that is why Taylor is clear that in his network the firms (the sub-nodal level) are the ‘prime agents’: ‘The key point in this assumption is that it is the firms that are creating the flows and therefore it is they who define the world city network’ (Taylor, 2003: 33) and ‘global service firms are the key agency in world city network formation’ (2004b: 297). Second, critics note the economism of Taylor’s interlocking world city network decrying its ‘narrow’ focus on advanced producer service firms. But again that is its very raison d’être: ‘the research is very big geographically – global – but very narrow in topic’ (Taylor, 2004a: 3). Taylor’s approach purports (see Taylor, 2001: 518) to build on the neo-Marxist argument that advanced producer service firms are central for the strategic control of the global economy5. Third, critics are again only telling Taylor what he already knows when they point out that his raw data is attributional (e.g. see Nordlund, 2004; Robinson, 2005: 758–9). Taylor knows this, he is not in the alchemic business of turning apples (attributes) into oranges (relations) as Nordlund (2004: 292) supposes. Taylor never claims to have actual flow data, but rather – working in the tradition of calibrated spatial interaction modelling – describes intercity relations derived from the measurement of his set of attribute data. Fourth, although Neal (2011) correctly identifies the structural determinism of Taylor’s INM, it is not so much the structural, but rather the determinism which concerns him. This is a mistake because it means he fails to see that it is the full blown structuralism – where the terms (the cities) are reduced to mere ‘nodes’ – of Taylor’s approach which is where the fundamental problem lies. Indeed, if all the extant critiques of Taylor’s approach listed above miss the target, what is so wrong with it? What doesn’t Taylor admit to himself? What critique is there that totally invalidates everything written on the interlocking world city network? Well let me explain why I think the INM which underpins Taylor’s specification of the world city network is a fallacy and impasse for global urban studies.

To attempt to address a paucity of data on inter-city relations across the world – rather than the dearth of intra-city data about relations between advanced producer firms within the few global cities that Sassen identifies – Taylor (2004a) has worked with numerous co-researchers to identify the cities in which 100 independent rival firms predominantly specialising in one of six business services (accountancy, advertising, banking/finance, insurance, law, management consultancy) have located their offices. He has done this so that by adding up those offices on a city-by-city basis he can score and then rank those cities to ascertain which of them are in command and control of the global economy (because they contain the most advanced producer service firms)6. But this approach has two inevitable and fatal corollaries.

First, in Taylor’s INM individual offices from the 100 corporate service firms are aggregated or ‘lumped together’ within individual cities. Indeed, it is that summation which is the essential in situ ingredient to making, to defining, a world city within the world city network. In other words, the logic of Taylor’s methodology is one of adding up the presences of individual ‘global’ service firm offices in a city until one has, at a certain quantity, a world city of a particular rank or type. Now, the empirical basis of Taylor’s numerous writings is that the data he has collected concerns the urban locations of the intra-firm office networks of globalized business-service firms, not the inter-firm networks between firms in cities. Thus, independent unrelated service firm offices are added together when there is no evidence that they can be to define and rank the importance of any world city in a network of intercity relations7.  Reductio ad absurdum, Taylor’s is a methodology that ensures that any world city is never more than just a container of a quantity of unconnected, that is to say discrete, ‘global’ service firms. Taylor is not interested in, and has no data on, the relations, connections, and networks between the different 100 service firms8, but crucially it would be evidence of those inter-firm relations within ‘nodes’ (i.e. cities) that would constitute empirical evidence for Sassen’s global city concept in whose concept cities do add-up to be more than just mere containers of unconnected rival advanced producer service firms (see Figures 1 and 2 to understand how the concepts of Taylor and Sassen are fundamentally contradictory and incompatible; also see Smith & Doel, 2011)9.

Second, mindful of the neo-Marxist tradition in which he is working, and which he did much to encourage (e.g. see Knox & Taylor, 1995), Taylor is undone by his need to locate command and control in some cities. Taylor is caught-out by his ambition to want to tell a big story, to classify some of the world’s cities into ranked types (‘Mega’, ‘Major’, ‘Medium’, and ‘Minor’) to reveal a hierarchy of command and control in a ‘world city network’ (see Figure 3), because if cities (the ‘nodes’) are mere containers of independent in vacuo commercial service firms (i.e. they are unrelated/unconnected to one another) any attempt to rank them is inevitably bedevilled by a sorites paradox problem.

Figure 3: Taylor’s typology of global command centres

fig 3

fig 4
Source: Taylor (2004a: pages 90 & 72)

The classic telling of the sorites paradox concerns the difficulty of answering the seemingly simple question: how many pebbles make a heap? Two or three are not a heap; 1,000 clearly are. There is no agreed threshold for ‘heapness’, a ‘heap’ is intrinsically vague, but we think we know one when we see one. The sorites (from the Greek soros, meaning a ‘heap’) paradox highlights how little-by-little arguments, such as that which Taylor adopts to define, and then subsequently classify and rank world cities, are a flawed form of reasoning. To reveal a hierarchy within the ‘world city network’ Taylor has to first of all decide how many service firm offices a city must contain to make it a ‘world city’. That is to say that there is a point where the addition of one office makes a difference. A precise number of offices equates to a world city, to a ‘Mega’, a ‘Major’, or such-and-such a rank of world city, but crucially because Taylor’s is a little-by-little argument (adding together independent offices of rival commercial service firms to make a world city) the distinction/boundary between a world city and not a world city, a ‘Mega’ world city and say a ‘Major’ world city, is a count of one office that can never on its own have that effect. In other words, the paradox of Taylor’s modus ponens methodology looks like this:

1 office does not make a world city
If 1 office does not make a world city, then 2 offices do not.
If 2 offices do not make a world city, then 3 offices do not.
[and so on until ...]
If 9,999 offices do not make a world city, then 10,000 do not.
So, 10,000 offices do not make a world city.

Or conversely this:

10,000 offices is a world city.
If 10,000 offices are a world city, so are 9,999 offices.
So, 9,999 offices is a world city.
If 9,999 offices are a world city, so are 9,998 offices.
So, 9,998 offices are a world city.
If 2 offices are a world city, so is one office.
So, one office is a world city.

The addition of each individual independent office is not fully constitutive of the overall process of defining a world city and consequently Taylor’s definition of a world city – and his ranking of them as ‘Mega’, ‘Major’, ‘Medium’, or ‘Minor’ according to a numerical value – is not only hopelessly arbitrary, but is also a fallacy as one independent office has an effect (defines the rank of a world city) that it would never do on its own.

In nuce, Taylor’s work is founded on an INM which serves as the subjective and subjunctive determining ground on which the very existence of a meta-geographical world city network depends. But it is ironic that in pursuing his analysis of transnational service-producing firm-oriented geographical networks to gain an insight into the function of cities – in striving for more and more quantitative specification to score and rank ‘nodes’ (i.e. cities) – all urban specificity has vanished so that the world city network is a fully structuralist contradictio in terminis. And there is no way out of this. The flaw is fundamental so that the INM and world ‘city’ network is of no use so far as understanding the role of cities in the global economy is concerned.

Beyond the global city concept

'de omnibus dubitandum' (Said to be Karl Marx’s favourite motto)

‘On the word of no one’ (Motto of The Royal Society)

We have learnt how the global city concept (Sassen, 1991), – because of its neo-Marxist heritage – is extremely specific: it is not, was never intended to be, a universal urban theory for all cities from New York to the proverbial ‘nowheresville’. The global city concept is not a catch-all applicable to all cities, for everything that has or is being written on the general topics of cities in globalization. The global city concept contends that only the handful of cities that specialize in international financial and commercial services are necessary for the control and management of the global economy, other cities are of a different modality – of either lesser or no importance for that particular end. In short, the global city concept has a singular concern for identifying the locatedness of urbancommand and control within a global economy: with how strategic control is a prerogative of just a few power-towns from the more advanced economies.

So, how can we truly move beyond the global city concept, the INM-world city network, and the neo-Marxist tradition as a whole? Well my answer is that a new start can be made by drawing on literatures that throw into question the deep-seated idea in urban studies that economic command and control comes from centralization, concentration, clusters, and agglomeration in cities (also see Amin & Thrift, 2002). A number of sociologists (e.g. Knorr Cetina et al., 2000, 2004; Knorr Cetina and Bruegger, 2001, 2002a, 2002b, 2004), anthropologists (e.g. Zaloom, 2003, 2004, 2005) and organizational researchers (e.g. Beunza and Stark, 2003, 2004, 2005) have produced social studies of finance that, amongst other things, are attentive to the everyday practices (see Schatki et al., 2001) of financial firms in global cities. And what is pertinent is that their findings suggest that there is no meaningful command and control of the global economy in or by global cities as such. Inspired by ANT, and especially Callon’s (1998) insistence that purportedly ‘economic’ practices are always already social and cultural practices, SSF researchers have carried out ethnographies of some international financial services. This emphasis on practice, accomplishment, and cultural economy is consistent with attempts in urban studies to establish a new conceptual architecture for globalization and city research (Beaverstock et al., 2002; Beaverstock and Doel, 2001; Doel and Hubbard, 2002; Smith, 2003a, 2003b, 2005, 2007a, 2007b; Smith and Doel, 2011). From a growing SSF literature let’s consider just a few examples that challenge the notion that command and control resides in the function of global cities as international financial centres.

Beunza and Stark (2003, 2004, 2005) conducted ethnographic research in an international investment bank based in New York’s World Financial Center to study how the trading strategy of arbitrage is achieved through the formation of what Latour (1991) calls ‘socio-technical networks’, a form of ‘cognition that is socially distributed across persons and things’ (Beunza and Stark, 2003: 141). Amongst other findings, Beunza and Stark observe that physical proximity to other companies is important only for the practice of some arbitrage strategies (e.g. merger and convertible bond arbitrage). Being in a global city is not an important factor for investment banks engaged in arbitrage. Competitive advantage lies in producing a ‘community of interpretation’ – made up of interactions between technologies, humans, and ideas – within individual banks, not between competing banks. There is no summation. Manhattan’s investment banking does not add up. Moreover, Beunza and Stark (2003) discuss the changing urban geography of finance in Manhattan in the wake of the attacks on 11th September 2001. Their findings go some way to supporting an investment banking decentralization thesis10 because, despite city and state programmes of economic incentives to keep companies in Lower Manhattan, ‘an exodus seems to be in place, with companies such as Lehman Brothers, Aon, Pillsbury Winthrop, Dresdner Kleinwort Wasserstein and ABN Amro leaving the area to more expensive locations in Midtown or more distant offices in Brooklyn and Jersey City’ (Beunza and Stark, 2003: 154). This decentralization process is not simply dispersal for security reasons because the process was already underway before September 2001. The attacks have simply accelerated the decentralization trend: ‘[T]he real locus of modern finance is not the Exchange but the trading rooms’ (2003: 158)11. Unlike Kotkin’s (2005) general and therefore flawed decentralization critique of Sassen’s global city concept, Beunza and Stark’s (2003) observations are relevant because the observed decentralization is by an advanced producer service sector that is essential to the assumptions of the global city concept.

Zaloom (2003, 2004, 2005, 2010a, 2010b) also emphasizes the socio-technical in her ethnographic research on the trading floors and dealing rooms of Chicago and London. What is interesting is that she inadvertently throws into doubt any idea of the global economy being controlled, let alone commanded. Zaloom’s ethnography focuses on the discipline of speculators on trading floors and in dealing rooms. With reference to Foucault she observes how speculators are trained, and train themselves, to block out all but the rhythms of ‘the market’. Speculators discipline themselves to shed ego, affect, and individuality, so as to be no more than embodied instruments, responsive to the whims and twitches of the market: which is, after all, the highest authority. The traders are constantly disciplined and reminded not to delude themselves that through their own powers of reason they can outwit the market. Zaloom’s ethnography shows that the market is boss, and that the market is an event, a situation. Indeed, what all of these SSF findings demonstrate is a need to move away from the lingua franca of neo-Marxism; not towards a revival of empiricism, but towards a form of poststructuralism that is attentive to the hesitant, uncertain, and inconclusive taking place of events (see Smith & Doel, 2011).

And, moving on from Zaloom’s studies, where she explained how human control of trading is circumvented by the market, to the present where because of advances in the computerization of international financial centres the role of the trader is to design trading patterns to instruct computers on how to do the actual trading12, it is clear that received understandings of financial centres as places where male traders (as described by authors such as McDowell back in 1997) shout down several telephones at once is totally out-dated. The ability of machines to run data and execute trades very quickly means that the vast majority of share trading is now done through the use of computers pursuing ‘high-frequency strategies’ where shares may only be held for a matter of milliseconds (traders are just too slow at processing information (see Anon., 2007)), and algorithms based on statistics that are designed to predict market behaviour (i.e. prey on fear) and which require relatively few people – data scientists or ‘Quants’ (quantitative analysts with PhDs in mathematics or theoretical physics) – for programming and oversight purposes. Indeed, whilst the systemic risk – that which is not yet fully actual – to today’s global economy has perhaps moved away from bad lending practices and perverse incentives (salaries and bonuses) for traders (although there is not much evidence of this), the financial derivatives which triggered the 2008 crisis – which turned billions into trillions by repackaging subprime mortgages into attractive financial products – are actually increasing in importance as a source of revenue and fees for banks despite their evidential role in producing the 2008 financial crisis with its devastating consequences for the world’s economies and societies. Banking now resides primarily with the machines, programmes, algorithms, software, and trading methods – ‘black-box trading’, ‘algo trading’, ‘robo trading’, ‘high-frequency trading’ – that make financial centres fly-by-wire and markets much more volatile. Some twenty five years since the ‘Big Bang’ in the City the world’s international financial centres are now cyborgs13. For example, on the 6th of May 2010 some 19.4 billion shares were high-frequency ‘traded’ in one day, more than in the whole of the 1960s, producing what came to be known as a ‘Flash Crash’ because the NYSE temporarily froze causing panic and indexes to drop sharply. In fact, the global financial crisis – heralded by the bankruptcy of the investment bank Lehman’s Brothers in 2008 – and the subsequent recessions, stagnations, cut-backs and social uprisings across many advanced economies due to accumulated private, corporate and sovereign debt, perhaps now already only serves to remind us that capitalism is beyond command and control per se because systemic risk and evental crises are innate to capitalist assemblages, and even more so in this era of hyper-connection (Haldane & May, 2011).

In sum, the emphasis on practices, performativity, and socio-technical networks in SSF is important because it forces an understanding that just because cities have large central business districts, the headquarters of some of the world’s major MNCs (e.g. Stephen Hymer and John Friedmann), identifiable ‘clusters’ of corporate and financial services (e.g. Robert Cohen and Saskia Sassen), innovative cultural industries (e.g. Mark Abrahamson (2004), and countless others), does not mean that ipso facto such cities are ‘organizing nodes’ controlling and commanding the world’s capitalist economies. In short, by focusing on the acts accomplished by financial services in global cities, and by considering these active accomplishments to be performed (as events) rather than preformed (as functions), SSF scholars have inadvertently challenged the modus operandi, and consequently the locus operandi, of the global city concept (and the interlocking world city network model). What is more, the ethnographic fieldwork of SSF scholars suggests that if power is not centralized, clustered or agglomerated, this is because it is decentralized and distributed across socio-technical networks composed of humans, non-humans, and ideas. In line with this kind of thinking, highly connected cities such as international financial centres need to be reconceptualised as networks and assemblages (Smith, 2003a, 2003b, 2007a, 2007b, 2010), where power is always diffuse, always circulating, always mobile, always in-between, and always co-related (Smith, 2007b).

Finally, because extant research in neo-Marxist global-urban studies is analogous to a Greek tragedy, where the ending is known so that the drama of the play is not what will happen but rather how it will happen, this paper has prepared the ground for an alternative approach – a new theoretical framework that builds on the work of poststructuralism, SSF, cultural-economy, ANT, and post-continental philosophy – to urban studies that is attentive to how cities are multiplicities, complex assemblages of multiples, alive with both events (not just structures) and that which is not yet fully actual. As Smith and Doel (2011) have argued the city is not ‘One’ as so many have supposed, but is rather a multiple of multiples, a multiplicity that only becomes ‘One’ when it is assembled, fixed and lent consistency. However, a multiplicity only gains consistency on the basis of a ‘founding’ element that does not belong to the multiplicity. For example, the computerisation of trading through algorithms by ‘Quants’ in the City (and other financial centres) is not founded on value (worth): ‘Quants use past market patterns to fashion signature deals based on mathematically obscure relationships among securities. They develop strategies without reference to underlying debts or to companies that generate value’ (Zaloom, 2010a: 23). Indeed, it is profit that legitimates the practice of algorithmic trading, not its relationship to value per se. All the programs, models and software are baseless because they are only concerned with the estimation of value, with pricing, with the price a stock buys and sells for. In other words, it is not only the endemic contradictions of capitalism (i.e. the fact that booms and busts are, as Marx knew only too well, inherent to the logic of capitalism) and the hundreds of crises of modern finance since the 1973 global property market crash and oil crisis (e.g. the 1987 global stock market crash; the Japanese asset price bubble of the 1980s; the emerging market debt crises of the early 1980s and then again in 1997–8; the dot-com bubble, and so on) that escape the algorithms, but also what companies themselves actually do and own. Only the way a company’s stocks move is of interest (of profit) to the ‘Quants’, their managers, and ultimately much of modern banking. And this is why another event such as the financial crisis of 2008 can occur, for what does not belong in the ‘Quants’ algorithmic models, what is not included (the ‘founding’ element) in moving all the money around to ‘[Let’s] make nothing but money’ (Bear Stearn’s infamous self-description), is precisely that which is inconsistent and incalculable – people’s lives, their jobs, their workplaces, their homes, their families and communities – and which can destabilize and unbind the networks of financial capitalism (also see Smith & Doel (2011: 12) on sub-traction and the 2008 financial crisis). And the ‘Quants’ and the banks know this, but they carry on anyway because of how they are re-numerated (annual bonuses), they have already been paid when – following their mathematical acrobatics – several years later a company goes bankrupt, a home is repossessed, and so on.


The neo-Marxist tradition behind the global city concept and world city network INM is guilty of an idealisation of capitalist economic globalization as structured and in need of urban potentates that serve as kingpins holding power and control in place. In other words, it is a tradition that affords a systemic and strategic significance not to the process of urbanization under globalization, but rather to a few quintessential financial and knowledge service-based cities which are assumed to be in control of an economic globalization that is conceived as a structured totality: Los Angeles, Chicago, New York, Rio/São Paulo, London, Paris, Singapore, Tokyo (Friedmann, 1986) … New York, London, Tokyo (Sassen, 1991), Los Angeles and Frankfurt (Sassen, 2003) … London, New York, Hong Kong, Paris, Tokyo, Chicago, Frankfurt, Miami (Taylor, 2004a). Little wonder then that with the mondialisation of the urban (Lefebvre, 2009) the global city concept is not de rigeur for those wishing to research the developing world’s rapid urbanization – exemplified by such countries as China and India – or even the globalization of the world’s thousands of other cities.

As it stands two strands of contemporary urban studies have been discussed which capture much of the research effort about understanding cities in globalization. However, as this paper has explained contemporary urban studies is at a critical juncture because both strands are flawed. First, the global city concept as it has developed within a neo-Marxist tradition envisions the global economy as a structured totality with the unproven assumptions that, first, globalization has a strategic need for global control through certain cities, and second that, inter-firm co-operation between rival service firms within individual cities occurs and can be squared with a capacity for ‘global management’ (also see Jones, 2002). Second, the structuralist ‘interlocking world city network’ of Taylor and co-authors contradicts Sassen’s global city concept, and bedevilled by a sorites paradox, is a non-sequitur and impasse for the neo-Marxist tradition. Consequently the basis for a third research direction has been proposed by this paper. It has been shown that SSF literatures can be utilized as a foreground to develop a broadly poststructuralist understanding of cities in advanced economies as practices and performances, as actor-networks and assemblages (see Smith, 2003a, 2003b; Farías & Bender, 2010), and as multiplicities (Smith & Doel, 2011). In sum, this paper has demonstrated the need for an alternative ex novo theorisation of cities as globalization that has no investment in the long-held and pablum neo-Marxist myth that the global economy is under urban strategic control and command.


This paper was presented as a keynote address to the Russian city of Kazan in the September of 2011. I would like to thank the Kazan City Municipality and Ilsur Metshin (the Mayor of Kazan City) for their invitation to participate in the 2011 ‘Made in Kazan’ event [See:]. Thanks are also due to the politicians, civil servants, business leaders, architects, journalists, academics and students for their interest, questions and feedback on the talk. Thanks also to Nadir Kinossian, and to Vladimir Gritskikh and Sergey Golovin at the Urban Development Directorate of Kazan. Thanks also to Jim Leaviss of M&G Investments (London) for his clarifications.


Abrahamson M (2004) Global Cities (Oxford University Press, Oxford)

Abu-Lughod JL (1995) ‘Comparing Chicago, New York, and Los Angeles: testing some world cities hypotheses’, in Knox P and Taylor P eds. World cities in a world-system (Cambridge University Press, Cambridge), pp. 171–91

Alger C (1990) ‘The world relations of cities: closing the gap between social science paradigms and everyday human experience’, International Studies Quarterly, 34, 493–518

Amin A & Thrift N (2002) Cities: Reimagining the Urban (Polity, Cambridge)

Anon. (2007) ‘Ahead of the tape: The best newsreaders may soon be computers’, The Economist, June 21st.

Bassens D, Derudder B and Witlox F (2010) ‘Searching for the Mecca of finance: Islamic financial services and the world city network’, Area, 42 (1), 35–46

Bassens D, Derudder B and Witlox F (2011) ‘Setting Shari’a standards: On the role, power and spatialities of interlocking Shari’a boards in Islamic financial services’, Geoforum, 42, 94–103

Beaverstock J V and Doel M A (2001) ‘Unfolding the spatial architecture of the East Asian financial crisis: the organizational response of global investment banks’, Geoforum, 32(1), 15–32

Beaverstock J V, Doel M A, Hubbard P J and Taylor P J (2002) ‘Attending to the world: competition, cooperation and connectivity in the World City network’, Global Networks, 2(2), 111–32

Beaverstock JV, Smith RG and Taylor PJ (1999) ‘A roster of world cities’, Cities, 16 (6), 445–58

Beaverstock JV, Smith RG and Taylor PJ (2000) ‘World City Network: A New Metageography?’, Annals of the Association of American Geographers, 90 (1), 123–34

Braudel F (1984) The Perspective of the World (Collins, London)

Brenner N and Keil R eds. (2006) The Global Cities Reader (Routledge, London)

Buenza D and Stark D (2003) ‘The organization of responsiveness: innovation and recovery in the trading rooms of Lower Manhattan’, Socio-Economic Review, 1, 135–64

Buenza D and Stark D (2004) ‘Tools of the trade: the socio-technology of arbitrage in a Wall Street trading room’, Industrial and Corporate Change, 13(2), 369–400

Buenza D and Stark D (2005) ‘Resolving identities: successive crises in a trading room after 9/11’, in Foner N ed. Wounded City (Russell Sage Foundation, New York), pp. 293–320

Callon M ed. (1998) The Laws of the Markets (Blackwell, Oxford)

Coghlan A and MacKenzie D (2011) ‘The Hard Core of Power’, New Scientist, 22nd October, 8–9

Cohen R B (1981) ‘The new international division of labour, multinational corporations and urban hierarchy’, in Dear M and Scott A J eds. Urbanization and Urban Planning in Capitalist Society (Methuen, New York and London), pp. 287–315

Daniels PW (1993) Service Industries in the World Economy (Oxford, Blackwell)

Doel M A and Hubbard P J (2002) ‘Taking world cities literally: marketing the city in a global space of flows’, City, 6(3), 351–68

Farías I and Bender T eds. (2010) Urban Assemblages: How Actor-Network Theory Changes Urban Studies (Routledge, London)

Friedmann J (1986) ‘The world city hypothesis’, Development and Change, 17, 69–83

Friedmann J (1995) ‘Where we stand: a decade of world city research’, in Knox P and Taylor P eds. World Cities in a World-System (Cambridge University Press, Cambridge), pp. 21–47

Friedmann J and Wolff G (1982) ‘World city formation: an agenda for research and action’, International Journal of Urban and Regional Research, 6(3), 309–44

Geddes P (1915) Cities in Evolution (Benn, London)

Glaeser E (2011) The Triumph of the City (The Penguin Press, New York)

Gottdiener M and Budd L (2005) Key Concepts in Urban Studies (Sage, London)

Haldane A and May R (2011) ‘Systemic risk in banking ecosystems’, Nature, 469, 351–55

Hall P (1966) The World Cities (World University Library, London)

Hall P (1977) The World Cities (World University Library, London) 2nd ed.

Hall P (1984) The World Cities (World University Library, London) 3rd ed.

Hall P (1998) Cities in Civilization (Weidenfeld & Nicolson, London)

Hanssens H, Derudder B, Taylor PJ, Hoyler M, Ni P, Huang J, Yang X and Witlox F (2011) ‘The changing geography of globalized service provision, 2000–2008’, The Service Industries Journal, 31 (14), 2293-2307

Heenan DA (1977) ‘Global cities of tomorrow’, Harvard Business Review, 55, 79–92

Houellebecq M (2005) The Possibility of an Island (Weidenfeld & Nicolson, London)

Hymer S (1972) ‘The multinational corporation and the law of uneven development’, in Bhagwati J ed. Economics and World Order (Macmillan, London), pp. 113–40

Johnston R (2000) ‘World city’, in The Dictionary of Human Geography Johnston R, Gregory D, Pratt G and Watts M eds. (Blackwell, Oxford), pp. 900–1

Jones A (2002) ‘The “global city” misconceived: the myth of “global management” in transnational service firms’, Geoforum, 33, 335–50

King AD (1991) Global cities: post-imperialism and the internationalization of London (Routledge, London)

Knorr Cetina K and Bruegger U (2001) ‘Transparency regimes and management by content in global organizations: the case of institutional currency trading’, Journal of Knowledge Management, 5(2), 180–94

Knorr Cetina K and Bruegger U (2002a) ‘Inhabiting technology: the global life form of financial markets’, Current Sociology, 50(3), 389–405

Knorr Cetina K and Bruegger U (2002b) ‘Global microstructures: the virtual societies of financial markets’, American Journal of Sociology, 107(4), 905–50

Knorr Cetina K and Bruegger U (2004) ‘Traders’ engagement with markets: a postsocial relationship’, in Amin A and Thrift N eds. Cultural Economy Reader (Blackwell, Oxford), pp. 121–42

Knorr Cetina K, Karin D and Bruegger U (2000) ‘The market as an object of attachment: exploring postsocial relations in financial markets’, Canadian Journal of Sociology, 25(2), 141–68

Knorr Cetina K, Karin D and Preda A (2004) The Sociology of Financial Markets (Oxford University Press, Oxford)

Knox P and Taylor P eds. (1995) World Cities in a World-System (Cambridge University Press, Cambridge)

Kotkin J (2005) ‘Eclipse of the city’, Forbes Global, July 4th, 76–81

Latour B (1991) ‘Technology is society made durable’, in Law J ed. A Sociology of Monsters: Essays on Power, Technology, and Domination (Routledge, London), pp. 103–30

Lefebvre H (2003) The Urban Revolution (University of Minnesota Press, Minneapolis)

Lefebvre H (2009) State, Space, World: Selected Essays (University of Minnesota Press, Minneapolis)

McDowell L (1997) Capital Culture: Gender at Work in the City (Wiley-Blackwell, London)

Neal Z (2010) ‘Refining the Air Traffic Approach to City Networks’, Urban Studies, 47, 2195–2215

Neal Z (2011) ‘Structural Determinism in the Interlocking World City Network’, GaWC Research Bulletin 379 (forthcoming in Geographical Analysis)

Newman P and Thornley A (2011) Planning World Cities: Globalization and Urban Politics (Palgrave Macmillan, 2nd ed.)

Nordlund C (2004) ‘A Critical Comment on the Taylor Approach for Measuring World City Interlock Linkages’, Geographical Analysis, 36(3), 290–96.

Palloix C (1975) L’internationalisation du capital (Maspero, Paris)

Perec G (2008) Species of Spaces and Other Pieces (Penguin, London)

Pereira R and Derudder B (2010) ‘Determinants of Dynamics in the World City Network, 2000-2004’, Urban Studies, 47(9), 1949–1967

Robinson J (2002) ‘Global and world cities: A view from off the map’, International Journal of Urban and Regional Research, 26(3), 531–54

Robinson J (2005) ‘Urban geography: world cities, or a world of cities’, Progress in Human Geography, 29 (6), 757–76

Sassen S (1991) The Global City: New York London, Tokyo, original edition (Princeton University Press, Princeton)

Sassen S (1994) Cities in a World Economy (Pine Forge, Thousand Oaks)

Sassen S (2001) The Global City: New York, London, Tokyo, revised edition (Princeton University Press, Princeton)

Sassen S ed. (2002) Global Networks-Linked Cities (Routledge, London)

Sassen S (2003) ‘More Than CitiBank: Who belongs in the global city?’, Topic Magazine, Issue 3, 110–17

Sassen-Koob S (1984) ‘The new labour demand in global cities’, in Smith M P ed. Cities in Transformation: Class, Capital, and the State (Sage, Beverley Hills), pp. 139–72

Schatzki T, Knorr-Cetina K and Savigny E eds. (2001) The Practice Turn in Contemporary Theory (Routledge, London)

Short J, Kim Y, Kuss M & Wells H (1996) ‘The Dirty Little Secret of World Cities Research: Data Problems in Comparative-Analysis’, International Journal of Urban and Regional Research, 20 (4), 697–717

Smith DA (2000) ‘Urbanization in the world-system: a retrospective and prospective’, in Hall T ed. A World-Systems Reader (Rowman and Littlefield, Lanham), pp. 143–68

Smith MP (2001) Transnational Urbanism: Locating Globalization (Blackwell, Oxford)

Smith RG (2003a) ‘World city actor-networks’, Progress in Human Geography, 27(1), 25–44

Smith RG (2003b) ‘World city topologies’, Progress in Human Geography, 27(5), 561–82

Smith RG (2005) ‘Networking the city’, Geography, 90(2), 172–76

Smith RG (2007a) ‘Place as network’, in Douglas I, Huggett R and Perkins C eds. Companion Encyclopedia of Geography, 2nd Edition (Routledge, London), pp. 57–69

Smith RG (2007b) ‘Poststructuralism, Power and the Global City’, in Taylor P, Derudder B, Saey P and Witlox F eds. Cities in Globalization: Practices, Policies and Theories (Routledge, London), pp. 260–72

Smith RG (2010) ‘Urban studies without “scale”: localizing the global through Singapore’, in Urban Assemblages: How Actor-Network Theory Changes Urban Studies edited by Ignacio Farías and Thomas Bender (Routledge, London), pp. 73–90

Smith RG and Doel MA (2011) ‘Questioning the theoretical basis of current global-city research: structures, networks and actor-networks’, International Journal of Urban and Regional Research, 35 (1), 24–39

Storper M (1997) The Regional World (Guilford Press, London and New York)

Storper M and Venables A (2004) ‘Buzz: the economic force of the city’, Journal of Economic Geography, 4, 351–70

Surborg B (2011) ‘World Cities are just “Basing Points for Capital”: Interacting with the World City from the Global South, Urban Forum, 22, 315–30

Taylor PJ (2001) ‘Specification of the World City Network’, Geographical Analysis, 33(2), 181–94

Taylor PJ (2003) ‘Generating Data for Research on Cities in Globalization’, in A Borsdorf and C Parnreiter eds. International Research on Metropolises: Milestones and Frontiers (Wien: Verlag der Österreichischen Akademie der Wissenschaften), pp. 29–42

Taylor PJ (2004a) World City Network: A Global Urban Analysis (Routledge, London)

Taylor PJ (2004b) ‘Reply to “A Critical Comment on the Taylor Approach for Measuring World City Interlock Linkages” by C. Nordlund’, Geographical Analysis, 36(3), 297–98

Taylor PJ (2011) ‘Measuring the world city network: new results and developments’, in ICTs for Mobile and Ubiquitous Urban Infrastructures: Surveillance, Locative Media and Global Networks (IGI Global), pp. 15–23

Taylor PJ and Walker D (2001) ‘World cities: a first multivariate analysis of their service complexes’, Urban Studies, 38(1), 23–47

Yeoh B (1999) ‘Global/globalizing cities’, Progress in Human Geography, 23(4), 607–16

Zaloom C (2003) ‘Ambiguous numbers: trading technologies and interpretation in financial markets’, American Ethnologist, 30(2), 258–72

Zaloom C (2004) ‘The Productive Life of Risk’, Cultural Anthropology, 19(3), 365–91

Zaloom C (2005) ‘The Discipline of Speculators’, in Ong A and Collier S eds. Global Assemblages (Blackwell, New York), pp. 253–69

Zaloom C (2010a) ‘The Derivative World’, The Hedgehog Review, Summer, 20–27

Zaloom C (2010b) Out of the Pits: Traders and Technology from Chicago to London (University of Chicago Press, Chicago)


* Richard G. Smith, Centre for Urban Theory, College of Science, Swansea University, UK, e-mail:

1. A common misconception is that the concept and term global city was first coined by the sociologist Saskia Sassen (e.g. see Robinson, 2002: 535). Sassen first referred to the global city in a 1984 publication (Sassen-Koob, 1984), but in fact the global city concept can be traced to Cohen (1981), and the termglobal city to Heenan (1977) (and even further back to Mao(ism), see Lefebvre, 2003: 169)).

2. The purpose of Heenan’s article was to inform two groups: (1) Chief Executive Officers (CEOs) on where to locate their corporate headquarters and regional offices; and (2) urban planners as to the benefits of attracting corporate tenants to their cities. Heenan used multivariate analysis (cf. Taylor and Walker (2001) who claim to have conducted the first study adopting this method) to reveal which cities rate highest amongst CEOs as regional office locations in Latin America and the Asia-Pacific.

3. Cohen (1981: 311 and 313) cites Hymer but fails to mention Heenan.

4. The origin of the world cities literature is commonly attributed to Geddes’ (1915) work on cities in evolution (e.g. see Johnston, 2000). However, Geddes’s (1915) and Hall’s (1966, 1977, 1984) works are merely the thematic origins of world cities research. The theoretical basis for many of the so-called ‘landmark’ contributions to the world cities literature (Friedmann and Wolff, 1982; Friedmann, 1986, 1995; Knox and Taylor, 1995, Taylor, 2004a) is the structural history of Braudel, the world-systems analysis of Wallerstein, and Fröbel’s idea of a New International Division of Labour.

5. And incidentally that is why: (1) post-colonial critics are misguided when they decry Taylor for ‘dismissing’ important cities around the world just because they are not service centres. When Robinson (2005: 759) criticises Taylor for having ‘passed over’ (2005: 759) evidence on other forms of globalization such as NGO locations in cities she is ignoratio elenchi: ‘[Taylor] asserts that even though “clearly cities in globalization involves more than financial and business services … the latter are the dominant networkers and I continue to focus on them” (Taylor, 2004a: 100)’ (Robinson, 2005: 759, my emphasis). In contrast, research on Islamic financial services does at least constitute a valid rejoinder to the ‘Western-centred’ bias of Taylor’s data set (e.g. see Bassens et al., 2010, 2011); and (2) critics such as Neal (2011) are similarly missing the point when they suggest other types of data to describe a world city network. Neal (2010, 2011) suggests airline traffic data as a source of relational data because techniques have now been developed to ameliorate the well-known biases found in those data sets. But frankly, this is to miss the point that airline traffic is not part of the neo-Marxist argument about how strategic command and control is achieved through advanced producer services.

6. What has not been noted is that, whilst the original data (e.g. see Beaverstock et al. 1999, 2000) is correct, the ‘GaWC 100’ data set developed and used by Taylor (2004a: 215–17) and his new co-researchers is erroneous because the researchers only extended the original data set from four to six sectors (incidentally, more producer service sectors were included in Sassen’s (1991) thesis and there are many others that could also be included (e.g. see Daniels, 1993)) forgetting to up-date it and keep it up-to-date. For example, Chicago’s Arthur Andersen LLP remains in the data listed under ‘Accountancy’ even though – after the 2001 Enron scandal – it effectively ceased operating in 2002. Odder still is that under ‘Management consultancy’ Andersen consulting is listed which is a company that became ‘Accenture’ at the beginning of 2001.

7. In his reply to Nordlund’s (2004) misjudged critique Taylor (2004b) insists that ‘All global service firm’s networks are different: they are idiosyncratic depending as they do on such matters as a firm’s geographical origin, its agglomeration history, its clientele, etc.’ and ‘That is why the method of deriving intercity relations depends upon aggregating a large number of office networks to iron out the idiosyncratic’ (page 298). Thus, working in the spatial interaction tradition – with its assumptions such as the ‘multiplicative element’ – Taylor adds together the ‘sub-nodal’ intra-firm urban office locations of individual service firms to weigh the significance of any ‘node’ (city). In other words, Taylor admits to adding together the urban office locations of different idiosyncratic advanced producer service firms to subsequently weigh, score and rank cities according to a scale of world city-ness.

8. The subprime financial crisis spread because of linkages between banks. Thus, it is inter-firm data, not Taylor’s intra-firm data, which matters when trying to understand connectivity in the global economy.

9. Researchers continue to make the error of believing Sassen’s and Taylor’s concepts to be compatible (e.g. see Pereira and Derudder, 2010; Neal, 2011). They make this mistake because they look no further than seeing advanced producer services as a theme spanning and linking both authors’ works.

10. Also see Zaloom (2010a: 21): ‘Geographically … Wall Street now lacks a single address. The sites are dispersed across the metropolitan landscape. Banks and trading floors have mostly moved to midtown Manhattan, clustered together with the high-end law firms that service their needs … Hedge funds cluster in Greenwich’. Indeed, Deutsch Bank is the only major bank left on Wall Street proper as – except for the iconic NYSE – the Street is no longer the cradle of American finance.

11. The latest evidential attempt to entangle the location of control in the global economy comes from complex systems analysis. A study analysed the share ownerships linking the world’s 43,060 transnational corporations together. The study showed how corporations are highly interconnected in that regard revealing a dominant ‘super entity’ core of 147 firms all of whom own each other and who consequently ‘control’ 40% of the total wealth in the network. Less than 1 per cent of firms (mainly financial institutions) own and control 40 per cent of the network (see Coghlan & MacKenzie, 2011). However, it would be a flawed exercise to build on that research to ascertain the urban headquarter locations of those transnational firms in the ‘super entity’ (such as Barclays Bank, JP Morgan Chase & Co, The Goldman Sachs Group, and so on). Complexity does not lead to simplicity. Finding out that those TNCs are located in New York, London, and the other usual suspects, does not get around the fact that ownership cannot easily be equated with control, precisely because ‘Most company shares are held by fund managers who may or may not control what the companies they part-own actually do’ (2011: 9). Indeed, not even held by professional fund managers as increasingly their stock-picking services are being replaced by tracker funds managed by computers which are transparent, low-cost and often afford better returns.

12. In excess of 70% of shares on the NYSE are computer traded (on ‘auto-pilot’) whether it is to do with the timing, price, and even quantity of an order.

13. Whilst trading is being enacted by algorithms, urban geographies are now also being designed for algorithms. The competitive advantage of ‘algo-trading’ depends on speed: milliseconds and microseconds really do matter as trading occurs at the speed of light. Consequently the practice of ‘algo-trading’ is actively shaping the internal and external urban geographies of contemporary financial centres. For example, in Manhattan the Western Union Building at 60 Hudson Street is just one ‘carrier hotel’ where firms locate for faster connectivity. The cluster of firms located there reduce ‘latency’ (the delay between order and execution) to gain an 8 millisecond advantage over rival firms based around Wall Street. And it is not only within Manhattan that a new economic geography is emerging. An 852 mile trench, to run a fibre-optic cable between Chicago and New York with a 13.3 millisecond round-trip, has been constructed by the company Spread Networks® to enable inter-city algorithmic trading through the reduction of latency.


Edited and posted on the web on 6th December 2011

Note: This Research Bulletin has been published in International Journal of Urban and Regional Research, 38 (1), (2014), 98-115