GaWC Research Bulletin 126

GaWC logo
  Gateways into GaWC

This Research Bulletin has been published in Urban Forum, 15 (1), (2004), 36-47.


Please refer to the published version when quoting the paper.


The Global Cities of Sub-Sahara Africa: Fact or Fiction?

I.J. van der Merwe*


Present studies of globalization and global cities have been biased towards Western core countries. Considering differences in urbanization tendencies and urbanism structures in industrialised First World countries compared to those in the developing Third World, the question is how applicable current global city definitions and associated methodologies are outside of the Western world? African cities in particular have received scant attention in global city research.  It will therefore be the specific objective of this exploratory paper to firstly compile a theoretical overview of the definitions and indicators of global cities as the concept may impact on Sub-Sahara Africa; and secondly, to apply some of the measuring instruments to lay the foundation for possible categorising of Africa's cities in a world and regional context. The results of the provisional analysis reveal an interesting picture. When the ranking of the 14 possible cities are executed according to the connectivity indices, four categories of cities emerge on an arbitrary basis.


In recent years there has been an explosion of academic interest in the notion of globalization. This interdisciplinary concept involves a complex set of related processes that have served to enrich the integration of socio-cultural, political and economic life in a post-modern world. Technological innovation and other global forces have brought about a more informed population in a shrinking world. Definitions of globalization usually register the growing integration of the world into an interconnected global economy, political structure and cultural network. World cities fulfil an important role as apexes and ordering centres in this global system. When one thinks of such cities, only a few classical names normally come to mind, e.g. New York, London and Tokyo. They are classified as such because of the intensity of their interaction as hubs of international finance and business, corporate and communication services, as well as carriers of mass information and culture. However, present studies of globalization have been biased towards Western core countries. Considering differences in urbanization tendencies and urbanism structures in industrialised First World countries compared to those in the developing Third World, the question is how applicable current global city definitions and associated methodologies are outside of the Western world?

African cities in particular have received scant attention in global city research. It will therefore be the specific objective of this exploratory paper to set the scene for a larger ongoing international project1 launched recently. This will be achieved, firstly, by compiling a theoretical overview of the definitions and indicators of global cities as the concept may impact on Sub-Sahara Africa; and secondly, by applying some of the measuring instruments to lay the foundation for possible categorising of Africa's cities in a world and regional context. By implication, the question is whether the integration of Sub-Sahara Africa into the global city network is an empirical fact or only a theoretical fiction? However, to analyse this issue we first need some clarity on what the 'global city' concept entails.


A variety of definitions, criteria and theories have developed over the course of time while many researchers have tried to interpret the phenomenon of the global city. In the process the notion of a world/global city2 became a fuzzy and sometimes speculative construct with more than one meaning. However, literature reviews by Beaverstock & Taylor (1999), Short & Kim (1999), Douglas (2000), Scott (2001), Taylor (2001), Doel & Hubbard (2002), and Smith (2003) help to structure the complex development of the various interpretations. A major compilation and review of the global cities literature was recently produced by the 'Globalization and World Cities Study Group' at the University of Loughborough, led by Peter Taylor and John Beaverstock (

Although the term 'world city' was probably coined as early as 1915 by Patrick Geddes and reintroduced by Peter Hall in 1966, Saskia Sassen (1994) observes that these earlier uses of the notion could not effectively incorporate the transnationalisation process of globalization. Hall (1966) mainly focused at that stage on certain great cities in which a disproportionate share of the world's most important business and political power was concentrated. Over the intervening period between the sixties and eighties the meaning of the concept and the list of world cities have been substantially revised.

Friedmann's (1986) world city hypothesis is viewed by many as an appropriate starting-point for contemporary developments on both the theory and empirical research associated with a special set of urban agglomerations as sites for global accumulation. He identified a 'world city hierarchy', where each city is defined in terms of its role as a major financial, manufacturing, transport and cultural centre. He argued in typical institutional fashion that a city's supreme position in a world system of cities is in the first place related to the relative degree of global economic centralization and power it hosts. Friedmann and Wolff (1982) further introduced the concept of a 'global network of cities' and identified command centres to control and articulate the new international division of labour being created by the information revolution and by the formation of multinational corporations. The result of this urban labour division was the appearance of service orientated cities in the developed world; a higher share of manufacturing cities in the developing middle-income world; and informal-economy cities, which were in the least-developed countries only weakly connected to global networks. The institutionalised global network idea persisted up to the present time, e.g. Sassen's (1994) transnational urban system, Short and Kim's (1999) global urban network and Taylor (2001), who identified the world city system as a network operating on different structural levels. World cities have become centres for the production and consumption of advanced services and information in support of the organization of global capital (Beaverstock, Smith & Taylor, 2000).

However, in the course of time a wider understanding of the concept became necessary. The notion that a world city is a bounded place plugged into a global space of nodes was rejected, in favour of a perspective on world-cityness as the achievement of flow performances globally decentralized and connected with varying degrees of dispersion and clustering. We no longer live and work in a bounded space of places, but in a globalized open system of flows through a multiplicity of sites and networked time. The shift from the idea of a hierarchy of world city nodes, developed by pioneers such as Friedmann (1986) and Sassen (1991; 1994) to the idea of a world city network of links and relations suggested by Beaverstock and Taylor (1999) represents a significant modified perspective. The flows of capital, people, information, commodities and practices between the connected world cities created a networked platform for certain places to become global city role players (Taylor et al, 2002). Subsequently a new empirical agenda has been taken up by Beaverstock et al (2000) to generate new relational data that followed Castells's (1996) 'space of flows' idea and focuses on the relationships between and connectiveness of cities that make up a world network society.

The above brief conceptual framework enables the summary of global city criteria and associated indicators in Table 1. This presented a functional link for the subsequent empirical investigation of possible global cities in Africa. The contributions of Friedmann (1986), Thrift (1994), Sassen (1994; 2002), Short et al (1996), Taylor (1997; 2001), Short & Kim (1999), Beaverstock & Taylor (1999), Douglas (2000), Lo & Marcotullio (2000), Doel & Hubbard (2002), Yusuf & Wu (2002) were extremely valuable in this regard.


"Sub-Sahara Africa is revealed as a huge information lacuna: the world's poorest region is a 'nowhere place' for the US world cities" (Taylor, 1997). This suggests that the prospects for a positive answer to the question whether there really are global cities in Africa are bleak. When previous researchers implemented the criteria in Table 1, the literature revealed only a few African cities in the various global city rosters (Beaverstock & Taylor, 1999). Africa is by far the continent with the fewest world cities in any of the classifications. The six cities most often indicated in the literature are Cairo and Casablanca in the North, Lagos in the West, Nairobi in the East, and Johannesburg and Cape Town in the South.

Globalization in Africa, however, should be interpreted within a specific historical, economic and cultural context (Stock, 1995). The apparently poor performance of the continent on the global city rosters cannot be understood without reference to its marginalized position in the global economy. This situation stems from the distinctive indigenous urban traditions and the urban imprints of the strong colonial heritage. In the twentieth century Africa's role in the world economy was mostly to supply raw materials and import consumer goods, leaving countries vulnerable to the policies of its colonial powers. Even today Africa is an extreme case of peripheral globalization. Africa's exports (2,3%) and imports (2,2%) in 1998 represent a small proportion of the world's total trade (Ajayi, 2001). African countries are isolated from the core markets and dependent on trade relations over which they have little control. In terms of trade and investment, their relative position in the world economy is declining rather than growing. Although foreign direct investment in developing countries increased in recent years, Africa's share of the total investment package has remained as low as 3 percent (Ajayi, 2001). No wonder many economists regard Africa as not really open to the globalization system. Further, the hierarchical nature of globalization has tended to give millions of Africans a raw deal as the most likely groups to be excluded anywhere in the world, if they attempt to cross the boundaries of poverty and powerlessness.

Notwithstanding this bleak picture, one cannot rule out the possibility that Africa's cities in a specific way do take part in the global network process. In a developing peripheral environment like Africa, the conceptualization and operationalisation of the global cities concept should be handled with some discretion and suppleness. Some of the negative factors that necessitates a differential look at the phenomenon relates to the continent's unique colonial history, its disadvantageous geography, its relatively low urbanization level in combination with a high population growth rate, the management capacity and sustainability of the cities, the proliferation of poverty and accompanying social disorders in the cities, the economic dependence on exports of primary products and subsequent marginal economic positioning of the different countries in the global system. From this profile it is obvious that Sub-Saharan Africa could not be expected to feature high on the global city rankings. According to the roster of Taylor et al (2002) and others, only Johannesburg features as a relative strong contender on the list (a regional command centre), with Cape Town, Lagos and Nairobi in a secondary position. Nevertheless, there is a prospect for more cities to play a specific role in the global setting as emerging world cities. These cities' fundamental role in the continental, regional or local context must be analysed in a systematic way to find their global niche.

A useful point of departure for an empirical evaluation is, first, to determine which countries have the best capacity to house global cities. Second, the largest cities in these countries will be selected for further analysis according to a specific set of criteria and indicators (Table 1). The raw data for the classification of the countries was extracted from the World Bank website ( For every quantitative indicator a proportional unit value (highest rank with a value of 1,0) were assigned according to the magnitude of the difference between the respective countries (Table 2). In the final ranking of the countries, the average of the six indicators was determined according to equal weights. The accompanying map (Figure 1) shows the 13 Sub-Saharan countries that seem the most promising to house global cities and should therefore be investigated further. South Africa and Nigeria are the top scorers, with Kenya in a strong third position. Ten other countries probably have an outside change to house emerging world cities in future.

On the basis of the rationale that a world city has to be a certain size to support world city formation, only the biggest city in each country was identified for further investigation. South Africa's superior position in the ranking justifies two cities. In Table 3 the 14 cities are listed and an indicator framework was constructed for each centre. The GaWC website ( of the University of Loughborough research group acts as the main data resource for this exercise. As a starting point we focused on the criterion of corporate/producer services and their function of connectivity between cities: Accounting; Advertising; Banking; Insurance; Law; and Management. Taylor's (2001) rationale in this regard is vital for this pilot study. The world city network operates on three levels: cities as the nodes, the world economy as the supra-nodal network, and advanced producer service firms forming a critical sub-nodal level. The latter represents the corporate service firms operating through cities and are the prime actors in world city network formation. The point of departure is each city's connectivity index. From this value the proportional percentage that each service sector contributes to that of the city's overall connectivity performance is calculated. Taylor et al. (2002) derives this index by multiplying the service value of every service firm in the city (size and referrals) by its service value in every other city and summing the product.

The results of the provisional analysis reveal an interesting picture. When the ranking of the 14 possible cities are executed according to the connectivity indices, four categories of cities emerge on an arbitrary basis. Johannesburg (1,0) stands out as Sub-Sahara Africa's best candidate for a 'global city', although it figures at a level less than half the strength of London (2,42), the world's number one global city. Cape Town (0,58) and Nairobi (0,55) operate on a connectivity level of about half that of Johannesburg and shows signs of secondary world city status. The next tier incorporates five cities with a notable potential to develop minor global city status in future (Lagos, Abidjan, Harare, Accra and Lusaka). The relative gap to the last six cities is quite significant and suggests that global city capacity might still be in the distant future for them. This does not imply that certain elements of global city activities are not already present in these cities. This is illustrated when the general connectivity indicator is disintegrated into the six relevant corporate service components. Except for Maputo and Addis Ababa, all the cities performed better than the world average in two or more services. Johannesburg even excelled on the whole spectrum, except for accounting. As the case in most developing countries, advertising and accounting are ubiquitous in most cities.


Although it is premature to draw any conclusions from this pilot analysis, it appears that the existence of global cities in Africa is not only a theoretical dream, but can also be established on the basis of empirical fact. Specific traits of global cityness seem present in the Sub-Sahara Africa city system. However, this scenario should be further investigated in greater depth through follow-up studies. In doing so the following situations and recommendations will have to acknowledge the continent's unique Africanness in a Third World context:

  • Any conceptualisation, analysis and interpretation of global city development in Africa should always take place within the unique context of its historical, geographical, cultural and developmental setting described in this paper.
  • Africa is the least urbanised of the continents, yet exhibits the greatest variety of urban forms, e.g. European, Colonial, Indigenous, Dual and Apartheid city structures. This makes global city research very complex.
  • African cities should not always be considered on the same competitive level as in the developed industrialised world, but on a relative basis within the norms of its unique local and regional context. Although not many African cities qualify as global cities gauged by the present international norms, quite a number may play a significant role as information disseminators, service providers and economic cores in their respective mega-regions.
  • The links between globalization, urbanization patterns and regional structures should be incorporated more extensively via the core-periphery framework.
  • Continued research on this theme in Africa should operate in a more interdisciplinary mould when both subjective/qualitative and objective/quantitative methodologies are integrated.
  • The new institutional economy approach, which allows for more effective decentralised decision making and network formation (North, 1990), could perhaps supplement the command/power framework generally used so often in global city research.
  • Fully fledged global cities in Africa may be rare, but collective world cities consisting of fragmented parcels are an alternative possibility. On the basis that a single city complies only partly with the criteria for a global city, it may be possible that two or three cities collectively play the role of a global city.
  • Taylor's concept of 'cities in globalization' is quite relevant in this argument. Since globalization are ubiquitous across the world it follows that the process is not just a feature of global/world cities. Small towns and medium sized cities across the world also have links and flows that are part of the myriad of global relations. Hence most urban nodes currently exist within the globalization context. The important point is that some are more connected than others.

Like urbanization, globalization brings opportunities as well as problems. The challenge is to develop solutions to the problems associated with the phenomenon, while at the same time realizing its positive prospects for economic growth, social justice and environmental sustainability on the continent of Africa (Van Vliet, 2002). Global cities play a pivotal role in this process.


Arlene Davids rendered skilful assistance in the preparation and analysis of the data for this research project. Stan du Plessis made valuable comments on the first draft of the manuscript from an economist's perspective, while Peter Taylor did the same from a geographer's perspective.


Ajayi, I., 2001: What Africa needs to do to benefit from globalization. Finance and Development, 38(4), 6.

Beaverstock, J. and Taylor, P., 1999: A roster of world cities. Cities, 16(6), 445.

Beaverstock, J., Smith, R. and Taylor, P., 2000: Geographies of Globalization: United States law firms in world cities. Urban Geography, 21(2), 95.

Beaverstock, J., Smith, R., Taylor, P., Walker, D. and Lorimer, H., 2000: Globalization and world cities: some measurement methodologies. Applied Geography, 20(1), 43.

Castells, M., 1996: The Rise of Network Society. Oxford: Basil Blackwell.

Doel, M. and Hubbard, P., 2002: Taking world cities literally: Marketing the city in a global space of flows. City, 6(3), 351.

Douglas, M., 2000: Mega-urban Regions and World City Formation: Globalisation, the Economic Crisis and Urban Policy Issues in pacific Asia. Urban Studies, 37(12), 2315.

Friedmann, J., 1986: The World City Hypothesis. Development and change, 17(1), 69.

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

Hall, P., 1966: The world cities. London: Heinemann.

Lo, F-C. and Marcotullio, P., 2000: Globalisations and Urban Transformations in the Asia-Pacific Region: A Review. Urban Studies, 37(1), 77.

North, D., 1990: Institutions, institutional change and economic performance. Cambridge: Harvard University Press.

Sassen, S., 1991: The global city. Princeton: Princeton University Press.

Sassen, S., 1994: Cities in a world economy. London: Pine Forge Press.

Sassen, S., 2002: Locating cities on global circuits. Environment and Urbanization, 14(1), 13.

Scott, A., (ed) 2001: Global City-Regions: Trends, Theory, Policy. London: Oxford University Press.

Short, J. and Kim, Y., 1999: Globalization and the City. New York: Longman.

Short, J., Kim, Y., Kuus, M. and 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.

Smith, R., 2003: World city actor-networks. Progress in Human Geography, 27(1), 25.

Stock, R., 1995: Africa South of the Sahara. New York: Guilford Press.

Taylor, P., 1997: Hierarchical tendencies amongst world cities: a global research proposal. Cities, 14(6) 323.

Taylor, P., 2001: Specification of the World City Network, Geographical Analysis, 33(2) 181.

Taylor, P., Walker, D., Catalano, G. and Hoyler, M., 2002: Diversity and power in the world city network. Cities, 19(4), 231.

Thrift, N., 1994: Globalization, regulation, urbanization: The case of the Netherlands. Urban Studies, 31(3), 365.

University of Loughborough,, 26 May 2003.

Van Vliet, W., 2002: Cities in a globalizing world: from engines of growth to agents of change. Environment & Urbanization, 14(1), 31.

World Bank,, 22 August 2003.

Yusuf, S. and Wu, W., 2002: Pathways to a World City: Shanghai Rising in an Era of Globalisation. Urban Studies, 39(7), 1213.


* Department of Geography and Environmental Studies, University of Stellenbosch,

1. This international research project, initiated by Prof Manie Geyer (University of Potchefstroom, South Africa), focuses on core-peripheral population and economic flows inside and between global regions and the role of world cities as nodes within the global space network. Sub-Sahara Africa is one of seven regions being investigated in this way.

2. Although some academics identify a subtle difference between the terms "world city" and "global city", in this paper they will be used synonymously.

Figure 1: 

Figure 1

Table 1: Global City Criteria and Indicators 

Table 1

Table 2: Capacity of African Countries to House Global Cities (2000) 

Table 2

Table 3: Possible Global Cities in Sub-Sahara Africa 

Table 3

Edited and posted on the web on 27th November 2003

Note: This Research Bulletin has been published in Urban Forum, 15 (1), (2004), 36-47