GaWC Research Bulletin 68

GaWC logo
  Gateways into GaWC

This Research Bulletin has been published in International Journal of Urban and Regional Research, 26 (2), (2002), 389-403.


Please refer to the published version when quoting the paper.


Immigration and the Global City Hypothesis: Towards an Alternative Research Agenda

M. Samers*



During the 1990s, urban studies from a critical perspective seemed to experience a strange polarisation. At the same time that the analytical and normative strands of neo-Weberian and Marxist urban political economy were slowly eclipsed by cultural studies and a related gamut of 'post-ist' critiques (post-modern, post-Marxist post-structural, post-colonial and transnational), the more economistic 'Global City' hypothesis (henceforth GCH) began to set a certain agenda for research. But if this economistic discourse focused initially on the links between capital and labor, and especially migrant labor (Sassen, 1988, 1991), then it soon turned more conspicuously to the business and technological dimensions of the theory and less to the relationship between cities and low income/'low-skilled' immigration.1 For example, a now leading group of researchers in the United Kingdom (the Globalization and World Cities Network or GaWC, to which Sassen has been a regular contributor)2 has been preoccupied with identifying how putatively global cities are networked in terms of law, accountacy, consultancy, and other financial firms.

If the original theoretical and empirical focus of the GCH has been overshadowed by 'post-ist' urban research and a new propensity to concentrate on business and technological processes, then what value does Sassen's initial research (as well as similar studies in the 'global city vein') still hold? This paper responds to that general question, and in this sense, it has three aims. First, it seeks to redress the lack of focus on the relationship between immigration and the GCH. Second, it aims to evaluate critically the 'global city hypothesis' in relation to immigration in primarily (but not exclusively) Europe's large metropolitan regions. I do this initially by discussing Sassen's thesis, and then follow with an exploration of the subsequent literature that has sprouted from her arguments. Indeed, much of the research inspired by the GCH during the 1990s has either deployed her theoretical and conceptual arguments for the purposes of empirical verification or have questioned her framework altogether. In any case, I maintain that such a critical analysis of Sassen's on-going research project and the parallel work of others is useful insofar as the GCH can be modified to address issues of urban inequality. I call this a 'renewal' of the 'global city hypothesis'. Third and finally, it offers an alternative research path (what I call a 'reformulation' of the GCH) that combines the global city hypothesis with more recent debates around one element of the 'post-ist' urban research; that is, transnationalism.


In this section, I briefly outline Sassen's contribution in The Global City (1991) with respect to immigration. She offers at least two arguments here. First, Sassen explains why global cities are witnessing large-scale immigration and second, why they necessarily involve increasing income and occupational polarisation. Regarding the first argument, she claims that so-called 'producer services' (law, accountancy, management and financial consulting, and so forth) drive immigration through a demand for low-paid jobs. However, her argument is not exclusively demand-driven, since she insists that migration to rich countries is partly set in motion by FDI flows into poorer countries. Nonetheless, Sassen develops her argument further by focusing on the processes of 'informalisation' in global cities (see also Sassen, 1996). She claims that 'Third World' immigration does not lead to the informalisation of global city economies, but that informalisation has been a permanent feature of such economies. Instead Sassen insists that a crucial question is what role immigration plays or does not play in this process? She argues that increased earnings inequality together with the inability of some producers to compete for the requisite resources in the context of sharply inflated prices for commercial space, business inputs, related services, and labor, have led to informalisation, often under 'sweatshop conditions' - what she refers to as 'down-graded manufacturing'. And she argues that 'sweatshops and even industrial home-work have grown over the last decade, especially in large cities, such as New York and London" (1991: 218, my emphasis; see also Sassen, 1996). The fragmentation of markets means that immigrants and other ethnic minorities cannot afford the luxury goods that are offered in global cities. They then seek necessary goods from 'co-ethnic' producers, and/or from other low-cost immigrant-run shops - what she calls 'down graded mass consumer services'. Similarly, the type of niche-market small-batch goods aimed at more affluent consumers associated with the gentrification of large cities - what she calls upgraded 'nonmass consumer services', leads to labor-intensive, small-scale sub-contracting which itself is concentrated in these cities and dominated by migrant labor (Sassen, 1991, 1996, 2000, chapter 6; see also earlier Soja and Scott, 1986). But in Sassen (1996) her argument begins to adopt a more pronounced 'supply-side' perspective. She argues that the growth of the migrant population in global cities has led to an expansion of small-scale producers that can effectively compete with large chain stores and supermarkets, although competition is intense, returns are extremely marginal, and this in turn drives the demand for ever cheaper labor.

The above discussion is certainly an abbreviated and stylised account of Sassen's GCH, but it allows us to capture the main theoretical assumptions and tenets of her argument in order that we can understand some of the criticisms for which it has been a focus.


Sassen's arguments have felt the brunt of seemingly endless scrutiny. While it is not my intent to cover the broad spectrum of these criticisms (for reviews, see Short and Kim, 1999; Yeoh, 1999), below I outline four reservations that have considerable gravity in terms of immigration.

Global Cities and the Expansion of Labor Migration?

The first reservation regarding Sassen's work is its apparent assumption that because of 'globalization' (or at least a growth in FDI flows to Third World countries), global labor migration is expanding and therefore accompanying, if not necessarily driving the growth of global cities. Yet, it should be pointed out that there is unreliable evidence regarding the massive expansion of 'low-skilled' labor migration on a global scale since the 1970s (and by labor migration I mean migrating under the rubric of a formal job contract). This indeed may be happening, but again the international statistical evidence is difficult to assess (Zlotnick, 1998).3 In Europe, low-skilled labor recruitment has certainly been reduced to a trickle with a few notable exceptions such as construction workers in Berlin, or temporary labor migration into French and Spanish agriculture, for example. Migration under the category of family reunification continues, and spouses and dependants may certainly search for work in these large metropolitan areas, but it too is limited in European countries (IOM, 2000; SOPEMI, 2000). Thus, given the restrictions on labor recruitment, asylum, and even family reunification, the linking of labor migration to the growth of global cities implies that much of the migration is undocumented. The anecdotal and patchy statistical evidence suggests that a significant proportion of migration to, or immigration within Europe's largest cities is in fact undocumented (see e.g. Burgers and Engbersen, 1996; SOPEMI, 1999, 2000; Black, 2001; Samers, 2001). But again comparative data, especially across European countries is problematic given the legal dynamics of immigration policy. In short, if social scientists draw a link between the expansion of migration and the growth of global cities, then it rests on a certain amount of 'guess work' concerning the former.

Increasing Informal Employment?

There seems to be an implicit assumption (whether popular or academic) that globalization means a dangerous cocktail of de-regulation and increasing global competition, and hence the growth of informal employment. Thus, there is a considerable literature (mostly in the US and not in the European Union) which argues that global cities together with their respective immigrant/'ethnic minority' concentrations account disproportionately for the dynamism of these activities. And so emerges a second question about the GCH. As Williams and Windebank (1998) point out, "The inevitable result is that they [research in the global city vein] identify what they seek: that informal employment is closely associated with such groups [immigrants/ethnic minorities] and that these groups engage in organised forms of exploitative, low-paid, informal employment" (p. 83). Regardless of the merit of these studies, Williams and Windebank insist that informal employment is not necessarily increasing in the advanced economies (note that their arguments are not discussed uniquely in the context of global cities). They insist further that immigrants do not dominate this work, even though the majority of immigrants, according to some studies anyway, are concentrated in informal activity (see e.g. Reyneri, 1998; Solé, 1998; Wilpert, 1998 for evidence from the European Union). If one is to believe in the generality of the anecdotal evidence, it would seem that in many cases, immigrant employment in informal economic activity has been falling rather than growing throughout the 1990s in Europe (Solé, 1998; Marie, 1999, 2000). Nonetheless, other studies point to wide spread sexual trafficking/ prostitution and a growth in the number of unrecorded (female) domestic workers (Kofman, 1999; Anderson, 2001). In this sense, the contribution of immigrants to informal economic activity should not be seen as exclusive to 'global cities' as the research from Southern Europe, and especially Italy demonstrates (see e.g. Quassoli, 1999). Furthermore, there is little evidence to suggest that informal economic activity is confined to global cities. For example, building and maintenance work in the city of Liverpool (hardly in anyone's imagination, a global city) functions with a veritable army of informally employed builders, plumbers, electricians, and decorators. And El Paso, Texas with its concentration of undocumented Mexican workers constitutes the sixth largest garment district in the US (Spener and Capps, 2001).4

The Growth of Sweatshops?

A third and related reservation stems from the debate which surrounds the alleged growth since the 1970s of 'sweatshops'/'down-graded manufacturing' in global cities.5 On the one hand, Waldinger and Lapp (1993) use an innovative methodology to show that there is little evidence of the growth of such 'sweatshops' in the New York apparel industry, even if they acknowledge the presence of numerous garment contractors who violate a range of labor laws. Thus, the study of New York (by Waldinger and Lapp) and of the Randstad Holland (by Kloosterman et al, 1998) conclude that in these cities, there is little change in their 'post-Fordist economic structure'. Rather immigrants find employment in sectors where there are 'vacancy chains' (as in bakeries in Amsterdam or Rotterdam) or in garment manufacturing (in New York or Los Angeles). In other words, in some sectors, high rates of forced shop closure and/or voluntary exit among natives are combined with lower rates of 'start-up', thus offering a space for 'Third world' migrants to literally set up shop. In short, this would imply out-migration of natives from certain economic activities and migrant in-migration without a necessary expansion of sweatshop activity. On the other hand, Robert Ross' (2001) historical and contemporary study of sweatshops in the United States demonstrates that there has in fact been a steady increase in garment-related sweatshops since the 1970s in New York, Los Angeles/Southern California, and San Francisco.6 This is the result, he argues, of a combination of processes associated with a new phase of global capitalism. These processes include "the declining capacity of the state to enforce labor laws, the increasing market power of retailers through concentration of sales; the competitive pressure brought about by massive imports from low-wage export platforms; and finally, the availability of a large pool of vulnerable immigrant labor" (p. 28). Ross' thoughtful diachronic analysis seems more convincing than that of Waldinger and Lapp, but there is a glaring paucity of similar cross-national and inter-urban studies in the context of the European Union over the last decade. Thus, claims for the EU-wide growth of sweatshops since the 1970s require considerable more empirical demonstration.

The Explanatory Bias of the GCH

A fourth reservation is the exogenous explanatory bias within the GCH debate. That is, there is too much focus on economic globalization as happening to cities, and not enough on what might be considered endogenous processes - or for example, how immigrants themselves structure the labor markets of large metropolises.7 In this sense, there is an enormous literature on 'immigrant entrepreneurship', ethnic enclaves, immigrant/ethnic economies, migrant day laborers and so forth which runs in parallel to the GCH, but which has not been the subject of any sustained synthesis. And notwithstanding some major world cities that do not have comparatively high levels of immigration like Tokyo, it may in fact be the presence of such large-scale immigrant economic 'communities' (with their attendant global financial remittances and their ability to incubate small business growth, rather than simply their complementarity to producer services employment) which partially distinguishes mega-cities from other more nationally-oriented urban centers.


The 1990s witnessed a veritable explosion in global city research. In this paper, however, I want to focus on only one central aspect of it - that is the considerable theoretical, methodological, and empirical mêlée around Sassen's 'social polarisation thesis' (see inter alia Hamnett, 1994, 1996a, 1996b Baum, 1999; Body-Gendrot, 1996; Bruegel, 1996; Kloosterman, 1996; Gibson, 1998; Hamnett and Cross, 1998; Kofman, 1998; Rhein, 1998; Kesteloot and Meert, 2000; Wessel, 2000; Walks, 2001:).8 Such an analysis is necessary insofar as the polarisation debate is one of the chief attributes of the GCH that has direct relevance for immigrants/ethnic minorities (i.e. chronically low wages, underemployment, job insecurity, 'downward mobility', etc.).

Let us begin with Hamnett's (1994, 1996) request for clarity concerning Sassen's use of the term polarisation (is it absolute or relative?). Hamnett asks whether this is a matter of occupational or income inequalities? While he suspects that Sassen is referring to absolute occupational polarisation, he finds only limited evidence for such polarisation in his analysis of the Randstad Holland. Instead, he argues, what is happening are processes of 'professionalisation' and the growth of an unemployed population supported by the state. Similarly, Hamnett and Cross (1998) make us aware of the conflation of polarisation and inequality, and find that earnings inequality has increased between 1970 and 1995 in London and Great Britain as a whole. They also find limited support for Sassen's earnings (as opposed to occupational) polarisation thesis in the service sector. Thus, across a range of large metropolitan areas, the evidence for at least income polarisation is rather mixed, and thus by no means convincing. The absence of such polarisation may be explained by certain 'cultural particularities' such as active protest and other 'path-dependencies' that help sustain relatively strong local and national welfare systems, 'adequate' social housing and more worker-favourable labor market institutions. These then mediate the ravages of market processes (Body-Gendrot, 1996; Hamnett, 1996a, 1996b; Kofman, 1998).9 In other words, the nature of European societies precludes the kind of inequality found in US cities. Using Esping-Andersen's work as a guide, Hamnett (1996a) concludes that, "We have to bring the state, the structure of civil society and political strategy back into discussions of polarisation alongside economic restructuring" (p. 1429). Nonetheless, as Wacquant (1999) and certainly many feminist analyses are right to point out, states are also generative of inequalities, especially, but not exclusively, along the lines of gender.10


While I do not intend to evaluate empirically the evidence for social polarisation in large European cities, it may be useful to offer some critical comments on this literature. To this end, there are at least three glaring weaknesses that emerge from the polarisation debate with respect to immigration. The first weakness is that the relevant studies do not or cannot capture statistically those immigrants who are deemed to be undocumented. In no subsequent discussion of Sassen's work have I noted any comprehensive recognition of the problems of statistical evaluation of income or occupational polarisation in this regard. It is in fact Sassen (1991) herself who must be one of the few to point out this issue, especially in relation to informal labor markets (p. 245). In short, if we can concede that many of the immigrants living in these putative global cities are undocumented (see the discussion below), then using the available statistical data is not likely to be a very robust measure of inequality, whether occupational or income-based.11 Furthermore, the argument that European welfare states soften the hard edge of their market-oriented societies (at least relative to the United States) makes little sense for undocumented immigrants. In fact, evidence from London shows that undocumented immigrants are quite critical of the benefits system, and will do their most to avoid detection by not relying on social entitlements (Jordan, 1999).

The second weakness, which follows from Peter Taylor's critique of the 'evidential structure' of the world cities literature, is whether the evidence for especially income polarisation is really sufficient given the transnational nature of economic activity among migrants. In other words, Taylor claims that the global city literature is based on nationally-gathered statistics which are inadequate for assessing how cities are networked transnationally in the global economy. For example, in Chinese communities, many individuals and families rely on the huì (or the inter/intra-family pooling of money in the form of a savings bank) in order to facilitate business expansion, or just quotidien survival (White, Winchester, and Guillon, 1987). As the huì and certainly other potentially transnational sources of income beyond employment exist, this is not easily captured by existing urban or national-level data. Needless to say, the implications here for urban policy are legion.

A third conceptual and methodological limitation is the lack of specificity about the timing (to not mention the spacing) of these 'causal' (supply/demand) relationships as suggested by Sassen's original thesis (for exceptions, see Howell and Mueller, 2000; Ley and Smith, 2000). For example, Wacquant's (1999) otherwise wide-ranging overview of urban inequality seems to reference work from different moments of capital accumulation, job expansion and so on, without adequately addressing how the timing of these various studies might shape the relationship between cities, inequality, and immigration.12 Indeed, as Gordon and Richardson note about urban income data, "The most important insight to be sought from income distribution data is this: What are the odds that any individual will remain at the bottom (or at the top or anywhere in between) and for how long? Social mobility is the real news"13 (p. 577).


By marshalling together the criticisms of Sassen's GCH, as well as those of the polarisation debate, we can move towards five propositions for a modified GCH. The theoretical and empirical results might translate into different and more appropriate urban policies. Indeed, a major difficulty with the GCH is its policy relevance. In other words, following Smith's (1998) intervention, we can ask why 'global cities' should be an object of research? I would argue that the longevity of Sassen's approach rests on its ability to provide insights either into exposing inequalities, or ways of addressing these inequalities. Otherwise, I do not see the merit in pursuing this line of enquiry. We might ask more specifically then how Sassen's research and similar work contributes to urban policy? If Sassen is correct and the nature of immigration is related to just another round in the urbanization of global capitalist accumulation, what can or should be done about it?14 There are a number of axes in which urban governments can intervene. These include immigration policy, so-called 'integration' policy, and employment policy. Rather than discuss the contours of appropriate policy, which would require a much longer and different paper, let me add that since global cities are not isolated entities, urban governments might impose these together and uniformly through a network of global cities. A salient example is the 'Urban League' with its headquarters in New York, and which is concerned with the problems and possibilities of African-Americans. And I use the term 'network' expressly for reasons that I discuss in the section on 'Reformulating the GCH'.

Re-assess the Relationship between International Labor Migration and Urban Labor Markets

Comparative national-level data on migration is too incongruent and lacks sufficient disaggregation to draw a relationship between a growth in the labor supply and the expansion of global cities. And it is the ambiguity and hyperbole of statements about global labor migration that require us to relate the status of immigrants (undocumented, refugee/asylum-seeker, family member with or without the right to work, student, entrepreneur, etc.), and their background (skills, education, financial and commercial resources, and so forth) to labor market entry and the structure of urban labor markets. And this in turn must be combined with sensitivity to the timing of these migrations. In other words, one must be attentive to the relationship between changes in immigration policy and the growth of global cities.

Re-assess the Relationship between Informal Employment and Global Cities

By its very nature, informal employment is exceedingly difficult to capture in a quantitative sense, not least because businesses and employment are so ephemeral. And here we need a more precise notion of what constitutes informal economic activity (is it the 'drug economy', prostitution, domestic labor, or simply the illegal production of textiles?) (Samers, 2001). For example, it may be female prostitution and domestic work in the EU, rather than an illegal garment industry that marks the informal character of economic activity in European global cities at the beginning of the 21st century (Kofman, 1999; Anderson, 2000). Nonetheless, if there is a strong relationship between mega-cities, immigration, and the growth of informal employment, then this must be demonstrated rather than simply asserted. And it must be shown why informal economic activity is relatively unique to these cities. Otherwise, informal economic activity as a defining characteristic of global cities cannot be assumed. Thus, anecdotal qualitative evidence (and not quantitative estimations) is likely to provide valuable insights into the processes at work, and therefore help to sustain dynamic and appropriate urban policies with respect to potential employment regulations, job training, language training, and housing.

Evaluate Critically the Growth of Sweatshops and Down-graded Manufacturing

Like many forms of informal employment, statistical evidence across large metropolitan regions with regard to sweatshops/down-graded manufacturing is likely to prove elusive. Yet, again, qualitative case studies may offer a useful service in terms of public policy. They will not, on the other hand, help to assess whether these kinds of employment units are growing or declining, unless such studies are carried out en masse. And as suggested in the previous proposition, there has been a tendency to focus disproportionately on garment manufacturing (surprising given the diminished fortunes of this industry in the advanced economies of Europe (Iskander, 2000), and this may obscure the importance of other kinds of down-graded manufacturing. Again, the specific links between a growth in sweatshops/down-graded manufacturing (including how one might delineate such economic units) and how this is specific to global cities should be articulated.

Match Exogenous and Endogenous Procesess in Global Cities

There is a need to construct a sustained synthesis between processes of economic globalization (what I referred to earlier as exogenous processes) and the way in which these processes both provide the space for and constrain the economic activity of immigrants in urban labor markets (so-called endogenous processes). To this end, a dialogue between the GCH and the entire literature on 'immigrant entrepreneurship', ethnic enclaves, immigrant/ethnic economies, day laboring and so forth, could prove enormously fruitful. Perhaps a combination of Kloosterman, Rath, and Van der Leun's (1999) concept of mixed embeddedness15, an alternative conception of entrepreneurship based on day laborers as low-skilled entrepreneurs (Valenzuela, 2001), and a sophisticated labor segmentation theory for understanding wage labor (see e.g. Peck, 1996) might prove to be one valuable path for research? Yet the point is not to substitute one discourse for another (i.e. endogenous processes for exogenous), but to show specifically how the world's largest cities may be distinguishable from other urban centers. In other words, how do the disproportionately large numbers of immigrants help to shape the urban labor markets of these largest cities, and what implications does this have for urban policy.

Re-think Polarisation and Inequality

I have argued that the polarisation debate suffers from a number of conceptual and empirical weaknesses that have significant implications for both the GCH debate and urban policy. These include the 'fuzziness' of the terms polarisation and inequality, the national bias of statistics, the inattention to undocumented immigration, and the lack of temporal specificity. Thus, with regard to addressing these issues simultaneously, a starting point is an estimation of undocumented immigration. Notwithstanding the potentially insidious political and ethical implications of gathering or applying such statistics, and regardless of how paradoxical such measurements may seem (because they appear to officialize the unofficial), they are likely to provide a more robust measure of at least occupational, if not income polarisation/inequality. Yet this assumes that social scientists become more creative and look elsewhere for the relevant statistical sources. In fact, some complex statistical procedures have been adopted by various countries to measure the number of undocumented immigrants and/or undocumented immigrants working informally (SOPEMI, 1999).16 This data would have to be combined with more conventional occupational, income, employment and micro-census statistics, and less conventional (transnational) data such as measurements of remittances, capital flows, the intra-ethnic pooling of capital, and other financial activities between the largest cities and the countries of emigration (see e.g. Benbouzid, 1999). Finally, this would have to be analysed through the lens of changes in immigration policy over at least a census period (i.e. roughly 10 years) in order to tease out the temporal dynamics of inequality/polarisation.

In sum, I have pointed to 5 ways in which we can 'renew' the GCH and parallel studies. These propositions together represent a considerable challenge to social scientists working in this vein, but it is my argument that if these are not addressed, then future studies into global cities are likely to be inadequate assessments of urban inequalities in the world's largest cities.


There is something distinctively victimising about the global city literature with respect to immigrants, and there will be those who see such studies as empiricist, and the inequality of cities as an ineluctable feature of capitalist urbanization without any possible remedy. Perhaps the most ardent supporter of this view is Gibson (1998). I argue in this section that for the more skeptical observers, abandoning the GCH need not be a necessary consequence of this reasoning. In my move therefore from 'renewal' to 'reformulation', I shift roughly from a critical discussion of largely economistic assumptions within the GCH to thinking about how we might nurture the political implications of the GCH. This is less of a giant step than one might think, if, for example, we imagine that the enormous sums of money that migrants remit to their countries of origin from employment in global cities, allows for the development of what has been called political transnationalism. In other words, it would make more sense to view global cities not so much as the transnational loci of inequality (which they seem to be), as in looking at global cities as offering a different set of networks that are envisioned by the GaWC research project in the UK (for which Sassen herself is a regular contributor). These alternative 'networks' may contribute to 'new' forms of democratic participation in the countries of destination, as well as in the countries of origin. How might this be conceptualised?

I would defer here to Michael Peter Smith's research on 'trans-localism' (Guarnizo and Smith, 1998). This is an idea which grew out of his research on the movement of the entire inhabitants of villages in Mexico to their localised re-grouping in Los Angeles, San Diego, San Francisco, but also in many other smaller Californian suburbs and agricultural towns. Thus, we can speak of a network of global cities in two senses, one in the types of networks that we associate with transnational communities (largely south-north, but also intra-national and intra-European networks) and the kinds of networks which global cities afford through the concentration of NGOs and other political organisations. In other words, the concentration of immigrant communities in these global cities, together with the headquarters of transnational migrant associations, such as the Amicale des Travailleurs Marocains en France (ATMF) in Paris, may provide the opportunities for the mobilisation of immigrants/ethnic minorities in large metropolitan areas.17 This would transform the 'global city hypothesis' from one that concerns victimisation to one that concerns political hope - a sort of 'transnationalism from below' (Smith and Guarnizo, 1998; Vertovec and Cohen, 1999). True, research on immigrant communities in European cities has revealed that formal political representation remains the domain of migrant elites rather than strictly 'grassroots political mobilisation' (e.g. Bousetta, 1997). And more generally, there is little certainty as to whether such 'transnationalism' reconfigures or reaffirms existing class, racial, and gender power relations (Mahler, 1998), or that it does not also bring with it 'less progressive' processes to migrant communities both in the 'North' and in the 'South' (Mattei and Smith, 1998). In the end, 'transnationalism from below' may only be a very crude metaphor that begs the question as to what other kinds of 'spatialities' will be necessary for building a progressive immigrant transnational politics. I leave this as a potential question for research.

Thus, the two types of networks discussed above facilitate (although do not guarantee) the mobilisation of 'grassroots' migrant interests where their disadvantaged status (and often their undocumented status) may not ensure political participation and/or the delivery of rights. Such rights are extended to those migrants living outside global cities, but further research might seek to reveal how the largest metropolitan regions form the epicenter of the definition and delivery of migrant and ethnic minority rights. This is especially the case in Paris where the nature of the relatively centralised political system allows Paris-based Moroccan and other Maghrebin associations to wield considerable power over other non-Paris based organisations (Poinset, 1993). In contrast to the 'network-driven' research of GaWC, such research may then also reveal how hierarchies still matter. Thus, the specificity of the global city (in other words a new 'global city hypothesis') might involve thinking of global cities as the locus of trans-national political mobilisation and its effects on the welfare of migrants.


The GCH debate and particularly the research agenda of GaWC have been pre-occupied recently with the technological dimensions of so-called 'global cities'. This paper has sought to recover the role of immigration in large urban economies. Using mainly observations from European metropolises, I argued first that the GCH requires significant revision insofar as it can be used as a tool for addressing issues of inequality, and offered five propositions for a renewal of the existing contours of the GCH. Second, beyond these revisions, I suggested a complete reformulation of the debate, and sought to link it with ideas emanating from the literature on transnationalism.

Despite the move by researchers (such as the GaWC) to transcend the emphasis on 'hierarchies' and explore the question of global city networks, I argue that this emphasis is at best incomplete, at worst misplaced. Instead, I have a very different idea of the possibilities of a 'network paradigm' for global cities - one that focuses instead on cities as the locus for a 'transnational political mobilisation from below' (rather than above). The 'Urban League' in the United States or the ATMF (with its headquarters in Paris) are examples of the sorts of inter-urban networks that I have in mind. However, the possibilities for increased political mobilisation within a single global city, or indeed a network of such cities should not blind us to falling into the trap of what Drainville calls the 'feteshism of 'global civil society', nor what Mahler identifies as the weaknesses of the 'transnationalism from below' idea. Nonetheless, such a theoretical move seems a necessary task, lest our understanding of 'global cities' be confined to simply ranking them, outlining business networks, or categorising their characteristics.


Anderson, B. (2001) Why madam has so many bathrobes: demand for migrant domestic workers in the EU. Tijdschrift voor Economische en Sociale Geografie, 92,1: 18-26.

Baum, S. (1997) Sydney: social polarisation and global city status. Urban Studies 22: 21-36 .

Baum, S. (1999) Social transformations in the global city: Singapore. Urban Studies 36,7: 1095-1117.

Benbouzid, M. (1999) Réseaux financiers et marchés parallèles de devises: des Algériens dans l'économie informelle. Revue Europeéne des migrations internationales 15,2: 123-139.

Black, R. (2001) Breaking the convention: the study of 'illegal' migration. Antipode, forthcoming.

Body-Gendrot, S. (1996) Paris: a 'soft' global city? New community 22,4: 619-635.

Böhning, W.R., and Oishi, N. (1995) Is international economic migration spreading? International migration review 29,3: 794-799.

Bousetta, H. (1997) Citizenship and political participation in France and the Netherlands: reflections on two local cases. New community 23,2: 215-231.

Bruegel, I. (1996) Gendering the polarisation debate: a comment on Hamnett's 'Social polarisation, economic restructuring and welfare state regimes'. Urban Studies 33,8: 1431-1439.

Burgers, J., and Engbersen, G. (1996) Globalization, migration, and undocumented Immigrants. New community 22,4: 619-635.

Drainville,A.C. (1998) The fetishism of global civil society: global governance, transnational urbanism and sustainable capitalism in the world economy, in M.P. Smith, and L.E. Guarnizo (eds.), Transnationalism from below. Transaction Publishers, NewYork.

Esping-Andersen, G. (1999) Social foundations of postindustrial economies. Oxford University Press, Oxford.

Gibson, K. (1998) Social polarization and the politics of difference: discourses in collision or collusion?, in R. Fincher,., and J.M. Jacobs (eds.), Cities of difference, Guilford Press, New York.

Gordon, P., and Richardson, H.W. (1999) Review Essay: Los Angeles, City of Angels? No, City of Angles. Urban Studies 36, 3: 55-591.

Guarnizo, L.E, and Smith, M.P. (1998) The locations of transnationalism. In M.P. Smith, and L.E. Guarnizo (eds.), Transnationalism from below, Transaction Publishers, New Brunswick.

Hamnett, C. (1994) Social polarisation in global cities: theory and evidence. Urban Studies 31,3:401-424.

Hamnett, C. (1996a) Social polarisation, economic restructuring and welfare state Regimes. Urban Studies 33,8: 1407-1430.

Hamnett, C. (1996b) Why Sassen is wrong: a response to Burgers. Urban Studies 33, 1: 107-110.

Hamnett, C. and Cross, D. (1998) Social polarisation and inequality in London: the earnings evidence, 1979-1995. Environment and Planning C 16,: 659-680.

Howell, D., and Mueller, E.J. (2000) Immigration and native-born male earnings: a jobs-level analysis of the New York City metropolitan area labor market, 1980-1990. Journal of Ethnic and Migration Studies 26, 3: 469-493.

Hum, T. (2001) Mapping global production in New York City: the role of Sunset Park, Brooklyn's immigrant economy, Paper prepared for the conference: "Working on the Fringes: Immigrant Businesses, Economic Integration, and Informal Practices", Liverpool, UK, March 22-25, 2001.

Iskander, N. (2000) Immigrant workers in an irregular situation: the case of the garment industry in Paris and its suburbs. In OECD (ed.), Combating the illegal employment of foreign workers, OECD, Paris.

International Organization for Migration. (2000) World Migration Report. IOM and the UN, Geneva.

Jordan, W.J.O. (1999) 'End of Award Report to ESRC on Project 000236838, Undocumented immigrant workers in London'.

Kesteloot, C., and Meert, H. (2000) Segregation and economic integration of immigrants in Brussels. In S. Body Gendrot and M. Martiniello (eds.), Minorities in European cities: the dynamics of social integration and social exclusion at the neighbourhood level, Macmillan, London.

Kloosterman, R. (1996) Double Dutch: polarisation trends in Amsterdam and Rotterdam after 1980. Regional Studies 30: 467-476.

Kloosterman, R, van der Leun, J., and Rath, J. (1998) Across the border: immigrants, economic opportunities, social capital, and informal business activities. Journal of ethnic and migration studies 24,2: 249-268.

Kloosterman, R., van der Leun, J., and Rath, J. (1999) Mixed embeddedness: (in)formal economic activities and immigrant businesses in the Netherlands. International Journal of Urban and Regional Research 23,2: 252-266.

Kofman, E. (1998) Whose city? Gender, class, and immigrants in globalizing European cities. In R. Fincher, and J.M. Jacobs (eds.) Cities of difference, Guilford Press, New York.

Kofman, E. (1999) Female 'birds of passage' a decade later: Gender and immigration in the European Union. International Migration Review 33,2: 269-299.

Knox, P.L., and Taylor, P.J. (eds.) (1995) Word cities in a world system. Cambridge University Press, Cambridge.

Ley, D, and Smith, H. (2000) Relations between deprivation and immigrant groups in large Canadian cities. Urban Studies 37, 1: 37-62.

Lo F-C, Yeung, Y-M (eds) (1998) Globalization and the World of Large Cities. United Nations University Press, Tokyo.

Mahler, S. (1998) Theoretical and empirical contributions: Toward a research agenda for transnationalism, In M.P. Smith, and L.E. Guarnizo (eds.), Transnationalism from below, Transaction Publishers, New Brunswick.

Marie, C-V. (1999) Emploi des étrangers sans titre, travail illégal, régularisations: des débats en trompe-l'oeil, in P. Dewitte (ed.), Immigration et Intégration l'Etat des Savoirs, L'Harmattan, Paris.

Marie, C-V. (2000) Measures taken to combat the employment of undocumented foreign workers in France: their place in the campaign against illegal employment and their results. In OECD (ed), Combating the illegal employment of foreign workers, OECD, Paris.

Martiniello, M. (2000) The residential concentration and political participation of immigrants in European cities. In S. Body-Gendrot and M. Martiniello (eds.), Minorities in European Cities, Macmillan, London.

Matthei, L.M., and Smith, D.A. (1998) Belizean "Boyz 'n the hood"? Garifuna labor migration and transnational identity. In M.P. Smith, and L.E. Guarnizo (eds.), Transnationalism from below, Transaction Publishers, New Brunswick.

Peck, J. (1996) Workplace: the social regulation of labor markets. Guilford Press, New York.

Poinset, M. (1993) Competition for political legitimacy at local and national levels among young North Africans in France. New community 20,1: 79-92.

Quassoli, F. (1999) Migrants in the Italian underground economy. International Journal of Urban and Regional Research 23,2: 212-231.

Reyneri, E. (1998) The role of the underground economy in irregular migration to Italy: cause or effect? Journal of ethnic and migration studies. 24,2: 313-331.

Rhein, C. (1998) Globalization, social change and minorities in metropolitan Paris: the emergence of new class patterns. Urban Studies 35, 3: 429-47.

Ross, R.J.S. (2001) The new sweatshops in the United States: how new, how real, how many, and why?, Chapter 5, forthcoming. In G. Gereffi, D. Spener, and J. Bair (eds.), Globalization and regionalism: NAFTA and the new geography of the North American Apparel Industry, Temple University Press, Philadelphia.

Samers, M. (2001) Here to work: undocumented immigration and informal employment in Europe and the United States. SAIS Review Winter-Spring.

Sassen, S. (1988) The mobility of labor and capital: studies in international investment and labor flow. Cambridge University Press, New York.

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

Sassen, S. (1996a) New employment regimes in cities: the impact on immigrant Workers. New community 22,4: 579-594.

Sassen, S. (1996b) Rebuilding the global city: economy, ethnicity and space, in A. King (ed.), Representing the City: ethnicity, capital, and culture in the 21st century metropolis, Macmillan, London.

Sassen, S. (1996c) Analytic borderlands: race, gender and representation in the new city, in A. King (ed.), Representing the city: ethnicity, capital, and culture in the 21st century metropolis, Macmillan, London.

Sassen, S. (1998a) Swirling that old wine around in the wrong bottle. Urban Affairs Review 33,4: 478-481.

Sassen, S. (1998b) The impact of the new technologies and globalization on cities, in F-c Lo and Y-m Yeung (eds.), Globalization and the world of large cities, United Nations University Press, Tokyo.

Sassen, S. (2000) (2nd ed.). Cities in a world economy. Pine Forge Press, Thousand Oaks, Ca.

Short, J., and Kim, Y.H. (1999) Globalization and the city. Addison Wesley Longman, New York.

Smith, M.P. (1998) The global city - whose social construct is it anyway? Urban Affairs Review 33,4: 482-488.

Smith, M.P., and Guarnizo, L.E. (ed) (1998) Transnationalism from below. Transaction Publishers, New York.

Solé, C. (1998) Irregular employment amongst migrants in Spanish cities. Journal of Ethnic and Migration Studies 24,2: 333-346.

Soja, E., and Scott, A. (1986) Los Angeles, capital of the late twentieth century. Environment and Planning D: Society and Space 4: 249-254.

SOPEMI (1999) Trends in International Migration. OECD, Paris.

SOPEMI (2000) Trends in International Migration. OECD, Paris.

Spener, D., and Capps, R. (2001) North American free trade and changes in the nativity of the garment industry workforce in the United States. International Journal of Urban and Regional Research 25, 2: 301-326.

Taylor, P. (1999) So-called 'world cities': the evidential structure within a literature. Environment and Planning A 31: 1901-1904.

Valenzuela, A. (2001) Day laborers as entrepreneurs? Journal of Ethnic and Migration Studies 27,2: 335-352.

Vertovec, S., and Cohen, R. (eds.) (1999) Migration, Diasporas, and Transnationalism. Edward Elgar, Cheltenham.

Wacquant, L. (1999) Urban marginality in the coming millenium. Urban Studies 36,10: 1639-1647.

Waldinger, R., and Lapp, M. (1993) Back to the sweatshop or ahead to the informal sector? International journal of urban and regional research 17, 1: 6-29.

Walks, R.A. (2001) The social ecology of the post-Fordist/Global City? Economic restructuring and socio-spatial polarisation in the Toronto Urban Region. Urban Studies 38,3: 407-447.

Wessel, T. (2000) Social polarisation and socioeconomic segregation in a welfare state: the case of Oslo. Urban Studies 37,11: 1947-1967.

White, J.W. (1998) Old wine, cracked bottle? Tokyo, Paris, and the Global City Hypothesis. Urban affairs review 33,4: 451-477.

White, P., Winchester, H., and Guillon, M. (1987) South-east Asian refugees in Paris: the evolution of a minority community. Ethnic and racial studies 10,1: 48-61.

Williams, C. and Windebank, J. (1998) Informal employment in the advanced economies. London, Routledge.

Wilpert, C. (1998) Migration and informal work in the new Berlin: new forms of work or new sources of labor? Journal of ethnic and migration studies 24,2: 269-294.

Yeoh, B.A. (1999) Global/globalizing cities. Progress in Human Geography 23,4: 607-616.

Zlotnick, H. (1998) International migration 1965-96: an overview. Population and development review 24, 3: 429-68.


* Michael Samers, Department of Geography, Roxby Building, University of Liverpool, Liverpool L69 3BX, UK.

1.For exceptions, see the series of papers written by Sassen in 1996 (a,b,c), but also Body-Gendrot (1996), Gibson (1998), and Kofman (1998).

2. GaWC:

3. For an alternative view, see Böhning and Oishi (1995).

4. This then raises an important question about the deductive reasoning behind the GCH. In other words cities are often chosen as global cities a priori and then a list of specified processes or characteristics are evaluated. While the GaWC network is certainly a move towards reversing this sort of approach, it seems important (if only for urban policy) to define what is unique (if anything) about those cities deemed 'global'.

5. For a critical discussion of the term 'sweatshop', see Ross (2001). For 'down-graded manufacturing', see Sassen (1991).

6. But see most recently Hum (2001) for strong evidence of a decline in the New York garment industry, though not necessarily sweatshops.

7. I am not arguing for a crude analytical distinction between 'exogenous' (so-called global) and 'endogenous' (so-called 'local') processes, but I employ these terms as heuristic devices in order to underline some conceptual and empirical oversights. Clearly, how immigrants structure their economic worlds cannot be divorced from other putatively 'global' flows, such as financial or commercial capital, yet neither are cities and their inhabitants at the linear mercy of global capitalism. This is a well-known counter-thesis in urban geography that has served to dispel some of the myths about place-dependence and economic globalization.

8.Note that some studies in this vein involve explicit analyses of both socio-spatial polarisation and/or immigration, and others do not.

9. Yet it may be that Hamnett's criticisms of Sassen's work are now less powerful than they might have been, given the transition from post-war welfare societies to workfare-oriented states, as well as the reported increases in undocumented immigration.

10. See also Bruegel, 1996 and Gibson, 1998 for a 'feminist'/post-structuralist critique of Hamnett's analysis and the social polarisation debate more generally.

11. Even more broadly, Gibson (1998) has critiqued the polarisation debate as empiricist. Her argument is that it relies on an empiricist notion of class - either in Marxist and/or neo-Weberian terms and fails to situate itself among more post-modern/post-structuralist ideas of 'difference'. This clearly opens up a much larger debate as to whether one can quantify and therefore 'freeze' social categories (and to what extent this is actually the case in global city research), rather than explore how class and other categories are constructed and mobilised.

12. But see Hamnett and Cross, 1998, and Kloosterman, 1996, for more temporally sensitive analyses.

13. I do not necessarily agree with Gordon and Richardson, that 'social mobility is the real news', but their point about temporal dynamics is well-taken.

14. As Gibson (1998) points out, the polarisation debate completely fails to question whether polarisation is actually addressable under capitalism.

15. In the context of immigrant entrepreneurship, 'mixed embeddedness' concerns "the crucial interplay between the social, economic, and institutional contexts. In this view, the rise of the immigrant entrepreneurship is, theoretically, primarily located at the intersection of changes in socio-cultural frameworks on the one side and transformation processes in (urban) economies on the other. The interplay between these two different sets of changes takes place within a larger, dynamic framework of institutions on neighbourhood, city, national or economic sector level" (Kloosterman et al., 1999, p. 257).

16. An analysis of these procedures is beyond the scope of my analysis here. For a discussion, see SOPEMI (1999), pp. 231-235.

17. For an empirical assessment of this hypothesis in the context of Belgium, see Martiniello (2000). In his study, Martiniello argues that there is no definitive link between immigrant concentration in particular urban areas and the presence of an immigrant or 'ethnic vote'. Moreover, the political mobilization of immigrant 'communities' can not be restricted to only large metropolitan areas, as Boussetta's (1997) work on France and the Netherlands demonstrates. Nonetheless, further research is needed on the links between the concentration of immigrants/ethnic minorities in large metropolitan areas and the unique effect it has (if any) on their political mobilisation.


Edited and posted on the web on 21st November 2001

Note: This Research Bulletin has been published in International Journal of Urban and Regional Research, 26 (2) (June 2002), 389-403