The changing social structure of global cities: Professionalisation,proletarianisation or polarisation

Abstract

Chinese

This paper addresses a simple, but very important, question. How has the occupational class structure of major world or global cities changed in recent decades? This has major social and theoretical implications given the claims made by Friedmann (1986) and Sassen (1991) regarding social polarisation in world or global cities. The paper outlines and compares three positions regarding the changing occupational class structure of world cities and of Western societies in general: professionalisation, proletarianisation and social polarisation. The paper does not provide original empirical evidence. Instead, it provides a wide-ranging overview of evidence from existing studies in a range of cities in both Global North and South over the last 50 years. It concludes that whereas professionalisation is common to most global cities, there is little evidence for proletarianisation, and that polarisation is a contingent outcome in certain cities at certain times. The claims for a common universal pattern are rejected and variations in national economic, political and social structure and policy are argued to be more important.

Keywords

class diversity/cohesion/segregation inequality polarisation professionalisation

Introduction

The changing social structure of cities has intrigued social commentators, academics, politicians and the media for many years. There have been recurrent fears that cities are heading in a downward or negative direction and, during the period of industrialisation and rapid migration in the 19th century, there was a persistent middle-class fear in some Western cities of the rise of the mob or the underclass (Hamnett, 1998; Steadman-Jones, 1971). In the post-war period, there were similar fears regarding the changing social structure of American cities (Banfield, 1970; Beauregard, 1993a, 1993b; Dewar and Thomas, 2013; Doucet, 2017; Mollenkopf and Castells, 1991; Wilson, 1996). But see Glaeser (2011) for an alternative view. The last 50 years, however, have been characterised by a shift from industrialisation in most major Western cities to de-industrialisation and post-industrialism with associated changes in occupational class structure. This paper addresses three simple, but very important, questions. First, what are the key theoretical claims regarding occupational class change in world or global cities over the last 50 years? Second, what does available empirical research indicate regarding these changes and the wider theoretical claims? And third, to what extent, if at all, is it valid to claim as Friedmann (1986), Friedmann and Wolf (1982) and Sassen (1991, 2001) have done, that there are direct links between globalisation, the growth of world or global cities and their changing industrial structure and the growth of a polarised occupational structure? Both Friedmann and Sassen argued that these cities have seen the emergence of a ‘polarised’ class structure, dominated by a wealthy transnational elite and characterised by growth at both the top and bottom combined with a shrinkage in the middle. These claims have been criticised, however, for being over-generalised and failing to take into account significant national differences in global embeddedness, economic and political structures and history, the role of the state, international migration and welfare policies, amongst other things.

Taken overall, three main positions have emerged regarding the changing occupational class structure of world cities and of Western societies in general since 1970. First, the changes in industrial structure are said to be giving rise to an increasingly ‘professionalised’ occupational structure characterised by a growing number and proportion of professional, managerial and technical workers, some on high incomes. The second thesis is that of proletarianisation, whereby the changing structure of work has given rise to an increasing number of de-skilled jobs. The final thesis, which has become popular in recent years, is that of polarisation – a process whereby both the top and bottom ends of the occupational and income distribution are growing at the expense of the shrinking middle, creating what has sometimes been termed a ‘dual city’ (Mollenkopf and Castells, 1991). Although Robinson (2002, 2006) has questioned the focus on global cities compared with what she terms ‘ordinary’ cities, the debate about the direction of occupational class change in the world’s major cities is still valid, and it is important to ask which, if any, of these three trends have been dominant.

The structure of the paper is as follows. The second section looks briefly at the changing nature of the occupational class structure and the emergence of the three main theses about occupational class change: professionalisation, proletarianisation and polarisation. The third section looks at the main characteristics of both Friedmann’s and Sassen’s ideas about social polarisation and their criticisms, and the fourth, largest, section looks at a range of empirical findings from different cities across the world over the last 50 years. There is a short conclusion. The paper is wholly reliant on secondary sources and does not involve any new empirical analysis. Its main objective is a comparative global overview of existing literature. It is focused on the overall occupational class structure (not on income) and not on patterns of geographical differentiation or segregation (see Musterd, 2020). It argues that, contra Friedmann and Sassen, there is not a common pattern of social polarisation across global cities over the last 50 years and that the nature of change varies depending on the role of the state, the extent of state intervention and welfare policy, historic structures of accumulation, and the type and level of immigration among other things.

The changing occupational class structure: Professionalisation, proletarianisation and polarisation

Although this paper focuses on the major changes over the last 50 years it is important to briefly put them into a longer historical context. It is widely accepted that many old Western cities have undergone a transition from a pre-industrial past (Sjoberg, 1955) to the industrial era (Briggs, 1963; Cannadine, 1977; Ward, 1971) and, since 1970, to a post-industrial era. Pre-industrial cities were characterised by a pyramidal class structure with a very small ruling class, a small middle class and a large base (Allen, 2018). The onset of the industrial era from about 1840 saw large-scale migration to the cities, rapid urban growth and the creation of a large, new, urban labour force, accompanied by the growth of a large underclass or ‘residuum’ of casual or sporadically employed workers (Engels, 1846; Green, 1995; Harvey, 1985, 2003; Steadman-Jones, 1971; Ward and Zunz, 1992). The class structure retained a very pyramidal structure with a small elite of landed, industrial and financial capital, and a small professional class and bourgeoisie (Marsh, 1965; Rubenstein, 1977). The big change arguably began in the early 20th century with the advent of mechanisation and automation. The growth of Fordism saw a growing semi-skilled manual working class in factories and the growth of white-collar clerical work. At this stage the occupational class structure began to change from a pyramid to more of an onion shape (Goldthorpe and Lockwood, 1967; Pahl, 1988).

The second major change in occupational class structure concerns the growth of what Daniel Bell (1973) has termed the coming of post-industrial society (PIS). According to Bell, the PIS is characterised by the decline of manufacturing industry, the growth of what is termed a quaternary or information-processing economy, the growing importance of professional and technical knowledge and growing demand for more highly educated workers and an increasingly professionalised society. This era saw the rapid expansion of the university sector in developed Western societies and a rapid expansion in the number and proportion of managerial professional and technical jobs. This analysis has been contested (Brick, 1992; Ferkiss, 1979; Kumar, 1978; Vogt, 2015), but the term is now widely used to refer to a shift from societies dominated by manufacturing employment to those dominated by the quaternary service sector and where manufacturing employment has shrunk dramatically.

At broadly the same time, Henry Braverman (1974) outlined his views on the changing nature of white-collar work in capitalist societies. In his view, the massive growth in white-collar work after WW2 was misleading in that most of these jobs were repetitive with a high degree of labour control and the dominant process was the de-skilling of labour. Braverman’s Marxist analysis is directly contrary to that of Bell. While Bell saw a future characterised by rising professionalisation, Braverman saw a future of increasing proletarianisation, a view supported by Kumar (1978). The difference between the two positions was well summarised by Wright and Martin (1987):

post-industrial theorists generally expect a process of gradual de-proletarianisation characterised by an expansion in the semi-autonomous employee category and decline in the working class. Marxists, however, generally postulate an inherent tendency in capitalism for technical change to destroy high skilled, cognitively complex jobs … It would generally be predicted therefore that transformation of class structure should be characterised by an increasing process of proletarianisation. (pp. 2–3)

Most sociologists who have researched the class structure of post-industrial society generally argue that the evidence points unequivocally to an upwards shift, and this was the conclusion reached by Wright and Martin. Esping-Andersen (1993) in his study of trends in the changing class structure in Western countries also stated that:

Virtually all research rejects Braverman’s de-skilling and mass proletarianization thesis pointing, instead, to a pervasive skill upgrading and professionalisation. Despite the divergent evidence of the post-industrial hierarchy, there is little evidence to suggest strong polarisation. Everywhere, the trend favours the higher grade occupations such that the shape of the post-industrial occupational hierarchy is biased towards the top and the middle, rather than the bottom. (p. 53)

AH Halsey (2000), a British sociologist of class and mobility, observed of the overall changes in Britain that:

By the end of the century millions of children of manual workers had risen into non-manual jobs and many thousands had become the graduate grandchildren of butchers, bakers and candlestick makers, following professional careers. The occupational, and therefore the class structure, had shifted. In 1900 the vast majority of Britons were elementary school proletarians; by 1970 they were divided half and half between white and blue collar jobs. By 2000 the balance had been tipped decisively to form a white-collar majority. (Halsey, 2000: 17)

Similarly, Marshall and Rose (1988) argued on the basis of an empirical study of British class mobility that Britain had seen a net upgrading rather than a degrading of the occupational class structure and had not experienced proletarianisation. Admittedly these conclusions were reached 20–30 years ago, before the onset of the global financial crisis, which may have significantly affected market processes in the last decade, but the evidence to date is fairly clear. This is not to claim, however, that the working class have disappeared, that inequality has shrunk or that the rich have not got richer. On the contrary, growth of the professional and managerial class has been accompanied by an increase in income inequality with disproportionate gains at the top end (Dorling, 2014; Piketty, 2014). The bottom end has also seen an increase in precarious work. But the occupational class structure today is very different from what it was 100 or even 50 years ago. This is clearly shown by Elvery’s (2019) study of the changing occupational structure of the USA 1860–2015. But growth of the middle class is not equivalent to the triumph of the middle class, as some suggest.

Global cities, the social polarisation thesis and national variation

Looking specifically at cities, and more particularly at large world or global cities which have a key role in the organisation and operation of the global capitalist system, particularly in its financial and business systems, it is clear that many of them have experienced major changes in their occupational class structure since the 1970s as de-industrialisation has given way to a post-industrial economy (Savitch, 1988). What form have these changes taken? We have discussed the professionalisation and proletarianisation theses above but social polarisation has become influential in recent years.

Both Friedmann (1986), Friedmann and Wolf (1982) and Sassen (1991) have argued that the growth of globalisation and the global economy have led to the emergence of a new type of city, which they term, respectively, world or global cities. These cities which are the control and command centres of the global economy have a particular role in the international division of labour with an economy characterised by major corporate HQs and a large financial and business-services sector and are marked by a polarised class structure dominated by what Friedmann calls a transnational elite (see Walks, 2001). Both their approaches were broadly economically deterministic, attributing the major causal drivers of urban change to be global and economic, with endogenous national factors playing a secondary role. Friedmann (1986) argued that: ‘Contemporary urban change is, for the most part, a process of adaption to changes which are externally induced’ and ‘The form and extent of a city’s integration with the world economy, and the functions assigned to the city in the new spatial division of labour, will be decisive for any structural changes occurring within it …’ (pp. 70–71).

Transnational elites are the dominant class in the world city, and the city is arranged to cater to their lifestyles and occupational necessities … The contrast with the third or so of the population who make up the permanent underclass in the world city could not be more striking …The primary social fact about world city formation is the polarisation of its social class divisions. (p. 322, emphasis added)

Friedmann recognised that national and local factors could play a role, but he saw them as secondary.

Five years later, in her influential book The Global City, Saskia Sassen (1991) argued that these cities were seeing a shrinkage of skilled and semi-skilled manufacturing jobs and that:

[N]ew conditions of growth have contributed to elements of a new class alignment in global cities. The occupational structure of major growth industries characterized by the locational concentration of major growth sectors in global cities in combination with the polarized occupational structure of these sectors has created and contributed to growth of a high-income stratum and a low-income stratum of workers. It has done so directly through the organization of work and occupational structure of major growth sectors. And it has done so indirectly through the jobs needed to service the new high-income workers, both at work and at home, as well as the needs of the expanded low-wage work force. (p. 13, emphasis added)

Frustratingly, Friedmann never clearly defined what he meant by polarisation, using it as a synonym for growing inequality or segregation (1986: 76), but Marcuse (1989) noted: ‘The best image … is that of the egg and the hour glass: the population of the city is normally distributed like an egg, widest in the middle and tapering off at both ends; when it becomes polarised the middle is squeezed and the ends expand till it looks like an hour glass’ (p. 699).

In summary, we are faced by three quite different and fundamentally contradictory perspectives on the changing occupational class structure of the city after 1970. The key question is which, if any, of these propositions are correct? It is clearly impossible for all of them to be simultaneously correct in one city at one time, although it is possible for different processes to be at work in different cities at different times. Intriguingly, both Friedmann and Sassen made far-reaching general claims over the scope and applicability of their ideas and Sassen (1991) controversially suggested that global cities show similar processes of change, claiming that New York, Tokyo and London:

have undergone massive and parallel changes in their economic base, spatial organization and social structure … and … transformations in cities ranging from Paris to Frankfurt to Hong Kong and São Paulo have responded to the same dynamic. (p. 4, original emphasis)

But this conclusion has been increasingly questioned. As Fainstein et al. (1992) pointedly noted:

The images of a dual or polarised city are seductive, they promise to encapsulate the outcome of a wide variety of complex processes in a single, neat and easily comprehensible phrase. Yet the hard evidence for such a sweeping and general conclusion regarding the outcome of economic restructuring and urban change is, at best, patchy and ambiguous. If the concept of ‘dual’ or polarising city is of any real utility, it can serve only as a hypothesis, the prelude to empirical analysis, rather than as conclusion which takes the existence of confirmatory evidence for granted. (p. 13)

Fainstein’s warning was subsequently echoed by others. Hamnett (1994, 1996) for example, argued that the range of urban social outcomes is influenced by the existence, role and extent of welfare state regimes and also by variations in the level of immigration and the ability to provide a reserve of low-cost labour. Similarly, Burgers and Musterd (2008), Mishra (1999), Stryker (1998), Wessel (2000) and White (1998) have raised questions about generalisability of arguments regarding polarisation and Hill and Kim (2000) and Tai (2010) have raised similar questions regarding the role and influence of developmental states in Asia (see Sassen, 2001, for a defence). More recently, Therborn (2011) has strongly criticised the overall world city hypothesis, particularly what he terms the misconceived globalist and economistic ideas of ‘stateless cities’, arguing that the global financial crisis was only averted by the financial intervention of states on a massive scale. Wade (1996) and Weiss (1997) also strongly reject the idea that globalisation has led to a decline in the role of states.

It seems, after the first flush of excitement regarding the impact of globalisation and the alleged declining role of the nation state, that significant national variations are once again being recognised. The next section looks at the empirical evidence regarding occupational class change in global cities in recent years. It will first discuss evidence and findings for a wide range of cities in the Global North, before moving to examine evidence for some cities in the Global South, including SE Asia, China, South Africa and Latin America. It will argue that, contra Sassen, the available evidence does not point to a general pattern of polarisation and that there is a general increase in numbers and proportion of the managerial and professional groups across most, if not all, global cities over the last 20–50 years, while increases in the proportion of the less skilled only appear to occur in certain circumstances, notably where immigration is less regulated, where the economy is weaker and/or there is a weaker degree of state intervention in economy and social policy. Similarly, the middle of the occupational spectrum has not always shrunk despite Sassen’s suggestions to the contrary. Indeed, it appears that a common feature of most global cities is growth of a transnational, professional and managerial elite as Friedmann and Sassen suggest, but significant growth of the low-skilled service sector is a variable, contingent feature. Thus, despite the importance of globalisation, there are different processes at work and national variations in economic, social and political regimes can strongly influence local outcomes: sometimes very significantly. There is not a ‘one size fits all’ model of urban responses to globalisation and international economic processes.

An overview of the international empirical evidence

An important point needs to be made here regarding the difference between occupational class and income. Friedmann and Sassen have argued that the polarisation of occupational structure has generated greater income inequality. But while there is very clear evidence of growing income inequality in London and other cities, this is not necessarily linked to occupational polarisation as the two things are different (Esteban and Ray, 1994). Indeed, increasing occupational professionalisation may help generate increased income inequality. The evidence examined below is almost entirely about occupational class structure, not income. While there are major variations in the geographical scale and coverage of the data and precise time periods covered this is inevitable in this type of general comparative literature review.

Cities in Europe, North America and the Global North

The first author to take a clear view on social polarisation in London was Ruth Glass (1973) who stated almost 50 years ago (pre-Sassen) that:

The image conveyed by the term ‘polarisation’ (alias the ‘rout of the middle classes’) is, first, that both the top and bottom groups in the society of London are becoming larger, at the expense of the middle strata; second, that the extreme groups are, or will be, located in sharp juxtaposition to one another – on either side of the ‘tracks’. Both these images are false. In fact, the upper and middle strata are becoming stronger, numerically, in Greater London, and the skilled manual group has remained stable, while the proportion of non-skilled workers has decreased. (p. 423)

She continued in the same vein:

London is now being ‘renewed’ at a rapid pace – but not on the model about which we are so often warned. Inner London is not being ‘Americanised’: it is not on the way to becoming mainly a working class city, a ‘polarised’ city, or a vast ghetto for a black proletariat. The real risk for Inner London is that it might well be gentrified with a vengeance, and be almost exclusively reserved for higher class strata. (Glass, 1973: 426, emphasis added)

Glass’s interpretation is strongly empirically supported though London has not become ‘almost exclusively reserved for higher class strata’. London, like other global cities, still has a substantial low-skilled and low-income population, albeit a much smaller one than 50–60 years ago. One of the first empirical studies of urban occupational class structure was undertaken by Wilmott and Young (1973) who looked at London in the period from 1951 to 1966. They concluded:

There was a general upgrading. Both Greater London and the Outer Metropolitan Area showed an increase in the proportion at the top and a decrease at the bottom, with intermediate class staying much as it was. In this sense, the whole London region became more middle class over the 15 years. (p. 200, emphasis added)

Hamnett (1976) also found similar results in his analysis of socio-economic change in Inner London, 1961–1971, which showed absolute and percentage increase in social class 1 and absolute decreases in all the other socio-economic groups. He concluded that: ‘Inner London has not become increasingly polarised in the sense of having become increasingly bi-polar in its socio-economic distribution. The histogram of socio-economic groups has flattened in the middle and increased towards the higher end of the socio-economic distribution’ (p. 270).

His subsequent work (1994, 2003a, 2003b) shows that London did not become polarised in 1961–1991 but instead saw an increase in the number and proportion of its professional and managerial workforce combined with stability in the intermediate non-manual groups and decrease in the size of its skilled, semi-skilled and unskilled manual labour force from 1961 to 1991. He argued that the dominant process is one of professionalisation. This finding was reinforced by Butler et al. (2008) who argued on the basis of NS SEC data on occupational class change from 1981 to 2001 that the process of professionalisation had continued with extensive growth of the lower middle class as well as the top, and large increases in inner London. But these findings have been queried by three papers. May et al. (2007) argued that while London may have undergone a process of professionalisation in the 1960s to the 1990s, growing globalisation and internationalisation with weakening labour protection has more recently led to the growth of a large migrant labour force working in the low-wage hospitality, catering and security industries.

More recently, Davidson and Wyly (2012, 2015) have challenged Hamnett and Butler’s analysis and findings. They argued that it is wrong to claim that the working class has disappeared or that London has become a middle-class city. But Hamnett and Butler (2013) only claimed that the working class had shrunk and that London had become a more (not exclusively) middle-class city, albeit one with a higher degree of income inequality (Hamnett, 2003a, 2003b). More importantly, Davidson and Wyly queried the allocation of certain occupational class groups to the middle class, arguing that they are very heterogeneous and that some jobs are more clearly working class and that the middle class is a ‘non-class’. While there are legitimate questions about occupational classification, this latter claim would come as a major surprise to most non-Marxist contemporary class analysts including Savage (2015) who see the growth of the middle class as very clear cut across most Western countries.

Perhaps more importantly Manley and Johnston (2014) challenge Hamnett and Butler on the basis that their analysis of occupational class change 2001–2011 does not show a continued expansion of the professional and managerial middle class. Rather, it appears that the growth of the middle class stalled during this decade. Hamnett (2015) presents very similar figures to Manley and Johnston but argues that their analysis is flawed by virtue of their exclusion of a large group of self-employed from the middle class. Nonetheless, it seems that while London has seen a long-term process of professionalisation from 1961 to 2001, the process has now stalled, though whether temporarily or permanently is unclear. The changes may reflect the growth of low-wage immigration over the last intercensal decade as May et al. (2007) suggested.

In a somewhat similar analysis of Amsterdam and the Randstad in the Netherlands, Hamnett (1994) argued that the Randstad had seen a process of professionalisation in both its occupational and its educational structure from 1981 to 1991. Burgers (1996) challenged this on the grounds that it ignored the unemployed but Hamnett responded that they were not part of Sassen’s thesis. Kloosterman’s (1996) examination of polarisation in Amsterdam and Rotterdam found it was more a result of sectoral income differences than changes in sectoral composition but Burgers and Musterd’s (2008) comparison of the two cities reveals significant differences in terms of labour market and institutional history, and subsequent research on a number of other European cities has produced contradictory findings. Prteceille’s (1995, 2006) detailed work on Paris showed that Paris has undergone a very similar process to that of London with significant growth of the professional, managerial and technical groups and a decline in the share of other groups. Recent comparative work by Prteceille and Cardoso on Paris and Rio shows continuing middle-class growth to 2010. This is discussed below.

The pattern is more mixed in some other major European cities. In Oslo, for example, Wessel (2015) found that the city had experienced major de-industrialisation during the 1970s, 1980s and 1990s, with major shrinkage of manufacturing industry (25% to 5%) and a corresponding increase in producer services (from 5% to 24%). It is, says Wessel (2015), a ‘quintessential post-industrial city’, but he argues that the pattern of outcomes is complex and that it is not possible to simply read off changes in income and wealth or segregation from changes in the industrial structure and that polarisation, if it exists, is a ‘contingent outcome’ rather than a necessary consequence (p. 133). This is an important point.

For Helsinki, Vaattovaara and Kortteinen (2003) conclude that:

The bimodal urban differentiation is not regarded as a sign of polarisation, but rather as a new phase in economic development linked with a structural shift in the demand for labour that can result in very different outcomes in different cities and welfare regimes. In other words, both polarisation and professionalisation are interpreted as different versions or outcomes of a more fundamental structural shift combining both. (p. 2145)

So too, Andersen (2004) concludes that while Copenhagen has seen growing spatial polarisation between upper-class areas and other areas, it has not seen increased social polarisation.

So, while London, Paris and the Randstad show clear evidence of professionalisation, at least until 2001 in London and 2006 in Paris, the evidence is more ambiguous for some smaller European cities. But the debate is much wider than Europe. There have now been a number of studies in cities across the globe and the next section looks at other cities in the Global North. Surprisingly, there do not seem to have been many studies of occupational class change in New York or other American cities (see Nørgaard, 2003, on wage polarisation) but writing on New York, Ehrenhalt (1993) has highlighted the growth in managerial, professional and technical jobs, and Brint (1991) also noted that:

Where salaried professionals and managers … accounted for something under 5 percent of the working population throughout the nineteenth century and early twentieth century, they now represent nearly 30 percent of the labor force [of New York]. At the higher end of the professional and managerial spectrum, affluent professional people are now so numerous, visible, and influential that some observers have characterized them as the ‘new dominant class’ of the postindustrial city. (p. 155)

Similarly, Clark and McNicholas’s (1996) study of occupational change in Los Angeles, 1970 to 1990, found that: ‘The substantial increase in the professional, managerial and technical group stands out from all the other occupational changes … the massive decline is in the proportion reporting unskilled occupations’ (pp. 61–62). Recent work by Scott (2019) also shows a sharp growth in the number of skilled white-collar jobs in LA county, 2000–2015, though does not look at the overall occupational structure. Finally, Walks’ (2001) analysis of occupational change in Toronto 1971–1991 showed that professional occupations grew very significantly. He comments that:

Overall, the Toronto region would appear to mirror established occupational changes witnessed in many other North American cities with a similar industrial heritage such as New York, Chicago and San Francisco, where a reduction in factory jobs has been accompanied by an increase in professional employment. (p. 417)

Looking at industrial and occupational change in Tokyo and Osaka 1985–2005 and the links to inequality, Debnar et al. (2014) conclude that: ‘Our analysis has shown that the extended social polarization hypothesis regarding the structural change in particular occupations, industries and their causal connection to income inequality cannot be adopted in the case of Japan in general. Moreover our analyses strongly suggest that the social polarization hypothesis cannot be supported in its orginal non-extended form either’ (pp. 45–46).

Similarly, in his work on occupational class change in Sydney from 1986 to 1991, Baum (1997) showed that while the percentage point share of managers and administrators, professionals, para professionals and sales and personal service workers all grew slightly, ‘All the remaining occupational categories (trades persons, clerks, plant and machine operators and drivers and labourers and related workers) recorded declining numbers and overall occupational shares’ (p. 1885). But, while Baum acknowledges professionalisation at the top end, he also suggests that the growth at the bottom end may be evidence of polarisation. The period is abnormal, however, as it spans the early 1980s recession and he also finds a large increase in unemployment. His other work on Singapore suggests that there is a clear trend towards professionalisation rather than polarisation: ‘within Singapore, rather than the development of a polarised structure, there is a trend towards a professionalised occupation structure, and a growing upper-middle income group’ (Baum, 1999: 1095). Thus, we can see a broadly common pattern across most global cities in the Global North. Indeed, White (1998) concludes from his study of Paris and Tokyo that: ‘globalisation as defined in the model does not necessarily follow from the internationalisation of great cities in advanced capitalist countries nor does the socioeconomic dualisation presumed to follow from globalisation’ (p. 451).

Cities in the Global South

Moving to the Global South, there have been some interesting and rigorous studies of polarisation in Johannesburg and Cape Town by Borel-Saladin and Crankshaw. They have taken issue with one of the fundamental propositions regarding the service sector. As Borel-Saladin and Crankshaw (2009) note:

many scholars have misunderstood the significance of middle-income service-sector occupations for their interpretations of the post-industrial class structure of cities. Through a comparative study of deindustrialisation in Cape Town, evidence is presented to show that the growth of service-sector employment can produce a large middle-income occupational class of clerks, sales and personal services workers. The growth of this class can offset the decline of middle-income jobs caused by the loss of artisans, machine operators and drivers in the declining manufacturing sector. These results therefore suggest that many studies have overestimated the extent of occupational polarisation and underestimated the extent of professionalisation. (p. 645)

Their work on Johannesburg (Crankshaw and Borel-Saladin, 2014) confirms that occupational change within economic sectors has produced professionalisation alongside a higher level of unemployment. This parallels Burger’s arguments regarding the changes in the Randstad. Looking at Latin America, Marques’ (2016) detailed work on São Paulo has shown massive inequality between rich and poor but he found that: ‘Contrary to the expectations of studies formed by general descriptions of globalisation, economic restructuring and neo-liberalism, the recent economic transformations have not led to social polarisation’ (p. 3, emphasis added). In fact, he argues, the different situations and histories of different cities make generalisations about common patterns very problematic. Valuable comparative work on Rio de Janeiro, São Paulo and Paris (Préteceille and Cardoso, 2020) using the French catégorie socioprofessionnelle and data from the 2000 and 2010 censuses in Brazil, and 1999 and 2006 in France for the wider metropolitan areas, also produced important findings:

The major contrast between Rio and São Paulo on the one hand and Paris on the other is the relative weight of the working class as a whole: it is about half of the active population in the two Brazilian cities, and only one quarter in Paris. The trend, nevertheless, is similar in the three cities, with a decline of the working class, more intense for blue-collar workers …Contrary to the dual city model, in all three cities the middle class represent a substantial part of the active population, and an expanding one. (p. 309, emphasis added)

Social polarisation and professionalisation in China and SE Asia

The debate over polarisation and professionalisation in Asia introduces an important new dimension regarding the existence of what are termed ‘Developmental States’: that is, countries where the state has taken a leading role in economic and urban development. The term was initially formulated to encompass Japan and South Korea but it has been extended to embrace Hong Kong, Singapore and Taipei and should arguably be extended to include China. Thus, on the one hand, Chiu and Lui (2004) argue that: ‘Hong Kong experienced during the 1990s a process of occupational polarisation and widening income inequality as a result of its transformation from an industrial colony to a producer service-driven global city … The findings largely support Sassen’s hypotheses regarding the social consequences of global city development, but highlight the effect of local institutional contexts in mediating the impact of global forces’ (p. 1863).

But on the other hand, Hill and Kim (2000) argue that Friedman and Sassen’s claims regarding global cities do not fit cities in South-East Asia such as Tokyo and Seoul, where the state plays a leading role and the relationship between global economic processes and national urban outcomes is mediated or directed by the state. This point has been developed by Tai (2010). In her comparative study of Hong Kong, Singapore, Seoul and Taipei, she challenges the polarisation hypothesis and argues that the situation in developmental states is more complex:

The results show that, in all these cities, there is a common trend towards professionalization, but no accompanying growth in managers, clerks, and self-employed and household labour. Different scenarios of social polarization have been observed in two types of Asian world city, the ‘city state’ (e.g. Singapore and Hong Kong) and the ‘capital city’ (e.g. Taipei and Seoul). Singapore and Hong Kong are examples of a centralized development model that enhances the labour force and expands cross-border networks with polarized immigrant workers. In contrast, Taipei and Seoul demonstrate anti-centralizing tendencies and have had to prevent the emigration of native elites and control the immigration of low-skilled foreign workers. The social polarization hypothesis in Asian world cities needs to be re-examined in the local institutional context mediated by the social policies of developmental states. (p. 743, emphasis added)

We can thus see a consistent picture emerging of differences between different cities depending on, amongst other things, the type and level of immigration, the presence and strength of welfare state regimes and the scale and extent of development states. In this context it is useful to turn to China: a country with very strong central and local government. Gu’s (2001) early study of Beijing argued that the reforms from 1980 to 2000 led to a growing polarisation between large numbers of low-skilled rural to urban migrants at the bottom end of the social structure and growth of foreign managers and professionals working for multi-national or joint venture companies at the top.

But although the migrant population in Chinese cities has continued to grow and totals an estimated 200+ million people, the top layer of professionals and managers has also grown rapidly, and Shi et al. (2017) ask whether Beijing has become polarised or a professionalised society? They point out that, like the developmental state in Tokyo, the local state has had a very major role in globalising Beijing and has regulated: ‘the impacts of economic globalisation and national migration on its social structure, thus pushing the social structure of Beijing away from a polarised one’ (p. 160). Their detailed analysis of the changing occupational structure of Beijing from 2004 to 2103 reveals a sharp increase (+13 pp) in the proportion of professionals, administrators and managers from 31% to 44%, a big shrinkage of the middle group of manual workers (−11 pp) and a slight shrinkage (−2 pp) of the semi-skilled or unskilled workers. They thus conclude that:

Beijing, as a global city in China, has witnessed a significant occupational upgrading by the process of professionalization, in which the proportion of managers and professionals rose whereas the shares of employment in all the other occupational groups fell. (p. 160)

Their general conclusion is that social structural change in global cities is a result of a complex interplay between economic globalisation, large-scale migration and institutional factors and that, in the case of Beijing, professionalisation rather than polarisation has been a result of local institutional factors and state control of migration (p. 163). Indeed, the Beijing municipal government has recently initiated the demolition of informal peripheral settlements for migrant workers as part of a policy to upgrade and internationalise the economy and social structure of the city. This is state-directed social upgrading.

Conclusion

We have attempted in this paper to examine the question of what is happening to the occupational class structure of what are termed world or global cities. We have done this by adopting a broad comparative perspective examining what has happened to these cities over the past 50 years, since the onset of de-industrialisation in the late 1960s and early 1970s; the coming of a post-industrial society; and the expansion of finance and allied business services. The introduction posed the key question of whether these cities had become more professionalised, proletarianised or more polarised over time. It seems clear that global cities have not experienced a general process of proletarianisation at an aggregate level though this is not to say that individual occupations have not been proletarianised. On the contrary, the dominant process seems to have been an expansion of professional, managerial and technical occupations and a decline in the size of the less skilled manual groups. Where there has been polarisation it has generally been asymmetric with more growth at the top end than the bottom, though this depends in large part on the role of immigration and state policy and the level and form of state direction of the economy.

This is not to say, however, as some have argued, that the working class has disappeared or that inequality has reduced. On the contrary, income inequality has increased in many cities driven, in large part, by massive increases in incomes of the top 1–2%. While there has been growth in some parts of the low-wage service sector, this has also been offset by contraction in other sectors such as public transport. What is unquestionable is that there have been major changes in the occupational structure of large Western cities over the last 150 years and that most world/global cities of today have far fewer less skilled residents than they had 100 or even 50 years ago. It is also important, given the criticism of the economic base of world/global city theory and the importance of what are termed ‘ordinary’ cities, to critically examine the social changes taking place within these cities and not to simply label them as polarising. If it is possible to reach a general conclusion about professionalisation and polarisation it is that while a city’s position in the global division of labour is extremely important, economically deterministic mono-causal theories which fail to incorporate differences in local welfare state regimes, political priorities, social and historical structures and planning systems are generally likely to be insufficient in explaining local variations in outcomes. This point is strongly made by Marques (2016: 7), and Maloutas (2007) states that irrespective of the validity of the polarisation thesis, its widespread diffusion:

gives the impression that polarization and segregation are growing throughout the world even if in most major urban areas of the western world the socially polarizing mechanism assumed for global cities is not present. The social polarization thesis has thus created a dominant way of seeing urban society extending far beyond the global city context. (p. 736)

We hope that this paper may go some way towards challenging the growing uncritical acceptance of the idea of social polarisation. It does exist in some places but its scale, extent and importance vary over space and time and depend on a variety of factors both global and national. Similarly, and not withstanding growing inequality, the dominant trend in global city social structure over the last 50 years has broadly been one of growing professionalisation, often combined with growing inequality, not one of proletarianisation. Although some academics may regret the decline (not disappearance) of the working class and the growth of the middle class, it appears to be an empirical reality in most global cities, and this squares with the well-documented growth of urban gentrification. So, although the three Ps of professionalisation, proletarianisation and polarisation are useful lenses to investigate occupational class change, it is important not to let the lens obscure the empirical evidence, which suggests that the decline of manufacturing industry and the rise of a post-industrial economy have primarily generated growth in the upper part of the occupational structure rather than at both ends. This seems to hold in a wide range of countries and contexts although no evidence is presented for cities in the Indian subcontinent or Africa, apart from South Africa, and this could change in the future. While the notion of polarisation may be politically and ideologically attractive this does not mean that it is empirically correct, as Mollenkopf and Castells (1991) noted 30 years ago. It also suggests that theories of urban change which primarily rely on globalisation need to be critically assessed vis-à-vis a range of other factors including national historic, economic, political and social variations.

Footnotes

Acknowledgements

I would like to acknowledge the valuable comments of three anonymous referees.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Chris Hamnett

References

Allen

(2018) Class structure and inequality during the industrial revolution: Lessons from England’s social tables, 1688–1867. The Economic History Review 72(1): 88–125.

Andersen

(2004) Spatial – not social polarization: Social change and segregation in Copenhagen. The Greek Review of Social Research 113(113): 145–165.

Banfield

(1970) The Unheavenly City: The Nature and Future of Our Urban Crisis. Boston, MA: Little Brown.

Baum

(1997) Sydney, Australia: A global city? Testing the social polarisation thesis. Urban Studies 34(11): 1881–1902.

Baum

(1999) Social transformation in the global city: Singapore. Urban Studies 36(7): 1095–1117.

Beauregard

(1993a) Voices of Decline: The Postwar Fate of US Cities. Oxford: Basil Blackwell.

Beauregard

(1993b) Representing urban decline: Postwar cities as narrative objects. Urban Affairs Review 29(2): 187–202.

Bell

(1973) The Coming of Post-Industrial Society. New York: Basic Books.

Borel-Saladin

Crankshaw

(2009) Social polarisation or professionalisation: Another look at theory and evidence on deindustrialisation and the rise of the service sector. Urban Studies 46(3): 645–664.

10.

Braverman

(1974) Labour and Monopoly Capital. New York: Monthly Review Press.

11.

Brick

(1992) Optimism of the mind: Imagining post-industrial society in the 1960s and 1970s. American Quarterly 44(3): 348–380.

12.

Briggs

(1963) Victorian Cities. Harmondsworth: Penguin.

13.

Brint

(1991) Upper professionals: A high command of commerce, culture and civic regulation. In: Mollenkopf

Castells

(eds) Dual City: Restructuring New York. New York: Russell Sage, pp. 155–176.

14.

Burgers

(1996) No polarisation in Dutch cities? Inequality in a corporate country. Urban Studies 33(1): 99–105.

15.

Burgers

Musterd

(2008) Understanding urban inequality: A model based on existing theories and an empirical illustration. International Journal of Urban and Regional Research 26(2): 403–413.

16.

Butler

Hamnett

Ramsden

(2008) Inward and upward: Marking out social class change in London, 1981–2001. Urban Studies 45(1): 67–88.

17.

Cannadine

(1977) Victorian cities: How different? Social History 2(4): 457–482.

18.

Chiu

SWK

Lui

T-l

(2004) Testing the global city – Social polarisation thesis: Hong Kong since the 1990s. Urban Studies 41(10): 1863–1888.

19.

Clark

WAV

McNicholas

(1996) Re-examining economic and social polarisation in a multi-ethnic metropolitan area: The case of Los Angeles. Area 28: 56–63.

20.

Crankshaw

Borel-Saladin

(2014) Does deindustrialisation cause social polarisation in global cities? Environment and Planning A: Economy and Space 46(8): 1852–1872.

21.

Davidson

Wyly

(2012) Class-ifying London: Questioning social division and space claims in the post-industrial metropolis. City 16(4): 395–421.

22.

Davidson

Wyly

(2015) Same but different: Within London’s ‘static’ class structure and the missing antagonism. City 19(2–3): 247–257.

23.

Debnar

Yasui

Tarohmaru

(2014) Global cities and social polarization in Japan: Industries, occupations and inequality in comparison with other regions. Kyoto Journal of Sociology 22: 23–48.

24.

Dewar

Thomas

(2013) The City after Abandonment. Philadelphia, PA: University of Pennsylvania Press.

25.

Dorling

(2014) Inequality and the 1%. London: Verso.

26.

Doucet

(2017) Why Detroit Matters: Decline, Renewal and Hope in a Divided City. Bristol: Policy Press.

27.

Ehrenhalt

(1993) Economic and demographic change: The case of New York. Monthly Labour Review 116: 40–50.

28.

Elvery

(2019) Changes in the occupational structure of the United States: 1860 to 2015. Economic Commentary (Federal Reserve Bank of Cleveland), pp. 1–8.

29.

Engels

(1846) The Condition of the English Working Classes in 1845. London: Fontana.

30.

Esping-Andersen

(ed.) (1993) Changing Classes: Stratification and Mobility in Post-industrial Societies. London: Sage.

31.

Esteban

Ray

(1994) On the measurement of polarization. Econometrica 62(4): 819–851.

32.

Fainstein

Gordon

Harloe

(1992) Divided Cities: New York and London in the Contemporary World. Oxford: Basil Blackwell.

33.

Ferkiss

(1979) Daniel Bell’s concept of post-industrial society: Theory, myth and ideology. The Political Science Reviewer 9: 61–102.

34.

Friedmann

(1986) The world city hypothesis. Development and Change 17(1): 69–83.

35.

Friedmann

Wolf

(1982) World city formation: An agenda for research and action. International Journal of Urban and Regional Research 6(3): 309–344.

36.

Glaeser

(2011) Triumph of the City: How Urban Spaces Make us Human. London: Pan Macmillan.

37.

Glass

(1973) The mood of London. In: Donnison

Eversley

(eds) London: Urban Patterns, Problems and Processes. London: Heinemann Educational, pp. 405–432.

38.

Goldthorpe

Lockwood

(1967) The Affluent Worker. London: Allen Lane.

39.

Green

(1995) From Artisans to Paupers: Economic Change and Poverty in London, 1790–1870. Aldershot: Scolar Press.

40.

(2001) Social polarisation and segregation in Beijing. China Geographical Sciences 11(1): 17–26.

41.

Halsey

(2000) A hundred years of social change. Social Trends 30: 15–20.

42.

Hamnett

(1976) Social change and social segregation in Inner London, 1961–71. Urban Studies 13: 261–271.

43.

Hamnett

(1994) Social polarisation in global cities: Theory and evidence. Urban Studies 31(3): 401–424.

44.

Hamnett

(1996) Social polarization, economic restructuring and welfare state regimes. Urban Studies 33(8): 1407–1430.

45.

Hamnett

(1998) Social segregation, social polarisation and the underclass. In: Paddison

(ed.) The Urban Studies Handbook. London: Sage, pp. 162–176.

46.

Hamnett

(2003a) Gentrification and the middle class remaking of inner London, 1961–2001. Urban Studies 40(12): 2401–2426.

47.

Hamnett

(2003b) Unequal City: London in the Global Area. London: Taylor and Francis.

48.

Hamnett

(2015) The changing occupational class composition of London. City 19(2–3): 239–246.

49.

Hamnett

Butler

(2013) Reclassifying London: A growing middle class and increasing inequality. City 17(2): 197–208.

50.

Harvey

(1985) Paris, 1850–1870. In: Harvey D Consciousness and the Urban Experience. Baltimore, MD: Johns Hopkins University Press, pp. 63–220.

51.

Harvey

(2003) Paris: Capital of Modernity. London: Routledge.

52.

Hill

Kim

(2000) Global cities and developmental states: New York, Tokyo and Seoul. Urban Studies 37(12): 2167–2195.

53.

Kloosterman

(1996) Double Dutch: Polarization trends in Amsterdam and Rotterdam after 1980. Regional Studies 30(5): 467–476.

54.

Kumar

(1978) Prophecy and Progress: The Sociology of Industrial and Post-Industrial Society. Harmondsworth: Penguin.

55.

Maloutas

(2007) Segregation, social polarization and immigration in Athens during the 1990s: Theoretical expectations and contextual difference. International Journal of Urban and Regional Research 31(4): 733–758.

56.

Manley

Johnston

(2014) London: A dividing city, 2001–11? City 18(6): 633–643.

57.

Marcuse

(1989) Dual city: A muddy metaphor for a quartered city. International Journal of Urban and Regional Research 13(4): 697–708.

58.

Marques

(2016) São Paulo in the Twenty-First Century: Spaces, Heterogeneities, Inequalities. London: Routledge.

59.

Marsh

(1965) The Changing Social Structure of England and Wales, 1871–1961. London: Routledge.

60.

Marshall

Rose

(1988) Proletarianisation in the British class structure. The British Journal of Sociology 39(4): 498–518.

61.

May

Wills

Datta

, et al. (2007) Keeping London working: Global cities, the British state and London’s new migrant division of labour. Transactions of the Institute of British Geographers 32(2): 151–167.

62.

Mishra

(1999) Globalization and the Welfare State. Cheltenham: Edward Elgar.

63.

Mollenkopf

Castells

(eds) (1991) Dual City: Restructuring New York. New York: Russell Sage.

64.

Musterd

(ed.) (2020) Handbook of Urban Segregation. Cheltenham: Edward Elgar.

65.

Nørgaard

(2003) The global city thesis – Social polarization and changes in the distribution of wages. Geografiska Annaler: Series B, Human Geography 85(2): 103–119.

66.

Pahl

(1988) Some remarks on informal work, social polarisation and the social structure. International Journal of Urban and Regional Research 12(2): 247–267.

67.

Piketty

(2014) Capital in the Twenty First Century. Cambridge, MA: Harvard University Press.

68.

Préteceille

(1995) Division sociale de l’espace et globalisation: Le cas de la métropole Parisienne. Sociétés Contemporaines 22(1): 33–67.

69.

Préteceille

(2006) La ségrégation sociale a-t-elle augmenté? Sociétés Contemporaines 62(2): 69–93.

70.

Préteceille

Cardoso

(2020) Socio-economic segregation and the middle classes in Paris, Rio de Janeiro and São Paulo: A comparative perspective. In: Musterd

(ed.) Handbook of Urban Segregation. Cheltenham: Edward Elgar, pp. 270–288.

71.

Robinson

(2002) Global and world cities: A view from off the map. International Journal of Urban and Regional Research 26(3): 531–554.

72.

Robinson

(2006) Ordinary Cities: Between Modernity and Development. London: Routledge.

73.

Rubenstein

(1977) The Victorian middle classes: Wealth, occupation and geography. Economic History Review 30(4): 602–623.

74.

Sassen

(1991) Global City: New York, London and Tokyo. 2nd edition. Princeton, NJ: Princeton University Press.

75.

Sassen

(2001) Global cities and developmentalist states: How to derail what could be an interesting debate: A response to Hill and Kim. Urban Studies 38(13): 2537–2540.

76.

Savage

(2015) Social Class in the Twenty First Century. London: Penguin Books.

77.

Savitch

(1988) Post-Industrial Cities: Politics and Planning in New York, Paris and London. Princeton, NJ, and London: Princeton University Press.

78.

Scott

(2019) Residential adjustment and gentrification in Los Angeles, 2000–2015: Theoretical arguments and empirical evidence. Urban Geography 40(4): 506–528.

79.

Shi

Liu

Musterd

, et al. (2017) How social structure changes in Chinese global cities: Synthesising globalisation, migration and institutional factors in Beijing. Cities 60: 156–165.

80.

Sjoberg

(1955) The pre-industrial city. American Journal of Sociology 60(5): 438–445.

81.

Steadman-Jones

(1971) Outcast London. Harmondsworth: Penguin.

82.

Stryker

(1998) Globalization and the welfare state. International Journal of Sociology and Social Policy 18(2–4): 1–49.

83.

Tai

(2010) Beyond social polarization: A test for Asian world cities in developmental states. International Journal of Urban and Regional Research 34(4): 743–761.

84.

Tammaru

van Ham

Marcińczak

, et al. (2015) Socio-Economic Segregation in European Capital Cities. East Meets West. London: Routledge.

85.

Therborn

(2011) End of a paradigm: The current crisis and the idea of stateless cities. Environment and Planning A: Economy and Space 43(2): 272–285.

86.

Vaattovaara

Kortteinen

(2003) Beyond polarisation versus professionalisation? A case study of the development of the Helsinki region, Finland. Urban Studies 40(11): 2127–2145.

87.

Vogt

(2015) Post-industrial society: From utopia to ideology. Work, Employment and Society 30(2): 366–376.

88.

Wade

(1996) Globalization and its limits: Reports of the death of the National Economy are greatly exaggerated. In: Berger

Dore

(eds) National Diversity and Global Capitalism. Ithaca, NY: Cornell University Press, pp. 60–88.

89.

Walks

(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.

90.

Ward

(1971) Cities and Immigrants: A Geography of Change in Nineteenth Century America. New York: Oxford University Press.

91.

Ward

Zunz

(eds) (1992) Landscape of Modernity: Essays on New York City, 1900–1940. New York: Russell Sage.

92.

Weiss

(1997) Globalization and the myth of the powerless state. New Left Review 225: 3–17.

93.

Wessel

(2000) Social polarisation and socio-economic segregation in a welfare state: The case of Oslo. Urban Studies 37(11): 1947–1967.

94.

Wessel

(2015) Economic segregation in Oslo: Polarisation as a contingent outcome. In: Tammaru

Marcińczak

Van Ham

, et al. (eds) Socio-Economic Segregation in European Capital Cities: East Meets West. London: Routledge, pp. 132–155.

95.

White

(1998) Old wine, cracked bottle: Tokyo, Paris and the global city hypothesis. Urban Affairs Review 33(4): 451–477.

96.

Wilmott

Young

(1973) Social class and geography. In: Donnison

Eversley

(eds) London: Urban Patterns, Problems and Policies. London: Heinemann, pp. 190–214.

97.

Wilson

(1996) When Work Disappears: The World of the New Urban Poor. New York, NY: Random House.

98.

Wright

Martin

(1987) The transformation of the American class structure, 1960–1980. American Journal of Sociology 93(1): 1–29.