THE SPACE DYNAMICS OF INEQUALITY IN THE SÃO PAULO METROPOLITAN AREA

Transcription

1 THE SPACE DYNAMICS OF INEQUALITY IN THE SÃO PAULO METROPOLITAN AREA Suzana Pasternak Lucia Maria Machado Bógus 1 Introduction The Observatório das Metrópoles (Metropolis Observatory) organization has been developing studies and surveys with a view to contributing theoretically and methodologically to the debates, in the academia sphere, of government institutions and non-government organizations (NGOs) and social movements, about the social impacts produced by the economic transformations taking place in Brazil since the midnineteen-eighties. Those impacts are externalized particularly in large cities and in metropolitan areas, where transformations acquire a deeper meaning. The controversy that feeds the debate focuses on the effects of productive labor market restructuring, with significant changes in the opposition between social classes that marked the Fordist industrial era, and the emergence of a new social structure, marked by an increasing polarization between the upper and lower strata of society. Those issues are the focus of discussion on the global city (Sassen, 1998), whose main assumption is the existence of structural nexuses between changes in progress in the economy and the intensification of social duality. In that process, where the Tertiary Sector would be taking predominance against simultaneous modernization processes and a relative retraction in Secondary Sector jobs, there would equally be a reconfiguration and shrinking of middle classes, in view of the changes in the productive structure and in organizational and technological standards. Some occupations, which are typical of the middle classes would be declining; others would be of-qualified, and new professions connected to the expansion of management functions would emerge (Sassen, 1998). The social structure, backed on the information-technology industry, especially in large cities (metropoles), would be supported, on the one hand, on the existence of highly qualified and well paid professionals and, on the other hand, on a less qualified and equally important contingent of workers such as secretaries, cleaners and maintenance workers, configuring a social structure in the shape of a sandglass. That image, which is not unanimous among surveyors, would be in contraposition to that of an egg, that represents the predominance of middle and working classes in the social structure and the reduced presence of higher and lower strata of that structure. Based on those assumptions, and intending to check the applicability or not of the global city theses to the Brazilian reality, a socio-occupational hierarchy was prepared with the construction of a set of categories, based on the census occupational variables according to the CBO Classificação Brasileira of Ocupações (Brazilian Classification of Occupations), established according to International Labor Organization (ILO) guidelines of International Standard Classification of Occupations (ISCO). Census data are the only data available in Brazil with simultaneous time and space comparability capacity, contemplating data of the labor sphere. As a reference point, we used a sociooccupational category classification system, partially inspired by the French CS (adopted by the Institut National d Économie et Statistique (INSEE) and partly derived from Brazilian efforts on professions in France). The first comparative study was performed between Paris and Rio of Janeiro (Preteceille, Ribeiro, 1998). 1

2 Those surveys have as their starting point a multidimensional conception of social space structuring, which enables us to reach a more refined understanding of possible social positions occupied by groups of individuals and to detect the multiple hierarchical scales in the social space. The social structure [...]is understood, simultaneously, as a space of social positions and a space of individuals occupying those positions and provided with social attributes unequally distributed and connected to their histories (Ribeiro Lago, 2000, p 112), within an articulation that reminds the thinking of Bourdieu (1989). The author develops the notion that individuals or agents occupy relative positions in the social space, which are opposite. It is possible to empirically classify those relative positions according to different social groupings that may be identified by the volume of capitals (economic, social and symbolic capitals) held by them and by the structure of those capitals. Placed in similar positions and being subject to similar conditionings, those agents or individuals are likely to develop similar attitudes, interests and practices. The incorporation of that scheme to the survey on Brazilian metropoles is backed on the methodological assumption of labor centrality as a social relations structuring category (Ribeiro, Lago, 2000, p 112). The socio-occupational categories, through which it is possible to capture the social segmentation in Brazilian metropoles, were built based on some general principles that are contradictory and lie on the basis of the capitalistic society organization, such as: capital and labor, large and small capital, salaried and self-employed work, manual versus non-manual work, and control and execution activities. Also taken into account was the differentiation between production sectors, such as the Secondary and Tertiary sectors; finally, among those of the Secondary Sector, a distinction was made based on the insertion of workers in the modern or traditional industrial segments (Ribeiro, Lago, 2000). (Mammarella, Rosetta, 2007, p 157). From 1991 to 2000, the Brazilian Census changed its method of defining both unemployment and type of occupation, which makes a comparison of 1980, 1991, and 2000 difficult. In 1991, the reference period to check the status of employment was 12 months, as it was in And the occupation condition referred to three possible statuses: if a respondent worked habitually, or occasionally, in that reference period, or if a respondent did not work at all. In 2000, the reference period was one week, and the question was more detailed: the Census asked if the respondent had worked in a paid activity or not; if the answer was no, the question was if the respondent was temporarily unemployed; if the respondent had performed any non-paid activity or if, in the period of one month on the date before the Census, the respondent had taken any action to get work. Therefore, the figures of employed workers in 1980, 1991 and 2000 are not comparable: the adoption of the one-week period, instead of 12 months, may lead to a broader unemployment magnitude. On the other hand, home activities such as helping another individual, working for self-consumption, etc. reduce unemployment rates, because they start to be computed. In 2000, the occupational classification manner was also changed by using the CBO Classificação Brasileira of Ocupações (Brazilian Classification of Occupations) and the CNAE Classificação Nacional da Atividade Econômica (National Classification of Economic Activity). In the survey, an adjustment was made in the 1991 occupational classification with the 2000 Census methodology, which enables a comparison between those two dates. For 1980, however, this was not done. Therefore, any comparison 2

3 involving the last 20 years of the twentieth century may only be performed using large groups, and the percent of employed (occupied) population in relation to the total between those three dates is impossible to compare. For this reason, the analysis of the development of socio-occupational categories will focus on 1991 and It should be pointed out that the absolute number of employed workers in 1991 is not strictly obtained in the same manner as in Even so, we opted for analyzing such development, in spite of this comment. The relative analysis of percents may clarify occupational development. Socio-occupational categories Managers Higher Level Professionals Small Employers Medium Level Occupations Tertiary Sector Workers Secondary Sector Workers Non-specialized Tertiary Sector Workers Agricultural Workers Major Employers Public Sector Managers Private Sector Managers Self-employed higher level professionals Employed higher level professionals Statutory higher level professionals Higher level professors Small Employers Office Occupations Supervision Occupations Technical Occupations Healthcare and Education Occupations Security, Justice and Postal Occupations Artistic and Similar Occupations Commerce Workers Specialized Services Workers Modern Industry Manual Workers Traditional Industry Manual Workers Auxiliary Services Manual Workers Civil Construction Manual Workers Non-specialized Service Providers Domestic Employees/Help Peddlers and Improvised Workers Agricultural Workers 2. Issues The basic issue, as stated by Marcuse and van Kempem (2000, p 2) is: Is there something distinctive in the spatial standard of current cities that differentiates them from the cities of yesterday? What would be new in the cities and particularly in the intra- urban structure of today? Has social segregation increased? (Preteceille, 2006, p 69) A frequent question regarding the economic development of the city of São Paulo is its repercussion on the social structure and on a possible increase in inequality that is probably happening. As a starting point, the main question would be: What is the 3

4 impact of restructuring the productive systems induced by technological innovations and by the demands of a globalized world? (Leal Maldonado, 2000, p 177). As already mentioned, the assumption that the globalization and tertiarization processes cause an increase in the social duality so that extreme positions in the social hierarchy get more apart and are aggravated in global cities has been challenged by authors such as E. Preteceille (1995, 2006, 2008) or C. Hammett (1995). Those authors point out that in Paris and London the development occurred in a different manner. Maldonado (2000) states that in Madrid the result of social changes as a whole was a higher professionalization of social categories. According to Maldonado, if, taking into account society as a whole, there was a certain inequality reduction in income distribution in the period and a growth of middle classes, the difference seen in the groups (...) point to a diversified result of those advances in inequality reduction (Leal Maldonado, 2000, p 177). Preteceille and Cardoso (2008) compare the social structures of São Paulo and Rio of Janeiro with that of Paris. They comment that urban experts usually place São Paulo and Rio of Janeiro as metropoles with high segregation and inequality levels, whereas Paris is characterized as having a complex social and space structure, very different from the dualistic view, both in social terms (importance and growth of medium middle classes) and in space terms (the mixed middle-class zones represent 45% of the total population). (Preteceille and Cardoso, 2008, pp ). Our conclusion is that in socio-economic terms Rio of Janeiro and São Paulo are more segregated than Paris, mainly because of the space concentration of higher level layers, but they would not be dual cities either. (p 639) Cities undergo constant internal change processes. City centers decline or change in shape and function, new business centers arise; in Brazilian metropoles, new poverty clusters are formed, serving as dwelling for pauperized population segments; others rise in position and move to other areas. Spatial divisions are nothing new, but different authors have questioned in the last decades if they keep the causes, the scale, the shape, the location. There is a consensus in the literature that significant changes in the intraurban space have been occurring since the nineteen seventies. The globalization process changing manners of production, reduced government intervention, new technologies, and different power relations influence urbanization standards and the cities morphology. Marcuse and van Kempen, in the introduction to their book Globalizing Cities (2000, p 2 and 3), formulate a number of specific questions that help us guide the analysis. Some of those questions are not reflected in the São Paulo survey, but served as guidelines for us to raise some assumptions: Are there direct and visible globalization impacts on the cities internal spatial standard? Is there any urban form which is characteristic of globalization, be it a radial, border, gentrified city, or any other city? Or is it precisely the absence of a conventional form that characterizes the current city, as suggested by the Los Angeles post-modern model discussion? Do spatial divisions found in the cities of today reflect a dual, fragmented or plural city? Has segregation increased? The withdrawal of industrial plants to 4

5 smaller non-capital cities and the entry of services in the metropolitan space is not a new phenomenon. Is the impact of its effect on the intra-urban space now distinctive? Has the old doughnut standard with the middle classes in the white suburbia, surrounding lower black classes in the central areas become obsolete in the United States? Has the inverted standard, with the higher classes in the central areas and the worker layers in the suburbs, similarly to the standard that characterizes European cities (and most Latin American cities), also become obsolete? Or is the current standard simply a continuation of the old and wellknown real-property market, with variations in its magnitude only? Do municipal and national public policies impact the intra-urban structure? Therefore, the purpose of this paper is to answer two major questions: In relation to the social structure, which have been the transformations in the last decade in the São Paulo Metropolitan Area (SPMA)? What is the relationship between the transformations in the social structure and changes in the spatial segmentation standard of São Paulo s intra-metropolitan space? 3. Trends of Metropolitan Socio-Occupational Categories in the Nineteen Nineties As already mentioned, the labor market in the metropolis of São Paulo will be examined based on a hierarchical structure of 25 socio-occupational categories, relating to years 1991 and Looking at the structure of three major groups (higher level, gathering managers, higher-level professionals and employers; medium level and popular level, with manual workers: commerce, specialized and non-specialized services and industrial workers), one may conclude that in the nineteen nineties the metropolis of São Paulo did not present any significant change in its hierarchical structure (Table 1). Higher-level categories had a small relative increase, going from 10.85% of the total categories to 11.84%, as well as manual workers, who went from 56.25% to 59.50%. However, the participation of medium-level categories had a slight relative decrease, from 32.03% to 28.15%. That first look would be suggesting a slight trend to social duality. 5

6 Table 1 Distribution of Socio-Occupational Categories. São Paulo Metropolitan Area (SPMA) CATEGORIES SPMA NUCLEUS PERIPHERY Major Employers 1.54% 0.77% 1.79% 0.96% 1.06% 0.47% Public Sector Managers 0.07% 0.17% 0.08% 0.13% 0.05% 0.24% Private Sector Managers 0.56% 0.43% 0.73% 0.50% 0.23% 0.32% Managers 2.16% 1.37% 2.59% 1.59% 1.34% 1.03% Self-Employed Higher Level Professionals 1.00% 1.96% 1.23% 2.46% 0.55% 1.18% Employed Higher Level Professionals 2.06% 3.66% 2.49% 4.55% 1.23% 2.29% Statutory Higher Level Professionals 0.93% 0.44% 1.21% 0.57% 0.39% 0.24% Higher Level Professors 1.56% 1.77% 1.71% 1.89% 1.28% 1.58% Higher Level Professionals 5.54% 7.83% 6.65% 9.48% 3.46% 5.29% Small Employers 3.14% 2.65% 3.55% 3.08% 2.37% 1.98% Artistic and Similar Occupations 0.95% 1.10% 1.14% 1.31% 0.60% 0.78% Office Occupations 14.01% 10.58% 14.94% 11.50% 12.27% 9.17% Supervision Occupations 6.51% 5.33% 6.79% 5.89% 6.00% 4.46% Technical Occupations 6.39% 6.35% 6.58% 6.66% 6.03% 5.87% Medium Level Healthcare and Education 2.87% 3.26% 2.89% 3.35% 2.84% 3.11% Occupations Public Security, Justice and Postal Occupations 1.29% 1.53% 1.40% 1.58% 1.09% 1.46% Medium Level Categories 32.03% 28.15% 33.73% 30.29% 28.82% 24.85% Commerce Workers 8.68% 9.72% 8.82% 10.03% 8.39% 9.22% Specialized Service Providers 6.51% 9.62% 6.31% 9.71% 6.89% 9.49% Tertiary Sector Workers 15.19% 19.34% 15.14% 19.74% 15.29% 18.71% Modern Industry Workers 8.98% 7.10% 7.38% 5.49% 12.01% 9.59% Traditional Industry Workers 9.19% 4.59% 8.59% 4.45% 10.32% 4.80% Auxiliary Services Workers 3.17% 5.99% 2.94% 4.83% 3.61% 7.79% Civil Construction Workers 6.01% 6.32% 5.08% 5.33% 7.78% 7.85% Secondary Sector Workers 27.35% 24.01% 23.99% 20.11% 33.71% 30.02% Domestic Workers/Help 6.35% 7.22% 6.47% 6.94% 6.12% 7.66% Peddlers 2.02% 2.78% 2.10% 2.78% 1.87% 2.77% Improvised Workers 0.12% 0.94% 0.09% 0.84% 0.17% 1.09% Non-specialized Service Providers 5.22% 5.22% 5.04% 4.95% 5.58% 5.64% Non-specialized Tertiary Sector Workers 13.71% 16.16% 13.69% 15.51% 13.73% 17.15% Agricultural Workers 0.87% 0.50% 0.66% 0.21% 1.27% 0.97% Total % % % % % % Source: Demographic Censuses of 1991 and 2000 Relative decrease of 20% or more Relative increase of 20% or more However, it would be important to examine in more detail inside each one of the three major socio-occupational groups to obtain a deeper view of the labor market behavior. In regard to the higher-level world the higher-level professionals were the major responsible for the relative increase in the decade, both in the metropolis as a whole and in the nucleus and in the other metropolitan municipalities. Both self-employed professionals and higher-level employees had relative increases of over 20% in the decade, both in the nucleus and in the other metropolitan municipalities. Statutory workers had a relative decrease of over 20%, whereas professors had a small relative increase, however stressed in the metropolitan periphery. On the other hand, the so-called managers had a strong relative loss of over 20%, both in the nucleus and in the other metropolitan municipalities. Such loss is owed mainly to 6

7 the loss of major employers (with over 10 employees). The decrease in absolute numbers was almost 36,000 people, totaling 90,000, that is, a reduction of practically 40%. Some assumptions are outlined for such an incredible reduction in major employers, whose assumptions are not mutually exclusive; their leaving the metropolis, their reduction for company bankruptcy (in the nineteen nineties many companies closed their doors, mainly those of a medium size), reduction of companies for reasons of mergers or acquisitions (a process that had already started in the nineteen nineties). Another process that was evidenced was the withdrawal of managers from the private sector in the nucleus, where there was a relative reduction of over 20%, and a slight increase of that category in other municipalities of the SPMA. Therefore, the managing elite was not only seeing a decreasing process but also moving their homes outside the nucleus of the SPMA. Gated condominiums spreading through the metropolis, in more bucolic municipalities show the new residential choice of wealthier layers, escaping from the metropolis pollution, noise and violence. And, although the volume is reduced (in 1991, 4,000 employed individuals and in 2000, 12,300), the increased number of public-sector managers in the metropolis, mainly in the nucleus, is significant. Comparing the social structure of São Paulo with that of Rio of Janeiro, some singularities may be noted: Although the proportion of higher level categories is greater (in % of employed workers) in Rio than in São Paulo (in 2000, 11.85% of employed workers), trends were similar, that is, loss of managers and gain of higher level professionals in both cities. The internal composition of higher level categories is different: the proportion of manager categories is greater in São Paulo than in Rio, whereas in Rio higher level professionals exceed those of São Paulo by almost 1 percent point. Both cities show a proportion close to 30% for the middle class layers that decreased slightly during the decade. Both cities show a percent of popular urban categories close to 60%, with a small increase during the decade. In both cities there was also a significant increase of specialized tertiary sector workers. São Paulo has a much higher proportion of secondary sector workers, although both in São Paulo and in Rio that proportion has decreased between 1991 and A peculiarity of São Paulo: it was the only Brazilian metropolis under study where there was an increase of over 20% in the proportion of non-specialized tertiary sector workers. Paris profile is quite different: company owners and higher-class professionals represent 23.59% of employed workers. That profile is very different from the profile of the two Brazilian cities where the higher level layers hardly represent 12% of total employed workers. On the other hand, the proportion of workers in the secondary sector in São Paulo in 2000 was 24%, whereas industrial workers represent only 16.4% of total employed workers in Paris, in Another substantial difference occurs in the percent of domestic workers: only 1.2% in Paris, whereas in São Paulo such proportion attains 7.2% of employed workers in Finally, Paris is a city with a strong middle class, but with 50% of employed workers and a substantial presence of higher level layers (23.6% of employed workers). São Paulo and Rio have a middle class representing close to 30%, higher level layers on the order of 11% and a proportion of the so-called popular urban classes much higher than that of France. It is not surprising, then that the 7

8 middle-class spaces in Paris represent the majority and gather a significantly big population, as shown by Preteceille, stating that the mixed middle class zones represent 45% of the total Paris population (Preteceille, 2003, 2006, 2008). However, in spite of those singularities, the general characteristics of the change observed in the two large Brazilian cities are similar to some characteristics seen in large cities of industrialized countries, with a decrease in industrial workers in favor of service workers. The total population employed in formal jobs both in Rio of Janeiro and São Paulo show losses of industrial jobs between 1991 and In São Paulo, formal manufacturing industry jobs represented 35.22% of total jobs in 1991, going to 25.34% in That process is connected to: Tertiarization of some activities not relating to production as a consequence of a corporate reengineering process started in the mid nineteen eighties, when activities previously computed as manufacturing industry (manual) activities such as security, food preparation, cleaning and maintenance were outsourced. Increase in the manufacturing industry s productivity owing to technological improvement and later dismissal of labor. Emigration of industrial plants to the surrounding areas of the metropolis and even to other areas in the country. Leal Maldonado (2000) shows the same phenomenon of a decrease in industrial jobs in Madrid. According to the Spanish Censuses, in 1981 the proportion of industrial workers in the urban region of Madrid was 25.8%, and fell to 24.7% in 1991 and to 18.9% in Therefore, Madrid would take an intermediary position between Paris, with 16.4% of regular industrial workers and São Paulo, with 24% of employed workers and 25% of regular manual workers. However, interesting points emerge both in Madrid, a developed country metropolis, and in São Paulo, in an emerging country: increase in tertiary sector activities and decrease in secondary sector workers. Another common point is the strong increase in higher level professionals. In Madrid there was no increase in non-specialized workers, which apparently happened in São Paulo, but not in Rio of Janeiro. An interesting point to discuss would be that specificity of the city of São Paulo. Who would be the components of the several social layers in São Paulo? The members of the higher level layers managers, higher level professionals and small employers are mainly white, male, largely born outside of Brazil (31% in 2000). The entry of women in those higher level layers is observed, considering that the percent of men decreased between 1991 and That is also a specificity of São Paulo, considering that such fact is not observed in other large Brazilian cities. That rate of 30% probably reflects the presence of top executives of multinational companies headquartered in the metropolis of São Paulo, the preferred place for large corporations to install their head offices. The higher level layers in São Paulo have on average about 12 years of schooling, and a household income of over 20 minimum wages. They live in more and more verticalized units, with satisfactory infrastructure. Medium-level layers represented 32.03% of employed professionals in 1991, going to 28.15% in It is the majority group in both dates for the metropolis as a whole. The greatest change in its composition is owed to the reduction, of over 20%, of office 8

9 occupations. Employed professionals in the medium-level layers are no longer basically men: in 2000, 47% of them were females, a greater proportion than for employed professionals as a whole (41%). The percent of whites is still dominant, but with less intensity than in the higher level layers: about 75% are white. Few of those professionals were born outside of Brazil, and the household income was 14 minimum wages in An important point to note is the growing precariousness of jobs among employed professionals of medium level layers: the proportion of regular workers was 77% in 1991, going to 65% in Medium level layers also compensated for that growing precariousness of jobs with an increase in household income per capita of 38%. Employed professionals in the medium level layers live in homes with basic infrastructure, but with a lower rate of verticalization than the higher level layers; only 25% resided in apartments in In the lower level layer, specialized service providers (including waiters, security guards, cooks and communication, TV and telephone workers, woodworkers, carpenters and similar professionals) were the main responsible for the increase observed. That increase was followed by a rise in the number of workers of auxiliary industrial services (transportation, water, gas, electricity, etc.), peddlers and improvised workers. Domestic workers had an increase from 6.35% to 7.22% of the total occupied workers in the metropolis. In contrast, there was a decrease in traditional industry workers from 9.2% to 4.6% and, to a lesser degree, of modern industry workers, from 9% to 7%. The gender proportion is not uniform for the 3 large categories involved in the lower level layers : specialized tertiary sector workers, non-specialized tertiary sector workers and secondary sector workers. Males prevail in the last category, with over 80%, whereas the proportion of women is 45% in the specialized tertiary sector workers and 64% in the non-specialized tertiary sector workers. The percent of whites is also smaller in the specialized tertiary sector workers in relation to the non-specialized tertiary sector workers, attaining 50% (the proportion of whites in the total employed professionals was 66% in 2000). The presence of Brazilian northeastern migration is strong, totaling more than 705 of lower level layer employed professionals in The average household income ranges from 8 to 10 minimum wages, but it is worth noting that it decreases in the decade in the less favored group, that of non-specialized tertiary sector workers. The growing precariousness of employment relations is more significant in the medium level layers, representing 60% of employed professionals among nonspecialized workers. The average years of schooling increased in the period, but it is still around 6 years of formal schooling. Dwellings are basically horizontal. 4. Socio-Occupational Structure Considerations The metropolis of São Paulo became tertiarized, losing industrial activities and increasing services activities, both among specialized service providers and, to a lesser degree, among non-specialized service providers. The growth of specialized service providers was more significant in the capital city than in the other cities of the metropolitan area, reflecting its transformation in a tertiary-sector city. The changes observed for the whole group of socio-occupational categories are part of a set of economic and social transformations that affected the SPMA in the nineteen eighties and nineteen nineties. Actually, those transformations emerge from a strong 9

10 crisis marked by important falls in the economic production and unemployment, combined with the loss of industrial dynamism, production restructuring and in increase in informal jobs (in 1991, 62% of employed workers had a regular job, whereas in 2000 only 51% did). Throughout that period, the SPMA lost industrial activities (the proportion of employed professionals among secondary-sector workers was 27.35% in 1991 and fell to 24.01% in 2000) and saw an increase in services activities (both among tertiary-sector workers, from 15.19% to 19.34%, and among non-specialized tertiary sector workers, from 13.71% to 16.16%). Tabulation of formal jobs (data provided by RAIS/MTE Annual Records of Social Information, from the Brazilian Ministry of Labor) shows the same trends of industrial job loss and services sector job gain. Among the 4,749,100 formal jobs in the metropolis in 1991, 35.22% belonged to the industrial sector, 11.03% in commerce and 45.23% in services. However, in 2000, with 4,630,809 formal jobs, those percents are 25.34%, 15.12% and 59.26% data (5,721,804 formal jobs) confirm such industrial job loss trend. Table 2 Formal jobs per sector of activity. São Paulo Metropolitan Area, 1991, 2000 and 2006, in percent Sector of activity Industry Commerce Services Crop and Livestock Farming Others/Unknown Total Source: RAIS/MTE data In spite of the increase in tertiary sector activities, the metropolis did not solve its problems of unemployment that continued high (in 2000, for the municipality of São Paulo, the unemployment rate among the population of 15 years of age and over was 17.78%, according to F Seade data). In relation to the socio-occupational structure, there are signs of duality, but not as Sassen stated. The top of the pyramid diminished (shrinking of managing elites), but the immediately lower level had a strong increase (higher-level professionals). The medium level layers had a slight decrease, accompanied by a reduction in secondary-sector workers and an increase in tertiary-sector workers and in the basis of the pyramid. There was an income increase among employers (small and large employers), managers and higher-level professionals. Among employed workers in medium level layers, the income increase was small. Among secondary-sector workers, the income was practically stable, whereas among non-specialized tertiary sector workers there was a loss. The income gain accompanied the social hierarchy. On the other hand, labor relations in all categories became more precarious. Labor instability increased. Those two factors may be reflected, as stated by Lago (2008, p 6), in a concomitant increase in the consumption power in the short run and a reduction in the indebtedness capacity in the long run. Home-living conditions improved, as well schooling levels. There is a 10

11 certain femininization of activities classified as hierarchically inferior. It is in those activities that labor relations are more precarious and where there was income loss. The socio-occupational structure basically keeps its diversity, however presenting: Tertiarization, with a strong increase of specialized tertiary sector workers Professionalization, with a strong increase in higher-level professionals Loss of office occupations Gain of auxiliary services workers and loss of modern and traditional industry workers; therefore, a relative de-proletarization Loss of managers, mainly of major employers Gain among peddlers, improvised workers, and, to a lesser degree, domestic workers Therefore, the result is neither a sandglass-shaped nor an egg-shaped structure. It resembles more a Christmas tree. 5. Social Segmentation of the São Paulo Intra-Metropolitan Space in the Nineteen Nineties One of the objectives of this paper was to examine the relation between the trends reviewed above and the changes in the social segmentation pattern of the intrametropolitan space, in the nineteen nineties. As a background for our analysis, we used a socio-spatial typology for the year 2000 that hierarchically classifies the 812 areas in which the metropolitan space was segmented. Those 812 spatial segments are the so-called AEDs Áreas de Expansão e Disseminação da Amostra (Areas of Sample Expansion and Dissemination) of The AEDs are areas of demographic expansion, that is, spatial segments for which the IBGE Instituto Brasilieiro de Geografia e Estatística (Brazilian Institute of Geography and Statistics) provided micro-data, and later used them for the 1991 spatial analysis. That hierarchy results from the hierarchical grouping of about 400 occupations used by the IBGE, as described above. Therefore, based on 812 AEDs, a factorial analysis was performed using a binary correspondence, followed by a hierarchical classification of the resulting conglomerates, considering the socio-occupational categories. In that factorial analysis, we managed to group homogenous areas in conglomerates of areas with predominance of the same socio-occupational categories, resulting in a typology of areas. Therefore, one can identify the principles whereby the social space of São Paulo is divided, that is, 11 types of areas in 1991 and 12 areas in Each type gathers a set of areas that are considered as socially homogenous. For this study, we opted for gathering different types in large worlds, namely: Higher Level Areas Medium Level Areas Manual Worker (working-class) Areas Popular Areas Agricultural Areas 11

12 Table 3 - Employed Population and areas of review, by area type, 1991 and 2000 Areas/Clusters AEDs Employed Pop. AEDs Employed Pop. Higher Level 183 1,641, ,455,065 Medium Level 212 1,742, ,250,812 Manual Workers 283 1,716, ,013,311 Popular , ,271,396 Agricultural , ,674 Total 812 5,877, ,114,258 Areas/Clusters AEDs Employed Pop. AEDs Employed Pop. Higher Level 22.54% 27.88% 20.57% 20.45% Medium Level 26.11% 29.60% 30.91% 31.64% Manual Workers 34.85% 29.16% 29.80% 28.30% Popular 13.67% 11.36% 16.75% 17.87% Agricultural 2.83% 1.99% 1.97% 1.74% Total % % % % Socio-Spatial Typology of the São Paulo Metropolitan Area in Detailed Maps grouped by types of areas 12

13 13

14 The maps above show a clear social segmentation of the metropolitan territory, a higher-profile stain in the center of the metropolis, superposed to the capital, and, as we get a distance from that stain, we see a decreasing hierarchy of socio-spatial types. That hierarchy is broken by some blots of a higher-level type, in the west zone of the Greater São Paulo, in addition to a small stain in the east area. The central higher level areas are surrounded by medium level areas, also located predominantly in the capital city. The expansion of the medium and higher income level classes in the west axis leads to the emergence of higher level stains and a large medium level stain on the west. And in the east axis, where there is the higher level blot, there is a medium level blot at its side, a pattern: side by side with higher level areas a medium level area generally emerges. Preteceille and Cardoso (2008), when reviewing the space distribution of social groups in São Paulo by using French social categories, conclude in a similar manner: in São Paulo the continuity of the higher level group is even more stressed than in Rio. The higher level types are organized in concentric circles starting from the central area of São Paulo and extending throughout all directions before starting to mix with other groups, especially medium level groups. At the center of this social space virtually all the higher level type is confined to the limits of the municipality of São Paulo...) (Preteceille and Cardoso, p 628) There is a decrease in the employed population living in areas of the higher level and medium level types from 1991 to 2000: therefore, in 1991 that proportion represented 57.48% of the employed population, and in 2000 it decreases to 52.09%. Such reduction is mainly a result of the decrease in the population of higher level areas, considering that the averages showed the same relative increase. 14

15 Those maps show large manual workers stains in areas traditionally connected to industrial activities, such as the southeast axis (Diadema, Santo André, Mauá, São Bernardo, Rio Grande da Serra, Ribeirão Pires), east axis (Suzano, Mogi das Cruzes, Itaquaquecetuba, Guarulhos), and north axis (Cajamar, Franco da Rocha, Caieiras). The largest part is occupied by areas of the popular and agricultural types. The general impression of the socio-spatial typology of São Paulo also corresponds to a doughnut pattern, where the higher level types are still located in more central areas, surrounded by hierarchically lower level types: first medium level, then industrial workers, and finally popular and agricultural workers. The medium level type areas in 2000 (251, with 2.25 million residents) were the most populated areas, although not having the largest territorial dimension. In 1991, the type with the largest number of areas was the industrial workers type, with 283 areas. The areas of the intra-metropolitan space in 2000 internally present a reasonable degree of social heterogeneity. The higher level type areas have 31% of higher level categories (employers+ managers + professionals), 37% of medium level occupations and 32% of manual workers (industrial workers, service providers, commerce workers). One sees the heterogeneity inside them, although in 1991 such heterogeneity was even slightly greater (24% of higher level categories, 38% of medium level occupations and 38% of manual workers). It may also be noted that the higher areas concentrate 54% of the higher level categories in 2000, whereas in 1991 such concentration represented 61%. As a matter of fact, among the higher level categories, only major employers are more concentrated in the higher level areas in 2000: 71% of major employers that still reside in the metropolis live in the higher level areas. At the other end, the areas of the popular type in 2000 present 4% of higher level categories, 20% of medium level occupations, and 75% of manual workers. In 1991, those areas had a greater weight of medium level categories (22%) and a lower weight of manual workers (72.5%), being, therefore, a little more heterogeneous. Table 3 shows a significant percent increase of concentration in all socio-occupational categories (with the exception of major employers) in popular areas. Likewise, in 2000 the medium level areas had 11% of their employed residents in the higher level categories, 33% in medium level occupations and 55% employed as manual workers. The profile of the medium level areas in 1991 was also more varied, with 8% of higher level categories, 37% of medium level occupations and 54% of manual workers. A great percentage difference is observed in higher level categories between the areas of the higher and medium level type, a difference that became more significant in the year In the medium level areas, the concentration of higher level occupations increased significantly, with the exception of small and major employers, and also of domestic workers. In 2000, the industrial workers areas have a smaller proportion of medium level occupations than in 1991 and a greater proportion of manual workers. Therefore, they concentrate more manual workers than in 1991 and have less heterogeneity in the sociooccupational structure. Finally, the agricultural areas of 2000 have 12% of agricultural workers, a greater proportion than 9% in However, they also have an internal social diversity, 15

16 although slightly lower than in They concentrate more agricultural workers in Therefore, socio-occupational heterogeneity was greater in 1991 than in 2000, although such heterogeneity persisted at that date. The segmentation trends between 1991 and 2000 may be assessed based on Table 4. The main metropolitan labor market trends, seen above, were found in almost all sociospatial types. The medium level profile of each large typology had a gain of higher level professionals and specialized service providers, peddlers and improvised workers, and loss of office occupations, major employers and workers in the traditional and modern industry. However, the intensity of gains and losses was different in the diverse areas. Meaning of the following expressions: Elitization increase in higher level categories, combined with a considerable loss of manual workers. Professionalization increase in higher level professionals Tertiarization increase in tertiary sector workers De-proletarization relative loss of secondary sector workers; by analogy, proletarization is when there is a relative gain of those workers Popularization relative gain of non-specialized tertiary sector workers The higher level areas had a relative elitization process. The term elitization refers to the combination of a high relative increase of higher level categories with a considerable loss of manual workers. It is relative because there was a loss at the top of the pyramid, that is, a loss of major private-sector employers and managers. In the medium areas, the greatest concentration occurs among medium level occupations, but is less intense than in They also show a significant increase of higher level professionals. They have a clear tertiarization, with a strong increase in specialized service providers. The decline in secondary sector workers is caused by a reduction in traditional and modern industry workers, in spite of the relative increase in workers of auxiliary services. And in those areas there is also a relative increase of nonspecialized tertiary sector workers, caused mainly by the relative increase in domestic workers and peddlers. In sum, the medium level spaces present: tertiarization, professionalization de-proletarization and popularization. Those are spaces that are gathering a number of trends and require a more detailed study of their distinctive typologies. Industrial workers spaces show a certain professionalization, in addition to the gain among workers of the tertiary sector. However, such features are not those distinguishing those areas from the other areas. They also show a change in the profile of secondary sector workers, with a large loss of workers in the traditional and modern industries, and gain among auxiliary services workers. There is also a strong popularization, given the relative increase among non-specialized tertiary sector workers. Those are spaces that became popularized. Popular spaces, as is the case of all other spaces, also show a loss of managers and gain of higher-level professionals, but the concentration of those categories in those spaces is 16

17 small. There is a relative increase in popular spaces of some medium level occupations, such as occupations connected to healthcare and education, public security, justice and postal areas, as well as a relative increase in tertiary sector workers. On the other hand, there is a relative loss of workers in the modern and traditional industries, and a gain among peddlers and improvised workers. Those are spaces that are also mixed, with a little entry of some medium level occupations, strong tertiarization, de-proletarization and relative popularization. Agricultural spaces show signs of proletarization, with the relative increase of auxiliary services workers, in addition to popularization, with the relative increase of peddlers and improvised workers. And, curiously, they also show a relative increase in agricultural workers. The number of agricultural workers increased from 10,322 to 14,736 in the period. A possibility is that those spaces include false agricultural workers that are dedicated to the care of weekend leisure areas (chácaras), such as housekeepers and defined themselves as workers connected to agriculture. Those are areas that became proletarized and popularized. Therefore, the socio-occupational transformations of the decade loss of managers, strong tertiarization, professionalization, relative decrease of medium level occupations with strong loss of office occupations, relative de-proletarization, with loss of workers in the modern and traditional industries and gain of auxiliary service workers, popularization, with a relative gain of workers in the non-specialized tertiary sector workers occupied the space in a differentiated manner: In the higher-level spaces, the loss of the elites is restricted to the loss of major employers. There is professionalization, that is, a gain of higher-level professionals in a percent well above the other areas. It is the only area with a significant loss of manual urban workers. By comparing the profiles in 1991 and 2000, at the end of the century, the higher level areas as a whole are less mixed, but with a profile more connected to higher-level professionals, rather than to managers. The preferred housing locus of professionals was in that typology. Major employers are strongly concentrated in the higher level spaces: in 2000, 71% of total major employers resided in those spaces (Table 5). In higher level spaces there is an average income increase from minimum wages in 1991 to in 2000, that is, a 55% increase. The income increase is associated to a 17% increase in schooling years, leading the average schooling years from 9.86 to in the decade. The greatest income increase was observed in the higher level space. The higher level spaces are those presenting the highest level of verticalization with 48% of apartment housing units. In the medium level spaces, the heterogeneous profile of 1991 remained stable in 2000, but with some changes. In those spaces, the loss of office occupations is significant, reflecting, at the end of the decade, a smaller proportion of medium level occupations. The increase of tertiary sector workers is significant in the medium level spaces, which does not occur in higher level spaces. Higher level professionals occupy medium level spaces that also present a relative increase of popular categories. The concentration of higher level categories in medium level spaces increased from 1991 to The medium level spaces mixed more. Medium level income in medium level spaces also increased, going from 5.01 minimum wages in 1991 to 6.65 in 2000, a slightly lower increase than the medium income of higher spaces (33%). Schooling levels also increased, from 7.5 schooling years on average to 9.03 (20%). The income of higher 17

18 level spaces in 1991 was twice as that of medium level spaces; already in 2000 that relation was 2.4 times bigger. The medium level spaces presented a verticalization rate of 17% in 2000, with strong increase in relation to Industrial workers spaces had the most significant profile change of the decade. They gained higher level categories, although their proportion in that type of areas is small. They are spaces that became strongly tertiarized and popularized. Their profile followed the loss of industrial workers in the modern and traditional industries, and in 2000 they kept the industrial workers profile through the relative increase of auxiliary service workers. The relative increase of non-specialized tertiary sector workers between the socio-occupational categories in the decade makes those spaces be marked by the presence of this type of worker, with popularization. The average income of industrial workers spaces was 4.02 minimum wages in 1991 rising to 4.16 minimum wages in It was the spatial segment with the lowest income increase, only 3.5%. The fact is probably related to the change in the profile of industrial workers spaces, with a greater proportion of tertiary sector workers and non-specialized tertiary sector workers. The income of industrial workers spaces in 1991 represented 40% of the income of higher level spaces, and in 2000, that proportion falls to 26%. Schooling level increased more than income levels, showing a 20% increase (6.03 schooling years in 1991 to 7.25 in 2000). On the other hand, verticalization, although well below that of medium and higher level spaces, more than doubled in the decade. To a certain extent, popular spaces became even more popularized in 2000, as compared with They gained non-specialized tertiary sector workers, especially domestic workers and peddlers, and became tertiarized, with a strong increase in specialized service providers. Also significant is de-proletarization of those areas. The concentration of both higher and medium level categories and of manual urban workers increases in those spaces in the decade: if, in 1991, 15% of manual urban workers lived in that type of areas, in 2000 that proportion increases to 23%. The income of popular spaces increased more than that of the industrial workers spaces, with a 13% increase (going from 3.65 minimum wages in 1991 to 4.13 in 2000). That medium income in 2000 was also 26% of the income of higher level spaces, whereas in 1991 it represented 36%. One should note that the income differential between the higher level spaces and those of industrial workers and popular spaces increased. The 28% increase in the schooling level exceeds that of income. Verticalization of popular spaces that was a little bigger than that of industrial workers spaces in 1991, in 2000 is lower than in the previous years. There was an increase in the proportion of apartments, but lower than in the industrial workers spaces. Agricultural spaces, on the one hand, show a greater relative proportion of agricultural workers, and concentrate them more intensely: they are the residential location of 41% of agricultural workers in On the other hand, they attest the expansion to the peripheral areas of proletarization (represented by the relative increase of auxiliary service workers) and popularization, represented by the relative increase of domestic workers and peddlers. The only space where there was loss of medium level income was the agricultural space, with a loss of 24% in the decade (the medium level income went from 5.34 in 1991 to 4.07 in 2000). The schooling level, however, grew from 5.62 to 6.68 schooling years. Even in the agricultural areas, there was a minimum verticalization. 18