Gentrification processes : how many? how resistible? The case of metropolitan Paris ( )

Transcription

1 ID # : 2736 Submission within WS 26 : Urban housing in the time of crisis Gentrification processes : how many? how resistible? The case of metropolitan Paris ( ) Catherine Rhein, directeur de recherche CNRS U.M.R. n 8504 GéographieCités, C.N.R.S. and Universités Paris 1 et Paris 7 Abstract In metropolitan Paris, the housing market has been deeply transformed over the last 25 years. Upper middle classes have been literally invading working class districts within the city of Paris. However gentrification processes differ according to housing provision and to a diversifying housing demand. Indeed upper middle classes have been growing, as a result of ongoing economic changes. New social strata are emerging, linked to the rise of knowledge economy, but along far more complex, diversified lines than those developed by R. Florida for U.S. metropolis. In metropolitan Paris, creativity is definitively not the key concept defining those strata and integrating them as a social class. On the contrary, labor markets requirements, income levels and life styles induce differing housing demands within upper middle strata themselves. Consequently, housing markets have been evolving in unexpected ways over the last 25 years. Households structures have diversified mostly along two lines: households shrinking size, linked to ageing processes, and growth of bi active couples. Those two processes, which have reached suburban and even peri urban parts of metropolitan Paris, are important, although hidden, factors of social polarization, as data drawn from the 1982 and 2007 census files help demonstrate. Another major change found is the housing supply plasticity, even in inner Paris, at a far greater degree than expected in a context of ever increasing housing prices and of strong architectural and urban regulation. It was found that housing supply had been growing within Paris, although there was no visible change in the urban landscape: this process was induced by conversion of vacant lots and buildings into dwellings and studios The third part of the paper is dealing with the place left to the poor, to the migrant and to ongoing polarization processes in formerly working class districts. Pockets of old, ruined down dwellings are still existing in north eastern parts of the city of Paris, while population has been aging in social housing, due to building cycles effects in renovated districts. Are ghettoization processes at work within Paris as well as outside? Key words: gentrification, housing demand, housing provision, social housing, Paris, metropolisation 1

2 Introduction Over the last 40 years, in urban research, gentrification has been identified as a major process in metropolitan changes. Every large city in the world has been analyzed along such lines. Ways of life, consumption habits have been closely scrutinized, mostly among upper middle classes: innovative, creative habits would reflect gentrification at work in any king of inner cities districts. However the other side of the mirror, that is housing provision and its evolutions over time, has not been as closely examined as were changes in housing demand: such critics were raised by N. Smith (Smith, 1996). In this paper, we would contribute to Marcuse s argument, by providing some clues in one of the less investigated global cities, under gentrification s lights, metropolitan Paris. This focus upon housing provision raises both theoretical and methodological issues, since housing provision and its production are strongly embedded within national technical, socioeconomic and cultural histories, reflecting national specificities, including former social structures. The case of Haussmannian buildings in central Paris makes an interesting case, as explained by F. Loyer (2005) and by P. Pinon (2002). Apartments design can be understood as material objectification of bourgeois ways of life. At that time, indeed the middle of the XIXth century, the industrial bourgeoisie was in the making. Large flats, including large reception rooms, as well as the dominant household model, that is the nuclear family : a couple and children. Housewives didn t work at that time, and were helped by maids, poorly housed at the Six th or seventh floor, in special rooms. Since then, upper classes evolved a lot, in Paris, in France, more generally elsewhere in large metropolis. The strong relationships between housing conception and production and life styles are far more complex today. Let us present some of the most important changes. 1. Upper classes and their housing needs and demands What can be identified, today, as upper and middle classes? This point has been already examined under different theoretical lights (Butler and Robson, 2003). In fact, the bourgeoisie would include entrepreneurs, professionals working in private firms, banks, insurance companies, «professions libérales» (physicians, lawyers, engineers, architects, other experts working as nonsalaried workers). But the conception of bourgeoisie, as a social class, is still close to its meaning in the second half of the XIXth century, at times when salaried jobs would account for less than 10 % of overall jobs (Katznelson, 1982). Since then, most members of the so called bourgeois class would be salaried : in neo marxist terms, such changes opened large theoretical debates, as most members aren t no longer directly holders of the means of production, but couldn t be considered as members of the working class either. Positions held within job structures still remain major determinants for class affiliation, along with skill levels, diploma and revenues. In France, nature and volume of 2

3 patrimonies (portfolios, assets, houses, gold) would help identify the well off strata within the overall category called Cadres et Professions Intellectuelles supérieurs (CPIS, further on) as defined by the Institut National de la Statistique et des Etudes Economiques (INSEE, see Desrosières and Thévenot, 2005). Globalisation processes had contrasted outcomes upon job structures in France, as in many other European countries. On the one hand, it induced a strong decline of industry, therefore of industrial jobs, mostly those jobs directly linked to production, while, in a growing proportion, immigrant workers entered the labor market in unskilled services and remaining industrial unskilled jobs (Rhein, 2007). Simultaneously, conception, design, in other words research/development have been developing at a fast rate, providing jobs for engineers, highly skilled technicians, and inducing jobs in higher education and public research. This latter process of change can be presented as the rise of a knowledge economy, over the last fifty years : the diffusion of computers, of hard and software, induced changes in firms management, even in their very conceptions. Another dimension of knowledge economy would be the rise of cultural industries, including arts, information and tourism. Among key job characteristics, the type of job contract should be stressed, whether on a permanent, at least stable basis or not. This characteristic has a major role in housing choice, on the housing markets more generally. In France, in metropolitan Paris, in particular, the kind of job contract is a major variable differentiating between: jobs based upon the safest contracts, that is civil servants, including teachers from kindergarten to University and Grandes Ecoles, in State institutions and agencies, at different scales, and in hospitals and health services, contracts with no time limits (contrats à durée indéterminée : CDI) : mostly skilled workers, with seniority, contracts on a short time limit (contrats à durée determinée : CDD) : most likely to be held by foreign, immigrant unskilled workers, in services. jobs with specific contracts, such as interim, Intermittents du Spectacle (artists employed on irregular bases can get regular revenues through a specific public agency). Households, not workers as such, are the relevant unit of analysis for housing demand, as they are the significant socioeconomic unit for revenues formation (Aglietta, 1978 ; Rhein, 1998a and 1998b). In Metropolitan Paris, such changes induced changes in social strata (table n 1). Manual workers are the category whose decline has been the most impressive : from 21 % of total heads of households in 1982 to 13 %, at the regional level, but from 13 % to 6 %. within central Paris. On the reverse, households whose heads are members of the upper middle classes (Cadres et Professions Intellectuelles Supérieures) have been literally invading central Paris : but the increase was much higher for those working in the private sector than for those working in the public sectors. 3

4 Demographic changes should not be forgotten: they account for many effects upon housing demand. At this point, retired households account for 25 % of the housing stock, in France, 22 % in Paris, 23 % in metropolitan Paris. People are living much older, inducing a decrease in residential mobility rates: this is particularly clear in the Beaux Quartiers, namely the neo Haussmannian XVIe arrondissement where the ratio of the 75yrs old or more on the 0 18 yrs is the highest. Moreover, households life cycles are more complex and longer than they used to be four decades years ago: households began their life when (male economically active) heads, about years old; entered the housing market after getting married to an non economically active spouse, then a nuclear family would grow, then decrease then the kids left home: the couple s life expectancy was 30 years, household s life expectancy around 40 years. Today, couples life expectancies are shorter, while household s life expectancies may reached 70, even 80 years. This results in the rise of women s headed households, and more generally, of the increase of oneperson households, with more sequences in households life cycles due to separation and/or divorce. Retired people are former active people. Even if part of them leave metropolitan Paris at retirement times, most do not move when they retire: so as life expectancy has been increasing over the last decades, households are staying longer in dwellings: they can even spent the household s life cycle as active and the other half as retired households : about 30 years long each, for a total of 60 years. As demonstrated in table n 2, the majority of retired households are one person households, even if the proportion of couples has been increasing over years. Small households are, only partly, retired households ; it is also the dominant household s type in two social strata, namely semi skilled and unskilled employees, as well and intermediate professions, where more than half of the jobs are held by women. In these two strata, one parent families are more frequent than in any other social strata, even if they do not account for more than 15 % of the total households. Over the last four decades, the overall job structures changed a lot: the most prominent and general feature is the increase of skilled jobs, in any social strata, although this doesn t imply that creativity would be the unifying characteristic of these new middle class, contrary to Florida s thesis. Indeed, these new middle classes are internally stratified along lines drawn from type of work contract. These fractures are nicely objectified by patterns of residential location, as shown in fig.2.. How so different locational pattern could be explained? 2. Gentrification and its varying forms in metropolitan Paris 4

5 Let us first provide some clues about housing provision, as its differential uses by segments of the middle classes help understand, in turn, types of occupancy and of position in the housing market. Most gentrification studies focused upon gentrifiers as households, as social strata, much fewer upon gentrified dwellings and districts. However gentrifying or gentrified dwellings and districts are quite different between metropolis, and even within metropolis such as New York, London or Paris. In metropolitan Paris, various urban fabrics can be identified, as four cycles of urbanization and industrialization can be identified. The first cycle reached the city of Paris, within the Mur des Fermiers Généraux,built in 1784, mainly in the eastern parts ( Xe and XIe arrondissements), along the Canal Saint Martin. The cours industrielles are typical industrial settings built over this period ; they would include workers dwellings, small workshops (upholstery, metal work). Buildings are built with cheap material such as bricks and mortar, no basement, but provide today nice settings with open spaces. Over the second cycle (second half of the XIXth century), functional zoning would mostly concern non yet urbanized areas, at the city s outskirts. Zoning was then an informal process, developed mostly around railways. Large industrial infrastructures induced a clear cut spatial specialization between work/production and residence : La Plaine Saint Denis is the most emblematic industrial zone, and the largest : it was full at the end of the XIXth century, with many companies and redeveloped over the last thirty years in a complex of cultural industries. During the third cycle, from the end of the XIXth century up to the 1960s, industry was revolving around car production. However there is nothing such as the fordist city, at least in metropolitan Paris. Functional differentiation between production, conception and other types of firm functions (financial, managerial) was the rule. Production sites developed in large industrial sites, first within the city (Billancourt for Renault, Javel for Citroën, Tolbiac for SNECMA and Panhard), and since the 1950s in remote suburbs and small towns located along the Seine river (Les Mureaux, Mantes, Flins) while high tech industries and research centers were located mostly in suburban parts of metropolitan Paris, even in New Towns such as Renault s TechnoCenter, located in Guyancourt. The fourth cycle is mostly characterized by deindustrialization. This process reached most close working class suburbs (Red Belt) as well as formerly industrialised western suburbs (car and aeronautics). However, redevelopment was much quicker in the latter than in the former, due to opposite urban policies developed by local powers namely, municipalities. (Rhein, 2005). Overall embourgeoisement processes developed as middle classes were substituted to formerly working class households. In Paris, some segments of the working classes were highly skilled, moving back and forth between salaried jobs and craftsmanship. Housing conditions were rather diverse. In many parts of Paris, in particular in western arrondissements reached by the third industrialization cycle, urban redevelopment schemes, including renovation programs, were the dominant public policies over from the 1970s to the

6 Simultaneously, many smaller private programs took place on formerly industrial settings. To what extent are such programs parts of gentrification processes? This is a question raised by P. Marcuse also in the case of New York and Boston and which is raised for Paris, as such processes do not imply reinvestment, by middle classes, of formerly working classes dwellings, since there were no such dwellings preexisting to gentrification. In eastern Paris, the most working class districts underwent radical urban renewal schemes called renovation urbaine : entire districts were demolished, as they were diagnosed as îlots insalubres (unhealthy blocks), as demonstrated by L. Murard and P. Zylberberg (1996). Most urban renewal schemes did actually include large social housing projects: specially in the eastern, working class arrondissements such as the XIIIe, the XIXth and the XXth arrondissements. Today, in some settings,existing social housing projects prevent gentrification processes to develop at the very local level. Three major forms of gentrification have been identified. The first should hardly be named gentrification, rather embourgeoisement, in formerly industrial settings rebuilt and remodeled. The second form would directly affect housing provision : it is mainly located in the northeastern part of Paris, where few dwellings have been built over the last two decades, out of large renovation zones such as Orgues de Flandre, Cité de la Musique, La Villette : however, as shown on figure n 3, dwelling units have been created by remodeling former industrial settings. From an architectural viewpoint, such newly created if not built dwellings have few characteristics comparable to New York lofts, created in cast iron buildings, while their Parisian counterparts are of cheap material brick and mortar, not as strong as New York lofts (Clerval, 2004). Whatever their quality, newly created dwellings in old buildings are also large dwellings : this explains the reason why housing provision has changed over time in these districts, without any large and visible building operations. The third form of gentrification has been developing in some districts in north eastern Paris. It would correspond to new pioneer fronts, right in the middle of decaying neighborhoods, where migrant workers in the district of Barbès, for those coming from Algeria and Marocco, and in the district of Château Rouge for those coming from sub Sahelian Africa. Around those ethnic enclaves, some pockets of small dwellings are housing young members of the new middle classes, specially students, artists of any kind and journalists, accounting for more than 12 % of the total households (fig. n 2). Most are young, living alone, as intermittents du spectacle (short term work contracts), renting small dwellings : some may be highly skilled in arts and their way of life and work implies accessibility to public transportations, stations, in brief centrality. This is a functional requirement that they manage to find within Paris, despite skyrocketing housing prices. 3. The Red Belt under stress : the other side of gentrification Where are housed poor households today in metropolitan Paris? This is indeed one of the most important issues in French urban policy, one of the most puzzling ones. Most issues are deeply 6

7 intertwined with political issues, as the majority of working class households have been living either in eastern Paris, or in close suburbs, mostly in the Red Belt. The Red Belt evolved over time, since the birth of the French Communist Party in 1920 : it really appeared as a challenging force in the 1950s, right after the Second World War, when the Red Belt was a set of more than 40 communes, most of them strongly industrialised, where working class households were living either in social housing projects or in one family dwellings, and which have been managed by Communist municipalities for decades. Today, the Red Belt is still existing, although it changed a lot : the Communist Party lost influence over time, and lost half of these communes, but is still holding the majority in 38 municipal councils (Rhein, 2005). The other major change has been the rise of migrant workers and of migrant families living in eastern Paris and in the Red Belt. In every commune, immigration has had different origins and paces. Two communes Nanterre and Saint Denis played a prominent role, as many Algerian workers were already living in camps, between the two world wars, as Musulmans d Algérie, since Algeria was a French colony up to 1962, but Algerians were not fully French citizens. Most immigrants were male workers, going back and forth, up to the mid 1970s : this migration regime was called the noria. French migratory policy changed in 1977 : the noria system almost disappeared, as public policy encouraged families, rather than bachelors, to settle in France. Since the 1970s, many changes occurred in migratory origins: up to the 1980s, Southern Europe (Italy, Spain and Portugal) was sending large flows of workers and their families : most workers would be employed in construction and maintenance. The other major flows were stemming from Tunisia, Marocco and Algeria, three former French colonies or protectorats. At the end of the 1980s, flows from Southern Europe decreased, but new flows from Turquia, Sub sahelian Africa (Mali, Tchad, Senegal, Guinée, Burkina Faso), and from Asia. Immigrant generations have been slowly integrated, first in the housing market, getting access to social housing projects, but in the most remote suburbs (Rhein, 1998), then in closer suburbs ; and third, even fourth generations have been slowly reaching skilled jobs and integrating middle classes. Such processes are much slower for African families, trapped in decaying dwellings in the western part of the Seine Saint Denis where public services, and in the first place, the educational system has been put under stress over the last four decades (Rhein, 2007). At the end of the XIXth century, social housing s financing was secured by the Loi Bonnevay, although this law didn t produce much effects until the 1950s. Right after the Second world war, housing shortage was severe ; economic growth opened up opportunities for the State to provide financial help both to potential owners, as well s to municipal agencies in charge of building social housing projects. Urban forms which emerged at that time were mostly a mix of towers and bars. Such forms were somehow the exclusive architectural solution at that time, for public as well as private developments (Depaule, ). But most social housing projects were of poorer architectural quality and have been decaying since the end of the 1970s : some experts were aware of the physical, as well as of socioeconomic forthcoming crisis and launched a first set of measures, focused upon physical renovation and social life within the eldest projects, under the name of Habitat et Vie Sociale (HVS). Conclusions Gentrification research focused upon housing demand, more than upon housing provision. In other words, it did provide only one part of the story, while the entire housing market has been evolving along various lines over time, in Paris as well as in many other metropolis. On the one hand, focusing about the city of Paris limits the scope of analysis : indeed strong polarization processes 7

8 have been at work, resulting in concentrations of poor migrant households in former working class suburbs known as the Red Belt. On the other hand, it was also found that the Beaux Quartiers underwent strong functional changes, resulting in loss of dwellings : those districts were bourgeois districts, for expanding upper classes. Looking both at social strata and at their dwelling conditions helps decipher emerging faults within those upper and middle classes, among which types of work contract are playing a key role, for defining the individuals s place on the housing market. As far as public intervention is under scrutiny, it appears that its impacts have to be deciphered in the long term. At a very local level, public intervention is always impressive. At a higher scale, it reveals somehow more complex, inducing aggregate, when not unexpected, effects. 8

9 References Atkinson R., Bridge G., eds (2005) Gentrification in a global context, London, Routledge. Butler T., Robson G. (2003) London calling: the middle classes and the remaking of Inner London, London, Berg. Clerval, A. (2004) La cour de Bretagne, un cas de gentrification dans un quartier populaire, Paris APUR. Clerval, A. (2007) Evolution de la géographie sociale de Paris, : embourgeoisement et gentrification. Paris, APUR Desrosières A., Thévenot L. (2005) Les catégories socioprofessionelles, Paris, La Découverte. Katznelson, I. (1982) City trenches, urban politics and the patterning of class in the United States, Chicago, University of Chicago Press Loyer, François Paris, XIXe siècle, l'immeuble et la rue. Paris: Fernand Hazan. Murard N., Zylberman P. (1996) L hygiène dans la République ou l utopie contrariée ( ), Paris, Fayard. Pinon, P. (2002) Atlas du Paris Haussmannien, Paris, Parigramme. Rhein C. (2007) «Changements sociaux et transformations de l espace», chap. 6, p in T. Saint Julien et R. Le Goix, dir. La métropole parisienne, centralités, inégalités, proximités, Paris, Belin Rhein C., with E. Markou (2005) Tissu industriel, planification spatiale des activités économiques et rapports sociopolitiques dans la métropole parisienne ( ), rapport de recherche au PUCA Ministère de l Equipement, Paris, 220 p. Rhein C. (1998a) «The working class, minorities and housing in Paris, the rise of fragmentations», GeoJournal, n 46, Rhein C. (1998c) «Globalization, social change and minorities in metropolitan Paris : the emergence of new class patterns», Urban Studies, vol.35, n 3, Smith, Neil (1996) The new Urban Frontier : gentrification and the revanchist city, Routledge, London. 9

10 Annexes Graph n 1 : Households structures according to head s SES Graph n 2 : Couples according to head s SESS : spouse s SES 10

11 Table n 1 : Heads SocioEconomic Statuses, in 1982 and 2007 Heads SES Ile de France Paris Petite Couronne Grande couronne % Non salaried 5 % 4 % 5 % 4 % 5 % 4 % 6 % 4 % Prof. Managerial 15 % 22 % 18 % 31 % 12 % 19 % 15 % 18 % Intermediate 16 % 17 % 13 % 16 % 16 % 17 % 19 % 19 % Low skilled non 16 % 14 % 17 % 12 % 17 % 16 % 14 % 14 % manual Manual workers 21 % 13 % 13 % 6 % 24 % 14 % 24 % 15 % Retired 20 % 25 % 22 % 22 % 20 % 25 % 17 % 26 % Other non active 7 % 6 % 11 % 9 % 6 % 6 % 5 % 4 % Total 100% 100% 100% 100% 100% 100% 100% 100% Sources : INSEE, 1982 and 2007 censuses tables (MEN1 and FAM) Table n 2: Households composition according to SocioEconomic Status of the heads, Paris and Petite Couronne (2,94 millions households) Head s SES Non Prof. Intermediate Low skilled Manual Retired Total Households salaried Managers non manual workers No couple 32 % 48 % 58 % 70 % 34 % 58 % 55 % Couples 68 % 52 % 42 % 30 % 66 % 42 % 45 % 1 active 17 % 9 % 6 % 5 % 16 % 8 % 15 % 2 actives 51 % 43 % 35 % 24 % 50 % 30 % Same SES 6 % 23 % 13 % 14 % 7 % 34 % Sources : INSEE, original census file, 2007 census, tabulations by the author. Table n 3: major characteristics of the housing stocks in 2007 Size of Types of Building age dwellings housing % 1 or 2 rooms % Flats Before After 1975 Paris 48 % 99 % 61 % 21 % 18 % Petite Couronne 31 % 80 % 25 % 42 % 32 % Grande Couronne 16 % 49 % 18 % 37 % 45 % Ile de France 30 % 72 % 31 % 35 % 33 % Sources : INSEE, census 2007, on ligne table LOG ( Table n 4 : Housing stocks according to status of occupancy Owner Unfurnished Social Free Total Nb of occupied rented housing dwellings Paris 33 % 38 % 16 % 6 % 100 % 1,14 Petite Couronne 43 % 24 % 28 % 3 % 100 % 1,80 Grande Couronne 60 % 16 % 20 % 2 % 100 % 1,92 Ile de France 47 % 24 % 22 % 2 % 100 % 4,86 Sources : INSEE, original census file, 2007 census, tabulations by the author. Nb of dwellings in thousands. 11

12 Figure n 1: proportion of householdss whose heads professional/managerial strata in 1999 and 2007 (active and retired) belong to Sources : 1999 and 2007 Censuses, INSEE, tables PRINC4 and FAM. Figure n 2: the internal diversity of the upper middle classes within metropolitan Paris 12

13 Figure n 3: the hidden dynamics of the housing stock within Paris over the period. 13