Navigation – Plan du site

AccueilnumerosHors-sérieMigrations like any other? Intern...

Migrations like any other? Internal migrations of immigrants in the French countryside (2011-2015)

Des migrations comme les autres ? Les migrations internes des immigrés dans les campagnes françaises (2011 - 2015)
Julie Fromentin
Traduction de Richard Hillman
Cet article est une traduction de :
Des migrations comme les autres ? Les migrations internes des immigrés dans les campagnes françaises (2011 - 2015) [fr]

Résumés

Si le renouveau démographique des campagnes au cours des dernières décennies a notamment été analysé au prisme des arrivées de migrants internationaux, l'importance des migrations internes des immigrés vers les campagnes françaises a été très peu étudiée. A partir des données du recensement de 2013 réalisé par l'INSEE, cet article propose de porter un autre regard sur l'évolution de la géographie de l'immigration en France, en analysant de manière quantitative les migrations internes des immigrés en France métropolitaine entre 2011 et 2015, en particulier à destination des campagnes françaises. Les traitements successifs mettent en évidence la participation accrue des immigrés aux reprises démographiques des campagnes sur la période récente, en particulier aux processus de périurbanisation. Ils montrent également les nombreuses similarités dyears les profils de migrants entre immigrés et non-immigrés (âge, situation familiale, CSP), mais aussi l'existence de quelques profils atypiques dyears certaines campagnes.

Haut de page

Texte intégral

1The residential localisation of immigrants in metropolitan France is marked by a heavy concentration in the border and industrial regions, in the South-west of France, and especially in the urban centres [Desplanques, Tabard, 1991; Brutel, 2016]. This mainly urban spatial implantation is the result of successive waves of immigration since the 1960s. Strong inertia characterises this distribution, and newcomers generally settle where populations previously arrived are already concentrated. Taken overall, these elements outline a relatively stable geography of immigration, modified only at the margins by new settlements of recent immigrants. Most studies of the spatial distribution of immigrants, however, are conducted on a fine scale, and especially at the intra-urban level of the Parisian agglomeration in the framework of work on socio-spatial segregation [Guillon, 1996; Préteceille, 2009]. In a dynamic perspective, it has notably been shown that despite a strong spatial concentration, immigrants experience the trajectories of residential mobility within the urban space, leading often to the improvement of their housing conditions [Pan Ké Shon, 2009].

2In parallel to the evolution of work on immigration in France, the field of studies of geographic and residential mobility has itself considerably developed since the 1980s. Innovative methods of mobility analysis have made it possible to confirm the role of life-cycle, professional changes and evolving individual aspirations (access to ownership, desire to be close to a city or, on the contrary, to ‘nature’, etc.) in the decision to change place of residence [Bonvalet, Fribourg, 1990; Debrand, Taffin, 2005]. Among these studies, one finds in geography and sociology substantial literature on the ‘renouveau démographique’ (demographic renewal) [Pistre, 2012] of country regions linked to residential migrations to these spaces [Kayser, 1993]. Several studies set out in particular the diversity of profiles and motivations of the populations at the origin of this change: retired persons establishing themselves more permanently in their secondary residence; disadvantaged households relegated to the periphery of large urban agglomerations where the cost of living and housing is too high; young couples seeking to achieve ownership of an individual house, etc.

3The question of the contribution of immigrants to these demographic upturns has been approached, however, only through the study of arrivals of international migrants in the French countryside coming especially from northern Europe. The first studies record arrivals from the end of the 1960s, first in Dordogne, then in the South-west more generally, as well as in Brittany and Normandy, beginning in the 1980s, up to their implantation in Poitou-Charentes and Limousin during the 2000s [Buller, Hoggart, 1994; Cognard, 2011]. These studies are concerned to specify this diversification of localisations, to show the diversity of practices between occasional presence in secondary residences and permanent installation, or, further, to explain the motivations (economic, quality of life, etc.) at the origin of these migrations. By contrast, few studies have dealt with the possible residential moves within France of these immigrants following their arrival, in the countryside or elsewhere.

4In general, the internal migrations of immigrants in France remain very little studied: from the qualitative point of view, monographs have been produced describing the geographic and residential trajectories of specific immigrant groups. Villanova, for example, describes the residential space of the Portuguese in France in connection with their strong desire to become property owners [Villanova, 1997]. The study of the internal migrations of the immigrants, however, is not the principal object of this work. From a more quantitative perspective, Rathelot et Safi [2014] have studied, on the basis of the ‘Echantillon Démographique Permanent (EDP)’ (the Permanent Demographic Sample) the link between the geographic mobility of immigrants and the ethnic composition of French communes. In line with American studies dealing with the spatial assimilation of immigrants [Massey, 1985], this work shows that the mobility of immigrants is closely linked with the presence of persons of the same geographic origin in the commune of destination and qualifies the hypothesis of a ‘white flight’ in France. Again on the basis of the EDP, Solignac [2013, 2016] has enriched our knowledge concerning the internal movements of immigrants in France by showing that the mobility rates of immigrants and those born in France are similar, when the attrition linked to departures from France and the low death rate of immigrants is taken into account. The truly geographic dimension of these studies is limited, however, since geographic localisation is missing and only a variable of the population size of the communes is introduced in the models (more or less than 10,000 inhabitants for Rathelot and Safi, a variable in six modalities for Solignac).

5Apart from these few French studies, it is the Anglo-Saxon literature that today makes the most important contributions on the subject and makes it possible to put in place a framework for reflection for all the countries characterised by long-standing immigration. Descriptive studies have shown the greatest mobility of immigrants, compared with non-immigrants [Belanger, Rogers, 1993], and suggest that changes in localisation take place relatively late in the life-cycle of individuals. In an approach making use of theories of spatial assimilation, numerous studies deal with the causes and constraints linked with the internal migrations of immigrants: human capital, economic factors and social capital constitute the explanatory factors most often brought to bear [Gurak, Kritz, 2000; Reher, Silvestre, 2009]. Most of the studies concerned with the internal migrations of immigrants seek to make the connection between the literature dealing with international migrations and those dealing with internal migrations – subjects which, in the United States as in France, still remain quite widely separate [King, Skeldon, 2010]. Finally, a more geographic perspective is being adopted today by some American researchers, who study the ‘suburbanisation’ of immigrants and show that many immigrants migrate to the peripheries of large cities, indeed to rural spaces. These dynamics contribute to the creation of a new geography of immigration in the United States and also raise new questions concerning the integration and segregation of immigrant populations in these spaces [Lichter, Johnson, 2006; Farrell, 2016].

6The present article takes up the line of these last studies and proposes an analysis of the internal migrations of immigrants to or within the French countryside in recent years, based on census data from 2013. It aims first at providing another view of the evolution of the national geography of immigration in France, and at showing that the changes within it are not uniquely linked to arrivals of international migrants. Thus we posit the hypothesis that the recent internal migrations of immigrants participate in a ‘new geography of immigration’ marked by more sizable implantations outside the large cities, and that there exists a differential geography between internal migrations and the international migrations of immigrations in France. In other words, this article raises the question of the participation of immigrants in the contemporary processes of the socio-demographic re-composition of rural areas, which are closely associated with positive migratory impacts, and in particular with the processes of peri-urbanisation and demographic upturns in the isolated countryside.

7This study, then, seeks to interrogate, through the prism of geographic mobility, the pertinence of the category of immigrant in studying residential migrations to the countryside. If the spatial establishment of immigrants is highly specific on their arrival in France, does this also apply to the subsequent stages of their geographic progression? Are the socio-demographic characteristics of immigrants similar to those of non-immigrants in the context of internal migration to rural areas? We advance the hypothesis that the determining factors of internal migration to rural areas are, for the most part, common to immigrants and non-immigrants. We also hypothesise that there nevertheless exist important differences among the immigrants according to their migratory characteristics (country of birth, length of time since arrival in France, etc.) and the type of countryside of their destination.

8A first section presents the data used and the definitions chosen in the framework of the article. A second section deals with the internal migrations of immigrants in metropolitan France between 2011 and 2015. A final section is concerned with the socio-demographic determinants of internal migrations, especially those whose destination is the French countryside.

I. Data and definitions

Geographic mobility and internal migrations

  • 1 This work has benefitted from French government assistance managed by the Agence nationale de la Re (...)
  • 2 Certain biographical surveys precisely record the changes of residence and/or commune of individual (...)

9This study is founded on the use of the data of the 2013 population census.1 It assembles the annual census surveys from 2011 to 2015, and constitutes a primary source for analysing the geographic and residential movements of immigrants. Residential mobility has, in fact, been the object of numerous studies in France, notably quantitative, for some decades. But the sources of available data concerning geographic and residential mobility of populations nevertheless remain limited, all the more so when one is concerned with a statistical sub-population, such as immigrants, and sparsely populated spaces, such as the rural areas.2 The revised version of the census, implemented beginning in 2004, has especially aimed at recording residential mobility more precisely, with a time lag of only one year [Pan Ké Shon, 2007]. Recording the place of residence one year prior to the survey has only been in effect, however, since the census of 2011, the question concerning housing and the commune of residence on 1 January of the previous year having been asked for the first time during the annual census survey of 2009. An internal migrant is therefore considered here to be any person who has changed commune of residence in the year preceding the census, and an international migrant to be any person resident in France at the moment of the census, but not a year earlier. This definition, then, does not take into account the fact that individuals could have undertaken one or more additional migrations besides that declared during the year under consideration. Since the census is based on the resulting movements, outward migrations of persons leaving France are also not collected. A final limitation is that recent studies show an underestimation of internal mobility among immigrants linked to incomplete declaration of previous places of residence in France [Solignac, 2013]. Despite all these limitations, the population census – and in particular its principal exploitation – remains the source enabling access to the most closely detailed geographical analysis, and as such constitutes an invaluable source for geographers. It should be noted that the estimations of the number of migrations may vary slightly according to whether one refers to the principal exploitation or to the complementary exploitation (‘quarter sample’) of the census. Finally, unless otherwise stated, all the results concern the population resident in metropolitan Frence in 2013, and the internal migrations concern movements within the metropolitan perimeter.

Definition of immigrants

  • 3 In the population census, two questions permit the identification of immigrants: one concerning nat (...)

10The term ‘immigrant’ became a category of statistics and analysis only at the beginning of the 1990s, designating every person born as a foreigner outside France and residing in France.3 It has progressively taken the place of the term ‘foreigner’, considered inappropriate for the demographic additions over the long term of immigration in France. This category has nevertheless had the effect of fixing the populations it designates in a double immobility: a geographic immobility, which the term connotes, and an immobility linked to the variables used to construct the category, with being an immigrant corresponding to a category of classification from which one cannot extract oneself – contrary to that of ‘foreigner’, for example. The immigrant internal migrants thus designate in this article all those persons born outside France who reside in metropolitan France at the moment of the census and who declared another commune of residence in metropolitan France a year earlier. This group is characterised by its youth, a great diversity as to the length of their presence in France, a larger number of foreigners than of those who acquired French nationality, and a very high proportion of immigrants born in an African country or a country of the European Union (document 1).

Document 1. The characteristics of immigrants having effected an internal migration in metropolitan France

 

Number of individuals (weighted)

Percentage (%)

Age

 

18-29 years

56179

23.9

30-44 years

105471

44.9

45-59 years

45615

19.4

60-74 years

18608

7.9

75 years and over

8833

3.8

 

 

Length of time since arrival in France

 

Less than 3 years

16095

6.9

3 to 5 years

29285

12.5

6 to 10 years

39186

16.7

11 to 20 years

43783

18.7

More than 20 years

64225

27.4

Not specified

42131

18.0

 

 

Nationality

 

French by acquisition

86907

37.0

Foreign

147797

63.0

 

 

Country of birth (grouped by continent)

 

Africa

109446

46.6

America

12952

5.5

Asia

25240

10.8

Europe ouside European Union

18482

7.9

Asia

322

0.1

European Union (27)

68262

29.1

Number of individuals

234705

 

Sources: Population Census INSEE 2013, complementary exploitation
Field: totality of immigrants aged 18 years or more, employed or retired, having effected an internal migration in metropolitan France in the year preceding that of the census (2011-2015)

Definition of rural areas

11To characterise the French rural areas, we rely on the Zonage en Aires Urbaines (ZAU) of 2010, spatial nomenclature proposed by the INSEE based on a functional approach to space by employment centres and measurement of movements between home and work. The various disadvantages of using this breakdown for the study of rural areas have already been described: the ZAU is based on a definition of rural by default; it reinforces the urban-centred vision of space already present in the previous divisions of the INSEE, and tends to produce a homogeneous image of French rural areas by failing to take into account the variety of the socio-demographic characteristics of the rural territories [Pistre, 2012; Depraz, 2013]. Since the objective of this article is to study the place of the rural areas in the aggregate for a given phenomenon (the internal migrations of immigrants), rather than to describe the diversity of those areas in relation to that phenomenon, this last criticism of the ZAU does not seem to us too great an obstacle to its utilisation here. Moreover, we apply the categories of this breakdown with flexibility, considering that the peri-urban ‘rings’ of the large urban centres enter into a broad definition of the French peri-urban countryside.

12The rural areas such as we identify them here designate, then, both the isolated rural spaces situated at a distance from the large urban centres (the ZAU category ‘Isolated communes outside influence of the centres’, here termed ‘isolated countryside’) and the peri-urban spaces with more intermediate population densities (the categories ‘Large centre crowns’ and ‘Multi-centre communes of the large urban regions’, here termed ‘peri-urban countryside’), as well as a more heterogeneous collection of small rural centres of employment and surrounding communes (categories ‘Small centres’, ‘Crowns of mid-sized centres’, ‘Crowns of small centres’ and ‘Other multi-centre communes’, here termed ‘other rural areas’). These contain 39 % of the entire population resident in France and 18 % of the immigrant population. This decision therefore rests on a deliberately extensive vision of the countryside, which does not fit into a resolutely ‘ruralist’ framework of interpretation but conforms to an extra-urban approach to immigration in France.

II. The place of the French countryside in the internal migrations of immigrants: between dynamic of peri-urbanisation and circulations in the isolated rural space

13This first section seeks to understand the specificities of the geography of the internal migrations of immigrants in the French countryside in three ways: first, it compares the distribution of internal migrations to that of immigrants overall; then, it concentrates on the differences in distribution between internal and international migrations of immigrants. It shows, lastly, the existence of a dynamic of peri-urbanisation which is stronger among immigrants than non-immigrants.

1. Internal migrations which reproduce the national distribution of immigrants

Document 2. Evolution of the distribution of immigrants in metropolitan France between 1968 and 2013

 

1968

1975

1982

1990

1999

2008

2013

Urban centres

2,481,845

3,055,761

3,236,075

3,345,690

3,487,481

4,254,963

4,664,165

76.7%

78.6%

80.2%

81.0%

81.0%%

81.3%

81.5%

Peri-urban areas

431,122

485,255

467,219

473,735

499,895

583,882

639,781

13.3%

12.5%

11.6%

11.5%

11.6%

11.2%

11.2%

Isolated rural areas

106,686

106,624

100,552

93,198

101,925

142,902

150,173

3.3%

2.7%

2.5%

2.3%

2.4%

2.7%

2.6%

Other rural areas

215,351

239,750

233,189

219,239

219,226

254,861

265,642

6.7%

6.2%

5.8%

5.3%

5.1%

4.9%

4.6%

Total

3,235,004

3,887,390

4,037,035

4,131,862

4,308,527

5,23,608

5,719,761

Sources: Population Census INSEE 1968-2013, principal and complementary exploitations
Field: Totality of immigrants residing in metropolitan France
Note: calculations performed with a constant geographic breakdown on the basis of the ZAU2010

  • 4 For each of the maps created, the scale chosen is that of the ‘bassin de vie’ (living area), a spat (...)

Document 3. Spatial distribution of immigrants in metropolitan France in 1975 and 20134

Document 3. Spatial distribution of immigrants in metropolitan France in 1975 and 20134

Sources: Population census INSEE 1975 and 2013, principal exploitation
Champ: totality of immigrants resident in metropolitan France

14If the spatial distribution of immigrants between the different types of spaces has changed little overall since the 1960s, some major tendencies can nevertheless be observed (documents 2 and 3). The historic implantation of immigrants in the large urban centres has been reinforced over time, with nearly 82 % of immigrants currently residing in these, as opposed to 77 % in 1968, while the tendency is towards a decline for the non-immigrants over the same period. Further, 18% of immigrants, that is, more than a million persons, resided in 2013 in the peri-urban and rural countryside. The peri-urban rural areas of the large cities have seen the relative proportion of immigrants diminish since the 1960s before stabilising in the 2000s, and today one finds roughly 11 % of immigrants in these rural zones. In the more isolated rural areas, the proportion of immigrants diminished until the 1990s before rising significantly until the end of the 2000s. Today one finds somewhat less than 3 % of immigrants in these spaces. From a geographic point of view, this translates notably into a much larger proportion of immigrants in the countryside of the West of France. Finally, despite the relatively divergent evolutions according to the spaces considered, it is important to note that the number of immigrants has increased everywhere in France over the past 50 years, whatever the type of space considered, even as the number of non-immigrants has diminished in the isolated rural areas. In the French countryside, the number of immigrants has thus risen from around 754,000 in 1968 to 1,060,000 in 2013. This general increase of the immigrant population is linked to the continual arrivals of new population groups, but it also poses the question of the role of internal movements of immigrants in the transformations of the geography of immigration: what is the importance of internal migrations to the French countryside in the recent period?

Document 4. Migrations and immobility of immigrants and non-immigrants between 2011 and 2015

Link to migration

Internal migrations

International migrations

Immobility

Total

Immigrants

333,105

164,152

5,217,477

5,714,791

5.8%

2.9%

91.3%

100%

Non-immigrants

4,038,167

92,544

53,033,557

57,164,268

7.1%

0.2%

92.7%

100%

Field : Total of the population resident in metropolitan France in 2013
Source : Population census 2013, principal exploitation

  • 5 In order to work on a homogeneous field, the number of international migrations has been calculated (...)

15Document 4 permits us to appreciate the important of internal migrations of immigrants in the overall immigrant population of metropolitan France. On observes first that, of 5.7 million immigrants in metropolitan France in 2013, about 164,000 arrived from outside France in the course of the year preceding the census, and more than 333,000 changed their commune of residence within metropolitan France, or 5.8 % of immigrants.5 By comparison, 7.1 % of the 57 million non-immigrants changed their commune of residence within metropolitan France.

  • 6 Significativity of the differences verified according to a 95% confidence interval.[termes/expressi (...)

16Were these migrations distributed in the same way between rural and urban communes? The great majority of these migrations took place to or within the large urban centres (80%), while about 67,500 immigrants migrated to or within the French countryside (20%). The internal migrations of immigrants in France therefore reproduced the pattern of distribution of the immigrant population in France: urban implantation initially, then peri-urban and rural. One observes slight differences, however, between these two distributions in favour of the countryside for internal migrations. Thus 20% of the immigrants who were internal migrants undertook a migration to or within the rural areas of metropolitan France, while only 18% of all immigrants reside in the countryside. This slight over-representation6 of the rural areas in the internal migrations of immigrants stems chiefly from migrations to the peri-urban countryside (13% of the internal immigrant migrants as opposed to 11% of all immigrants), while the tendence would be rather the inverse for the isolated rural areas.

17If the majority of the internal migrations take place to the urban centres, the migrations to or within the French countryside are therefore far from being marginal, and even tend to develop further in the course of the period studied.

2. Internal and international migrations to the countryside

18The scientific literature generally insists on the role of international migrants in the demographic evolutions of rural areas, rather than the contribution of internally migrating immigrants [Pistre, 2012]. Here we wish to verify the following hypothesis: are the new immigrant presences chiefly linked to direct arrivals from outside France, or to internal migrations? Which rural areas are particularly affected by the arrivals of internal migrants?

Document 5. Internal and international migrations according to the type of space of arrival, between 2011 and 2015 (in %)

Document 5. Internal and international migrations according to the type of space of arrival, between 2011 and 2015 (in %)

Field: total number of immigrants residing in metropolitan France at the moment of the census and residing in another commune or outside France in the preceding year.
Source: RP 2013

19Document 5 details the volume of internal migrations in relation to international migrations according to the different kinds of spaces. Thus one observes that internal migrations constitute the essence of immigrant arrivals in the peri-urban countryside (73%), and that these are composed in large part of migrations stemming from urban centres. Specifically, it is mainly migrations from the urban centres, but also from the peri-urban rural areas themselves that supply these migrations. In the more isolated rural areas, international arrivals are more important, and internal migrations represent only 54% of the immigrant migrations to these spaces. The majority of these internal migrations take place this time within the countryside, and especially within isolated rural spaces (35%). Two results, then, should be underlined: first, the place of the peri-urban areas as a space for immigrant relocations in France, in particular those coming from cities. If considerable research since the 1970s has emphasised the residential migrations from cities to the peri-urban spaces for such motives as access to ownership of an individual house [Fruit, 1985], the present study contributes a complement by confirming the importance of this dynamic within the immigrant population. A second result is that the migratory dynamics are different in isolated rural spaces and marked by the predominance of direct arrivals from outside France. This result coincides with recent literature on the subject, which has notably documented the importance of North-European immigrant arrivals in the isolated countryside of the Southwest of France during the 2000s [Pistre, 2012]. But the high number of migrations within the isolated countryside also suggests the existence of more numerous local migrations. In certain isolated rural areas, therefore, international and regional or local movements may be combined.

Document 6. A differential geography of recent immigrant arrivals in the catchment areas of metropolitan France (2011-2015)

Document 6. A differential geography of recent immigrant arrivals in the catchment areas of metropolitan France (2011-2015)

Field: total number of immigrants residing in metropolitan France in 2013 having undertaken a recent migration.
Source: RP INSEE 2013, GEOFLA 2015
Reading: calculation of the index of specialisation for each catchment area i:
(immigrant population in i recently arrived by internal migration / total of the recently arrived immigrant population in i) / (Σi immigrant population recently arrived in i by internal migration / Σi total of the recently arrived immigration population in i)
An index superior to 1 indicates an over-representation of immigrants arrived by internal migration relative to those arrived by international migration; an index inferior to 1 indicates an under-representation.

20Of possible interest, too, is the differentiated localisation of these migration. Document 6 makes it possible to see those spaces in which the proportion of immigrants arrived by internal migration in relation to those arrived from outside France is superior or inferior to the French average. One observes first that, for a large part of the territory of France, the ratio of internal to international migrants is close to the French average, with an index close to 1. A significant proportion of rural spaces is characterised, however, by an over-representation of internal migrants (index superior to 1), in particular in the peri-urban countryside surrounding the great agglomerations of Paris and Lyons, or the region of Paris. Inversely, certain rural areas of southwestern France or central Brittany, which mostly involve isolated rural spaces in a situation of demographic and economic decline, are marked by an over-representation of international migrations and an under-representation of internal migrations. The geography of immigrant mobility in France, therefore, cannot be summed up by the map of localisations of new arrivals, in particular when one focuses on rural spaces. In part of the French countryside, it is the internal migrations of immigrants within the territories which dominate and which contribute to their socio-demographic evolutions.

3. A dynamic of peri-urbanisation more evident with immigrants than non-immigrants

21If differences appear between the destinations of international migrants and those of internal immigrant migrants, one must also compare these migrations to those of non-immigrants in France. Document 7 synthesises the number and percentage of internal migrations for immigrants and non-immigrants in metropolitan France between 2011 and 2015.

Document 7. Internal migrations according to the types of spaces (2011-2015)

Link to migration

Urban centres

Peri-urban countryside

Isolated countryside

Other countryside

Total countryside

Ensemble

Immigrants

Number

263,829

43,761

8,195

15,520

67,476

333,105

%

5.7

6.9

5.5

5.9

6.4

5.8

Average deviation from the total

-0.02

0.18

-0.06

0.01

0.10

 

Non-immigrants

Number

2,353,931

1,071,142

190,319

422,775

1,684,236

4,038,167

%

7.0

7.3

6.9

7.1

7.2

7.1

Average deviation from the total

-0.01

0.03

-0.03

0.00

0.01

 

Field: total of the population resident in metropolitan France in 2013 having undertaken one recent internal migration
Source: RP 2013
Reading: 6.9 % of the immigrants residing in the peri-urban countryside of metropolitan France in 2013 are immigrants arrived by recent migration. The average deviation from the profile of the total of recent immigrant migrants is 0.18 ((6.9 – 5.8)/5.8)

22As a reminder, 5.8 % of immigrants undertook an internal migration in metropolitan France between 2011 and 2015, as opposed to 7.1 % of non-immigrants. The immigrants are thus less mobile than the non-immigrants within the national territory. However, this mobility did not take place in the same proportions according to the spaces of destination: only 5.7 % of the immigrants in the urban centres arrived recently by internal migration, as opposed to 6.4 % of the immigrants in the countryside. This difference can be observed to a lesser degree in the non-immigrant population, with a slight under-representation of internal movements among residents of the urban centres and a slight over-representation among inhabitants of the countryside. If one focuses especially on the peri-urban countryside, one observes, however, that the average deviation from the total is distinctly greater for the immigrants than for the non-immigrants: 6.9 % of the immigrants in the peri-urban countryside are individuals arrived recently by internal migration (deviation of 0.18), as opposed to 7.3 % of non-immigrants (deviation of 0.03). This result shows the amplitude of the movement of peri-urbanisation for the immigrant population, and confirms by a national statistical approach the recent qualitative studies of the peri-urbanisation of immigrant populations of North-African origin in the region of Lyons [Lambert, 2015]. Inversely, in the isolated rural areas, internal migrations are under-represented for both immigrants and non-immigrants, but slightly more so for immigrants. The immigrants are therefore less mobile than the non-immigrants in terms of internal migrations, but they follow the same tendencies from one space to the other, and the dynamic of peri-urbanisation is today distinctly more evident among immigrants than non-immigrants.

III. What are the specific socio-demographic profiles of immigrants newly resident in the countryside?

23The broad tendencies of the spatial distribution of the immigrants who have undertaken an internal migration to the French countryside have been considered in the first section, but our analyses have so far not proposed differentiations according to the characteristics of individuals. This section aims to explore the socio-demographic profiles of those immigrants having effected an internal migration to or within the French countryside, following a double approach: do their profiles differ from those of non-immigrants who have undertaken the same type of migration? What is the internal diversity of the profiles of this population?

24To start with, the factors determining internal migration are compared between immigrants and non-immigrants. We posit the hypothesis that, in general, the internal migrations of immigrants to the French countryside respond to the same socio-demographic determinants as for non-immigrants: migrations associated mainly with the first stages of the life-cycle, which characterise individuals living in families and belonging to the middle classes. Age, composition of the household, Professions and Socio-professional Categories (PCS), diplomas held, as well as sex, are used as proxies for these social characteristics in models of logistic regression. A variable of geographic origin is also added to the models to determine if the fact of coming from the countryside or from urban centres has a differential effect on internal migration. Secondly, the characteristics specific to immigrants – namely, current nationality, country of birth and length of time since arrival in France – are added to the models. The hypothesis here is that significant differences exist in the profiles of immigrants according to the type of destination space of their internal migrations.

25This section is therefore based on the analysis of the results of a group of models of binary logistic regressions systematically concerning individuals present in metropolitan France at the moment of the census and in the preceding year, aged 18 years and over, working or retired. All of the studies are effected on the basis of the complementary exploitation of the census of 2013, applied in its entirety for models 2, 4 and 5, and on the basis of a random sample of a million individuals for models 1 and 3. A first series of models is based on analysis of the variable of interest ‘internal migration / absence of internal migration’ for the totality of individuals (1a), for non-immigrants (1b) and for immigrants (1c). A second series of models concerns the field of persons having undertaken an internal migration and analyses the destination ‘migration to the countryside / migration to the urban centres’ for the totality of individuals (2a), for non-immigrants (2b) and for immigrants (2c). A third series of models adds to the first two, with regard to immigrants, their migratory characteristics and nationality (3 et 4). Finally, a last series of models concerns immigrants having undertaken an internal migration to the countryside and analyses the determinants of different destinations, notably distinguishing peri-urban rural areas (5a) and isolated rural areas (5b). The descriptive statistics associated with the models are presented in appendixes (appendixes 1, 2 and 3).

1. Determinants common to immigrants and non-immigrants

26The first step is to analyse the determinants of internal migrations for immigrants and to compare them with those of non-immigrants. If the immigrants are distinguished by lodging characteristics and specific geographic distribution at the time of their arrival in France, one may nevertheless suppose that, following their installation, their geographic and residential evolutions correspond generally to the same determinants as for non-immigrants.

Document 8. Factors which influence the probability of undertaking, or not, an internal migration in metropolitan France, by link with migration (2011-2015)

Model

Model 1a

Model 1b

Model 1c

Field

Total Number

Non-immigants

Immigrants

 

Odds-ratio

Odds-ratio

Odds-ratio

Link to migration

 

 

 

 

 

 

Non-immigrant

Ref.

 

 

 

 

 

Immigrant

0.907

***

 

 

 

 

Sex

 

 

 

 

 

 

Male

Ref.

 

Ref.

 

Ref.

 

Female

1.020

1.026

*

0.956

 

Age at time of census

 

 

 

 

 

 

45-59 years

Ref.

 

Ref.

 

Ref.

 

18-29 years

5.560

***

5.733

***

4.126

***

30-44 years

2.836

***

2.916

***

2.290

***

60-75 years

0.536

***

0.534

***

0.540

***

75 years et plus

0.531

***

0.518

***

0.689

*

Origin of the migration

 

 

 

 

 

 

Urban centre origin

Ref.

 

Ref.

 

Ref.

 

Countryside origin

1.103

***

1.101

***

1.180

***

Professions and Socio-professional Categories

 

 

 

 

 

 

Labourer

Ref.

 

Ref.

 

Ref.

 

Farmer

0.394

***

0.388

***

0.924

 

Craftsperson, business person, entrepreneur

1.050

1.068

*

0.916

 

Manager, higher intellectual profession

1.008

 

1.012

 

1.009

 

Intermediate profession

1.078

***

1.081

***

1.081

 

Salaried worker

1.041

*

1.051

**

0.968

 

Retiree

0.902

*

0.926

0.745

*

Household situation

 

 

 

 

 

 

Couple

Ref.

 

Ref.

 

Ref.

 

Family

0.434

***

1.550

***

1.445

***

Single-parent family

0.552

***

0.423

***

0.549

***

Alone

1.022

 

0.540

***

0.705

***

Other

1.527

***

1.016

 

1.091

 

Educational level

 

 

 

 

Secondary school diploma

Ref.

 

Ref.

 

Ref.

 

Higher degree

1.141

***

1.138

***

1.160

**

No diploma

0.853

***

0.858

***

0.812

***

Constant

0.052

***

0.051

***

0.053

***

Number of individuals

685,308

 

610,113

 

75,195

 

Field: model 1a: totality of the population aged 18 years and over, working or retired, resident in metropolitan France at the time of the census and one year previous. Model 1b: totality of immigrants aged 18 years and over, working or retired, residence in metropolitan France at the time of the census and one year previous. Model 1c: totality of non-immigrants aged 18 years and over, working or retired, residing in metropolitan France at the time of the census and one year previous.
Source: RP 2013, complementary exploitation
Reading: a person aged from 18 to 29 years has 5.56 times more likelihood to undertake an internal migration than a person aged from 45 to 49 years, all other things being equal.
Significativity thresholds; ***: 0.1% ; **: 1%, *: 5%

27Document 8 gives us information on the determinants of internal migration in metropolitan France. Model 1a shows us first that, given equal socio-demographic characteristics, immigrants have less likelihood than non-immigrants of undertaking an internal migration (whatever the destination space) between 2011 and 2015. This first fact having been established, we then seek to know whether, despite this lesser likelihood of mobility, it is the same determinants that induce immigrants and non-immigrants to undertake an internal migration. Comparison of models 1b and 1c shows us that some of the determinants commonly associated with internal migration are present for the two populations: a greater propensity to migrate in younger than in older persons, no significant difference regarding sex, greater likelihood to undertake a migration for those belonging to educated classes. One also notes that, for the two populations, couples are considerably more likely to migrate than individuals in families. The only difference between the two populations consists in the effect of the Professions and Socio-professional Categories (PCS): while classically one finds a greater propensity to migrate among non-immigrants of the middle classes (defined roughly by the intermediate professions and salaried workers), as opposed to labourers, among immigrants all the PCS except retirees have a significantly greater likelihood of migrating than do labourers. Internal migration therefore appears at first less marked among immigrants than non-immigrants. This may also be linked to the existence of a greater diversity of specific profiles among the immigrants. On the whole, then, the socio-demographic determinants of internal migration are common to immigrants and non-immigrants.

Document 9. Factors influencing the probability of undertaking an internal migration to the countryside rather than to the urban centres in metropolitan France, by link to migration (2011-2015)

Model

Model 2a

Model 2b

Model 2c

Field

Total number

Non-immigrants

Immigrants

 

Odds-ratio

 

Odds-ratio

 

Odds-ratio

 

Link to migration

 

 

 

 

 

 

Non-immigrant

Ref.

 

Ref.

 

Ref.

 

Immigrant

0.402

***

 

 

 

 

Sex

 

 

 

 

 

 

Male

Ref.

 

Ref.

 

Ref.

 

Female

1.103

***

1.101

***

1.182

***

Age at time of census

 

 

 

 

 

 

45-59 years

Ref.

 

Ref.

 

Ref.

 

18-29 years

0.743

***

0.753

***

0.599

***

30-44 years

0.908

***

0.925

***

0.759

***

60-75 years

1.056

**

1.057

**

1.020

 

75 years and over

1.018

 

1.014

 

1123

 

Origin of the migration

 

 

 

 

 

 

Urban centre origin

Ref.

 

Ref.

 

Ref.

 

Countryside origin

3.673

***

3.574

***

5.641

***

Professions and Socio-professional Categories

 

 

 

 

 

 

Labourer

Ref.

 

Ref.

 

Ref.

 

Farmer

3.665

***

3.552

***

5.272

***

Craftsperson, business person, entrepreneur

0.869

***

0.844

***

1.151

**

Manager, higher intellectual profession

0.443

***

0.430

***

0.629

***

Intermediate profession

0.700

***

0.685

***

0.897

**

Salaried worker

0.715

***

0.707

***

0.746

***

Retiree

0.910

***

0.884

***

1.214

*

Household situation

 

 

 

 

 

 

Couple

Ref.

 

Ref.

 

Ref.

 

Family

0.599

***

0.593

***

0.702

***

Single-parent family

1.280

***

1.299

***

1.082

**

Alone

0.727

***

0.735

***

0.574

***

Other

0.497

***

0.495

***

0.565

***

Educational level

 

 

 

 

Secondary school diploma

Ref.

 

Ref.

 

Ref.

 

Higher digree

0.800

***

0.799

***

0.828

***

No diploma

1.000

***

1.031

***

0.811

***

Constant

0.832

***

0.840

***

0.310

***

Nomber of individuals

911,573

 

828,392

 

83,181

 

Field: model 2a: totality of the population aged 18 years and over, working or retired, resident in metropolitan France at the time of the census and having undertaken an internal migration in the year preceding the census. Modèle 2b: totality of immigrants aged 18 years and over, working or retired, resident in metropolitan France at the time of the census and having undertaken an internal migration in the year preceding the census. Modèle 2c: totality of non-immigrants aged 18 years and over, working or retired, resident in metropolitan France at the time of the census and having undertaken an internal migration in the year preceding the census.
Source: RP 2013, complementary exploitation.
Reading: among the totality of persons having undertaken an internal migration, a woman is 1.103 times more likely than a man to undertake an internal migration to the countryside rather than to the urban centres, all other things being equal.
Significativity thresholds ; ***: 0.1%; **: 1%, *: 5%

28To follow up, one might focus more specifically on the persons having undertaken an internal migration and ask whether the determinants of an internal migration to the countryside, rather than to the urban centres, are similar for immigrants and non-immigrants. On the one hand, the participation of immigrants in the broad demographic dynamics of the countryside (peri-urbanisation and renewal of the isolated countryside), as shown in the preceding section, suggests the existence of determinants common to the two groups. On the other hand, the direct establishment of foreigners in the countryside since the 1990s is chiefly due to specific profiles. The possible internal mobility of these individuals within the countryside might therefore indicate the existence of determinants particular to immigrants. Model 2a shows, to start with, that once the differences in socio-demographic structures between the two populations are taken into account, immigrants are far less likely than non-immigrants to undertake a migration to the countryside rather than to the urban centres. Now, if one compares the determinants of migration to the countryside for immigrants and non-immigrants, one again finds numerous similarities. In both cases, one finds an age-factor, which this time comes into play in the opposite direction, young people being less likely to undertake a migration to the countryside (as opposed to the urban centres) than those from 45 to 59 years old. Women are slightly more likely to undertake this sort of migration, as are persons living in families, rather than in a couple. Coming from the countryside, moreover, greatly increases the likelihood of a migration to the countryside (by 3.6 for non-immigrants and 5.6 for immigrants). A portion of the classic determinants for migration to the countryside is thus common to the two groups. The factor of education, however, operates differently according to the link to migration: while for non-immigrants the lack of a diploma has a slightly positive influence, relative to possession of a secondary school diploma, the contrary is true for immigrants. For that population, possessing a higher degree or no diploma at all makes for less probability of migrating to the countryside than possession of a secondary school diploma. This effect calls for explanation, suggesting that internal migrations to the countryside are socially less selective that in cities. Finally, the PCS in this case have significant effects, but these are not exactly the same for immigrants and non-immigrants. For the non-immigrants, all the professions, except for farmers, have less likelihood than labourers of undertaking a migration to the countryside instead of to the urban centres. Managers in particular are characterised by a far lower probability of undertaking this type of migration. Among immigrants, one finds some differences: craftspersons, business-persons and entrepreneurs, as well as retirees, are more likely to undertake a migration to the countryside than are labourers. This difference suggests the existence of atypical profiles among the immigrants, concerning which one may posit hypotheses: if the establishment of craftspersons and retirees, especially British and Dutch, is qualitatively documented, most studies record direct migrations from outside France [Cognard, 2011]. This result leads to the conclusion that re-locations may also take place among these populations following their settlement in France. With respect to the retirees, an alternative hypothesis to the re-locations of North-European foreigners may also be developed: one might suppose that this concerns migrations of immigrants formerly established in France, who, upon arriving at the age of retirement, decide to undertake this residential migration. These two kinds of profiles combine, perhaps, while privileging different locations of settlement. One aspect of these hypotheses will now be explored by detailing the migratory characteristics of the immigrants and concentrating on the different types of countryside where they settle, especially peri-urban and isolated rural areas.

2. Specificities of migratory profiles and types of countryside

29We initially consider the effects of migratory characteristics and nationality on the undertaking of an internal migration (document 10). Model 3 shows that, all else being equal, the country of origin, grouping these according to large regions, has little effect on internal migration, except for immigrants born in Asia or America, who are more likely than immigrants from the European Union to undertake an internal migration between 2011 and 2015. The length of time since establishment in France has a more marked effect: those individuals recently arrived in France are more likely to undertake an internal migration than those who arrived some time ago.

Document 10. Factors which influence the probability of undertaking, or not, an internal migration (model 3) and undertaking an internal migration to the countryside (model 4) in metropolitan France, with regard to immigrants (2011-2015)

Model

Model 3

Model 4

Field

Internal migration / No internal migration

Internal migration to the countryside / to the urban centres

 

Odds-ratio

 

Odds-ratio

 

Country of birth

 

 

Countries of the European Union (28)

Ref.

 

Ref.

 

Africa

1.055

 

0.374

***

America

1.201

*

0.543

***

Asia

1.156

*

0.374

***

Europe outside the European Unionn

1.000

 

0.557

***

Oceania

1.140

 

 

 

Time since arrival in France

 

 

6 to 10 years

Ref.

 

Ref.

 

Less than 3 years

1.668

***

0.910

3 to 5 years

1.339

***

0.925

11 to 20 years

0.894

1.020

 

More than 20 years

0.690

***

1.187

***

Not recorded

0.874

*

0.979

 

Nationality

 

 

Foreign

Ref.

 

Ref.

 

French by acquisition

0.962

 

1.181

***

Sex

 

 

 

Male

Ref.

 

Ref.

 

Female

0.950

 

1.095

***

Age at time of census

 

 

 

 

45-59 years

Ref.

 

Ref.

 

18-29 years

3.172

***

0.717

***

30-44 years

1.955

***

0.910

**

60-75 years

0.591

***

1.013

 

75 years and over

0.780

0.993

 

Origin of the migration

 

 

 

 

Urban centre origin

Ref.

 

Ref.

 

Countryside origin

1.188

***

4.872

***

Professions and Socio-professional Categories

 

 

 

 

Labourer

Ref.

 

Ref.

 

Farmer

0.929

 

4.901

***

Craftsperson, business person, entrepreneur

0.940

 

1.096

.

Manager, higher intellectual profession

1.020

 

0.593

***

Intermediate profession

1.132

*

0.851

***

Salaried worker

0.981

 

0.785

***

Retiree

0.751

*

1.066

 

Household situation

 

 

 

 

Couple

Ref.

 

Ref.

 

Family

1.406

***

0.811

***

Single-parent family

0.593

***

1.180

***

Alone

0.782

**

0.650

***

Other

1.126

*

0.605

***

Educational level

 

 

Secondary school diploma

Ref.

 

Ref.

 

Higher digree

1.084

0.891

***

No diploma

0.787

***

0.843

***

Constant

0.063

***

0.440

***

Number of individuals

75,195

 

83,181

 

Field: model 3 : totality of immigrants aged 18 and over, working or retired, resident in metropolitan France at the time of the census and one year previous. Model 4: totality of immigrants aged 18 and over, working or retired, resident in metropolitan France at the time of the census having undertaken an internal migration in the year preceding the census.
Source: RP 2013, complementary exploitation
Reading: among the totality of immigrants having undertaken an internal migration, un immigrant French by acquisition is 1.181 times more likely to undertake an internal migration to the countryside instead of to the urban centres, all other things being equal.
Significativity thresholds; ***: 0.1%; **: 1%, *: 5%

30Now, if one focuses on the effect of these variables on the persons who have undertaken an internal migration, whether to the countryside or the urban centres (model 4), one observes a more significant effect of geographic origin: immigrants from the European Union have greater likelihood than all others of undertaking an internal migration to the countryside. The length of time since establishment is a less important factor, but one observes nevertheless that more internal migrations to the countryside are undertaken by immigrants long established in France than by immigrants who arrived between 6 and 10 years previous. Finally, one also observes a strong effect of nationality: those persons having acquired French nationality have a greater probability of undertaking an internal migration to the countryside than to the urban centres. These two last elements open lines of thought on the effects on the internal migrations of immigrants of the duration of their presence in France. The results show, in fact, that length of time resident in France and acquisition of French nationality – which is also closely related to the time elapsed since settlement in France – differentiate the forms of internal mobility of immigrants. This suggests that internal migration to the French countryside corresponds to a selective temporal – indeed generational – process among immigrants. If one can imagine migration to the countryside of older individuals, at the end of their working lives, who arrived in France some decades before, one may also suppose that younger immigrants, belonging to the ‘generation 1.5’, conform to this profile and undertake this migration as part of a residential project involving accession to ownership of an individual dwelling, in keeping, for example, with a new stage of their life-cycle.

Document 11. Spatial distribution of immigrants having undertaken an internal migration to the countryside in metropolitan France, according to birth-place (2011-2015)

Document 11. Spatial distribution of immigrants having undertaken an internal migration to the countryside in metropolitan France, according to birth-place (2011-2015)

Field: totality of immigrants having undertaken an internal migration to the less dense and denser catchment areas of metropolitan France.
Source : RP 2013
Note: the typology of densities by catchment area is constructed on the basis of the grid of communal densities of the INSEE.

31The geography of migrations to the French countryside is, moreover, differentiated according to the geographic origin of the immigrants (document 11). While the migrations of individuals originating in the European Union are largely over-represented in the Southwest of France, in Brittany and in the frontier zones of eastern France, the migrations of immigrants from outside the Union are numerous in the East of France but especially over-represented in the Parisian region. The African immigrants, for their part, are very numerous in the Parisian region but especially over-represented in the Rhône-Alpes region, as well as in the Southwest of France.

Document 12. Factors influencing the probability of immigrants undertaking an internal migration to peri-urban (modèle 5a) or isolated rural areas (modèle 5b), rather than to other rural areas in metropolitan France (2011-2015)

Model

Model 5a

Model 5b

Field

Internal migration to the peri-urban rural areas / to other rural areas

Internal migration to isolated rural areas / to other rural areas

 

Odds-ratio

 

Odds-ratio

 

Country of birth

 

 

Countries of the European Union (28)

Ref.

 

Ref.

 

Africa

1.293

***

0.707

***

America

1.047

 

0.989

 

Asia

1.298

**

0.949

 

Europe outside European Union

1.068

 

0.890

 

Oceania

0.550

 

3.436

**

Time since arrival in France

 

 

6 to 10 years

Ref.

 

Ref.

 

Less than 3 year

0.812

*

1.219

 

3 to 5 years

0.895

 

1.161

 

11 to 20 years

1.111

 

0.958

 

More than 20 years

1.302

***

0.805

*

Not recorded

1.077

 

0.866

 

Nationality

 

 

Foreign

Ref.

 

Ref.

 

French by acquisition

0.994

 

0.806

**

Sex

 

 

Male

Ref.

 

Ref.

 

Female

0.958

 

1.099

 

Age at time of census

 

 

 

 

45-59 years

Ref.

 

Ref.

 

18-29 years

1.054

 

0.747

**

30-44 years

1.171

**

0.727

***

60-75 years

0.699

**

1.076

 

75 years and over

1.005

 

0.877

 

Origin of the migration

 

 

 

 

Urban origin

Ref.

 

Ref.

 

Isolated rural origin

0.094

***

10.243

***

Peri-urban rural origin

1.704

***

0.635

***

Other rural origin

0.147

***

1.828

***

Professions and Socio-professional Categories

 

 

 

 

Labourer

Ref.

 

Ref.

 

Farmer

0.503

*

2.237

**

Craftsperson, business person, entrepreneur

0.889

 

1.365

*

Manager, higher intellectual profession

1.764

***

0.696

*

Intermediate profession

1.247

**

0.956

 

Salaried worker

1.183

*

1.116

 

Retiree

0.937

 

1.378

.

Household situation

 

 

 

 

Couple

Ref.

 

Ref.

 

Family

0.999

 

1.053

 

Single-parent family

1.316

***

0.738

***

Alone

1.134

 

0.947

 

Other

0.907

 

1.125

 

Educational level

 

 

Secondary school diploma

Ref.

 

Ref.

 

Higher degree

1.033

 

1.128

 

No diploma

1.022

 

1.097

 

Constant

1.880

***

0.125

***

Number of individuals

13,042

 

13,042

 

Field: totality of immigrants aged 18 years and over, working or retired, resident in metropolitan France at the time of the census, having undertaken an internal migration to the countryside during the year preceding the census.
Source: RP 2013, complementary exploitation
Reading: among the total number of immigrants having undertaken an internal migration to the countryside, an immigrant born in an African country has a 1.293 times greater probability of undertaking an internal migration to the peri-urban rural areas than an immigrant born in a country of the European Union, all other things being equal.
Significativity thresholds; ***: 0,1%; **: 1%, *: 5%

32Finally, the last two models explore the variety of determinants of internal migration to the countryside, according to the type of countryside considered. The migrations to the peri-urban rural areas (model 5a), then the migrations to the isolated rural areas (model 5b), are studied in turn. By comparison with migrations to the other rural areas, the individuals migrating to the peri-urban rural areas are more likely to be of African or Asian than of European Union origin. If most of the effects relating to duration of establishment are not significant, one nevertheless observes a significant effect among recent arrivals, who have less likelihood of undertaking a peri-urban migration, compared with individuals established in France for 6 to 10 years. On the other hand, the situation concerning French national has no effect. With regard to the other socio-demographic variables, those individuals belonging to categories of age relatively young (30-44 years) have a higher probability of undertaking a peri-urban migration than those of 45-59 years, and those of 60-74 years far less. As for the PCS, one notices that, all other things being equal, more migrations of the upper and middle classes (managers, intermediate professions, salaried employees) take place that among labourers. There are also more migrations of families than of couples. Finally, as concerns the origin of the migration, one finds the greatest propensity to migrate within the same type of countryside rather than to change the type of space.

33Inversely, migrations to isolated rural areas are characterised by a greater probability to be undertaken by immigrants from Oceania or the European Union, rather than Africa. The factor of length of time since settlement also operates in the contrary sense, with less likelihood of immigrants long established in France (for more than 20 years) to undertake this type of migration, as opposed to immigrants established for from 6 to 10 years. In contrast with the preceding model, one now observes an effect of nationality, and foreigners have a greater propensity than those who are French by acquisition to undertake this type of migration. As for the remainder of the socio-demographic characteristics, one observes that there is a negative effect for the youngest ages; that there are significantly more farmers and craftspersons and fewer managers than labourers; and, further, that more migrations involve couples than families. One also finds that persons who are originally from isolated rural areas are more likely to undertake a migration to this type of space than persons originally from urban centres.

34There are indeed, therefore, specific determinants affecting migration to the different types of countryside, and particularly to the peri-urban and isolated rural areas. The individuals undertaking a migration to the peri-urban countryside do not have the same characteristics as those migrating to the isolated countryside. Thus Lambert [2015] has shown that the establishment of immigrants in the peri-urban countryside of the agglomeration of Lyons was especially due to individuals who originated in the Maghreb, arrived in France long before and relatively old. Our results confirm the existence of this profile quantitatively and on the national scale. That the peri-urban migrations should be due more to immigrants both long established in France and belonging to relatively young age-groups reinforces, however, the hypothesis of a strong participation of immigrants who grew up in France in these dynamics, notably at the point of developing a family. Migrations to the isolated rural areas correspond to very different profiles: involved here are migrations of individuals from the European Union, older but recently arrived in France – a fact which leads to the supposition that these involve mainly relocations of North-European retirees who had established themselves in the countryside from their arrival in France. Moreover, the fact that migration within the same type of space is much more pronounced for migrations in isolated rural areas (factor 10) than for peri-urban migrations (factor 1.7) reinforces this hypothesis of local movements within the isolated countryside.

Conclusion

35The ambition of this article has been to explore and quantify the internal migrations of immigrants to the French countryside and to propose some interesting lines of thought for further research. It should first be emphasised that the internal migrations of immigrants in France have been substantial, contributing to the re-composition of the geography of immigration in France, particularly in the peri-urban rural areas. The participation of immigrants in the process of peri-urbanisation is important today, and they share with non-immigrants the same social characteristics: migrations of the young working middle classes in families with children. It is mainly immigrants of African or Asian origin, long established in France, who fuel this dynamic. Taken overall, the determinants of internal migrations by immigrants to the countryside are very close to those of non-immigrants.

36This resemblance between immigrants and non-immigrants should not, however, obscure the great diversity of the migrations – socio-demographic diversity of the profile of immigrants migrating to the countryside, but also diversity of the territories in which these migrations take place. Indeed, the migrations to the isolated rural areas, of they are less numerous, appear very different from those to other countryside and are notably marked by profiles more atypical: these are migrations of European immigrants, recently arrived in France and of foreign nationality, older, among them craftspersons or entrepreneurs, and they most often involve relocations within isolated rural areas.

37In this article, only two large types of territories have been explored – peri-urban and isolated rural areas – but the diverse types of countryside would deserve to be more meticulously taken into account in a more detailed study. With respect to the socio-demographic characteristics of individuals, deeper investigations remain to be conducted to understand the meaning and the place of these internal migrations in the trajectories of the lives of individuals. The question of the effect of length of presence on the probability of undertaking an internal migration in the countryside especially deserves to be explored. Finally, it would be interesting to include the characteristics of households, such as the type of union (to identify mixed couples) or occupational status, in order to verify hypotheses concerning access to ownership and go beyond the exclusive framework of the individual in appreciating the determinants of internal mobility.

Haut de page

Bibliographie

Belanger A., Rogers A., 1993, « The Internal Migration and Spatial Redistribution of the Foreign-Born Population in the United States: 1965-70 and 1975-80”, International Migration Review, Vol. 26, n°4, pp.1342-1369.

Bonvalet C., Fribourg A.-M. (dir.), 1990, Stratégies résidentielles, Paris, INED, 459p.

Brutel C., 2016, « La localisation géographique des immigrés. Une forte concentration dyears l’aire urbaine de Paris », in Insee Première, n°1591, Avril 2016, 4p.

Buller H., Hoggart K., 1994, International counterurbanization : British migrants in rural France, Aldershot : Avebury 150p.

Cognard F., 2011, « Les migrations résidentielles des Britanniques et des Néerlandais : une figure originale de la nouvelle attractivité des moyennes montagnes françaises », Espace populations sociétés [En ligne], 2011/3 mis en ligne le 31 décembre 2013, consulté le 31 décembre 2013. URL : http://eps.revues.org/4672

Courgeau D., 1973, « Migrants et migrations », Population, 28(1), pp. 95-129.

Courgeau D., Lelièvre É., 2004, « Estimation des migrations internes de la période 1990-1999 et comparaison avec celles des périodes antérieures », Population, 59(5), pp. 797-804.

Debrand T., Taffin C., 2005, « Les facteurs structurels et conjoncturels de la mobilité résidentielle depuis 20 years », Economie et Statistique, N°381-382, pp.125-146.

Desplanques G., Tabard N., 1991, « La localisation de la population étrangère » In: Economie et statistique, n°242, Avril 1991, Dossier : Les étrangers en France / Conditions de travail et santé des ouvriers. pp. 51-62.

Depraz S., 2013, « Mais où est donc passé l'espace rural ? », Les cafés géographiques de Lyon (en ligne), Sep 2013, Lyon, France. 2013, http://cafe-geo.net/mais-ou-est-donc-passe-l-espace-rural/

Donzeau N., Pan Ké Shon J.-L., 2009, « L'évolution de la mobilité résidentielle en France entre 1973 et 2006 : nouvelles estimations », Population, 2009/4 (Vol. 64), pp. 779-795. DOI : 10.3917/popu.904.0779. URL : https://www.cairn.info/revue-population-2009-4-page-779.htm

Farrell C., 2016, « Immigrant suburbanisation and the shifting geographic structure of metropolitan segregation in the United States », Urban Studies, Vol. 53, n°1, pp.57-76.

Fruit J.-P., 1985, « Migrations résidentielles en milieu rural péri-urbain : le Pays de Caux Central », Espace, Populations, Sociétés, Vol. 1, pp.150-159.

Guillon M., 1996, « Inertie et localisation des immigrés dyears l'espace parisien », In: Espace, populations, sociétés, 1996-1. Hommage à Daniel Noin. pp. 55-63.

Gurak D.T., Kritz M.M., 2000, « Context Determinants of Interstate Migration of U.S. Immigrants”, Social Forces, Vol. 78, n°3, pp.1017-1039.

INSEE, 2011, « Le nouveau zonage en aires urbaines de 2010 : 95 % de la population vit sous l'influence des villes », in Insee Premiere, n°1374, octobre 2011

INSEE, 2012, « Le nouveau zonage en bassins de vie de 2012 : trois quarts des bassins de vie sont ruraux », in Insee Premiere, n°1425, décembre 2012

INSEE, 2014, « Estimer les flux d’entrées sur le territoire à partir des enquêtes annuelles de recensement  », Documents de travail, n° F1403, mai 2014

Kayser B. (dir.), 1993, Naissance de nouvelles campagnes, Éditions de l’Aube

King R., Skeldon R., 2010, « ‘Mind the Gap !’ Integrating Approaches to Internal and International Migration », Journal of Ethnic and Migration Studies, Vol.36, n°10, pp.1619-1646.

Lambert A., 2015, « Tous propriétaires ! ». L’envers du décor pavillonnaire, Paris, Seuil, collection Liber, 279p.

Lichter D., Johnson K., 2006, « Emerging Rural Settlement Patterns and the Geographic Resdistribution of America’s New Immigrants », Rural Sociology, Vol. 71, n°1, pp.109-131.

Massey D., Denton N., 1985, « Spatial Assimilation as a Socioeconomic Outcome », American Sociological Review, Vol. 50, n°1, pp.94-106.

Pan Ké Shon J.-L., 2007, « Le recensement rénové français et l’étude des mobilités », Population, 62(1), pp. 123-141.

Pan Ké Shon J.-L., 2009, « Ségrégation ethnique et ségrégation sociale en quartiers sensibles », in Revue française de sociologie, 50-3, pp. 451-487.

Pistre P., 2012, Renouveaux des campagnes françaises : évolutions démographiques, dynamiques spatiales et recompositions sociales, Thèse de doctorat de géographie, Université Paris Diderot (Paris 7), 407p. Disponible sur : http://halshs.archives-ouvertes.fr/tel-00764869/.

Préteceille E., 2009, « La ségrégation ethno-raciale a-t-elle augmenté dyears la métropole parisienne ? », in Revue française de sociologie, 2009/3, Vol. 50, pp.489-519.

Rathelot R., Safi M., 2014, « Local Ethnic Composition and Natives’ and Immigrants’ Geographic Mobility in France, 1982-1999 », in American Sociological Review, vol. 79, n°1, pp.43-64.

Reher D., Silvestre J., 2009, « Internal migration patterns of foreign-born immigrants in a country of recent mass immigration: Evidence from new micro data for Spain », International Migration Review, Vol. 43, n°4, pp.815-849.

Solignac M., 2013, « La mobilité résidentielle des immigrés en France : l’intérêt d’une approche longitudinale », Disponible sur : https://www.casd.eu/publication/la-mobilite-residentielle-des-immigres-en-france-linteret-dune-approche-longitudinale-par-panel/

Solignac M., 2016, « L’émigration des immigrés, une dimension oubliée de la mobilité géographique », <halshs-01422323>

Villanova R de., 1997, « L'espace résidentiel des Portugais de France », in Hommes et migrations, n° 110, pp. 32-42.

Haut de page

Annexe

Appendix 1. Descriptive statistics of individuals having undertaken an internal migration in metropolitan France, by link to migration (2011-2015)

Model

Model 1a

Model 1b

Model 1c and model 3

Field

Total

Non-immigrants

Immigrants

Link to migration

 

 

 

Immigrant

8.18

 

 

Non immigrant

91.82

 

 

Age

 

 

 

18-29 years

37.35

38.43

25.19

30-44 years

35.16

34.38

43.87

45-59 years

14.76

14.37

19.10

60-74 years

7.67

7.70

7.40

75 years and over

5.06

5.11

4.44

Sex

 

 

 

Male

50.63

50.20

55.42

Female

49.37

49.80

44.58

Professions and Socio-professional Categories

 

 

 

Farmer

0.36

0.38

S.S.

Craftsperson, business person, entrepreneur

4.08

3.98

S.S.

Farmer, craftsperson, business person, entrepreneur

 

 

5.41

Manager, higher intellectual profession

13.60

13.56

14.07

Intermediate profession

24.15

24.68

18.18

Salaried employee

26.56

26.57

26.41

Labourer

19.48

18.90

26.02

Retiree

11.77

11.93

9.90

Educational level

 

 

 

Secondary school diploma

43.12

44.08

32.42

Higher degree

39.29

39.51

36.84

No diploma

17.59

16.41

30.74

Household situation

 

Couple

30.16

30.64

24.80

Family

29.34

28.87

34.66

Single-parent family

7.16

7.14

7.32

Alone

22.54

22.85

19.09

Other

10.80

10.51

14.13

Origin of the migration

 

 

 

Urban centre

37.06

38.75

18.04

Countryside

62.94

61.25

81.96

Length of time since arrival in France

 

 

 

Less than 3 years

 

 

7.47

3 to 5 years

 

 

13.03

6 to 10 years

 

 

16.48

11 to 20 years

 

 

18.78

More than 20 years

 

 

26.56

Not recorded

 

 

17.68

Nationality

 

French by acquisition

 

 

36.81

Foreign

 

 

63.19

Country of birth (grouped by continent)

 

 

 

Africa

 

 

45.24

European Union (28)

 

 

29.12

Europe outside European Union

 

 

8.44

Asa

 

 

11.35

America + Oceania

 

 

5.85

Number of individuals

15,2612

140,121

12,491

Field: model 1a: totality of individuals aged 18 years and over, working or retired, resident in metropolitan France at the time of census and having undertaken an internal migration in the year preceding the census. Model 1b: totality of non-immigrants aged 18 years and over, working or retired, resident in metropolitan France at the time of census and having undertaken an internal migration in the year preceding the census. Models 1c and 3: totality of immigrants aged 18 years and over, working or retired, resident in metropolitan France at the time of census and having undertaken an internal migration in the year preceding the census.
Source: RP 2013, complementary exploitation
Reading: 8.18% of the persons having undertaken an internal migration in metropolitan France in the year preceding the 2013 census are immigrants.

Appendix 2. Descriptive statistics of individuals having undertaken an internal migration to the countryside in metropolitan France, by link to migration (2011-2015)

Model

Model 2a

Model 2b

Model 2c & model 4

Field

Total

Non-immigrants

Immigrants

Link to migration

 

 

 

Immigrant

4.29

 

 

Non-immigrant

95.71

 

 

Age

 

 

 

18-29 years

33.35

34.07

17.25

30-44 years

36.13

35.89

41.48

45-59 years

15.90

15.55

23.64

60-74 years

8.89

8.76

11.74

75 years and over

5.73

5.73

5.89

Sex

 

 

 

Male

50.43

50.24

54.65

Female

49.57

49.76

45.35

Professions and Socio-professional Categories

 

 

 

Farmer

0.73

0.73

0.62

Craftsperson, business person, entrepreneur

4.64

4.53

7.19

Manager, higher intellectual profession

8.32

8.26

9.49

Intermediate profession

21.91

22.16

16.38

Salaried employee

26.64

26.82

22.65

Labourer

23.87

23.70

27.74

Retiree

13.89

13.79

15.92

Educational level

 

 

 

Secondary school diploma

48.94

49.50

36.43

Higher degree

31.07

31.10

30.37

No diploma

19.99

19.40

33.20

Household situation

 

 

 

Couple

31.96

32.15

27.75

Family

35.19

34.97

40.03

Single-parent family

7.07

7.14

5.67

Alone

16.40

16.48

14.71

Other

9.37

9.26

11.84

Origin of the migration

 

 

 

Urban centre

55.78

56.26

45.28

Countryside

44.22

43.74

54.72

Length of time since arrival in France

 

 

 

Less than 3 years

 

 

6.41

3 to 5 years

 

 

10.20

6 to 10 years

 

 

14.18

11 to 20 years

 

 

16.70

More than 20 years

 

 

35.06

Not recorded

 

 

17.45

Nationality

 

French by acquisition

 

 

39.88

Foreign

 

 

60.12

Country of birth (grouped by continent)

 

 

 

Africa

 

 

30.15

European Union (28)

 

 

49.20

Europe outside European Union

 

 

8.38

Asia

 

 

7.00

America

 

 

5.06

Oceania

 

 

0.21

Number of individuals

1,199,248

1,147,772

51,477

Field: model 2a: totality of individuals aged 18 years and over, working or retired, resident in metropolitan France at the time of the census and having undertaken an internal migration to the countryside during the year preceding the census. Modèle 2b : totality of non-immigrants aged 18 years and over, working or retired, resident in metropolitan France at the time of the census and having undertaken an internal migration to the countryside during the year preceding the census. Models 2c and 4: totality of immigrants aged 18 years and over, working or retired, resident in metropolitan France at the time of the census and having undertaken an internal migration to the countryside during the year preceding the census.
Source: RP 2013, complementary exploitation
Reading: 4.29% of the persons having undertaking an internal migration to the countryside in metropolitan France during the year preceding the census of 2013 are immigrants.

Appendix 3. Descriptive statistics of immigrants having undertaken an internal migration to the peri-urban (model 5a) or isolated (model 5b) rural areas in metropolitan France (2011-2015)

 

Model 5a

Model 5b

Age

 

 

18-29 years

17.13

15.51

30-44 years

44.49

31.87

45-59 years

23.68

25.90

60-74 years

9.41

18.90

75 years and over

5.29

7.83

Sex

 

Male

54.84

51.70

Female

45.16

48.30

Professions and Socio-professional Categories

 

Farmer

0.39

1.63

Craftsperson, business person, entreprener

6.69

9.32

Manager, higher intellectual profession

11.06

6.26

Intermediate profession

17.76

12.88

Salaried employee

23.87

20.45

Labourer

27.18

23.99

Retiree

13.07

25.47

Educational level

 

 

Secondary school diploma

36.13

34.53

Higher degree

31.94

29.21

No diploma

31.93

36.26

Household situation

 

Couple

25.80

34.01

Family

43.95

27.55

Single-parent family

5.85

5.33

Alone

13.23

19.31

Other

11.18

13.79

Origin of the migration

 

 

Urban centre

61.19

37.02

Peri-urban rural area

31.56

12.13

Isolated rural area

2.22

35.01

Other rural area

5.03

15.84

Length of time since arrival in France

 

 

Less than 3 years

5.54

8.86

3 to 5 years

9.52

12.68

6 to 10 years

13.83

15.54

11 to 20 years

17.17

15.60

More than 20 years

36.99

29.82

Not recorded

16.94

17.51

Nationality

 

French by acquisition

41.90

30.69

Foreign

58.10

69.31

Country of birth (grouped byt continent)

 

Africa

33.41

18.56

European Union (28)

44.95

62.94

Europe outside European Union

8.53

7.56

Asia

7.92

5.33

America and Oceania

5.19

5.62

Number of individuals

34,420

6,024

Field: model 5a: totality of immigrants aged 18 years and over, working or retired, resident in metropolitan France at the time of the census and having undertaken an internal migration to the peri-urban countryside during the year preceding the census. Model 5b: totality of immigrants aged 18 years and over, working or retired, resident in metropolitan France at the time of the census and having undertaken an internal migration to the isolated countryside during the year preceding the census.
Source: RP 2013, complementary exploitation
Reading: 17.13% of immigrants having undertaken an internal migration to the peri-urban countryside during the year preceding the census of 2013 are between 18 and 29 years old.

Haut de page

Notes

1 This work has benefitted from French government assistance managed by the Agence nationale de la Recherche (National Resarch Agency) as part of the programme Investissements d’Avenir (Investments for the Future) designated by the reference ANR-10-EQPX-17 (Centre d’accès sécurisé aux données [Centre for secure data access] – CASD) [equivalents anglais – si nécessaires – à vérifier]

2 Certain biographical surveys precisely record the changes of residence and/or commune of individuals, such as the survey ‘Triple Biographie: familiale, professionnelle et migratoire’ (unpublished, Insee, 1981), the survey ‘Biographies et Entourage’ (unpublished, 2001), as well as the survey ‘Histoire de Vie’ (Insee, unpublished, 2003). Nevertheless, the sample or the field of these surveys is generally very limited: only 8403 are covered by the survey ‘Histoire de Vie’, a population surveyed uniquely from the Île de France for ‘Biographie et Entourage’, etc. The measure of the residential and geographic mobility of populations has therefore been estimated instead on the basis of investigations using larger samples, such as the survey ‘Logement et Emploi’, or indeed on the basis of the population census. These surveys, however, furnish information only on immigrants (change of place of residence between two dates), and not on the migrations themselves. Since the studies of Courgeau [1973], a number of models for estimating residential migrations have thus been developed to attempt to correct the underestimation of movements originating with this kind of information collecting [Courgeau and Lelièvre, 2004; Donzeau, Pan Ké Shon, 2009].

3 In the population census, two questions permit the identification of immigrants: one concerning nationality at present and at birth; one concerning place of birth in or outside France. These questions are not asked, however, with regard to the parents of the individual surveyed, so that the descendants of immigrants cannot be identified. Use of the census therefore limits comparison of the characteristics of immigrants to comparison with non-immigrants only.

4 For each of the maps created, the scale chosen is that of the ‘bassin de vie’ (living area), a spatial statistical unit created by the INSEE in 2003: each living area has been determined as a function of accessibility to current services and areas of attraction [INSEE, 2012]. These living areas were grouped together with a minimum threshold of 5000 inhabitants between 1968 and 2013 in order to guarantee respect for statistical secrecy.

5 In order to work on a homogeneous field, the number of international migrations has been calculated here exclusively on the basis of the census question dealing with the place of residence one year previous. This method tends to underestimate by some tens of thousands of individuals the flow of entries into the territory [INSEE, 2014]. It is therefore appropriate to undertake analysis only of the structural differences between the different kinds of spaces studied and not the number of international migrations in themselves.

6 Significativity of the differences verified according to a 95% confidence interval.[termes/expression à vérifier]

Haut de page

Table des illustrations

Titre Document 3. Spatial distribution of immigrants in metropolitan France in 1975 and 20134
Crédits Sources: Population census INSEE 1975 and 2013, principal exploitation Champ: totality of immigrants resident in metropolitan France
URL http://journals.openedition.org/eps/docannexe/image/12208/img-1.png
Fichier image/png, 174k
Titre Document 5. Internal and international migrations according to the type of space of arrival, between 2011 and 2015 (in %)
Légende Field: total number of immigrants residing in metropolitan France at the moment of the census and residing in another commune or outside France in the preceding year. Source: RP 2013
URL http://journals.openedition.org/eps/docannexe/image/12208/img-2.png
Fichier image/png, 22k
Titre Document 6. A differential geography of recent immigrant arrivals in the catchment areas of metropolitan France (2011-2015)
Légende Field: total number of immigrants residing in metropolitan France in 2013 having undertaken a recent migration. Source: RP INSEE 2013, GEOFLA 2015 Reading: calculation of the index of specialisation for each catchment area i: (immigrant population in i recently arrived by internal migration / total of the recently arrived immigrant population in i) / (Σi immigrant population recently arrived in i by internal migration / Σi total of the recently arrived immigration population in i) An index superior to 1 indicates an over-representation of immigrants arrived by internal migration relative to those arrived by international migration; an index inferior to 1 indicates an under-representation.
URL http://journals.openedition.org/eps/docannexe/image/12208/img-3.png
Fichier image/png, 216k
Titre Document 11. Spatial distribution of immigrants having undertaken an internal migration to the countryside in metropolitan France, according to birth-place (2011-2015)
Légende Field: totality of immigrants having undertaken an internal migration to the less dense and denser catchment areas of metropolitan France. Source : RP 2013 Note: the typology of densities by catchment area is constructed on the basis of the grid of communal densities of the INSEE.
URL http://journals.openedition.org/eps/docannexe/image/12208/img-4.png
Fichier image/png, 469k
Haut de page

Pour citer cet article

Référence électronique

Julie Fromentin, « Migrations like any other? Internal migrations of immigrants in the French countryside (2011-2015) »Espace populations sociétés [En ligne], Hors-série | 2021, mis en ligne le 06 novembre 2021, consulté le 30 novembre 2021. URL : http://journals.openedition.org/eps/12208 ; DOI : https://doi.org/10.4000/eps.12208

Haut de page

Auteur

Julie Fromentin

Université Paris 1 Panthéon-Sorbonne, INED
julie.fromentin@ined.fr

Haut de page

Droits d’auteur

Licence Creative Commons
Espace Populations Sociétés est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Université de Lille
  • Logo Laboratoire TVES
  • Logo Laboratoire Clersé
  • DOAJ - Directory of Open Access Journals
  • Revue soutenue par l’Institut des sciences humaines et sociales du CNRS
    CNRS - Institut national des sciences humaines et sociales
  • OpenEdition Journals
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search