CAIRN-INT.INFO : International Edition

1Residential migration can be analysed in various ways, focusing either on the behaviour of individuals or households, on their residential choices and strategies, or on the spatial aspects of migratory flows that reveal interactions between different localities (Desplanques, 1994; Baccaïni, 2006).

2Inter-regional migration in France has been widely described and analysed, generally whenever new census results are published. The first round of the new census took place in 2004. It provides an opportunity to examine changes in the spatial organization of inter-regional flows over the last fifty years.

3In each census from 1962 to 1999, a question was asked about the place of residence on 1 January of the year of the previous census. Hence in 1999, respondents were asked about their place of residence on 1 January 1990.

4This question has been modified in the new revised census. Annual census survey respondents are now asked to give their place of residence five years previously. So for the 2004 census, respondents were asked about their place of residence on 1 January 1999. From now on, the reference period will be of constant duration (five years) and shorter than previously (between six and nine years), making it easier to study residential mobility (Baccaïni, 2001b).

5This question serves to enumerate migrants, i.e. persons whose place of residence has changed between the beginning and the end of the period. However, because repeat and return migrations are not counted, the number of migrants is smaller than the actual number of migrations. We will see that models are needed to estimate annual mobility rates (frequency of change of dwelling, commune, département, region) and the annual arrival and departure rates for each region on the basis of these census data.

6With a question on the place of residence on 1 January 1999, the results of the 2004 annual census survey provide sufficient data for accurate analysis of inter-regional migratory flows, since sampling is balanced at this geographical level (Baccaïni, 2005). Note, however, that only internal migration is concerned here, and that only flows between the 22 metropolitan regions are considered. Migration to and from the French overseas départements and territories is not taken into account.

1 – Increased mobility over the last decade

Method

7The census results are used to determine the numbers of migrants, i.e. individuals who lived in another dwelling, commune, département or region at the date of the previous census or, with the new census, five years previously. They cannot be used directly to estimate annual mobility (migration frequency). As individuals may move several times during the period, or move away and then back to their previous place of residence, the number of migrants is smaller than the number of migrations.

8To measure this mobility, we use a formula, based on various parameters, to estimate an instantaneous migration rate (close to the annual rate) from the number of migrants enumerated in the census. This model was developed by Daniel Courgeau (Courgeau, 1973, 1988). The parameters were recently re-estimated by INSEE using labour force surveys from 1991 to 1999 and the 1997 young people and careers survey (L’Hospital, 2001).

After two decades of decline, residential mobility has been increasing since the 1990s

9Until the early 1970s, mobility increased steadily (Figure 1), with the annual inter-communal migration rate rising from below 5% between 1954 and 1962 to more than 6% between 1968 and 1975. The inter-regional rate rose from 1.3% to 1.8% over the same period.

Figure 1

Residential mobility in France

Figure 1

Residential mobility in France

Sources: INSEE, population censuses.

10This increase was followed by a downward trend that continued until the early 1990s and that affected all types of migration, both short- and long-distance, but local migration especially.

11Since the early 1990s, residential mobility has started rising again, and this rise, revealed by the 1999 census, is confirmed by the 2004 annual census survey.

12Between 1982-1990 and 1990-1999, the increase primarily concerned short-distance migration, while between 1990-99 and 1999-2004, it was long-distance migration that increased most sharply.

High mobility among young adults

13Between 1999 and 2004, as in the previous periods, mobility was highest among young adults in their twenties, and started falling beyond that age (Table 1). This is logical, since it is between the ages of 20 and 30 that the causes for residential mobility are most numerous: university enrolment, first employment, union formation, parenthood. This higher mobility of young adults is more pronounced for long-distance migration than for short-distance migration.

Table 1

Residential mobility in metropolitan France between 1999 and 2004, by age in 2004

Table 1
Annual mobility rate (%) Change of Total population (over age 4) 20-29 30-39 40-49 60+ Dwelling 12.0 21.4 18.7 8.5 4.7 Commune 7.3 14.0 11.3 4.9 2.7 Département 3.1 6.9 4.6 2.0 1.2 Region 1.9 4.2 2.7 1.2 0.8 Source: Insee, 2004 annual census survey.

Residential mobility in metropolitan France between 1999 and 2004, by age in 2004

14Between the last two periods (1990-1999 and 1999-2004), the general increase in mobility has only affected individuals under age 40, and has been stronger for long-distance than for short-distance migration. The annual inter-regional migration rate of persons aged 20-29 has risen by 60%, while the mobility of the over-40s has declined, with short-distance migration registering the sharpest fall.

2 – Net migration: a turning point in the 1970s

Major changes in the patterns of net migration over the last fifty years

15The correlation between the periods 1954-1962 and 1962-1968 is large and positive (+0.62). Between 1968 and 1975, the degree of correlation with the period 1954-1962 drops sharply and becomes non-significant (0.16). It becomes negative during the periods which follow. From the 1970s, the pattern of net migration rates thus differed radically from that prevailing in the 1950s and early 1960s. Since the 1970s, the pattern has been more stable, though variations are still observed from one period to the next (Pumain, 1986; Baccaïni, 1993, Baccaïni, 2001a; Baccaïni, 2005).

16During the 1950s (Map 1), Île-de-France (Paris region) was the most attractive region in terms of net migration, and only three other regions had a positive internal migration balance: Provence - Alpes - Côte d’Azur (Paca), Rhône-Alpes and Alsace. The largest losses were in rural western regions: Basse-Normandie, Bretagne, Poitou-Charente.

Map of the regions

figure im3

Map of the regions

Map 1

Net internal migration, 1954-1962

Map 1

Net internal migration, 1954-1962

Sources: INSEE, 1962 population census.

17During the 1960s (Map 2) no real trend reversal was observed, but the Paris region, while maintaining a positive balance, lost much of its appeal. It was overtaken by regions such as Provence-Alpes-Côte d’Azur, Centre, Rhône-Alpes and Alsace. The number of regions with a positive balance increased to seven. At the same time, migration deficits grew larger and started affecting regions in north-eastern France (Lorraine, Nord–Pas-de-Calais) which were facing severe economic problems.

Map 2

Net internal migration, 1962-1968

Map 2

Net internal migration, 1962-1968

Sources: INSEE, 1968 population census.

18The pattern of net migration changed radically in the early 1970s (Map 3). Heralding a new long-term trend, net migration to the Paris region became negative for the first time, with the number of departures exceeding the number of arrivals. Between 1968 and 1975, a majority of regions had a positive balance, and only six regions lost more migrants than they gained. These regions were Lorraine and Nord–Pas-de-Calais, followed by Champagne-Ardenne, Basse-Normandie, Île-de-France and Auvergne. Provence-Alpes-Côte d’Azur was by far the most attractive region in terms of net migration. The entire western side of France, from Bretagne to Aquitaine also had a positive balance, and joined the high-attraction group of regions in the south-east.

Map 3

Net internal migration, 1968-1975

Map 3

Net internal migration, 1968-1975

Sources: INSEE, 1975 population census.

19This pattern has persisted up to the present day, though some notable changes have occurred (Maps 4, 5, 6 and 7). Little by little, the negative balances have become concentrated in a group of regions in northern France, from Basse-Normandie to Franche-Comté, and including Île-de-France. Since the early 1990s, the negative net migration rate has been highest in the Paris region, while the appeal of the south-western and western regions has increased greatly. Between the periods 1990-1999 and 1999-2004, the differences between regions have tended to widen. All the regions with a negative balance in the 1990s (eight regions of northern France, from Basse-Normandie to Franche-Comté, excluding Alsace) became even less attractive in the early 2000s, while those with a positive balance became even more so. This dividing line between the strong draw of southern and western France on the one hand, and the weak draw of northern and north-eastern France on the other, is especially pronounced today.

Map 4

Net internal migration, 1975-1982

Map 4

Net internal migration, 1975-1982

Sources: INSEE, 1982 population census.
Map 5

Net internal migration, 1982-1990

Map 5

Net internal migration, 1982-1990

Sources: INSEE, 1990 population census.
Map 6

Net internal migration, 1990-1999

Map 6

Net internal migration, 1990-1999

Sources: INSEE, 1999 population census.
Map 7

Net internal migration, 1999-2004

Map 7

Net internal migration, 1999-2004

Sources: INSEE, 2004 population census.

The regions have unequal appeal for different age groups

20A generally attractive region may, in fact, only attract certain age groups, or some more than others. The same applies to unattractive regions, which may have a positive balance for certain age groups. Indeed, young adults, families or older people are attracted by very different criteria. The first tend to be drawn towards regions with a good university infrastructure and plentiful first employment opportunities, while families and older people give priority to the living environment.

21The net migration rates by age between 1999 and 2004 divide the regions into six main types (Map 8). Among the eleven regions with an overall deficit, Île-de-France and Alsace are characterized by a positive balance of young adults aged 20-29, and a negative balance at all other ages. The Paris region (and to a lesser extent Alsace) still attracts young adults who come to study or find their first job. It is for the populations aged 30-39 and 60+ that the Paris region has the least appeal. This situation was already visible in the Paris region during the previous periods.

Map 8

Status of regions in terms of net internal migration by age, 1999-2004

Map 8

Status of regions in terms of net internal migration by age, 1999-2004

Sources: INSEE, 2004 annual census survey.

22Three regions to the south and west of Paris – Bourgogne, Centre and Basse-Normandie – are attractive for people aged over 40, though their overall balance is negative. The positive balance for the highest age groups in these regions is due mainly to a net inflow of migrants from the neighbouring Paris region.

23For the six north-eastern regions with overall deficits (excluding Alsace), ranging from Haute-Normandie to Franche-Comté, departures outnumber arrivals at all ages, and at ages 20-29 especially.

24The same type of age profile, with a dip at ages 20-29, is observed in the western regions, though internal net migration remains globally positive. From Bretagne to Aquitaine, via Limousin and Auvergne, there are more arrivals than departures at all ages, except for young adults aged 20-29. These regions are attractive mainly for families and older people, and much less so for young adults who move away to study or work in regions offering more opportunities. Corsica has a similar profile.

25Only three southern regions, Provence-Alpes-Côte d’Azur, Languedoc-Roussillon and Midi-Pyrénées, have positive balances at all ages. They are especially attractive to the 30-39 age group.

26Though overall net migration is positive in the Rhône-Alpes region, it has the same type of profile as the Paris region and Alsace in terms of age: the region attracts migrants in the 20-29 age group above all. However, unlike the Paris region, Rhône-Alpes also has a positive balance for the 30-39 age group and, to a lesser extent, for 40-59 year-olds. Only among the over-60s do departures outnumber arrivals.

27It is the migration balance of young adults (age 20-29) which displays the most pronounced differences with respect to the population as a whole (Map 9). They prefer to head for the Paris region and, to a lesser degree, to the south (Midi-Pyrénées, Provence-Alpes-Côte d’Azur, Rhône-Alpes, Languedoc-Roussillon) and to Alsace. Young adults are drawn to urban regions with good university infrastructure and a large and varied labour market. Moreover, as this age group is also more mobile, these movements have a considerable impact on volumes and flows.

Map 9

Status of regions in terms of net internal migration by age, 1999-2004

Map 9

Status of regions in terms of net internal migration by age, 1999-2004

Sources: INSEE, population censuses.

3 – Arrivals and departures: “profiles” that vary from one region to another

28The migration balance of a region may correspond to a varied range of arrivals and departures. Likewise, its variation over time may be the consequence of changes that affect these two components differently. For example, an improved migration balance may be due to an increase in arrivals and/or fewer departures. Several types of “profile” exist therefore. Some representative examples are given below.

29The Paris region has a very specific profile that sets it apart from the other regions (Figure 2a). The drop in net migration from 1962 to 1975 is due mainly to a sharp increase in departures, with arrivals remaining practically stable. Between 1975 and 1982, departures levelled off, but the number of arrivals decreased. The less negative migration balance between 1982 and 1990 is due exclusively to fewer departures, since the number of arrivals remained constant, while the downswing since 1990 reflects a sharp rise in departures not counteracted by any significant change in arrivals. Over these fifty years, net migration primarily reflects the Paris region’s ability or inability to keep its inhabitants rather than its capacity to attract provincial migrants.

Figure 2

Arrivals and departures from selected regions, 1954-2004

Figure 2

Arrivals and departures from selected regions, 1954-2004

Figure 2 (cont’d)

Arrivals and departures from selected regions, 1954-2004

Figure 2 (cont’d)

Arrivals and departures from selected regions, 1954-2004

30Among the 3.8 million persons aged above 4 in 2004 who moved to a different region between 1999 and 2004, almost 40% entered or left the Paris region. In the early 2000s, though the Paris region has a negative inter-regional migration balance in overall terms, the number of arrivals from five French regions – Nord–Pas-de-Calais, Lorraine, Franche-Comté, Champagne-Ardenne and Alsace – is higher than the number of departures to these same regions (Figure 3). In other words, the regions of north-eastern France send more migrants to Île-de-France than they receive in the opposite direction. In fact, flows between these regions are relatively small in both directions. For the Paris region, the balance with the sixteen other regions is negative. Generally speaking, the regions achieving the highest positive balance with respect to the Paris region are in the south, south-west and west of France. In terms of intensity, the largest flows from the Paris region are to the neighbouring regions of Centre and Picardie. These flows reflect a pattern of periurbanization beyond the regional boundaries.

Figure 3

Migration balance between Île de France and other regions, 1999-2004

Figure 3

Migration balance between Île de France and other regions, 1999-2004

Sources: INSEE, 2004 annual census survey.

31The regions of northern France also have a specific profile (Figure 2b). A sharp upsurge in departures between 1962 and 1975 had a negative impact on the migration balance, only partially offset by a slow increase in arrivals. Between 1975 and 1982, fewer departures were recorded and the balance improved. Between 1982 and 1990 departures started rising again while arrivals remained stable. Net migration fell as a result. During the 1990s, the migration balance improved thanks to an upturn in arrivals, but since the end of the 1990s, departures have been increasing once more. Generally speaking, periods of rapid decline in net migration correspond to periods of rising departures rather than decreasing arrivals, since arrivals have increased over the last fifty years.

32The western regions, of which Bretagne is an example, have followed an opposite pattern to that of the Paris region (Figure 2c). From 1962 to 1975, the migration balance swung rapidly from negative to positive, due entirely to an increase in arrivals. Departures remained stable. Between 1975 and 1982, arrivals levelled off and departures decreased. The slight fall in net migration between 1982 and 1990 is due to a smaller number of arrivals. The sharp increase in net migration since 1990 can be attributed to an increase in arrivals, as in the 1960s.

33Languedoc-Roussillon (Figure 2d) has a profile similar to that of Bretagne, though the level of net migration is much higher, as is the case in many southern regions. The rapid rise in net migration between 1962 and 1975 is due to a sharp increase in arrivals, with departures rising much more slowly. Between 1975 and 1982, on the other hand, the increase can be explained by a substantial drop in the number of departures. A higher increase in departures than in arrivals accounts for the small drop in net migration observed during the 1990s. Since 1999, on the contrary, the number of arrivals has risen faster than the number of departures, resulting in higher net migration.

34The regions surrounding the Paris region, such as Centre for example, have yet another profile (Figure 2e). Net migration increased rapidly between 1962 and 1975 because the number of arrivals grew faster than the number of departures. From 1975 to 1982, arrivals and departures declined simultaneously, leaving net migration unchanged. In the 1980s, the balance fell slightly due to a drop in the number of arrivals while departures remained stable. Since the early 1990s, net migration in this region has fallen steadily, becoming negative since 1999-2004, due to a faster increase in departures than in arrivals.

35A final example of a specific profile is that of the Rhône-Alpes region (Figure 2f). Since the 1950s, the mean net migration rate has remained remarkably stable. Between 1962 and 1975, the number of arrivals and departures increased, but practically in parallel, so that while total migration increased, net migration did not change. Between 1975 and 1982, the number of arrivals and departures decreased, but again in parallel. Between 1982 and 1990, net migration rose slightly, due to slightly fewer departures and larger number of arrivals. Since the early 1990s, departures and arrivals have been increasing, but not always at the same rate, resulting in lower net migration between 1990 and 1999 (a faster increase in departures), followed by an upturn in recent years (a faster increase in arrivals).

Conclusion

36The internal migration trends observed in France between 1954 and 2004, for a change of dwelling, commune, département or region, all followed the same pattern, i.e. increased mobility between 1954 and 1975, followed by a trend reversal from 1975 to 1990. Since the early 1990s, residential mobility has been increasing again.

37If we observe net migration in the different regions, and analyse patterns of arrival and departure separately, we see that major changes have taken place since the 1950s, following a more complex logic than that of overall mobility. The direction of migration flows has changed radically over these fifty years, with previously attractive regions losing their appeal, and vice versa. The most striking change is probably the reversed situation of the Paris region which, over half a century, has fallen from the most attractive to the least attractive region in terms of net migration. We have seen that declining net migration in the Paris region is due primarily to a growing number of departures to other regions, with arrivals remaining practically stable. Leavers are mainly retired people and families, who are departing the region in growing numbers, while young adults are still drawn to the capital to study or find their first job. This change testifies to the appeal of the southern and western regions of France, now the country’s most attractive migrant destinations.

English

Abstract

Data from successive censuses give a picture of internal migration in France over the last fifty years. After twenty years of decline, residential mobility has been increasing over the last ten years, among the under-40s especially. The patterns of regional migration have changed radically since the 1950s. The Paris region, once the most attractive region of France (with strongly positive net migration) is now the least attractive, with many more departures than arrivals. The western regions, on the other hand, which posted a negative migration balance in the 1960s, are attracting an increasing number of migrants. The north-east remains unattractive, while the south has maintained a strongly positive migration balance. This overall picture varies by age group however, and the Paris region is still the most attractive region for the 20-29 age group. Net migration varies as a result of changes in the numbers of arrivals and departures, and these two components follow different patterns from one region to another. In the Paris region for example, the decrease in net migration over the last fifty years is due mainly to a sharp increase in the number of departing Parisians, since the number of migrants arriving from the provinces has fallen only slightly.

Français

Résumé

L’analyse des recensements successifs permet de dresser un tableau des migrations internes en France depuis cinquante ans. Après vingt ans de baisse, la mobilité résidentielle augmente depuis une dizaine d’années, en particulier pour les personnes âgées de moins de 40 ans. Des changements importants sont intervenus depuis les années 1950 dans la configuration des soldes migratoires régionaux. L’Île-de-France est ainsi passée du statut de région la plus attractive (ayant un fort excédent migratoire) à celui de région la moins attractive (caractérisée par un fort déficit migratoire). À l’inverse, les régions de l’Ouest, qui enregistraient un solde migratoire fortement négatif dans les années 1960, ne cessent de gagner en attractivité. Le Nord-Est reste peu attractif, alors que tout le Sud conserve un excédent migratoire élevé. Ce schéma global varie toutefois selon les classes d’âges. En termes de solde migratoire, l’Île-de-France reste ainsi la région la plus attractive pour les 20-29 ans. L’évolution des soldes migratoires résulte de modifications intéressant les entrées et les sorties, ces deux composantes jouant différemment dans les évolutions régionales. Ainsi, en Île-de-France, la baisse du solde migratoire interne depuis cinquante ans est principalement le résultat d’une forte hausse des départs de Franciliens, les arrivées de provinciaux ayant, pour leur part, peu diminué.

Español

Resumen

El análisis de los censos sucesivos permite elaborar un cuadro de las migraciones internas en Francia desde hace cincuenta años. Tras veinte años de baja, la movilidad residencial viene aumentando desde hace unos diez años, en particular para las personas de menos de 40 años. Han tenido lugar cambios importantes desde los años 1950 en la configuración de los saldos migratorios regionales. Es así como Île-de-France pasó de tener un estatuto de región más atractiva (con un gran excedente migratorio) al de región menos atractiva (caracterizada por un elevado déficit migratorio). A la inversa, las regiones del Oeste, que registraron un saldo migratorio muy negativo en los años 1960, no dejan de ganar atractivo. El Noreste sigue siendo poco atractivo mientras que todo el Sur conserva un excedente migratorio elevado. Este esquema global varía sin embargo según la edad. En términos de saldo migratorio, Île-de-France sigue siendo así la región más atractiva para las personas de 20-29 años. La evolución de los saldos migratorios resulta de modificaciones que interesan a las entradas y salidas, las cuales intervienen de modo distinto en las evoluciones regionales. De ese modo, en Île-de-France, la baja del saldo migratorio interno desde hace cincuenta años es principalmente el resultado de una fuerte alza de las salidas de los habitantes de Île-de-France, habiendo disminuido poco, por su parte, las llegadas de los habitantes de provincias.

References

  • OnlineBaccaïni B., 1993, “Régions attractives et régions répulsives entre 1982 et 1990. Comparaison avec la période 1975-1982 et spécificités des différentes classes d’âges”, Population, no. 6, pp. 1791-1812.
  • OnlineBaccaïni B., 2001a, “Les migrations internes en France de 1990 à 1999 : l’appel de l’Ouest”, Économie et Statistique, no. 344, pp. 39-79.
  • Baccaïni B., 2001b, “Première approche de l’étude des migrations avec le recensement rénové”, Journal de la Société Française de Statistique, no.142-3, pp. 31-39.
  • Baccaïni B., 2003, “Les migrations interurbaines entre 1990 et 1999” in Marie-Flore Mattei and Denise Pumain (eds.), Données urbaines, 4, pp. 323-340, Paris, Anthropos, 433 p.
  • Baccaïni B.,2005, “Enquêtes annuelles de recensement : résultat de la collecte 2004 – Des changements de région plus fréquents qui bénéficient aux régions du Sud et de l’Ouest”, Insee Première, no. 1028, 4 p.
  • Baccaïni B.,2006, “Observation et concepts en matière de migrations internes”, in Caselli G., Vallin J., Wunsch G. (eds.), Démographie : analyse et synthèse, Vol. VIII : Observation, méthodes auxiliaires, enseignement et histoire de la discipline, Paris, Ined, pp. 65-88.
  • OnlineCourgeau D., 1973, “Migrants et migrations”, Population, no. 1, pp. 95-130.
  • OnlineCourgeau D., 1988, Méthodes de mesure de la mobilité spatiale : Migrations internes, mobilité temporaire, navettes, Paris, Ined, 301 p.
  • OnlineDesplanques G., 1994, “Connaître les migrations”. Espace – Populations – Sociétés, no. 1, pp. 31-39.
  • L’Hospital F., 2001, “Les migrations internes en France. Estimation des paramètres du modèle ‘migrants-migrations’ de Daniel Courgeau”, Document de travail, Insee Rhône-Alpes, 61 p.
  • OnlinePumain D., 1986, “Les migrations interrégionales de 1954 à 1982 : directions préférentielles et effets de barrière”, Population, no. 2, pp. 378-388.
Brigitte Baccaïni [*]
Insee Provence-Alpes-Côte d’Azur, 17 rue Menpenti, 13387 Marseille Cedex 10, tel: +33 (0)4 91 17 58 15
  • [*]
    Institut national de la statistique et des études économiques (France).
    This paper was presented at the Entretiens Jacques Cartier in December 2005 (Choix résidentiels conference).
    Translated by Catriona Dutreuilh
Translated by
Catriona Dutreuilh
This is the latest publication of the author on cairn.
Latest publication on cairn or another partner portal
Uploaded on Cairn-int.info on 03/03/2014
Cite
Distribution électronique Cairn.info pour I.N.E.D © I.N.E.D. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit.
keyboard_arrow_up
Chargement
Loading... Please wait