Navigation – Plan du site
2000
123

Impact of migration, natural population change and age composition on the redistribution of the population in Sweden 1970 - 1996

Effets respectifs des migrations, de l'accroissement naturel et de la composition par âge sur la redistribution de la population en Suède en 1970-1996
Johan Håkansson

Résumés

Les objectifs de cette étude sont d'analyser dans un premier temps l'importance de la migration et des variations géographiques en termes de fertilité et de mortalité sur la redistribution de la population entre les municipalités de 1970 à 1996 en Suède ;puis les effets de la composition par âge des communes relativement aux effets des taux de fertilité, de mortalité et de migration sur les changements dans la répartition de la population . La redistribution actuelle de sa population est comparée à la redistribution générée par un ensemble de scenarii. Les scenarii s'appuient sur une projection statique annuelle de la population pour chaque commune. Dans chaque scénario, les effets de l'accroissement naturel et de la migration sont calculés en premier. Puis les effets des différences régionales en termes de composition par âge ainsi que les effets des différences de fertilité et de mortalité et de taux de migration par classe d'âge ont été calculés. L'analyse spatiale est basée sur l'indice de Hoover (Hoover, 1941) et sur une classification des communes suédoises. Un des résultats est que la population en Suède se concentre spatialement pendant la période considérée. Un autre résultat est que le facteur majeur derrière la concentration en Suède repose sur les effets des variations géographiques en termes de composition par âge sur les taux de mortalité et de fertilité.

Haut de page

Notes de l’auteur

Age-specific fertility and mortality rates are not documented before 1974 by Swedish Statistics. An ex-traction back to 1970 was necessary in order to enable the use of 70 as the starting point for the study

Texte intégral

Introduction

1Since the 1970s major demographic changes such as decreased fertility rates, ageing populations and changed migration patterns have swept various European countries. As a consequence of this politicians, city planners and the general public have become more concerned about possible effects on the economy and on the quality of life for people living in different places. Demographical issues have begun to regain an important place on political agendas (Blotefogel 1997). This study focuses on what impact migration, natural population change and age composition has had on the population redistribution at the municipality level in Sweden between 1970 and 1996.

2In the 1970s, the changing migration patterns in many western developed countries included the counterurbanisation process with a net in-migration to small towns and to rural areas and net out-migration from larger cities (Beale 1975, Berry 1976, Vining & Kontuly 1978, Fielding 1982, Ahnström 1986). In the 1980s and 1990s the counterurbanisation process was retained in some countries but in others it was once again a net in-migration to the larger urban areas (Champion 1989, 1998, Long & Nucci 1997, Kontuly 1998). Migration patterns in Sweden have in a similar way followed these international trends in the 1970s (Ahnström 1980, Forsström & Olsson 1982). In the first half of the 1980s the larger urban areas had a net in-migration but they had a net out-migration in the later half of the decade (Borgegård, Håkansson & Malmberg 1995). In the 1990s net in-migration to larger cities started to increase once more (statistics Sweden).

3As a consequence of the counterurbanisation process, the population should become more dispersed in its geographical distribution. However, concerning the overall population redistribution there is variation between countries. In the US the distribution of the population has become more dispersed (Vining & Strauss 1977, Long & Nucci 1997), whereas the opposite has been true in Sweden (Borgegard & Håkansson 1997). These differences between countries are due to the fact that other processes, such as international migration and natural population change, affect the redistribution of people as well.

4Migration, rather than fertility and mortality are often seen as the main ingredient of regional demographic change. However, the immediate impact of migration on the redistribution of people depends on the time perspective chosen. In a short-term perspective annual variation in-migration tends to be big. But over longer time these fluctuations tend to outweigh each other’. The immediate importance of migration therefore often seems to be overestimated. Regional rates of fertility and mortality are primarily a result of regional variation in age composition. Distribution of age groups in different areas at a certain point in time is to a large extent the result of in- and out-migration during former years. If these variations are large the natural increase could very well have a large impact on the redistribution of the population for a long time.

5Studies on gross migration emphasise that the frequency and destination of migration is highly correlated to age (Thomas 1938, Lowry 1966, Rogers 1979). It is well known that young people tend to migrate more often and choose urban destinations whereas families with children and the elderly move more seldom and have counterurban preferences. As a result of this geographical variation in age composition has to be seen as an important precondition for the migration pattern. Since migration frequencies are age related it could be expected that the age composition has an impact on the redistribution of the population. Several studies also show how changes in the size of cohorts in some countries affect the short-term redistribution of the population since 1970 (Plane 1992, Ishikawa 1999). These studies analysed temporal effects of the change in age composition on the redistribution of people. However, what is unclear is the long-term outcome of regional differences in age composition.

6The aims of this study are to analyse : (1) the importance of migration and geographical variations in fertility and mortality for the redistribution of population between municipalities in Sweden from 1970 to 1996 ; and (2) the impacts of age composition in municipalities versus local age-specific fertility, mortality and migration rates on changes in population distribution.

7The following questions will be addressed in this study :

  • 1.a To what extent does the redistribution of the population depend on natural population change ?

  • 1.b To what extent are regional fertility and mortality rates a result of age composition ?

  • 2.a To what extent does the redistribution of the population depend on migration ?

  • 2.b To what extent is in- and out-migration a result of age composition ?

8The paper consists of five parts. In the following section the data, analytic structure of the study and methodology for calculating the redistribution of migration, natural increase and age composition are presented. In section three some major population trends in Sweden are described. Results of the analyses are shown in the fourth part and finally some conclusions on the redistribution of the Swedish population between 1970 and 1996 are presented.

Method

9This study examines the redistribution of the Swedish population at the municipal level between 1970-96. The population change for the period 1970-96 was also calculated as if it was only affected by one demographic factor at the time. The results of these counterfactual scenarios were compared to the actual population change and the effect of fertility, mortality, in- and out-migration and age composition was thereby isolated.

10All analyses are based on the population data of the 284 Swedish municipalities. For each year between 1970 and 1996 data on population size, fertility and mortality, in- and out-migration, in five age groups (0-15, 16-24, 25-44, 45-64, 65+) by sex are used. Statistics Sweden has produced the data.

11The scenarios rely on a static annual population projection. An alternative analysis would be to also scrutinise the impact of for instance migration on age distribution and the effect of age distribution on fertility etc. However these cumulative dynamic effects are not considered.

12To isolate the effect of in- and out-migration the population change in each municipality is forecasted as if it was only generated by net migration. In order to isolate the effect of fertility and mortality, the population trend in each municipality is forecasted as if it was only generated by natural population change.

13The following two basic scenarios have first been calculated for the population trends of the examined municipalities in this study.

  • Population trends without natural increase to provide information about the impacts of in- and out-migration.

  • Population trends excluding migration showing merely the importance of fertility and mortality patterns.

14To isolate the effects on changes in the geographical distribution of the population due to spatial differences in age composition, the age-specific national average fertility, mortality and migration rates were assigned to each age group in all municipalities. The outcome, the single effect caused by geographical variations in age composition was then used to calculate counterfactual population trends for the municipalities.

15In order to isolate the effects caused by geographical variations in age-specific fertility, mortality and migration rates an average age structure, equal to the national, was assigned for each municipality. Then age-specific fertility, mortality, in- and out-migration rates equal to national average were also assigned to each municipality. As a result of this, each municipality had figures that correspond to a situation in which both regional differences in age composition and other non-age compositional conditions were controlled. The outcome, the single effect caused by regional variations in non-age compositional conditions was also used to calculate counterfactual population trends for the municipalities.

16Based on this, eight more counterfactual scenarios (scenarios 3-10) have been calculated that provide information about the impact of age-composition alone versus other conditions alone on fertility, mortality and migration and furthermore scrutinise their impact on the changing geographical distribution of population.

  • Population trends only considering the effect of geographical variations in age composition on fertility rates.

  • Population trends only considering the effect of geographical variations in age-specific fertility rates.

  • Population trends only considering the effect of geographical variations in age composition on mortality rates.

  • Population trends only considering the effect of geographical variations in age-specific mortality rates.

  • Population trends only considering the effect of geographical variations in age composition on in-migration rates.

  • Population trends only considering the effect of geographical variations in age-specific in-migration rates.

  • Population trends only considering the effect of geographical variations in age composition on out-migration rates.

  • Population trends only considering the effect of geographical variations in age-specific out-migration rates.

The locational analysis

17In order to analyse the consequences of the counterfactual scenarios for municipal population trends two methods have been used in the study, the Hoover index (Hoover 1941), which measure the degree of concentration in the geographical distribution of the population, and a classification scheme of the municipalities. The Hoover index is estimated according to the following formula :

18Ht = 0.5 ∑ ni=1 | pit - ai | 100

19where Ht is Hoover index at the point t in time, pit, the relative share of the population in the country living in area i at time t, and ai is area i relative areal share of the whole country’s area. Thus the index measures the difference between the municipality’s share of the population and share of the country’s area. The index gives values between 0-100, where 0 equals an evenly spread population and 100 equals a total concentration. A value of 50 would imply that 50% of the population would have to live elsewhere if the population should be evenly distributed. A change from 50 to 40 shows that the population had increased more in more sparsely populated regions than in more densely populated regions.

20How the municipalities were categorised is shown in Figure 1. Municipalities belonging to the three big city regions of Sweden (on average more than 316 inhabitants/km2) are classified as ‘metropolitan’ regions. Municipalities with Universities or University Colleges with more than 2.000 students in 1992 belong to the category university municipalities (approximately 49 persons/km2). Medium sized municipalities is the classification term for municipalities with cities with populations larger than 20.000 (about 36 people/km2). These are three densely populated categories. Industrial municipalities are those with more than 20% of their populations of working age employed in manufacturing or heavy industry in 1980 (19 inhabitants per km2). Municipalities belonging to EU Objective 6 programs are categorised as Inner Norrland, "the non-coastal areas of Northern Sweden" (with a maximum of 2 people/km2). Those municipalities that fall outside these mentioned categories sort under other municipalities (about 20 persons/km2). These latter three classifications are all sparsely populated municipalities, in particular Inner Norrland whilst the other two are pretty similar to one another in this respect.

Figure 1 : The Swedish municipalities divided into six different categories.

An empirical background to the redistribution of the population of sweden in 1970-1996

Regional population change until 1970

21Sweden experienced a relatively late urbanisation process. In the 1930s approximately 50% of the population lived in the countryside. After the 2nd World War the restructuring of manufacture industry attracted labour to urban areas. In-migration to cities culminated in the 1960s (Swedish National Atlas 1991, Falk 1976). By 1970 the process of urbanisation had resulted in nearly 80% of the population living in cities.

Table 1 : The share (in per cent) of total population in different categories of municipalities in 1970

22In 1970, 56% of the Swedish population was localised in the most densely populated regions (metropolitan, university, and medium-sized municipalities) whereas as few as 5% lived in the most sparsely populated area, Norrland’s inland (Table 1). The other two categories, the industry municipalities and the other municipalities, accounted for about 40% of the population.

Figure 2 : The age composition in different categories of municipalities compared with the national age composition in 1970 (index 100 equals the national age composition)

23In the general migration pattern important age related variations are found. The higher migration propensity among the young and the tendency to migrate urbanward resulted in regional variations in age composition (Figure 2). In 1970 there was an over-representation of elderly people in the more sparsely populated areas, whereas there was an overrepresentation of young and middle-aged people in the metropolitan areas and the University municipalities.

Migration and natural population change

24The migration propensity has decreased in Sweden as in many other countries for a long time (Bengtsson & Johansson 1992, Fischer et. alia 1998). This trend is evident in all age categories (Bengtsson & Johansson 1992). In total, the number of people migrating across municipal borders decreased from 5.1% (about 400.000) to 4.5% (about 400.000) of the population between 1971 and 1996.

25Fertility rates increased in the 1980s marking a departure from previous trends. In 1990 this baby boom reached its peak with 124.000 births. Since then birth rates have been on decline in the country. Mortality figures have been stable with about 95.000 deaths a year. Recently this number has increased slightly. The average age is however increasing, as in many other European countries.

Figure 3 : The net-redistributional effect (absolute sums) of net migration and natural change across municipalities in Sweden between 1970 and 1996.

26Only a small proportion of the gross migration has an impact on the redistribution of the population. Out of 400.000 migrants in 1970 only about 90.000 (approximately 1% of the population) of these had an impact on population redistribution (Figure 3). The net-redistribution due to migration decreased during the following decades to approximate 50.000 people in 1996. As shown above, the gross number of migrants in 1970 equalled the number in 1996. Since the gap between the gross migration and net-migration increase during the period a conclusion is that an decreasing part of the migration in Sweden during the period had effects on the geographical distribution of the population.

27In 1970 the net effect on the population distribution of natural population change constituted of approximately 30.000 persons (0.4% of the total population). Consequently, fertility and mortality rates had a smaller impact on the redistribution of the population than did migration. This figure was slightly reduced further between 1970 and 1996. Since the decrease in number of net migrants has been by far larger, it can be concluded that the natural population change has increased its impact on the redistribution of the population in the country between 1970 and 1996.

Results

The natural population change and migration as concentrating and diffusive variables on the population

28The Swedish population redistribution among municipalities was characterised by an increased concentration between 1970 and 1996 (Figure 4). Measured with the Hoover index this increase is about 2%. The effects of migration and natural population change (scenarios 1 and 2) are also shown in Figure 4. It reveals that the concentration of the population is primarily related to regional differences in fertility and mortality rates. The increase in the Hoover index by natural population change is about 1.5% while the migration only increases the index by about 0.6%.

Figure 4 : The actual degree of population concentration in Sweden measured with Hoover index at the municipality level between 1970 and 1996 as well as the counterfactual degree of concentration generated by natural population change and migration respectively (scenarios 1 and 2).

29It is also revealed that the natural population change has concentrated the population at a constant rate during the whole period, whereas the impact of migration on the Hoover index fluctuated over time. In 1971-1981 and 1988-1990 migration contributed to a dispersion of the population. However this was followed by a concentrated migration pattern during 1982-87 and 1991-96. In the 1990s the index increased by 0.5%, which approximately is equal to the effect of migration on the redistribution of the population over the whole period. The strong concentration of the population in the 1990s is thereby highly correlated with changes in the migration patterns. However an overall conclusion is that natural population change affects the long term trend towards concentration, and migration affects the short term fluctuations over time.

30The population changes in the different municipality classes are illustrated in Table 2. As in Figure 3 it also shows the impact of natural population change and migration (scenarios 1 and 2). It is shown that in municipality classes with the fastest population change, migration as well as natural population change are either negative or positive. However when these two processes counteract one another the population change is limited.

31It is also shown that in both the most densely populated categories and the most sparsely populated ones the effect of natural population change is largest. Natural population change accounts for 73% of the population increase in metropolitan regions and for 63% of the decrease in Norrlands inland. Migration is of special importance for the population change in the industrial and the "other municipalities".

The impact of age composition on the population redistribution

32Figure 5 reveals that age composition has stronger effect on natural population change than on migration. It shows that the age composition has the major effect on fertility and mortality rates (scenarios 3 and 5), whereas geographical variations in age specific migration are most important to the regional in- and out-migration rates. The geographical differences in age composition are consequently more important to the regional fertility and mortality rates whereas differences in age-specific attraction among municipalities are more important to the regional in- and out-migration rates.

Figure 5 : Counterfactual degree of population concentration between 1970 and 1996 generated by the isolated effects of age composition and other non-age compositional conditions divided into the demographic components (scenarios 3-10) measured with Hoover index.

33In Figure 5 it is shown that the effects of age composition on fertility and mortality rates in both cases concentrate the population (scenarios 3 and 5). This is due to the fact that there is a large share of young females in the more densely populated municipalities, whereas there is a large share of old people in the more sparsely populated municipalities. This leads to higher birth rates and a population increase in the more densely populated municipalities, and to higher death rates and a population decrease in more sparsely populated ones. The effect of variations in age-specific fertility rates on the population redistribution is dispersing (scenario 4). The effect of age-specific mortality rates on the population redistribution is concentrating (scenario 6). This means that even though women in more sparsely populated municipalities have more children this is counteracted by the fact that people die earlier in these areas.

34In Figure 5 it is shown that the geographical variations in in-migration rates resulted in concentration while variations in out-migration resulted in dispersion (scenarios 7-10). Both the age-specific attraction and the age-specific out-migration propensity are higher in more densely populated municipalities as compared to more sparsely populated municipalities. The concentrating effect of in-migration is approximately as large as the dispersing effect of out-migration. In- and out-migration therefore more or less counteract one another in the population redistribution. This was not the case for the fertility and mortality. The conclusion drawn from this is that the major explanation for the concentration of the population between 1970 and 1996 is the regional differences in age composition.

Changes in the effects of age composition on the population redistribution

35The changing impacts of age composition and age-specific variations in fertility, mortality and migration rates on the redistribution of the population over time are illustrated in Figure 6. This is demonstrated by the annual change in the Hoover index for scenarios 3-10, where an increase in index is seen for concentration whereas a decline illustrates the opposite. This figure thereby illustrates change and stability as well as the long term effects of age composition and age-specific variations in fertility, mortality and migration rates on the redistribution of the population.

Figure 6 : The annual change in age composition and non-age compositional conditions on the redistribution divided into the demographic components (The annual change in the Hoover index for scenarios 3-10).

36Figure 6 shows that the effects of geographical variations in fertility and mortality1 rates on the spatial distribution of the population are stable over time and that neither variations due to age composition nor age-specific fertility and mortality rates are changing over time. However geographical variations in age-specific in-migration and out-migration rates are unstable factors, which seem to explain a large share of the variations in the population distribution over time. The figure also shows that the effect age composition and age-specific conditions have on the different demographical components over time in most cases are either constantly concentrating or dispersing. There are two exceptions. Both the effect of age-specific out-migration rate and fertility rate alternates during the period between concentration and dispersion.

37The increasing concentration of the population is to a large extent caused by the effects of age composition on the natural population change. Therefore the change in effects of age composition on the fertility and mortality rates is of importance to long-term population redistribution. In the early 1970s regional differences in age composition had its largest impact on the fertility and mortality rates. Between 1970 and 1996 the impact has declined slightly. Neither migration nor changes in the effects of age-specific fertility and mortality rates have thereby changed the regional age composition in any significant way. The regional differences in age composition already in 1970 have therefore affected the redistribution of the population during the period. Consequently, such processes that changed the age composition before 1970 mainly caused the increased concentration of the population during the period. Subsequently, the urbanisation process during previous decades continues to affect the redistribution of the population between 1970 and 1996.

Discussion and conclusions

38The aims of this study are to analyse : (1) the importance of migration and geographical variations in fertility and mortality for the redistribution of population among municipalities in Sweden from 1970 to 1996 ; and (2) the impacts of age composition in municipalities versus local age-specific fertility, mortality and migration rates on changes in population distribution.

39Empirical results presented in the study showed that the redistributional effect of migration and the natural population change had declined over the period. Since the gross migration did not decline as much as the net migration this implies that migration over the period has had a decreasing effect on the geographical distribution of the population. Natural population change has instead become increasingly important for the changing geographical distribution between 1970 and 1996.

40One objective was to find out whether migration and natural population change contributed to concentration or dispersion of the population. It was concluded that both led to a concentration between 1970 and 1996. However, the natural population change affected the long-term trend towards concentration. Migration on the other hand varied between dispersion and concentration and is therefore the major determinant to the fluctuations over time. Migration led to dispersion of the population in the 1970s and the early 1980s, and in this respect it was counterurbanising. However, this effect by migration was counteracted by concentration due to natural population change in the 1970s, and the total effect on the population redistribution was just a concentration at a slower rate.

41The importance of natural population change and migration for the population change varied among different categories of municipalities. In metropolitan and university municipalities and in inner Norrland natural population change has had the largest impact on population change. In all other municipalities migration has been the most important process behind the population change.

42Additional empirical findings were that age composition has played a major role in affecting the regional fertility and mortality rates while non-age compositional conditions have been of greatest importance to the regional in- and out-migration rates. Since the regional in- and out-migration rates have counteracted one another, whereas regional fertility and mortality rates both have affected the redistribution towards concentration, age composition is the major determinant behind the redistribution of the population between 1970 and 1996.

43The annual effect by which age composition concentrates the population through natural population change has only decreased slightly during the period. Neither migration nor natural population change during the period investigated has changed the geographical variations in age composition much. Consequently, the regional patterns of fertility and mortality rates were mainly attributed to the period before 1970. The changing geographical distribution of the population between 1970 and 96 was thus highly effected by the urbanisation process in the period before. This means that we have not yet seen the full consequences of the previous urbanisation on the population’s distribution. Since the dispersion through migration has not changed the regional age composition much, it can be concluded that the population most likely will become further concentrated at the municipality level, even for a relatively long period to come.

44It should be emphasised that this study was conducted with a static method, excluding the impact of cumulative dynamic processes. The dynamics, which the redistribution of the population involves, were therefore not captured. This study revealed the impacts of age composition compared to other non-age compositional conditions on population redistribution. In order to enable more dynamic analyses on this, data on individuals or groups over time are necessary. This would be a valuable complement to static studies such as this one.

45The indication that age composition was a major and stable determinant behind the redistribution of the population points to the necessity of further investigations into the links between migration and natural population change. Another issue that needs deeper investigation is the decreasing redistributional effect of migration.

Haut de page

Bibliographie

Ahnström, L. (1980) : Turnaround-trenden och de nordiska huvudstadsregionernas utveckling efter 1950, NordREFO 1980, no. 3-4.

Ahnström, L. (1986) : The turnaround trend and the economically active population of

seven capital regions in western Europe, Norsk Geografisk Tidskrift 40 : 50-64.

Beale, C. L. (1975) : The revival of population growth in non-metropolitan America, : US Department of Agriculture, ERS 605.

Bengtsson, T. & Johansson, M. (1992) : Befolkningens geografiska rörlighet –Minskar den ?, in L. O. Persson and M. Wångmar (eds.) 90-talets Befolkningsfrågor, Stockholm, Arbetsmarknadsdepartementet, Ds 1992 :16), 41-71.

Berry, B. (1976) : The counterurbanisation process : urban America since 1970, in B. Berry (ed.) Urbanisation and counterurbanisation, Beverly Hills, Sage, 17-30.

Blotevogel, H. H. (1997) : Introduction : Population and Economic Restructuring, in H. Blotevogel and A. J. Fielding (eds.) People, Jobs and Mobility in the New Europe, London, Wiley, 3-21.

Borgegård, L. E., Håkansson, J. & Malmberg, G. (1995) : Population Redistribution in

Sweden –Long Term Trends and Contemporary Tendencies, Geografiska Annaler 77B : 31-45.

Borgegård, L. E. & Håkansson, J. (1997) : Population Concentration and Dispersion in

Sweden since the 1970s, in L. E. Borgegård, A. M. Findlay & E. Sondell (eds.) Population, Planning and Policies, Umeå University, : Cerum Report, no. 5, 9-29.

Champion, A. G. (ed.) (1989) : Counterurbanization –The Changing Pace and Nature of Population Deconcentration, London, Edward Arnold.

Champion, A. G. (1998) : Studying Counterurbanisation and the Rural Population

Turnaround, in P. Boyle and K. Halfacree (eds.) Migration into rural areas –Theories and issues, Chichester, Wiley, 21-40.

Falk, T. (1976) : Urban Sweden. Changes in the Distribution of Population –the 1960s in Focus,

Stockholm , The Economic Research Institute.

Fielding, A. J. (1982) : Counterurbanisation in Western Europe, Progress in Planning 17 : pp 1-52.

Fischer, P. A., Holm, E., Malmberg, G. & Straubhaar, T. (1998) : Why do People Stay ?

The insider advantages Approach –Empirical Evidence from Swedish Labour Markets, Centre for Economic and Policy Research : Discussion paper 1952.

Forsström, Å. & Olsson, R. (1982) : Boende i glesbygd. Göteborg, BFR, R100.

Hoover, E. (1941) : Interstate redistribution of population, 1850-1940, Journal of

Economic History 1 : 199-205.

Ishikawa, Y. (1999) : Contribution of the Demographic Factor to the Migration

Turnarounds in Japan, Sweden and Canada, International Journal of Population Geography 5 : 1-17.

Kontuly, T. (1998) : Contrasting the counterurbanisation experience in European nations, in P. Boyle and K. Halfacree (eds.) Migration into rural areas –Theories and issues, Chichester, Wiley, 61-78.

Long, L and Nucci, A. (1997) : The Hoover index of Population Concentration : A

Correction and Update, Professional Geographer 49 : 431-440.

Lowry, I. S. (1966) : Migration and Metropolitan Growth : Two Analytic Models, San Francisco, Chandler.

National Atlas of Sweden (1991) : The Population, Höganäs, Bra Böcker.

Plane, D. A. (1992) : Age Composition Change and the Geographical Dynamics of Interregional Migration in the U.S., Annals of the association of American Geographers 82 : 64-85.

Rogers, A. (1979) : Migration Patterns and Population Redistribution, Regional Science and Urban Economics 9 : 75-310.

Thomas, D. (1938) : Research Memorandum on Migration Differentials, New York, Social Science Research Council.

Vining, D. R. and Strauss, A. (1977) : A demonstration that the current deconcentration of population in the United States is a clean break with the past, Environment and Planning A 9 : 751-758.

Vining, D. R. and Kontuly, T. (1978) : Population Dispersal from Major Metropolitan

Regions : An International Comparison, International Regional Science Review 3 : 49-73.

Haut de page

Pour citer cet article

Référence électronique

Johan Håkansson, « Impact of migration, natural population change and age composition on the redistribution of the population in Sweden 1970 - 1996 », Cybergeo : European Journal of Geography [En ligne], Systèmes, Modélisation, Géostatistiques, document 123, mis en ligne le 29 février 2000, consulté le 17 octobre 2019. URL : http://journals.openedition.org/cybergeo/5615 ; DOI : 10.4000/cybergeo.5615

Haut de page

Auteur

Johan Håkansson

johan.hakansson@geography.umu.se
Department of Social and Economic Geography, Umeå University, Sweden

Haut de page

Droits d’auteur

Licence Creative Commons
La revue Cybergeo est mise à disposition selon les termes de la Licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 3.0 non transposé.

Haut de page