untitled Contemporary population changes in north Swedish rural areas ÖRJAN PETTERSSON Pettersson, Örjan (2001). Contemporary population changes in north Swedish rural areas. Fennia 179:2, pp. 159–173. Helsinki. ISSN 0015-0010. During the 1990s most municipalities and rural areas in northern Sweden have experienced renewed depopulation. At the same time, many regional centres (mainly university cities), but also some rural areas, have shown significant population growth. This article focuses on the latter rural areas and describes their geographical location and the socio-demographic characteristics of their populations. Three types of rural areas with population growth have been iden- tified. Firstly, there are the rural areas within daily commuting distance from regional centres. Secondly, there is a group of rural areas that has benefited from the tourist industry. Most of these tourist resorts are mountain villages close to the border with Norway. Finally, there are also a few rural areas char- acterised by attractive residential environments and leisure housing. Örjan Pettersson, Department of Social and Economic Geography, Umeå Uni- versity, SE-90187 Umeå, Sweden. MS received 19th June, 2001 (revised 29th October, 2001) Rural areas going against the stream of depopulation In Sweden, the 20th century has implied a major shift in the population distribution, away from the rural areas to the urban areas. This urbanisation process has primarily taken place within the re- gions, i.e. from surrounding rural areas to nearby localities and cities, and to much less extent meant a redistribution of the population between different parts of the country (Andersson 1987; Borgegård et al. 1995). However, the patterns concerning population concentration and disper- sion observed at different scales of spatial resolu- tion have shifted several times since the begin- ning of the 19th century (Håkansson 2000). Al- though the main trend in Sweden has been to- wards population concentration and urbanisation, there have also been shorter periods characterised by dispersion and counterurbanisation, in Sweden especially manifested during the ‘green wave’ of the 1970s. Furthermore, while most rural areas have experienced population decline in recent years, there have nevertheless been rural areas with substantial in-migration and population growth. There are also examples of expanding ru- ral areas in depopulating regions, and vice versa. However, such tendencies are difficult to trace with statistical data for counties and municipali- ties. By using socio-demographic data with a higher degree of spatial resolution it is though possible to analyse changes within the municipal- ities. The purpose of the article is to identify and de- scribe the rural areas in northern Sweden that have experienced population growth during the past decades and also to emphasise some of the characteristics of these rural areas. The main fo- cus is on the period after 1985, in particular the first half of the 1990s. Earlier studies of population changes in Swed- ish rural areas have often dealt with the whole country or focused on rural areas within compar- atively densely populated metropolitan regions. By contrast, northern Sweden exhibits substantial variation between on the one hand moderately large cities and densely populated rural areas mainly along the coast and on the other hand large tracts of extremely sparsely populated areas in the interior parts of the region. Even though the article primarily deals with the situation in north- ern Sweden the results should be of some rele- vance in other sparsely populated regions in the Nordic countries, especially regarding Finland with many similarities in settlement patterns and population trends (see for instance Tervamäki 160 FENNIA 179: 2 (2001)Örjan Pettersson 1987; Vartiainen 1989a; Karjalainen 1989; Räisänen et al. 1996; Rusanen et al. 1997; Wi- berg 2000). The tendencies towards a population turna- round, i.e. counterurbanisation, in many devel- oped countries all around the world have gener- ated an extensive literature. The counterurbanisa- tion phenomenon has been studied and discussed in numerous publications. Berry (1976) was one of the pioneers. Other often cited contributions have been written by Fielding (1982), Cochrane and Vining (1988), Vartiainen (1989b), Geyer and Kontuly (1993) and Champion (1989, 1992, 1994). More recent publications include for in- stance Halfacree (1997), Champion (1998), Kon- tuly (1998), Lewis (1998), Dahms and McComb (1999), Ford (1999) and Stockdale et al. (2000). There have also been a number of contributions dealing with the situation in the Nordic countries, for instance Ahnström (1980, 1986), Court (1989), Hansen (1989), Vartiainen (1989a), Borgegård et al. (1993), Illeris (1994), Borgegård and Håkans- son (1997), Amcoff (1997, 2000) and Westlund and Pichler (2000). In the article’s first section fo- cus is on Swedish literature relating to the phe- nomenon and various explanations to these ten- dencies. Main emphasis is on Swedish house- holds’ motives for moving to the countryside. In the next section there is a presentation of contem- porary population changes in northern Sweden. Finally, we take a closer look on certain rural ar- eas in the two counties of Dalarna and Västerbot- ten (see Fig. 1). The county of Dalarna was chosen since it contains some traditional rural municipal- ities with surprisingly high in-migration and rela- tively steady population growth since the 1970s. The choice of Västerbotten was due to the coun- ty’s large internal differences in population den- sity and development. Within the county there is also great variety in residential environments; from cities and villages in agricultural districts along the coast to scattered towns and settlements in forest dominated landscapes or even mountain- ous surroundings in the interior parts of the re- gion. Counterurbanisation in Sweden During the late 1980s in Sweden there were signs of a renewed interest in living in the countryside. Based on data from Statistics Sweden it is possi- ble to conclude that the 1980s was the first dec- ade in a very long time that had an absolute in- crease in the numbers of inhabitants living in ru- ral areas (Carlquist 1992). This trend shift led to speculations about the occurrence of a ‘new green wave’ similar to the ‘green wave’ of the 1970s and similar to the counterurbanisation trend observed in several other countries. It was in the metropolitan regions that these tendencies were most obvious, but there was also growth in rural population in some sparsely populated counties. The last delimitation of urban and rural areas was performed in 1995 (Statistics Sweden 1997). In Sweden urban areas are defined as localities with at least 200 inhabitants. By definition all ar- eas outside these localities are rural areas. Since 1990 these rural areas can be further subdivided statistically into rural agglomerations (with 50– 199 inhabitants) and other rural areas (i.e. very sparsely populated areas and scattered settle- ments). The last delimitation of urban and rural areas seemed to indicate that there had been a return to rural depopulation for the period 1990– 1995. This change, however, was mainly due to reclassification of rural areas into localities, thus in reality implying continued population growth in villages and small towns (Statistics Sweden 1997). Some researchers claim that there is a ten- dency towards polarisation with increasing pop- ulation growth in the major cities and in rural ar- eas, whereas small and medium-sized towns are growing at a slower rate or even losing inhabit- ants (Amcoff 2000; Westlund and Pichler 2000). The renewed interest for living in the country- side during the 1980s has been explained in Swe- den by changes related to economic development and various structural phenomena, for instance the incentive to move to the metropolitan regions diminish during recessions (Johansson and Pers- son 1991). Furthermore, the expansion of the pub- lic sector and various regional policy measures resulted in improved service accessibility and new job opportunities in sparsely populated regions (Borgegård et al. 1995). Other researchers main- tain that the changes are due to a shift in the gen- eral sense of values and residential preferences (Eriksson 1989; Forsberg and Carlbrand 1993). In a post-industrial society it becomes more impor- tant with a pleasant housing environment than closeness to place of work (Bengtsson and Johans- son 1992), whereas technological advances and commuting make a more dispersed settlement pattern possible (The Population 1991). Other as- pects are that different generations have different FENNIA 179: 2 (2001) 161Contemporary population changes in north Swedish rural areas Fig. 1. Counties, municipalities and centres of municipalities in northern Sweden. 162 FENNIA 179: 2 (2001)Örjan Pettersson sets of values (Johansson and Persson 1991) and that certain age groups more often choose to live in the countryside (Amcoff 1997). The increase in the population living outside urban areas could largely be assigned to a process whereby leisure housing in metropolitan regions has been turned into permanent dwellings (Nyström 1989; Carl- qvist 1992). Swedish households’ motives for moving to the countryside The tendencies towards an increased rural popu- lation during the 1980s also led to a number of studies of the motives of households for moving to the countryside (Borgegård et al. 1987; Ny- ström 1989; Ennefors 1991; Borgegård et al. 1993; Forsberg and Carlbrand 1993; Kåks and Westholm 1994; Amcoff 2000; Garvill et al. 2000; Stenbacka 2001). In this short overview a few of all the possible reasons are emphasised. In the studies many households gave priority to what could be called countryside features in their resi- dential environment. Besides the actual house and the immediate surroundings, other environmen- tal factors, such as access to nature and the open scenic landscape, are often mentioned as impor- tant. It is, though, often unclear what so-called environmental motives really are and how impor- tant such reasons are for in-migrants to rural areas. It has even been suggested by socio-biologists that the preferences for open landscapes are deeply rooted in our genes (for a discussion, see Amcoff 2000). Nonetheless, many households emphasise the improved possibilities to perform various out- door activities and other hobbies, for instance hik- ing, hunting, gardening and doing handicraft. The move to the countryside could also be a part of a self-realisation or choice of life-style. Many in-migrants to rural areas are so-called returnees who have kinship with local people or are in fact moving back to the homestead. High housing costs in newly built areas and city-regions could also be a driving force behind in-migration to rural areas. By purchasing a house outside ur- ban areas households with low incomes can ful- fil their wish for a house of their own. Families with children moving to rural areas often state that the countryside offers a more suitable environ- ment for children to grow up in. New and im- proved communication technology and less rigid working conditions, for instance working from a distance and flexible working hours, make it pos- sible for more workers to settle in rural areas. Some households mention that it is strategic to settle down in rural areas between two or more cities. In this way they have access to more than one local labour market within commuting dis- tance. There is also a quite large group of house- holds, mainly pensioners, who are not depend- ent on nearness to large labour markets. They are thus free from an important restriction concern- ing their choice of where to live. The relative importance of different motives for moving to the countryside varies from study to study. This could partly be explained by the fact that many Swedish studies are based on inter- views with a limited number of households. It is also likely that the in-migrants’ motives, at least to some extent, vary between different types of rural area. The situation is further complicated by the fact that many households mention more than one reason. Nevertheless, several studies highlight the importance of environmental and social mo- tives (such as housing, access to nature, family sit- uation or nearness to friends and relatives) behind the move to the countryside. Based on the an- swers from nearly 500 in-migrants to the coun- tryside all over Sweden, Garvill et al. (2000) con- clude that 70 per cent of the respondents gave emphasis to environmental or social motives. In comparison, only 20 per cent mentioned work- related reasons such as education, unemployment and finding a new job. Gentrification of the countryside and the rural idyll In recent years and especially in literature focus- ing on changes in the British countryside there has been a great deal of interest in the motives and socio-economic characteristics of the in-migrants to rural areas (see for instance Champion and Watkins 1991; Marsden et al. 1993; Murdoch and Marsden 1994; Clout 1996; Cloke et al. 1998; Fielding 1998). These studies show that a substan- tial part of the in-migrants belongs to a well-edu- cated and relatively prosperous middle-class. Of- ten they have settled in amenity-rich and pictur- esque villages and small towns. This gentrification process has also led to conversion of farm build- ings and rural land into new uses, such as exclu- sive housing and golf courses. Closely linked with gentrification of the countryside is the concept of rural idyll whereby the countryside is perceived as quiet, calm, clean, safe and inhabited by all- FENNIA 179: 2 (2001) 163Contemporary population changes in north Swedish rural areas white nuclear families with traditional values. Lit- tle and Austin (1996) argues that the notion of ru- ral idyll maintains traditional gender roles and emphasises women’s roles as mothers. The rural idyll concept is also associated with neighbourli- ness and sense of community. Based on case studies in the metropolitan re- gion of Stockholm, Amcoff (2000) claims that gen- trification of the countryside is a relatively mar- ginal phenomenon in Sweden. Nevertheless, such tendencies are though likely to be observed in some rural areas, such as Österlen in southern- most Sweden (Hjort and Malmberg 1996). Population redistribution in northern Sweden During the period between 1950 and 1974 sev- eral reforms were carried out in order to reduce the number of municipalities in Sweden from ap- proximately 2,300 to fewer than 300. One of the main purposes was to construct municipalities with populations large enough to secure the pro- vision of services, mainly in the public sector. Christaller’s central-place theory (1966) provided the theoretical basis for a hierarchical central- place system. The centre of the municipality was supposed to provide local people with everyday services, while the major cities, the so-called re- gional centres, were supposed to provide high- level services to the inhabitants in larger regions. However, many genuine rural areas continued to lose inhabitants while the centres of the munici- palities, primarily the major cities, experienced population growth. During the 1990s it was main- ly municipalities in the metropolitan regions and regional centres, in particular university cities, that had substantial population growth, whereas many rural municipalities, industrial communities and small and medium-sized cities outside the metropolitan regions experienced decreases. The study area, northern Sweden, is a very sparsely populated region (Fig. 1). The region ex- tends from the Gulf of Bothnia to the mountain- ous border with Norway. In the north-eastern part, the region borders on Finland. In this article north- ern Sweden is defined as the six northernmost counties. The area consists of 69 municipalities covering approximately 270,000 square kilome- tres. Northern Sweden has 1.5 million inhabitants; thus the average density is 5.5 inhabitants per square kilometre. This means that only 16 per cent of the Swedish population live on two thirds of the total land area. The population is unequally distributed over the region, however, and most people are in fact concentrated in cities and the coastal areas. Within the interior parts of the re- gion most inhabitants live in the river valleys and administrative centres at different levels. Many of the municipalities, especially those in the interi- or parts of the region, have had a negative popu- lation trend during the last decades. Since 1970 more than every second munici- pality in northern Sweden have lost more than ten per cent of their population. At the same time, many of the major cities with nearby municipali- ties have experienced strong population growth. Table 1 shows that during the 1990s there were more municipalities with population losses and fewer with a growth rate above the national aver- age. The maps in Figure 2 illustrate that popula- tion growth is increasingly concentrated to a few regional centres, mainly those with universities. This process has accelerated in recent years and, at least at the municipality level, it seems as if the 1990s has implied a shift back to strong urbani- sation. According to Persson and Nygren (2001) this trend is likely to continue at least during the first decades of the 21st century. Contemporary changes in rural areas in northern Sweden In Sweden, however, county and municipality Table 1. Population changes by municipalities 1970–2000. Northern Sweden (Statistics Sweden 1995, 1996, 1997). Number of municipalities with: 1970–1974 1975–1979 1980–1984 1985–1989 1990–1994 1995–2000 Population decrease 39 32 43 42 39 66 Increase less than country average 10 08 05 14 23 01 Increase more than country average 20 29 21 13 07 02 164 FENNIA 179: 2 (2001)Örjan Pettersson Fig. 2. Population changes by municipalities 1970–2000. Northern Sweden (Statistics Sweden 1995, 1996, 1997). Note: Municipalities with population growth above the country average during the period are shaded with black. The municipalities with population increase below the average are shaded with grey, while those exhibiting population de- crease are white. averages often conceal a great deal of the geo- graphical variations in population change and so- cio-economic development within the regions. For this reason, we have in several research projects employed data with a higher degree of spatial resolution than municipalities, mostly elec- toral wards or even smaller geographical entities. These geographical subdivisions are by Statistics FENNIA 179: 2 (2001) 165Contemporary population changes in north Swedish rural areas Sweden considered to represent homogeneous housing environments. This has also made it pos- sible to group the wards according to various fea- tures, for instance by distance to localities of dif- ferent sizes. The empirical data utilised in this article is mainly derived from three research projects (Pet- tersson et al. 1996; Johansson and Pettersson 1997; Pettersson and Westholm 1998). The data has been analysed in a rather straightforward way. Firstly, we identified the rural areas with popula- tion growth. Secondly, by combining statistical analyses with visual interpretations of the geo- graphical patterns we were able to generalise the rural areas with population growth into three broad categories. These are presented in the final section of the article. It should though be men- tioned that we have only had access to data con- cerning the number of inhabitants in the wards (together with some demographic and socio-eco- nomic characteristics) at different times. For this reason we are not able to analyse whether a pop- ulation increase is due to natural increase or net in-migration. However, most rural areas have a long history of net out-migration of young people and therefore nowadays exhibit skewed demo- graphic profiles with many elderly. Subsequently these rural areas are usually characterised by nat- ural decrease and therefore necessitate net in-mi- gration in order to avoid further population losses. Figure 3 shows that the population growth rate decreases with distance from the centres of the municipalities. Nevertheless, rural areas can have substantial growth, both in Sweden as a whole and in northern Sweden. In northern Sweden, however, the growth zones do not extend as far out into the periphery as in other parts of Swe- den. A comparison between the six counties in northern Sweden also shows that there are obvi- ous variations within the region. During the period 1990–1995 relatively many municipalities in northern Sweden have had pop- ulation growth in rural areas near towns and cit- ies (i.e. within a distance of 5–15 kilometres from the centre of their own municipality). On the oth- er hand, few rural areas farther away from the centres show an increased population. Not only do rural areas at different distances show differ- Fig. 3. Average population change in wards grouped into distance zones from their local municipality centres 1990–1995 (Statistics Sweden 1995, 1996, 1997). Note: All distances are calculated as the crow flies (between the geographical centroid of the electoral ward and the centroid of the centre of the municipality). In this way, the distance measure is only a rough approximation of the actual road distance. 166 FENNIA 179: 2 (2001)Örjan Pettersson ing population trends but their populations also differ in various aspects. The rural areas within the 15-kilometre zone of the centre of the municipal- ity are characterised by relatively many families with children and households where both part- ners work outside the home. It is mainly in the peripheries of the municipalities that it is possi- ble to trace obvious signs of a far-reaching depop- ulation and a high proportion of elderly people. In general, the rural areas with a rapid popula- tion increase have relatively many young inhab- itants, families with children and households where both partners work outside the home (Ta- ble 2). Contrary to the depopulating rural areas with an ageing population, the average age is de- creasing and the high proportion of inhabitants below 45 years of age indicates that these areas are attractive for families with children. The county of Västerbotten – growing villages in the expanding Umeå-region The county of Västerbotten is characterised by large internal differences. In particular, the con- trast between the expanding city of Umeå and the extremely sparsely populated interior parts of the county is manifested. The municipality of Umeå has experienced dramatic population growth dur- ing the last decades. In 1950 Umeå had 46,000 inhabitants and in 2000 the total number of in- habitants had risen to almost 105,000 (i.e. an av- erage annual growth rate of almost 1.7 per cent). One important explanation to this development is the founding of Umeå University in the 1960s. At the same time other municipalities in the coun- ty have had a stagnating population or a decrease in the number of inhabitants. Within the county Note: Only areas with more than five kilometres to the centre of their own municipality are included in the table. Table 2. Population statistics for rural areas in northern Sweden (Statistics Sweden 1995, 1996, 1997). FENNIA 179: 2 (2001) 167Contemporary population changes in north Swedish rural areas there has been a concentration of people to the large localities, most of all to the centres of the municipalities and especially to Umeå, Skellefteå and Lycksele. However, Figure 4 shows that many rural areas close to the major localities also have had population growth during the period 1985– 1995. During the first half of the 1990s most of the centres of the municipalities in the interior parts of the county had decreasing populations (Fig. 5). At the same time most localities in the coastal re- gion showed population growth. Within commut- ing distance from the cities we also find many ru- ral agglomerations with growth. The spatial pat- tern indicates the importance of the major roads, thus demonstrating a kind of extended suburban- isation. However, there are some rural areas and villages with population growth lying outside the commuter zones of the major localities. Some of these are characterised by strong entrepreneurial traditions or expansive firms. Unexpectedly, rela- tively many rural agglomerations close to the bor- der with Norway have maintained or even in- creased their number of inhabitants. Often these are villages benefiting from an expanding tourist industry. In particular, the Hemavan area show increased population figures (see also Wiberg 2000). A research project concerning the geographi- cal distribution of welfare in the county of Väs- terbotten showed that there are substantial differ- ences in living conditions between various parts of the county (Pettersson et al. 1996; Pettersson 2001). Usually, the expanding rural areas sur- Fig. 4. Population changes in electoral wards 1985–1995. The county of Västerbotten (Statistics Sweden 1995, 1996, 1997). 168 FENNIA 179: 2 (2001)Örjan Pettersson rounding the cities have populations characterised by general well-being in terms of employment, educational level and incomes. Another common feature is the high proportion of families with chil- dren. Some of these rural areas have benefited from scenic surroundings and are located in tra- ditional agriculture landscapes, in river valleys, close to lakes or by the sea. The county of Dalarna – picturesque villages around Lake Siljan and ski resorts in Sälen Like Västerbotten, the county of Dalarna is a re- gion with large internal differences. For instance, there is a sharp contrast between the municipali- ties in the south-eastern part of the region and the municipalities around Lake Siljan (Fig. 6). The first mentioned region is part of Bergslagen, a large region extending over many counties united by the history of their mining industry. The Siljan re- gion is characterised by a scenic landscape and old traditions. The area around Lake Siljan is often regarded as a national symbol of the old rural Sweden as it was during the 19th century. In a historical per- spective, however, it is a little ironic that the ru- ral areas around Lake Siljan were considered as backward and underdeveloped during the indus- Fig. 5. Population changes in rural agglomerations and localities 1990-1995. The county of Västerbotten. Agglomerations with at least 50 inhabitants (Statistics Sweden 1995, 1996, 1997). FENNIA 179: 2 (2001) 169Contemporary population changes in north Swedish rural areas trial era of Bergslagen. Today, Bergslagen is one of the Swedish problem regions due to restruc- turing difficulties and depopulation. The Siljan region, on the other hand, has had an increas- ing population since the 1970s. Nevertheless, not only different parts of Dalarna but also vari- ous rural areas in the region have shown quite different population trends. In this section we shall take a closer look on two examples of ru- ral areas in the county of Dalarna, namely the countryside around Lake Siljan and the Sälen area, both with stable or increasing populations in recent years. During the last decades the countryside around Lake Siljan has had an inflow of migrants and a steady population growth. The number of inhabitants in the area has increased by five per cent during the period 1985–1995 and nowa- days more than 25,000 people live in the rural parts of the region. The in-migrants have moved in mainly from other municipalities in Dalarna and the Stockholm region. Some of these in-mi- Fig. 6. Case-study areas in the county of Dalarna. 170 FENNIA 179: 2 (2001)Örjan Pettersson grants have resided in the genuine countryside while others have settled down in large villages and small towns. The in-migration cannot be ex- plained by such factors as expansion in a cer- tain industry, a single company investment or a university establishment. Due to the long dis- tances to major cities in Dalarna, such as Falun and Borlänge, it is unlikely that the possibility for commuting is the sole cause of the popula- tion growth (see also Kåks and Westholm 1994). During the period 1985–1995 the growth rate in the rural areas around Lake Siljan has been above the county average. The area is character- ised by a relatively low proportion of families with children and instead the proportion of eld- erly people is higher. As shown before, this is unusual for those rural areas in northern Swe- den that are distinguished by substantial popu- lation growth. In recent years, several municipal- ities in the Siljan area have started losing inhab- itants, although at a slower rate than most rural municipalities in northern Sweden. Earlier studies have shown that, in particular, the people living in the municipality of Leksand generally have a high and evenly distributed liv- ing standard (Pettersson and Westholm 1998). Perhaps it is a combination of general well-be- ing, the image of being the original Sweden, the scenic landscape, tourism, small-scale business- es and self-employment that has generated both a renewal and in-migration to traditional rural areas where one would rather have expected depopulation. The second rural area of interest in Dalarna is the Sälen area. The Sälen area is situated in the mountainous region close to the border with Nor- way. There are only 2,000 people living in these ski resorts. Due to the in-migration of young peo- ple the area has experienced a ten per cent pop- ulation increase during the period 1985–1995. The ski tourism has led to a development that dif- fers in many ways from other peripheral rural ar- eas in northern Sweden. The large tourist estab- lishments provide job opportunities in a region where there are few alternatives and the employ- ment rate is, in fact, seven percentage units high- er than the county average. One third of the la- bour force works in the private service sector. In the Sälen area we find many young people and single persons, mostly single men. In comparison with the county average and most rural areas in northern Sweden, the Sälen area has very few families with children. Concluding remarks In recent years most municipalities and rural are- as in northern Sweden have experienced renewed depopulation. Nevertheless, some rural areas have shown significant population growth, at least during the first half of the 1990s. This population growth in rural areas in northern Sweden is con- centrated to the commuter belts surrounding the major cities, while more peripheral rural areas generally exhibit population decrease. Access to large labour markets and access to services are still factors that make the countryside close to cities attractive. This rather suggests suburbanisa- tion around major cities than a general and wide- spread counterurbanisation of the countryside in northern Sweden. However, the rural areas with growth that do not fit the description of being within the commuter belts of cities are often are- as that have benefited from the tourist industry. This means that some of the most peripheral ru- ral areas in northern Sweden have often managed better than more common types of rural areas. The redistribution of the population in north- ern Sweden since the 1970s together with the fact that many rural municipalities have experienced a renewed strong depopulation during the 1990s puts further pressure on the rural municipalities’ economic situation and possibilities for maintain- ing an acceptable standard of service provision for their inhabitants. Moreover, many municipal- ity centres in northern Sweden have started to lose inhabitants in recent years. Westlund and Pichler (2000) maintain that it is the small and medium- sized towns outside metropolitan regions that presently are facing the most severe difficulties in attracting in-migrants and maintaining their pop- ulations. Forthcoming delimitations of urban and rural areas will reveal if this tendency has become even stronger in the late 1990s. However, different rural areas have dissimilar preconditions, for instance in terms of job oppor- tunities, service accessibility, quality of infrastruc- ture, distance to major cities, housing supply, landscape features, recreational possibilities, so- cial networks, image and status. Due to that dif- ferent rural areas are likely to attract a wide range of combinations of in-migrants, for instance re- garding age, family situation, stage in the life-cy- cle, purchasing power and choice of life-style. Altogether this generates a rather complex geo- graphical pattern of rural areas with more or less unique characteristics and preconditions for fu- FENNIA 179: 2 (2001) 171Contemporary population changes in north Swedish rural areas ture development. Nevertheless, the rural areas in northern Sweden that have experienced popula- tion growth during the studied period can, in most cases, be grouped into three types according to their relative location and characteristics: i) Rural areas close to cities: Here we often find a combination of low housing costs (at least in comparison with housing costs in residen- tial districts within the cities) and possibili- ties for daily commuting to workplaces in the cities. Furthermore, these areas are character- ised by a relatively high proportion of fami- lies with children and, in general, high stand- ards of living. In some cases these rural areas are strategic locations between two or more cities. ii) Tourist resorts: In northern Sweden these ru- ral areas are mostly ski resorts found in the mountainous areas close to the border with Norway. Obviously, tourism and recreation have generated job opportunities in truly pe- ripheral areas. The effects, however, seems to be concentrated to relatively small geograph- ical areas such as single villages or specific mountain valleys. In a sense they can be seen as ‘urban satellites’ attracting visitors and sometimes even young in-migrants from cit- ies and metropolitan regions. iii) Other attractive rural areas: In this group we often find rural areas with a distinctive his- torical and cultural heritage combined with an attractive scenic landscape, but also areas with a large proportion of leisure housing (es- pecially along the coast). In these rural areas one can expect a high proportion of house- holds that are not dependent on proximity to large labour markets, for instance pensioners, distance workers, some self-employed per- sons and the like. Here we often find ‘life- style’ arguments among the in-migrants (Kåks and Westholm 1994). Perhaps it is even pos- sible to identify gentrification tendencies in this type of rural areas. Of course, there are rural areas where two or more of the above features are combined, for in- stance in Leksand in the area around Lake Siljan, and that to some extent fit into the above descrip- tion of all three types of rural areas. However, there are also some rural areas with population growth that cannot easily be placed in one of the categories mentioned before. In some cases we find single successful firms, often small-scale manufacturers, or a history of entrepreneurial tra- ditions in these rural areas. Finally, it would be interesting to take a closer look at the rural areas that have had a relatively strong development despite being located outside the commuter belts of the major cities. How and why are these areas managing this well? Are there successful local firms or industries, entrepreneurs, tourist businesses, local action groups or devel- opment projects that might have contributed? Or are there other explanations? And what about people’s preferences of where to live? Which groups of people are attracted to different types of rural areas and why? These questions require further research. ACKNOWLEDGEMENTS An earlier version of the article has been published in Gourlay D (ed) (1999). Marginal Rural Areas in the New Millennium. Proceeding from the PIMA98 International Conference at the Scottish Agricultural College, Aberdeen, June 26–29th 1998. REFERENCES Ahnström L (1980). Turnaround-trenden och de nor- diska huvudstadsregionernas utveckling efter 1950. NordREFO 1980: 3/4, 43–100. Ahnström L (1986). The turnaround trend and the economically active population of seven capital regions in western Europe. Norsk Geografisk Tid- skrift 40: 2, 55–64. Amcoff J (1997). Kontraurbanisering i Sverige? In Westlund H (ed). Lokal utveckling för regional omvandling, 13–29. Rapport 100, SIR, Stock- holm. Amcoff J (2000). Samtida bosättning på svensk lands- bygd. Geografiska regionstudier 41, Kulturgeo- grafiska institutionen, Uppsala universitet, Upp- sala. Andersson R (1987). Den svenska urbaniseringen. Kontextualisering av begrepp och processer. Geo- grafiska regionstudier nr 18, Kulturgeografiska institutionen, Uppsala universitet, Uppsala. Bengtsson T & M Johansson (1992). Befolkningens geografiska rörlighet – minskar den? In Persson LO & M Wångmar (eds). 90-talets befolknings- frågor, 41–71. Ds 1992: 16, Stockholm. Berry B (1976). The counterurbanization process: urban America since 1970. In Berry B (ed). Ur- banization and counterurbanization, 17–30. Sage, Beverly Hills, California. Borgegård L-E, U Fransson, J Håkansson & A Tollef- 172 FENNIA 179: 2 (2001)Örjan Pettersson sen (1993). Att flytta till glesbygden. Om be- skrivningsfaktorer bakom inflyttningen till Nor- danstigs kommun. GERUM nr 20, Geografiska in- stitutionen, Umeå Universitet, Umeå. Borgegård L-E & J Håkansson (1997). Population concentration and dispersion in Sweden since the 1970s. In Borgegård L-E, AM Findlay & E Son- dell (eds). Population, planning and policies, 9– 29. CERUM report No. 5, Umeå University, Umeå. Borgegård L-E, J Håkansson & G Malmberg (1995). Population redistribution in Sweden – long term trends and contemporary tendencies. Geografis- ka annaler B 77: 1, 31–45. Borgegård L-E, E Rhodes & O Siksiö (1987). Inflytta- re till Ljusdals kommun. Meddelande M:13, Statens institut för byggforskning, Gävle. Carlquist T (1992). Fler i glesbygd. Välfärdsbulletinen 1992: 3, 28–30. Champion AG (ed) (1989). Counterurbanization: the changing pace and nature of population decon- centration. Edward Arnold, London. Champion AG (1992). Urban and regional demo- graphic trends in the developed world. Urban Studies 29: 3/4, 461–482. Champion AG (1994). Population change and mi- gration in Britain since 1981: evidence for con- tinuing deconcentration. Environment and Plan- ning A 26: 10, 1501–1520. Champion AG (1998). Studying counterurbanisation and the rural population turnaround. In Boyle P & K Halfacree (eds). Migration into rural areas, 21–40. John Wiley & Sons, New York. Champion AG & C Watkins (1991). Introduction: re- cent developments in the social geography of ru- ral Britain. In Champion AG & C Watkins (eds). People in the countryside. Studies of social change in rural Britain, 1–21. Paul Chapman Pub- lishing, Great Britain. Christaller W (1966). Central places in Southern Germany. Englewood Cliffs, N.J. Cloke P, M Phillips & N Thrift (1998). Class, coloni- sation and lifestyle strategies in Gower. In Boyle P & K Halfacree (eds). Migration into rural areas, 166–185. John Wiley & Sons, New York. Clout H (1996). Changing countrysides. In Douglas I, R Huggett & M Robinson (eds). Companion encyclopedia of geography, 752–771. Routledge, London. Cochrane S & D Vining (1988). Recent trends in mi- gration between core and peripheral regions in developed and advanced developing countries. International Regional Science Review 11, 215– 243. Court Y (1989). Denmark: towards a more decon- centrated settlement system. In Champion AG (ed). Counterurbanization: the changing pace and nature of population deconcentration, 121–140. Edward Arnold, London. Dahms F & J McComb (1999). ‘Counterurbaniza- tion’, interaction and functional change in a ru- ral amenity area – a Canadian example. Journal of Rural Studies 15: 2, 129–146. Ennefors K (1991). Att leva mellan stad och land – livskvalitet i den urbaniserade glesbygden. CWP 1991:15, CERUM, Umeå University, Umeå. Eriksson O (1989). Bortom storstadsidéerna. En re- gional framtid för Sverige och Norden på 2010- talet. Carlssons, Stockholm. Fielding AJ (1982). Counterurbanisation in Western Europe. Progress in Planning 17: 1, Pergamon, Oxford. Fielding T (1998). Counterurbanisation and social class. In Boyle P & K Halfacree (eds). Migration into rural areas, 41–60. John Wiley & Sons, New York. Ford T (1999). Understanding population growth in the peri-urban region. International Journal of Population Geography 5: 4, 297–311. Forsberg G & E Carlbrand (1993). Mälarbygden – en kreativ region? En studie av Mälardalens lands- bygd i förändring. Forskningsrapport nr 107, Kul- turgeografiska institutionen, Uppsala universitet, Uppsala. Garvill J, G Malmberg & K Westin (2000). Värdet av att flytta och att stanna: Om flyttningsbeslut, platsanknytning och livsvärden. Statens Offent- liga Utredningar (SOU) 2000: 36 Rapport 2, Stockholm. Geyer HS & T Kontuly (1993). A theoretical founda- tion for the concept of differential urbanization. International Regional Science Review 15: 2, 157–177. Halfacree K (1997). Contrasting roles for the post- productivist countryside. A postmodern perspec- tive on counterurbanisation. In Cloke P & J Little (eds). Contested countryside cultures, 70–93. Routledge, London. Hansen JC (1989). Norway: the turnaround which turned around. In Champion AG (ed). Counter- urbanization: the changing pace and nature of population deconcentration, 103–120. Edward Arnold, London. Hjort S & G Malmberg (1996). Landsbygdsgentri- fiering – en projektskiss. A project proposal pre- sented at Forskarforum in Östersund, 13–14 No- vember 1996, unpublished. Håkansson J (2000). Changing population distribu- tion in Sweden – long term trends and contem- porary tendencies. GERUM Kulturgeografi 2000: 1, Kulturgeografiska institutionen, Umeå Universitet, Umeå. Illeris S (1994). Regional development and the ur- ban system in Denmark. In Illeris S (ed). Essays on regional development in Europe, 7–28. Re- search report 101, Department of Geography and International Development Studies, Roskilde Uni- versity, Denmark. Johansson M & LO Persson (1991). Regioner för generationer. Allmänna Förlaget, Stockholm. Johansson P & Ö Pettersson (1997). Glesbygdens utveckling och lantbrevbäringen – tendenser i FENNIA 179: 2 (2001) 173Contemporary population changes in north Swedish rural areas bosättningsmönster samt analys av marknad för en utvidgad serviceroll för lantbrevbäringen. GERUM Arbetsrapport 1997-05-15, Kulturgeo- grafiska institutionen, Umeå universitet, Umeå. Karjalainen E (1989). Migration and regional devel- opment in the rural communes of Kainuu, Fin- land in 1980–85. Nordia 23: 1, 1–89. Kontuly T (1998). Contrasting the counterurbanisa- tion experience in European nations. In Boyle P & K Halfacree (eds). Migration into rural areas, 61–78. John Wiley & Sons, New York. Kåks H & E Westholm (1994). En plats i tillvaron – Studier av flyttning till landsbygden. DFR-rapport 1994: 8, Dalarnas forskningsråd, Falun. Lewis G (1998). Rural migration and demographic change. In Ilbery B (ed). The geography of rural change, 131–160. Addison Wesley Longman. Little J & P Austin (1996). Women and the rural idyll. Journal of Rural Studies 12: 2, 101–111. Marsden T, J Murdoch, P Lowe, R Munton & A Flynn (1993). Constructing the countryside. Restructur- ing rural areas 1, UCL Press, London. Murdoch T & T Marsden (1994). Reconstituting ru- rality. Restructuring rural areas 2, UCL Press, Lon- don. Nyström J (1989). From the city to the city’s coun- tryside. Geografiska Annaler B 71: 3, 183–200. Persson LO & O Nygren (2001). Det enkelriktade Sverige. Tjänstesektorn och den framtida regio- nala befolkningsutvecklingen. TCO-rapport, Stockholm. Pettersson Ö (2001). Microregional fragmentation in a Swedish county. Papers in Regional Science 80: 4, 389–409. Pettersson Ö, LO Persson & U Wiberg (1996). När- bilder av västerbottningar – materiella levnads- villkor och hälsotillstånd i Västerbottens län. Re- gional Dimensions Working Paper No. 2, CERUM, Umeå universitet, Umeå. Pettersson Ö & E Westholm (1998). Gräddhyllor och fattigfickor. En mikroregional analys av välfärdens geografiska fördelning i Dalarna. DFR-rapport 1998: 1, Dalarnas forskningsråd, Falun. The Population (1991). Volume in The National At- las of Sweden. SNA Förlag, Stockholm. Rusanen J, A Naukkarinen, A Colpaert & T Muilu (1997). Differences in the spatial structure of the population between Finland and Sweden in 1995 – a GIS viewpoint. Research reports 221, Statis- tics Finland, Helsinki. Räisänen S, J Rusanen & A Naukkarinen (1996). So- cio-economic grid data and GIS for analyzing changes in the Finnish countryside. In Rumor M & R McMillan (eds). Geographical information: from research to application through cooperation, 651–660. IOS Press, Netherlands. Statistics Sweden (1995). Småorter 1990. Statistiska meddelanden Be16SM9401. Statistics Sweden (1996). Tätorter 1995. Statistiska meddelanden Be16SM9601. Statistics Sweden (1997). Småorter 1995. Statistiska meddelanden Be16SM9602. Stenbacka S (2001). Landsbygdsboende i inflyttarnas perspektiv. Intention och handling i lokalsamhäl- let. Geografiska regionstudier nr 42, Kulturgeo- grafiska institutionen, Uppsala universitet, Upp- sala. Stockdale A, A Findlay & D Short (2000). The repop- ulation of rural Scotland: opportunity and threat. Journal of Rural Studies 16: 2, 243–257. Tervamäki E (1987). Migration in Finland, a multi- level system of regions. Fennia 165: 1, 1–88. Vartiainen P (1989a). The end of drastic depopula- tion in rural Finland: evidence of counterurbani- sation? Journal of Rural Studies 5: 2, 123–136. Vartiainen P (1989b). Counterurbanisation: a chal- lenge for socio-theoretical geography. Journal of Rural Studies 5: 3, 217–225. Westlund H & W Pichler (2000). En ny grön våg? Glesbygdstillväxt under storstadsexpansionens 90-tal. Working Paper 2000: 16, SIR, Östersund. Wiberg U (2000). Norrland – glest, glesare, glesast. In Hjelmskog S (ed). Hemort Sverige, 88–123. Integrationsverket, Norrköping.