1

Growth and decline in rural Sweden Geographical distribution of employment and population

1960–2010

Martin Hedlund

Department of Geography and Economic History

Umeå 2017

2

This work is protected by the Swedish Copyright Legislation (Act 1960:729)

Dissertation for PhD

ISBN: 978-91-7601-769-2

ISSN: 1402-5205

GERUM 2017:3

Electronic version available at: http://umu.diva-portal.org/

Printed by: UmU Print Service, Umeå University

Umeå, Sweden 2017

3

Table of Contents

Acknowledgement ........................................................................... i

Abstract ..........................................................................................iii

Sammanfattning på svenska ......................................................... iv

Introduction .................................................................................... 1 Research questions ................................................................................................ 3

The concept of rural in this thesis ......................................................................... 4

Rural restructuring ........................................................................ 7 Technological development ................................................................................. 12

Social modernization ........................................................................................... 16

Globalization ........................................................................................................ 21

Settlement and employment development in Sweden ............... 27

Data and methods ......................................................................... 40 Methodological perspective................................................................................ 40

Methods ............................................................................................................... 41

Data ......................................................................................................................42

Limitations and delimitations ............................................................................ 46

Article summaries ........................................................................ 50 Summary of Article I .......................................................................................... 50

Summary of Article II .......................................................................................... 51

Summary of Article III......................................................................................... 52

Concluding discussion .................................................................. 54 Two phases of rural restructuring ....................................................................... 55

The geography of the second restructuring phase ............................................ 60

Combatting rural decline? ...................................................................................62

Suggestions for further research ........................................................................ 64

References ..................................................................................... 65

4

Appendix Article I: Hedlund, M., Lundmark, L., & Stjernström, O. (2017). Rural

restructuring and gendered micro-dynamics of the agricultural labour

market. Fennia, 195, 25-35.

Article II: Hedlund, M., & Lundholm, E. (2015). Restructuring of rural

Sweden–Employment transition and out-migration of three cohorts born

1945–1980. Journal of Rural Studies, 42, 123-132.

Article III: Hedlund, M. (2016). Mapping the socioeconomic landscape of

rural Sweden: Towards a typology of rural areas. Regional Studies, 50,

460-474.

i

Acknowledgement

Writing this thesis has been a fun and interesting project. Interesting as it

has developed my own thinking on the workings of society in directions I

never would have thought of on my own, and fun when progress were

made. Much more can of course be said about writing a thesis but easy is

not one of those words. In that respect I am glad for the help I have received

from others.

First of all I would like to thank my supervisors Linda Lundmark, Emma

Lundholm and Olof Stjernström. You have done a very good job reading

and commenting on my drafts and your comments have made substantial

contributions to shape this thesis. We have also co-authored some articles

that came out in nice shape, of which two are part of this thesis. At times

when I got stuck or lost confidence in my work you helped me to get back

on track. I am particularly grateful for the extra effort you put in during the

last six months of writing, with more frequent meetings and commenting

on drafts – thank you for that!

Apart from my supervisors there are other persons who should be men-

tioned. I would like to thank Örjan Pettersson for good discussions and for

reading and commenting on my drafts. It was Örjan’s idea to make a con-

centric model for the paper Mapping the socioeconomic landscape of rural

Sweden: Towards a typology of rural areas. Thanks for that suggestion!

I would like to thank my colleague and friend Magnus Strömgren who has

helped me with map design and questions regarding data. I would also like

to thank Jan Amcoff who commented on the thesis draft for my half-time

seminar. His comments helped to keep the thesis on track. At my pre-

dissertation seminar I received good and detailed comments on how to

continue working with the kappa, and, while they have kept me busy these

last few months, they were valuable for the final state of the thesis. I am

therefore grateful to Dean Carson, Charlotta Hedberg and Roger

Marjavaraa who read the draft for my pre-dissertation seminar. I am also

thankful to Magnus Lindmark for reading and commenting on the kappa

in a few months after the pre-dissertation seminar. Finally, I would like to

thank Einar Holm and Rikard Eriksson who did the last reading of the

thesis.

ii

A great deal of time working with this thesis has been spent in the computer

lab working with data. For the most time everything worked, but when it

did not I could always ask Erik Bäckström, the computer guy, and he would

fix the problem.

As a PhD-student I had fellow colleagues in the same predicament as me.

While I know some of you better than others, I think we have had a good

time together: discussing, laughing, bowling etc. I was also member of the

Graduate School in Population Dynamics and Public Policy that was

managed by Johan Lundberg. We had some good times together, not least

on our trips to Geneva and Brussels. The graduate school also funded a

conference trip to Tampa for which I am grateful.

Last but not least this thesis was funded by The Swedish research council

FORMAS as part of an effort to strengthen the research on rural Sweden. I

am grateful for the funding and for being given the opportunity to carry out

this project. Hopefully this thesis will be perceived as a welcome addition

to the current research on rural Sweden.

iii

Abstract

This thesis investigates the combination of changes in the population and

employment into sectors in rural Sweden for the period 1960-2010. The

aim is to describe and analyze the demographic changes together with the

labour market changes, and to account for the spatial outcome of these

changes by considering the heterogeneity of rural areas. The analysis

departs from the framework of rural restructuring, where changes in

employment and population in rural Sweden are interpreted as local

products of the global processes of technological development, social

modernization and globalization. Empirically, the analysis is based on a

combination of longitudinal censuses and register data on the Swedish

population covering the period 1960-2010.

The first part of the aim is achieved by applying a life-course perspective

and exploiting the longitudinal nature of the data. The life-course

perspective distinguishes between historical time and the age of

individuals, making it possible to situate changes in employment and

migration on the individual level. The second part of the aim is achieved

through developing a typology of rural Sweden by doing a cluster analysis

on SAMS-areas.

The results show that rural change after 1980 was characterized by de-

industrialization and the rise of the urban service sector. The period was

also characterized by regional urbanization rather than local urbanization.

Peripheral urban and rural areas based on industrial employment found

themselves with a declining economic motor, which meant that people had

to find their source of income elsewhere. The migration stream in this

period was thus increasingly directed towards metropolitan or large city

centers, and their rural surroundings within commuting distance.

However, the more fine-tuned spatial typology reveals that also a few areas

in the rural periphery have experienced growth, these areas are mainly

attractive places based on various kinds of tourism. It can thus be

concluded that different rural areas have experienced, and will continue to

experience, the shift from manufacturing to services differently, where

some areas have grown in both demographic and employment terms while

others have declined. In this sense the heterogeneity of rural areas are a

product of both growth and decline – of old development paths that is

reaching their end and of new development paths that will continue into

the future.

iv

Sammanfattning på svenska

Svensk landsbygd, i likhet med landsbygden i andra Europeiska länder, har

under de senaste decennierna genomgått omfattande förändringar. Många

väljer att lämna landsbygden och bosätta sig i större städer, och befolk-

ningen har minskat på många platser trots att Sveriges befolkning som

helhet ökat. Denna omflyttning kan kopplas till en rad problem för de

platser som upplever en minskande befolkning, inte minst då det främst är

ungdomar och personer i arbetsför ålder som väljer att flytta. Några

problem är t.ex. tomma hus som förfaller, överdimensionerad

infrastruktur, minskande skatteintäkter och en oproportionerligt stor

andel äldre i behov av vård. Anledningen till att människor väljer att flytta

från landsbygden kan variera från individ till individ, men en

utgångspunkt för den här avhandlingen är att försörjning och arbete utgör

en viktig del av det övergripande bosättningsmönstret: där människor

väljer att bo måste det också finnas arbete inom pendlingsavstånd.

Syftet med den här avhandlingen är att undersöka hur arbete och

befolkning har förändrats på svensk landsbygd 1960-2010 och hur denna

förändring tagit sig till uttryck mellan olika platser på landsbygden.

Betoningen är här på just kombinationen av arbete och befolkning snarare

än att dessa två undersöks var för sig.

Medan avhandlingen undersöker hur befolkning och arbete förändrats på

svensk landsbygd så kan dessa förändringar antas bero på större och mer

genomgripande förändringar som inte bara sträcker sig utanför

landsbygdens gränser utan också utanför Sveriges. I detta avseende har

teknologisk utveckling bidragit till att förändra villkoren inom en rad

näringar. Handkraft har ersatts med maskiner och avståndet mellan

bostad och arbete har ökat. Moderniteten har inneburit en rad

förändringar också värderingsmässigt, inte minst att kvinnors arbete

flyttat från hemmet ut till industrierna, kontoren eller den offentliga

sektorn. Den moderna idén om självförverkligande och att varje individ ska

uppnå sin fulla potential tar sig också till uttryck i att allt fler unga väljer

att läsa vidare på universiteten. Slutligen har världen förändrats i en

riktning mot globalisering sedan 1970-talet och framåt. Kombinationen av

allt mer öppna gränser och minskade transportkostnader har

revolutionerat världsekonomin. Industrierna i Väst har i ökande grad

kommit att omlokaliseras till andra delar av världen, inte minst

Sydostasien. Sveriges import av både industri- och jordbruksprodukter har

v

ökat markant. Trots att dessa tre övergripande processer – teknologisk

utveckling, modernisering och globalisering – inte studeras empiriskt i

avhandlingen så kan de antas vara avgörande för hur den Svenska

landsbygden förändrats.

Den empiriska undersökningen är uppdelad i tre delstudier. Dessa

delstudier bygger alla på svensk registerdata från SCB (Statistiska

Centralbyrån). De två första studierna bygger också på Folk- och

Bostadsräkningarna 1960-1990.

I den första studien undersöks yrkeslivet för två kohorter av jordbrukare.

Den första kohorten är född mellan 1938 och 1940 och den andra är född

mellan 1958 och 1960. För att kvalificeras in i undersökningen skulle

individerna ha jobbat inom jordbruket när de var 20 och 25 år gamla och

båda kohorterna följs tills individerna är 50-52 år gamla. Artikeln

undersöker hur yrkeslivet skiljer sig åt mellan kohorterna uppdelat på kön

för att därigenom kunna dra slutsatser om hur förutsättningarna för

jordbruksarbete förändrats över tid. Den huvudsakliga slutsatsen är att de

förändrade villkoren, till följd av teknisk utveckling och ökad konkurrens,

gjort det svårare att fortsätta jobba inom jordbruket. Detta gäller speciellt

för kvinnor som i långt högre utsträckning slutade inom jordbruket i

kohorten född 1958-1960. Varför kvinnor i högre grad slutade inom

jordbruket kan bero på att det i ett jordbrukarhushåll i första hand är

kvinnan som letar jobb utanför gården medan mannen i högre

utsträckning fortsätter driva bruket. Andra förklaring är att många sysslor

som var traditionellt ”kvinnliga”, som t.ex. mjölkning, ersatts av maskiner,

eller att den stora utbyggnaden av offentlig sektor runt om i landet gav

kvinnorna möjligheter till högre inkomst utanför jordbruket.

Den andra studien undersöker arbete och boendeort för tre kohorter unga

vuxna. Den första kohorten var född 1945, den andra 1960 och den tredje

1980. I denna studie ingick hela Sveriges befolkning födda respektive år.

Dessa kohorter definierades som ”landsbygd” respektive ”stad” baserat på

deras boendeort när de var 15 år gamla. Studien utgår från en regional syn

på landsbygd där t.ex. Stockholms och Umeås omland definieras som stad

medan samhällena Norsjö och Nordmaling definieras som landsbygd.

Analysen gick ut på att jämföra mellan kohorterna vilken sektor kvinnorna

och männen jobbade inom och hur dom flyttade mellan landsbygd och stad

när de var 20, 25 respektive 30 år gamla. Huvudresultatet av studien är att

det skedde stora förändringar för både kvinnor och män i

vi

landsbygdskohorten 1980 jämfört med de två tidigare kohorterna. I

landsbygdskohorterna 1945 och 1960 arbetade en stor andel av männen

inom industrin på landsbygden och en stor andel av kvinnorna återfanns

inom offentlig sektor och servicesektor på landsbygden. I landsbygds-

kohorten 1980 hade arbete inom industri minskat avsevärt, särskilt för

män, och istället flyttade en relativt större andel städerna för att studera

eller jobba inom servicesektorn. Även om kvinnorna i landsbygdskohorten

1980 jobbade kvar inom offentlig sektor så hade de i högre utsträckning

flyttat in till städerna. Detta sammanföll också med ett skifte inom offentlig

sektor från sjukvård till utbildning, inte minst till följd av den stora

expansionen av universiteten på 1990-talet. Den huvudsakliga slutsatsen

av den andra studien är att dagens urbanisering drivs av industrins

nedgång och den generella omvandlingen från en produktiv industriell

ekonomi till en konsumtionsinriktad serviceekonomi. Denna omvandling

kan i sin tur kopplas till de tre processer som diskuterades ovan:

teknologisk utveckling, modernisering och globalisering.

Den tredje studien undersöker hur de förändringar som beskrivits i den

första och, framförallt, i den andra studien tagit sig till uttryck i geografin.

Syftet med studien är därför att identifiera och kategorisera olika typer av

landsbygd utifrån vad folk jobbade med på olika platser år 2008 och vilka

platser som ökat och minskat i befolkning 1985-2008. Detta görs genom

att summera individernas karakteristik till områden. Den

områdesindelning som används kallas för SAMS (Small Areas for Market

Statistics) och är framtagen av SCB. Områdenas storlek kan variera över

landet men varje kommun är indelad i ett flertal områden. I den här

studien definieras landsbygd på ett lite annat sätt jämfört med studie två.

Medan studie två utgår från en regional definition av landsbygden så utgår

studie tre från en lokal definition av landsbygden. Det innebär t.ex. att

orter såsom Norsjö och Nordmaling inte räknas som landsbygd medan

omlandet runt Stockholm räknas som landsbygd. Denna

landsbygdsdefinition motiveras genom studiens syfte, som är att

undersöka geografiska skillnader mellan olika typer av landsbygd. SAMS-

områdena på landsbygden (3983 till antalet) kategoriseras med hjälp av en

hierarkisk klusteranalys. Klusteranalysen sorterar områden med liknande

karaktäristik i samma kategori och placerar områden med tillräckligt olika

karaktäristik i skilda kategorier. Huvudresultatet av analysen består av en

typologi där Sveriges landsbygd delas in i 5 huvudgrupper och 16

undergrupper. En tredjedel (34%) av landsbygdsbefolkningen återfanns i

medel- och arbetarklassområden inom pendlingsavstånd från större

vii

städer. Huvuddelen (53%) bodde i mer perifera områden där tillverknings-

industri utgjorde en viktig källa till arbete, dessa områden är

huvudsakligen belägna i södra halvan av landet. Endast en mindre del av

landsbygdsbefolkningen (12%) bodde på den perifera landsbygden där

utvinning av naturresurser utgjorde en viktig källa till arbete. När det

kommer till befolkningsförändringar har den stadsnära landsbygden runt

större städer ökat medan den avlägsna landsbygden minskat. Här återfinns

dock ett par intressanta undantag som består i avlägsen landsbygd som

ökat sin befolkning. Detta handlar om områden där näringsbasen grundar

sig på turism (t.ex. Åre och Hemavan) och fritidshusområden som är

attraktiva att flytta till på äldre dar (t.ex. Österlen och Stockholms

Skärgård).

Sammantaget visar avhandlingen att Sveriges landsbygd förändrats genom

övergången från ett industrisamhälle till ett servicesamhälle. Dessa

förändringar har tagit sig olika uttryck på olika platser. Medan vissa

områden upplever tillväxt så befinner sig andra i nedgång. Denna nedgång

måste samtidigt förstås som en strukturell kris för dessa orter, en kris som

inte enkelt kan lösas eller kanske inte ens kan lösas. En utgångspunkt för

avhandlingen var att boende och arbete måste följas åt då människor inte

kan leva utan en försörjning. Samtidigt har ekonomin förändrats och de

platser som tillhandahöll en försörjning för 50 år sedan kanske inte gör det

idag. Ur ett historiskt perspektiv har bosättningsmönster och ekonomiska

aktiviteter befunnit sig i ständig förändring. Av den anledningen är det

svårt bevara ett bosättningsmönster som tillkommit under en annan

ekonomisk ordning.

1

Introduction

Similar to other European countries, rural areas in Sweden face several

problems related to population decline and ageing, and to a declining

demand for labor. Furthermore, the shrinking population is connected to

a range of other problems, such as oversized infrastructure and empty

industrial and housing units (Sousa and Pinho 2015, Wiechmann and

Pallagst 2012), lower tax revenues from labor (Haase et al. 2012, Hollander

2011), and the selective out-migration of individuals with high human

capital (Weck and Beißwenger 2014). As a response to these problems,

many peripheral municipalities have formulated goals and strategies for

turning the population decline into growth and creating new jobs (Syssner

2016). On the other hand, “rural is not synonymous with decline”, as the

OECD (2006, 31) puts it, as there also are rural areas that are experiencing

growth, in both economic and demographic terms.

This thesis examines these processes of change in rural Sweden, focusing

on demographic and employment changes. The aim is to describe and

analyze the demographic changes together with the labor market changes

in rural Sweden in the period 1960-2010, and to account for the spatial

outcome of these changes by considering the heterogeneity of rural areas.

In furthering this aim it is important that rural areas are defined in non-

essentialist terms, recognizing them as places “where wider processes

manifest themselves” (Lovering 1989, 213) rather than as a closed social

system. In this sense, this thesis also implicitly examines how urban areas

have changed – as the social and economic processes driving demographic

and employment changes cut right across any rural-urban division

(Hoggart 1988). Another point of departure is to consider demographic

changes together with changes in economic activities, a common

perspective in the structural approach to societal change (Andersson 1987).

Empirically, this is partly achieved by applying a life-course perspective

(Elder Jr 1994) on societal change, situating employment in different

sectors and urban-to-rural migration on the individual level.

In order to apprehend demographic changes together with employment

changes, this thesis departs from a multidimensional theoretical frame-

work in which rural change is seen as the interrelated consequences of

larger technological, economic and social changes on a global scale; a

perspective that has come to be known under the term “rural restructuring”

(Hoggart and Paniagua 2001b, Ilbery 1998, Marsden, Lowe and Whatmore

2

1990, Marsden et al. 1993, Pettersson 2002, Woods 2005, Woods 2010).

The idea of rural restructuring is to view rural change as a multifaceted and

interdependent phenomenon, not limited to only a few aspects of reality.

In a sense, rural restructuring applies a holistic view on rural change,

whereby everything is seen as related to everything else, which blurs the

boundaries between different dimensions (Pettersson 2002). This view on

rural and, by extension, societal change has much in common with the per-

spective of Fernand Braudel and the Annales School of historiography, in

which the importance of large but isolated events is downplayed in favor of

slower changes in broader societal structures (Braudel 1958, Braudel

1982b, Nordin 1989).

Partly related to the framework of rural restructuring is the discourse of

post-productivism, which has received a great deal of attention in rural

geographic research. Post-productivism stresses the importance of

changes in agriculture and other natural resource sectors, while manufac-

turing is at best given a peripheral role (Evans, Morris and Winter 2002,

Ilbery 1998, Mather, Hill and Nijnik 2006, Shucksmith 1993, Woods 2005,

Woods 2010). Within this discourse, a common point of departure is that

contemporary rural change is characterized by a shift from primary natural

resource sectors towards tertiary service sectors, especially the tourism

sector. It is furthermore argued that agriculture has undergone a post-

productive transition in market logics, whereby farming increasingly

entails the small-scale production of high-quality products, and that this is

a central aspect of contemporary rural change (Evans et al. 2002, Ilbery

and Bowler 1998, Mather et al. 2006, OECD 2006, Shucksmith 1993, Ward

1993, Wilson 2001). Along with this “post-productive” economic change, it

has been proposed that the migration streams have changed from

urbanization towards counterurbanization, with rural areas being

“gentrified” by the urban middle class (Boyle and Halfacree 1998, Phillips

2002, Phillips 2005, Phillips 2009, Urry 1995, Woods 2005).

However, this description of rural change, which arguably comprises the

dominant discourse in contemporary research, has not been sufficiently

supported. Empirical studies find little evidence of a large-scale post-

productive transformation of the agricultural sector (Almstedt et al. 2014,

Brouder, Karlsson and Lundmark 2015, Evans et al. 2002, Wilson 2001),

of urbanization being replaced by counterurbanization (Amcoff 2000,

Amcoff 2003, Amcoff 2006), or of a repopulation of the rural periphery by

the urban middle class (Amcoff 2000, Hjort and Malmberg 2006). Instead,

3

the evidence for the existence of agricultural post-productivism and rural

gentrification stems mostly from case studies (Evans et al. 2002, Phillips

2005). In the case of counterurbanization, Amcoff (2006) makes a solid

case for understanding the migration patterns in Sweden during the 1990s

as urban sprawl and suburbanization rather than counterurbanization and

a new “green wave”. In light of the contested status of the post-productive

discourse, it is argued here that there is a need for an open approach that

examines contemporary demographic and labor market change in rural

areas. Empirically, the approach of this thesis is therefore open and

descriptive rather than hypothesis-testing.

Despite the discourses of counterurbanization, rural gentrification and

post-productivism, it is widely acknowledged in academic research and

public debate that rural areas are declining, both economically and demo-

graphically. In the 1970s and 1980s, during the heydays of the debate on

how to define rural areas, the perception that contemporary rural areas in

developed countries are declining was so taken-for-granted that Cloke

(1977) and Cloke & Edwards (1986) included indicators of decline in the

very definition of the concept of rural. However, these two contrasting

images of the rural – the rural idyll that is invaded by the middle class on

the one hand, and the rural periphery that is declining on the other – are

not mutually exclusive, since rural areas are heterogeneous. The evidence

speaking for rural gentrification, for example, stems primarily from case

studies on areas close to larger cities (Ghose 2004, Phillips 2002, Phillips

2005, Phillips 2009), while the evidence against rural gentrification also

covers more peripheral rural areas (Amcoff 2000, Hjort and Malmberg

2006). Therefore, when accounting for the recent rural change, it is also

necessary to account for the geography and not assume a homogenous

rural landscape. In this respect, it should also be noted that there are

structural and institutional differences between different countries, where

for example Sweden is relatively less densely populated than many other

Western countries.

Research questions

As mentioned, the aim of this thesis is to describe and analyze the

demographic changes together with the labour market changes in rural

Sweden in the period 1960-2010, and to account for the spatial outcome of

these changes by considering the heterogeneity of rural areas. The

description and analysis are based on three separate empirical papers, each

of which corresponds to specific research questions but that, taken

4

together, give a picture of long-term demographic and labor market change

in rural areas. The empirical articles are built on longitudinal data covering

the entire Swedish population, and the time span covered by the data

stretches from 1960 to 2010. The research questions corresponding to the

three papers are:

1. How did the structural changes in the economy in general, and

in the agricultural sector in particular, translate into the

employment trajectories of men and women working in

agriculture? (Article I)

2. How did employment in different sectors and rural-urban

migration change in rural Sweden between 1965 and 2010? How

does this local change relate to larger processes and general

societal change? (Article II)

3. To what extent are rural areas in Sweden heterogeneous in

regard to their socioeconomic and demographic structure?

(Article III)

The first paper compares the employment trajectories of two cohorts of

agricultural workers, the first born in 1943-45 and the second in 1958-60,

in order to estimate how the employment conditions in agriculture

changed during this period of rationalization and intensification of the

sector.

The second paper investigates both the migration and employment trajec-

tories of three cohorts of male and female youths living in a rural or urban

area at age 15. The cohorts, born in 1945, 1960 and 1980, were investigated

when they were 20, 25 and 30 years old. By comparing the cohorts with

each other, the article shows how the migration and employment patterns

changed between 1960 and 2010.

The third paper develops a spatial typology of rural Sweden by mapping

differences between rural areas on a range of economic and demographic

variables using data from 2008. The paper focuses on the heterogeneity

between rural areas and their spatial distribution.

The concept of rural in this thesis

The definition of rural areas has been the subject of a long academic debate

and numerous academic articles (Amcoff 2000, Cloke and Edwards 1986,

Cloke 1977, Halfacree 1995, Hoggart 1988, Hoggart 1990, Lavesson 2017,

5

Mormont 1990, Pahl 1966, Wirth 1938). Among other things, it was

discussed whether rural should refer to geographic areas or certain

mentalities or people, whether the concept should be seen as a continuum

or a dichotomy, whether it should be defined locally or regionally, and

whether agriculture or the extraction of natural resources should be seen

as a necessary condition for an area to be deemed rural.

The debate on how rural should be defined did not result in any consensus.

The simple reason for this is that the concept of rural is elusive, since it can

refer to different things (Amcoff 2000). When one is faced with an elusive

concept, there are three options: the first is to settle on a global definition

once and for all, as was the goal of many participants in the debate on how

to define rural areas; the second is to abandon the concept altogether, as

was suggested by Hoggart (1988, 1990); and the third is to accept that there

is no global or absolute definition of the concept, but that it can be used

anyway as long as its content is specified in each case (Westholm 2008).

Obviously, the third option is much more fruitful than the first two, since

it allows for research to be conducted instead of getting stuck in endless

discussions on how to define concepts.

The debate on how to define the concept of rural in any absolute sense has

silenced. Instead, a research practice has developed whereby the concept

is defined based on sparsity of population, infrastructure and/or housing

(rather than on employment in agriculture or similar work-related

criteria). These are very reasonable criteria that, as Halfacree (1995) has

shown, also correspond to how most people understand the concept of

rural. In contemporary research, the concept is more ambiguous when it

comes to the geographic scale on which to measure sparsity. The

classification of a town located in a sparse region varies depending on, for

example, whether rural is defined regionally or locally. However, this does

not have to be a problem as long as the method of operationalization is

clearly stated and the analysis acknowledges the areas to which the results

correspond. Rather, depending on the research question, one can be

interested in both sparse regions and local sparsity, and it would not be

fruitful to rule out either of the perspectives (Du Plessis et al. 2001). To be

fair, this was also Hoggart’s (1988) point; namely, that there is a

continuous reality out there which can be studied in its own right, no

matter how researchers define certain concepts.

6

In this thesis, rural is operationalized in different ways between Papers II

and III. Paper II has a regional view on rural areas since it is interested in

the labor market aspects of sparse social structures. In contrast, Paper III

applies a local definition of rural whereby all urban localities with a popu-

lation over 1,000 are classified as urban while the rest of Sweden is classi-

fied as rural. Since Paper III maps rural heterogeneity and develops a

typology of rural areas, the local operationalization is more appropriate in

that paper. In Article I, which looks at employment trajectories for

agricultural workers, rural is not defined explicitly; instead agriculture is

looked upon as a rural sector. As the discussion above suggests, this does

not mean that agriculture is a defining characteristic of rural, only that

agriculture mainly takes place in rural areas.

7

Rural restructuring

This chapter presents the framework of rural restructuring, and argues for

the necessity to connect rural change in Sweden to the larger historical

processes of technological development, social modernization and global-

ization. From this perspective, rural change should thus be viewed as the

local products of global processes.

The perspective of rural restructuring aims at analyzing the multifaceted

and interdependent nature of rural change. Instead of analyzing one type

of change in isolation from others, the aim is to describe the interconnect-

edness between different types of change. At its core also lies the realization

that local change is largely a part of regional and global processes, and that

it should therefore be analyzed in a larger context.

However, exactly what is meant by rural restructuring varies between

different studies, and the concept is often only vaguely defined. The

temporal perspective of rural restructuring also varies between studies;

some consider the whole post-war period, whereas others begin with the

economic crisis that affected the Western world in the 1970s (Pettersson

2002). It was in light of this situation that Hoggart & Paniagua (2001b)

argued that rural restructuring as a theoretical approach has been used too

flippantly, resulting in a devalued concept. Therefore, changes in, for

example, migration patterns or the agricultural sector should not be seen

as restructuring when isolated from other dimensions; instead, in Hoggart

and Paniagua’s (2001b) view, restructuring “involves fundamental

readjustments in a variety of spheres of life, where processes of change are

causally linked” (42). In their empirical study on the restructuring of rural

Spain, Hoggart and Paniagua (2001a) subsequently concluded that rural

Spain had not experienced restructuring even though large interconnected

changes had taken place; in their view, the changes were not sufficiently

profound.

On the other hand, instead of debating exactly how large or small changes

must be in order to qualify as restructuring, there is another approach to

rural restructuring that could be more fruitful. In this approach, rural

restructuring is used as a perspective of departure rather than an outcome

to be confirmed or falsified (Pettersson 2002). As previously mentioned,

this use of the term rural restructuring has many similarities with the

longue durée and the Annales School of historiography (Braudel 1958,

8

Braudel 1982b, Braudel 1995, Nordin 1989). In this view on rural restruc-

turing, Woods (2005, 41) argued that phenomena such as agricultural

change can “be interpreted as the local expression of inter-connected

processes of rural restructuring.”

The perspective of rural restructuring links back to the restructuring

debate in geography of the 1970s and 1980s (Pettersson 2002). This debate

was a response to the economic downturn that had begun in the 1960s, in

which many old manufacturing industries went bankrupt or had to

rationalize and/or relocate production. Drawing on Marxist theory,

proponents of the restructuring approach predicted, or interpreted,

fundamental shifts in the spatial distribution of production, in the organi-

zation of labor, and, ultimately, in the spatial distribution of capital

accumulation (Lovering 1989). Some authors went so far as to suggest that

the structural crisis of the 1970s was merely a precursor to a new world

order, or even the beginning of the end of the capitalist world system itself

(Wallerstein 1974, Wallerstein 2010). A central aspect of these shifts, it was

argued, was the rapid deindustrialization of the industrialized countries in

the West and Japan, and the rapid industrialization in other parts of the

world, including China (Anderson, Duncan and Hudson 1983, Fröbel,

Heinrichs and Kreye 1978, Hobsbawm 1981).

While the restructuring debate of the 1970s is seldom used as a starting

point in discussions on rural restructuring, perhaps due to the perceived

urbanity of the manufacturing sector, there are concepts from the debate

that could be valuable for understanding changes in rural areas as well.

Here the focus is on two key concepts from the restructuring approach that,

it is argued, are of particular relevance for rural restructuring: “the spatial

division of labor” and “layers of investment” (Massey 1979, Massey 1984).

The concept of the spatial division of labor refers to its geographical

distribution – both horizontally, between sectors (as in agriculture or

manufacturing), and vertically, within sectors (as in production and exec-

utive functions).1 The spatial division of labor is furthermore subject to

redistributive pressures from new layers of investment, whereby profita-

bility determines the location for capital investment. In this sense, the

economic activities in an area may be the product of both new and old

1 While Massey (1995) primarily focused on the spatial division of labor within sectors, the concept is used here to refer to both regional and sectoral specialization. It is, however, not uncommon that these overlap.

9

layers of investments, which form separate paths of development (Massey

1978). Thus, economic decline in an area, and thereby the redistribution of

the spatial division of labor, is related to the decline of new investments,

whereby one or several paths of local development decline due to a relative

decrease in the profitability of local investment (Lovering 1989). In turn,

the redistribution of the spatial division of labor leads to changes in the

migration pattern, which ultimately lead to a redistribution of the aggre-

gate settlement pattern (Fielding 1989).

This depiction is complicated by the inertia and the constraints that espe-

cially labor is subjected to. While capital is also subject to inertia and

constraints (for example due to physical investments or regulations), it is

relatively more mobile compared to labor, and this mismatch between

capital and labor may create structural imbalances in periods of great

economic change. The reason for this is that the individuals comprising the

local labor force are bound by their training and skills, which makes it hard

(but not impossible) to switch occupations, should their industry of

occupation shut down. Individuals are also bound by geography to a

certain extent. While they can migrate or even emigrate, there is a cost

connected to this (Lovering 1989). Both these types of immobility of labor

– occupational and geographical – increase with the age of the individual

(see discussion on the life-course perspective in the Methods section).

The local outcome of the dynamic depicted above relates to the concept in

economic geography of “regional portfolios”: economically diverse regions,

often larger urban areas, are characterized by many separate layers of

investment meshed on top of each other, while economically conforming

regions, often smaller peripheral areas, are characterized by one or a few

layers of investment. In periods of restructuring, it has been hypothesized

and also shown empirically, that regions with a diverse economic portfolio

are more resilient to changes than those with a conform portfolio (Brown

and Pheasant 1985, Chiang 2009, Markowitz 1991, Neffke, Henning and

Boschma 2011).

While the number of possible changes to the spatial division of labor that

can arise from the dynamic of layers of investment is very large, not to say

endless (Lovering 1989), it can still serve a purpose to consider a few ideal

types. Therefore, four stylized examples of how layers of investment can

relate to labor demand are considered in Figure 1, where each box repre-

sents a local labor market. It should be emphasized that these four

10

examples are in no way exclusive. The top left box in Figure 1 shows the

rise and decline of one layer of investment, but as the first layer of invest-

ment declines a second layer takes off and is able to absorb the unemployed

labor from the first layer. The top left box could thus be seen as a stylized

example of the process of creative destruction, whereby the labor force of

declining industries is freed in order to contribute to growing industries

(Bartelsman, Haltiwanger and Scarpetta 2004). The top right box shows a

similar dynamic, but in contrast to the top left box the labor force of the

declining first layer of investment lacks the appropriate skills to be

absorbed into the second layer. Thus, a new labor force with the

appropriate skills is needed. One example of this dynamic may be the

growth of the tourism sector in the Swedish mountain range, which largely

depends on a non-local labor force, presumably due to a mismatch between

the local labor force and the service-oriented skills required in the tourism

sector (Lundmark 2005). The bottom left box exemplifies the rise and

decline of a local labor market, whereby the declining layer of investment

is not replaced by a new layer. Typical examples of these labor markets are

one-company towns, in many cases built around mining activities or large

factories, where the enterprise shuts down without being replaced by other

activities. The bottom right box, finally, represents a more diverse labor

market with many layers of investments that are meshed on top of each

other. This box reflects a local labor market with a more diverse economic

portfolio that is often found in larger urban areas

11

Figure 1: Four stylized examples of the dynamics between layers of investments and the local labor market

12

Apart from adding a deeper theoretical understanding to the rural restruc-

turing approach, the concepts of “spatial division of labor” and “layers of

investment” may also delineate the relevant focus of the approach. While

the rural restructuring perspective explicitly strives for holism, for a com-

plete description of many dimensions of change (cultural, social, political,

economic, demographical, etc.) in both the temporal and spatial dimen-

sions, research must by necessity be delimited to a manageable number of

factors, and this is also the case for research from a rural restructuring

perspective. By focusing on the redistribution of the spatial division of

labor, the approach is delimited to the spatial redistribution of the

population, the spatial redistribution of economic activities, and the

broader factors that affect the propensity for local investment.

Woods’ (2005) division of rural restructuring into global drivers and local

outcomes can be applied here, with the redistribution of the population

and of economic activities sorted under the category of local outcomes, and

the broader factors affecting the local outcomes sorted under global

drivers. According to Woods (2005), local rural change is driven by the

processes of technological development, social modernization and

globalization. However, the division into global drivers and local outcomes

is primarily a schematic one, made in order to aid comprehension, which

omits many complexities of the real world. Technological innovations are,

for example, local products that are historically dependent on previous

local outcomes, which, in a sense, makes them endogenous to the process

of economic change rather than exogenous (Warsh 2007). On the other

hand, treating technological development as an exogenous factor can aid

the comprehension of economic change (Dicken 2003, Schumpeter 1943,

Solow 1956, Woods 2005). A similar argument can be made for the

schematic division between technological development, globalization and

social modernization. While these processes are interrelated, for example

with development in transportation technology being an important part of

both globalization and modernization, it fits the purpose here to present

them separately.

Technological development

Technological development played a central role in the Industrial

Revolution, and through the introduction of new consumer goods and new

or improved methods of production, transportation and communication,

continues to shape the modern world (Chesnais 1986, Dicken 2003,

13

Schumpeter 1943, Toffler 1990); and rural areas are certainly no exception

to this development (Woods 2005).

One aspect of the local impact of technological development is its

importance for new local investments and thereby its redistributive

pressure on the spatial division of labor, whereby new technology has

changed the conditions for local investment by expanding the potential

markets for products, and by increasing industrial productivity (Dicken

2003, Lovering 1989, Schön 2012, Smith and Marx 1994). In this regard,

refrigeration technology, which was commercialized around 1900, made it

possible to store food for longer periods of time and to transport it over

longer distances. Together with other developments, such as transporta-

tion and globalization (see below), refrigeration has thus contributed to

increasing the potential market of some agricultural products (Woods

2005). A larger market, in turn, increases the benefits of the division of

labor, including the spatial division of labor and economy of scale (Massey

1984, Smith 1986). Following this line of thought, but without arguing for

technological determinism (Smith and Marx 1994), the technology of

refrigeration has thus contributed to, or at least facilitated, the

development of farm specialization, large food-processing companies,

supermarkets and regional specializations (Woods 2005). Transportation

vehicles comprise another invention, or rather bundle of inventions, that

has changed the conditions for local investments through expansion of the

potential market (and which, thereby, has increased the benefits of the

divisions of labor and economy of scale) (Dicken 2003, Massey 1984, Smith

1986). In terms of expanding the potential market, transportation vehicles

have led to the faster and cheaper transportation of goods. Refrigeration

trucks have already been mentioned as an example of technology with

implications for the spatial reach of some agricultural products, but the

economic impacts of transportation vehicles have of course been much

broader, and basically no economic activity is unaffected by the technology.

In forestry, the introduction of trailers completely replaced log driving and

timber rafting as methods of transporting timber (Schön 2012). In

manufacturing (or any other sector), a product produced in China could be

delivered by aircraft to consumers in Europe a couple of days later and vice

versa. In The Wealth of Nations, Smith (1986, 31) argued that “the division

of [labor] is limited by the extent of the market”, and through the

introduction and continued development of transportation vehicles this

extent has certainly increased, at least in potential terms (Dicken 2003,

Smith and Marx 1994). Transportation vehicles have also expanded the

14

potential market for tourist consumers, with importance for both rural and

urban areas. In the past few decades both international and national tour-

ism have increased considerably, partly as a result of development in

transport technology (Page 1998). A third bundle of inventions with major

importance for expanding the potential market of goods is those in infor-

mation technology (Castells 2010, Dicken 2003), with for example the

telephone, or more recently the Internet, allowing contact with faraway

customers in seconds. The continued development of information

technology has also facilitated the spatial division of labor within

corporations, by allowing fast information exchange over long distances

between the executive organ and the producing organ. This has often

resulted in a spatial division, with the headquarters and sales offices

located close to the market (for example, developed countries or urban

areas) and the production units located in places where production costs

are lower (for example, developing countries or rural areas) (Castells 2010,

Dicken 2003, Woods 2005).

Apart from expanding the potential market, technological development

has also increased productivity, not least in manufacturing, agriculture and

forestry. Motor vehicles, for example, and their subsequent use in

agriculture and forestry, have had great implications on the production

process in these sectors. In agriculture, the large-scale introduction of

tractors and harvest machines in the second half of the 20th century

substantially increased productivity (Schön 2012). While they are not a

simple, unidirectional, cause behind the decline of employment in the

agricultural sector, tractors and harvest machines, along with other sector-

specific technology such as milking machines, new fertilizer techniques,

GMO, etc., have certainly reduced the demand for labor in the sector by

increasing productivity (Schön 2012, Woods 2005). Likewise, productivity

in the forestry sector also increased as a consequence of motor vehicles,

such as forestry machines, and other forestry-specific equipment. The

introduction of chainsaws in the 1950s, for example, coincided with a

decrease in employment (Hjelm 1991). Since then, chainsaws have been

replaced by more sophisticated forest machines, which has contributed to

reducing the workforce in forestry to an all-time low in relation to

production, a development that is likely to continue.

A third aspect of technological development relevant for local investment

and related to both the immobility of physical capital and the labor force

(see discussion above) is that it may be more profitable to invest in new

15

technology in new places, unaffected by old technology. In this regard,

regions based on old technology may, due to lock-in effects and the inertia

of physical and human capital, obscure investments based on newer

technology. To the extent that this is important, it may therefore favor

areas that are not burdened by old technology. One example of this is what

has been termed the Second Industrial Revolution, which started in 1870s,

with Germany developing relatively more quickly than Britain, partly

because Germany was less bound by the old steam-based technology and

could therefore adapt more quickly to a technology based on electricity

(Broadberry and O'Rourke 2010).

A different aspect of technological development is its importance for the

mobility of the labor force; for example, the large-scale introduction of the

automobile meant that it was no longer necessary to live within walking

distance of one’s workplace (Falk 1976). One implication of this was that

people could live in rural areas and work in urban areas, and vice versa. In

Sweden, the large-scale introduction of the automobile took off around

1950 and had reached 1 million cars by 1960, a development that

substantially increased the spatial reach of the city and, among other

things, spurred counterurbanization (Amcoff 2000, Falk 1976). However,

the daily distance people are willing to travel is limited, which means that

the majority of counterurbanization due to commuting was spatially

limited to the vicinity of urban areas, something that has resulted in

commuter belts around urban areas (Falk 1976, Nutley 1998).2

Information technology has also been important in increasing the mobility

of the labor force, with the telephone, television and Internet all

contributing to reducing the information isolation in rural areas. When the

Internet arrived there were expectations it would induce employment

increase in peripheral areas, and some scholars even argued that the new

modes of communication may result in the death of geography due to the

space-time compression of information travel (Cairncross 2001, Friedman

2005). More reasonable views, however, are that geography is not dead;

instead, new developments in the field of economic geography on clusters,

relatedness, labor mobility, etc. suggest that geography remains an

important factor despite new communication technologies (Castells 2010,

Eriksson 2009, Markusen 1996, Porter 1998). Nevertheless, developments

2 While it is to some extent a semantic question, the urban nature of commuter belts should discourage the term “counterurbanization” in favor of “suburbanization” or “peri-urbanization”.

16

in telecommunication technology have at least made it conceivable for

companies dealing with information to locate in peripheral areas and for

persons living in rural areas to work from home.

Social modernization

In many accounts, the age of modernity started with the Industrial Revo-

lution in Britain in the 1780s and the French Revolution in 1789

(Hobsbawm 2010, Nordin 2016). As such, the historical transformation

from a pre-modern to modern society was a multidimensional phenome-

non, ranging from technological and economic changes to political changes

and changes in the thoughts and ideas at the center of society (Hobsbawm

2010). The focus in this section on social modernization, however, is

narrowed down to a few aspects of modernity that are of particular

importance for contemporary economic and demographic changes in rural

Sweden. The aspects in question are the location of industries, the growth

in prosperity and the development of the welfare state, and the shift in

societal norms towards post-materialistic values.

While all these aspects are a part of modernity they are of course processes

or products and, as such, vary over time and in space. A conventional

periodization of modernity is Berman’s (1983) division into early

modernity (1500-1789), classical modernity (1789-1900) and late

modernity (1900 onward) (Osborne 1992). Using this division, industri-

alization and growth in prosperity can thus be placed in the periods of

classical and late modernity3, and the rise of mass parties (as opposed to

cadre parties) and the development of welfare states in the Western world

can be placed into late modernity, as a constituent part of that period (Flint

and Taylor 2011). While changes in norms towards post-materialistic

values are perhaps harder to pinpoint in history, it is sufficient for our

purposes here to consider Inglehart’s (1971, 2004, 2000, 2003) empirical

findings, showing that the changes in norms correlate with prosperity (and

thereby with industrialization and the welfare state), whereby the

population in richer countries increasingly holds more post-materialistic

values (gender equality, self-actualization, cosmopolitanism, etc.). In this

3 However, Sweden did not industrialize on a large scale until the latter part of the 19th century (see Section 3).

17

sense, the shift towards post-materialistic values can be seen as an ongoing

process.

Industrialization has had large effects on new layers of investment and the

redistribution of the spatial division of labor. In fact, industrialization

demanded a more spatially divided production process (Smith 1986). In

the context of this thesis, one important aspect concerns the urban/rural

dimension of the new layers of investment related to industrialization and

the growing manufacturing sector. Whereas the literature on rural

restructuring primarily focuses on the agricultural sector, the manufac-

turing sector is certainly not absent in rural areas (Lundmark and

Malmberg 1988). In this regard, Weber (1962) argued that factories locate

where the sum of production costs is as low as possible, whereby he

primarily considered factors related to the nature of the raw material,

transportation costs to the factory and to the market, and the location of

the labor force. A consequence of this reasoning is, for example, that

industries using “weight-losing materials” (such as minerals or fuel) have

a propensity to locate close to the source of the raw material, in order to

minimize the transportation of the waste material, and this location is not

necessarily an urban area (although it may give rise to one). When, on the

other hand, the raw material is “ubiquitous” the industry will be located as

close as possible to the market and to the labor force in order to minimize

transportation costs. While Weber’s model finds many exceptions in the

real world, the point here is merely to note that manufacturing was, or is,

not exclusively an urban phenomenon. At the same time, industrialization

gave rise to urbanization on a scale surpassing all previous epochs

(Andersson 1987, Lundmark and Malmberg 1988).

Apart from urbanization, industrialization led to economic growth on a

scale unprecedented in human history, with GDP per capita more than

doubling in Western Europe in the 19th century and increasing six times

over in the 20th century (Bolt and Zanden 2014). The growth in prosperity

during the 19th century resulted in extreme economic inequalities between

the working and upper classes, creating tensions in industrializing

countries. In his pessimistic account of the effects of industrialization in

England, Engels (1963, 92) wrote that those in the working class “have no

property whatsoever of their own, and live wholly upon wages, which

usually go from hand to mouth. […] Every working-man, even the best, is

therefore constantly exposed to loss of work and food, that is to death by

starvation, and many perish in this way.” While the working class lived

18

under poor conditions the “bourgeoisie” became very rich, and, according

to Engels (1963, 55), this inequality was because “manufacture centralises

property in the hands of the few.”

In the latter half of the 19th century the social tensions led to a new political

movement in Europe of “mass parties” that challenged the old cadre parties

and, eventually, won (Flint and Taylor 2011, Weber 1968). With the victory

of the mass parties a new era of social liberalism in Western politics began,

whereby the central idea was to combine liberal policies benefiting the

market (for economic growth) with social policies redistributing wealth

from the hands of the few to the hands of the many (for legitimacy) (Flint

and Taylor 2011). The rise of social liberalism and the welfare state, in turn,

has had great importance for the labor market in Western countries, espe-

cially in Sweden, where the welfare state and the public sector grew partic-

ularly large (culminating in the 1980s). Taking Sweden as an example, the

welfare state could be said to have had a homogenizing effect on the geog-

raphy, with each municipal center (after the municipal reform of 1971)

obligated to provide a required minimum of public services, including

schools, libraries, healthcare, and other service facilities (see also Section

3). In this respect, Kohli (1986) has argued that modernity and the welfare

state have had both a homogenizing and a diversifying effect on the life

course – homogenizing in the sense that the importance of social and

regional background decreased (but did not diminish; especially gender

remained important) while the importance of individual age increased, and

diversifying in the sense that individuals had more freedom to choose their

own paths in life (Shanahan 2000).

The growth of the public sector coincided with a shift towards more post-

materialistic values with, among other things, demands on increased

gender equality resulting in new generations of women who were expected

to enter the labor market. Again this was especially the case in Sweden,

where women’s participation on the labor market is among the highest in

the world (Lewis 1992). On the other hand, the welfare state demanded

greater participation of females on the labor market, and there remains a

strong gender division of labor even in modern Sweden. Another way of

interpreting the relationship is thus that the growth of the female-

dominated public sector enforced the differences between the sexes (Lewis

1992, Orloff 1996).

19

The change towards post-materialistic values was, according to Inglehart

(1971), the result of prolonged prosperity, whereby the generations

growing up without experiencing economic hardship would increasingly

adapt post-materialistic values, as they had never had to concern them-

selves with the lower steps on Maslow’s hierarchy of needs. In this sense

the welfare state, with its redistributive function, contributed to this shift

in values. In a slightly different view, the breakdown of the rigid identities

of the past under capitalism was already noted by Marx and Engels (2002,

16) in the Communist Manifesto, summarized by their thesis “all that is

solid melts into air”. Thus, while Inglehart’s (1971, 2004) empirical

findings suggest that the changes in values were caused by growth in

prosperity, both growth in prosperity and changes in values, in Marx and

Engels’ view, are a constituent part of modernity (Berman 1983).4

Apart from gender equality, the shift towards post-materialistic values put

increased emphasis on values promoting self-actualization as opposed to

those directly connected to the survival of the individual, the family and

the nation. With increased emphasis on self-actualization, an individual’s

social background became less important as a determining factor for his or

her future occupation (Zelinsky 1971). Again, it could be argued that the

welfare state was an important material basis for the reduced importance

of social background, as a central idea of the welfare state was to balance

out inequalities due to different social backgrounds (Berggren and

Trägårdh 2015, Kohli 2007, Magnusson 2006).

The continued development towards post-materialistic and self-actualiz-

ing values coincided with increases in higher education, and, without

stating that the changes in values are a cause behind the increase in higher

education, their co-occurrence fits neatly together. The increase of

4 However, as with all historical epochs, modernity encompasses contradictions. While capitalism under welfare regimes encompasses trends towards increased equality and the reduced importance of social and regional background (Zelinsky 1971), capitalism, in its more unrestricted form, may also reinforce such factors by exploiting pre-existing divisions along ethnic or gender lines (McDowell and Massey 1984). This strategy of “divide and rule” was an integral part of not least the colonial project (Flint and Taylor 2011). In this sense, capitalism entails both continuity and discontinuity with the feudal past. Furthermore, while Marx and Engels (2002, 15-16) argued that capitalism dissolved the “idyllic relations” of feudalism, their central point was that capitalism merely replaced these relations with an all-encompassing class relation: “Society as a whole is more and more splitting up into two great hostile camps, into two great classes directly facing each other – Bourgeoisie and Proletariat. […] In one word, for exploitation, veiled by religious and political illusions, it [capitalism] has substituted naked, shameless, direct, brutal exploitation.”

20

participation in higher education in the developed world has changed the

life pattern of many young adults, which has had consequences on periph-

eral areas. Not only does it mean that people in peripheral areas choosing

higher education have to relocate to a university town, but also that their

prospects of employment in peripheral areas decrease due to the smaller

number of qualified jobs, which makes it unlikely they will move back.

Increase in higher education is thus an important factor in explaining rural

out-migration (Lundholm 2007, McGrath 2001, Pettersson 2002, Woods

2005).

Apart from changes in values, the increase of participation in higher edu-

cation from the 1960s onward coincided with the more general shift from

a manufacturing-based economy towards a service-based economy – a

shift that alternately has been described as the Third Industrial Revolution

(Magnusson 2002, Schön 2012), the Knowledge Economy (Cooke 2001,

Porter 1998), the Network Society (Castells 2010) or Liquid Modernity

(Bauman 2000). The increase in higher education can thus, partly, be

viewed as a response to meet the demands of this economic change, with

economic success increasingly depending on specialized knowledge and

comparative (or competitive) advantages on a highly competitive global

market (see also Section 2.3 on globalization).

Furthermore, the welfare state, which can be seen as a political response to

the manufacturing economy (Flint and Taylor 2011), has entered a period

of crisis as a consequence of the transformation of its material base

(Magnusson 2006). The shift towards a service economy and the crisis of

the welfare state may also have larger implications for the changes towards

post-materialistic values observed throughout the post-war period. If the

welfare state has contributed to reducing the importance of social and

regional background, its decline could mean a renewed importance of such

factors. In looking back on the past decades of life-course research, Kohli

(2007, 257-258) suggested that the homogenization of the life course

under late modernity, the “institutionalized life course” in Kohli’s terms,

rather than a linear trend was a product of “the ‘Fordist’ model of social

structure”, which culminated in the 1960s and “was based on rapid and

seemingly stable economic growth, low unemployment, and expansion of

the welfare state.” However, while there is general agreement that the

institutionalized life course is changing towards increased heterogeneity

and stratification, it remains an open question as to exactly what this

21

change entails and whether it correlates with established

sociodemographic groups (Kohli 2007).

Globalization

Contemporary economic globalization has in many ways altered the

conditions for economic activities in localities in many parts of the world,

which among other things has changed, and continues to change, the

spatial division of labor (Dicken 2003). However, and as the above

discussion on the dynamics of the spatial division of labor suggests, this is

not to say that the processes of economic globalization have had the same

effects in these localities (Woods 2005).

Viewed as the increased interconnection of localities around the world,

globalization is a multidimensional process that includes political changes,

economic changes, changes in migration patterns, and cultural changes

(Dicken 2003, Marks 2006). As the focus in this thesis is on economic and

demographic changes in rural areas, the depiction below is delimited to

economic globalization and the globalization of personal mobility (Woods

2005). Likewise, the period referred to here is that of contemporary

economic globalization, which, according to many interpretations, took off

on a large scale in the 1970s (Castells 2010, Dicken 2003, Friedman 2014,

Pettersson 2002, Woods 2005). While globalizing phenomena such as

colonization, imperialism, foreign direct investment, international trade,

large-scale international migration, etc. undoubtedly have a long history

(Braudel 1982b, Friedman and Friedman 2008a, Marks 2006, Nordin

2006, Wallerstein 1979), it is arguably the case that something new

happened in the 1970s, which stood in contrast both to the development in

the preceding decades and, at least in terms of technology and the extent

of the time-space compression, to previous periods of internationalization

in history (Castells 2010, Harvey 1999).

According to many authors, technological development in transportation,

computation and telecommunication is a main driver behind contempo-

rary globalization (Castells 2010, Dicken 2003, Harvey 1999, Schön 2012).

Apart from new technology, another driver of contemporary economic

globalization is the deregulation and financialization of the economy, and

the increased emphasis on free trade policies, which has reduced national

control over the economy (Schön 2012, Woods 2005). According to Harvey

(2007), the deregulation of the 1970s marked the start of a neo-liberal

hegemony in world politics. In Harvey’s view, neo-liberalism, represented

22

by politicians such as Ronald Reagan and Margaret Thatcher, should be

viewed as an intentional attempt by the upper classes to regain dominance

over the lower classes. However, from another perspective, that which

Harvey calls neo-liberalism may be a consequence of the mechanics of

capitalism rather than an intentionality (Friedman 2014). In referring to

earlier periods of internationalization and financialization, Braudel

(1982a, 246) argued that the Italian investors of the 16th century and the

Dutch investors of the 18th century had no choice but to invest in productive

enterprises abroad.

Expanding on Braudel’s argument, Friedman (2014) argues that globali-

zation – or neo-liberalism – is a historical phase of hegemonic decline,

whereby the old economic centers of the world export large shares of their

productive capital and new centers rise to a hegemonic position in the

world system (Arrighi 1994, Arrighi 1999, Friedman 2004, Friedman and

Friedman 2013, Wallerstein 2010). In Friedman’s view the phase of glob-

alization, including the phenomena of deregulation, internationalization

and financialization, is thus the last phase of a longer cycle of capital

expansion and contraction (Friedman 2014).

Besides emphasizing the structural features of globalization, this view thus

suggests that globalization, and the related phenomena of deindus-

trialization, deregulation and financialization, is a process of economic

decline in the old economic centers of the world, in which production is

replaced by consumption and increase of debt. In this sense, the world

system view on globalization stands in stark contrast to the view proposed

by Castells (2010), Albrow (1996) and Appadurai (1996) – whereas these

authors see the birth of a new global era characterized by flows and

networks, Friedman & Friedman (2013), Wallerstein (2010) and Arrighi

(1999) see the end of an era of American and Western hegemony.

While the world system view on globalization is appealing, not least since

it offers an explanation to the historical shifts in economic power and world

hegemony (Arrighi 1994), it should be emphasized that contemporary

globalization consists of many new features, not least technologically

(Castells 2010, Dicken 2003, Harvey 1999). Furthermore, the historicism

implied by this view is certainly not unproblematic, as it may hide that the

contemporary situation in fact, and by definition, is new (Popper 2002,

Popper 2012).

23

No matter whether contemporary economic globalization is viewed as an

intentionality or as a structural feature hardwired into the capitalist world

system, as a historically unique phenomenon or as a reoccurring

phenomenon, or as the start of a new global era or as the end of an era of

American and Western hegemony, there are certain facts of the contem-

porary situation that are relevant for local investments, for the redistri-

bution of the spatial division of labor and, by extension, for economic

changes in rural areas in Sweden. One such fact is the increase in inter-

national trade, which tripled as a share of world GDP between 1960 and

2010, surpassing the level of trade in the more integrated period before

WWI in the first half of the 1970s (Feenstra, Inklaar and Timmer 2015,

Klasing and Milionis 2014). This was especially the case for the interna-

tional trade of manufactured goods, which intensified due to a number of

factors, including major cuts in tariffs for manufactured goods, more

effective transportation technology, and increased intra-firm trade (Abreu

1996, Bruinsma 2003). The growth in trade was slower for agricultural

products than manufactured products. The reason for this relatively slower

growth in international trade was due partly to policy measures, such as

higher tariffs on agricultural goods and subsidization of the national

agricultural sector in developed countries, and partly to the structure of the

agricultural sector with relatively less intra-firm and intra-industry trade

(Bruinsma 2003).

While international trade with manufactured goods increased the most,

the trade of both agricultural goods and raw materials saw significant

increases as well (Bruinsma 2003, WTO 2017). One implication of

increased international trade is that prices compete on an international

level, which, in turn, means that the conditions for local economic activity

are determined by factors external to the national demand to a larger

extent than before, partly removing local control over production.

Especially prices on raw materials are vulnerable to price changes on the

world market; for example, mining activity can become profitable or

unprofitable based on changes in world market mineral prices (Cutter,

Renwick and Renwick 1985, Knobblock and Pettersson 2010). But the

agricultural sector is also affected by changes in world market food prices,

despite attempts to minimize the effect via economic barriers, such as

tariffs and subsidies (Bruinsma 2003).

A related implication of increased international trade is that the market for

each individual producer expands which, via comparative advantages and

24

economy of scale, leads to increased specialization in the division of labor,

including the spatial division of labor (Heckscher and Ohlin 1991, Massey

1984, Ricardo 1817, Schön 2012, Smith 1986). Changes in local invest-

ments and the redistribution of the spatial division of labor due to expand-

ing markets are most clearly visible in the manufacturing sector, for which

the profitability of investing in countries with lower wages increases. This

has led to a pressure towards both a direct and indirect relocation of indus-

trial production – direct when companies relocate or outsource their

production from high- to low-wage countries to increase profit, and

indirect due to the competition advantages of lower wages and cheaper

production costs (Dicken 2003, Friedman and Friedman 2013, Massey

1984, Porter 1986). The direct relocation of production, through FDI

(Foreign Direct Investment), is mainly relevant for the manufacturing

sector, since for obvious reasons it is harder to relocate production in the

primary or tertiary sector (natural resources cannot be relocated, and only

some services can). However, the expanding market has also affected

agriculture by, among other things, increasing farm agglomeration and

specialization. As farms become less dependent on local demand, but can

export their products to distant places, the profitability of specialization

increases (Woods 2005). A new, productivist, system of agricultural

production thus developed, with farms specializing in single products to

maximize their output. One example of this is the farm clusters of

California, where one producer may specialize in cattle and another in

carrots, while a third produces fertilizer by buying raw material from the

cattle farmer and selling the refined product to the carrot farmer (Woods

2005).

With increased specialization in both the manufacturing and the natural

resource sectors, there has also been a redistribution of the spatial division

of labor within corporations, whereby multinational corporations with vast

resources have emerged on a large scale. In agriculture, for example, only

four corporations control large parts of the global seed market

(Hendrickson and Heffernan 2002). Many of the large agricultural

companies are furthermore cooperating with each other in order to

increase control of the production chain. Hendrickson & Heffernan (2002)

identify three global food clusters that are governed by the corporations

Cargill and Monsanto, ConAgra, and Novartis and Archer Daniels Midland

(ADM). These corporations work with gene modification and seed

production (Monsanto, ConAgra and Novartis) and the processing of food

from farmers (Cargill, ConAgra and ADM), which gives them control of the

25

production within the food cluster. Furthermore, their control over food

processing and their dominant role on the market gives them significant

power to decide the prices paid to farmers. Another example from Sweden

is the forestry company Svenska Cellulosa Aktiebolaget (SCA), which in the

1960s was a regional company operating in northern Sweden. Today the

company operates on a global scale, with units and production in many

parts of the world (Layton and Lindgren 1992, Pettersson 2002).

Expanding markets have also affected the food retail sector, leading to the

concentration of economic power (Woods 2005). One example of this is

Sweden, where three corporations accounted for 88% of food retail sales in

2007 (Swedish Competition Authority 2008). Similar to the large food-

processing corporations, the near monopoly in the food retail sector allows

the leading food retail corporations to have significant influence over the

price paid to farmers. Moreover, the development of own-brand products,

which has increased steadily since the 70s, has further increased the food

retail corporations’ influence over the production chain and the price paid

to farmers. Apart from their direct effect on agriculture, large

supermarkets in combination with increased commuting have forced many

local stores in rural areas out of business, reducing the service level and

attractiveness of these places (Nutley 1998, Woods 2005).

To summarize the above discussion on economic globalization and

concentrate its importance for local economic changes: the causal chain

started with deregulation and new technology (not least telecommunica-

tion and transportation) (Castells 2010, Dicken 2003, Harvey 2007). These

factors have in turn led to increased international trade, which has

expanded the market for producers (Dicken 2003, Feenstra et al. 2015,

Woods 2005). The expanding market has led to spatial shifts in the layers

of investment, whereby, among other things, low production costs present

a comparative advantage (Friedman 2014, Friedman and Friedman 2013,

Friedman and Friedman 2008b, Massey 1984). This has also led to

increased specialization, rationalization, and the rise of international

corporations (Dicken 2003, Woods 2005). Increased specialization and

the spatial shift in local investment have, in turn, had redistributive effects

on the spatial division of labor; for example, there has been a spatial

redistribution in manufacturing employment from high- to low-wage

countries (Dicken 2003).

26

Another aspect of globalization that is relevant for economic and demo-

graphic changes in rural areas is the globalization of personal mobility:

both permanent and temporary migration have increased substantially

since the 1970s (Woods 2005). In this regard, the tourism sector has grown

several times over in recent decades, and, not least, otherwise peripheral

areas have become attractive destinations for both national and interna-

tional tourists (Lundmark 2006, Müller and Ulrich 2007). For a few

peripheral areas, tourism even comprises an essential part of their regional

economy (Müller and Ulrich 2007). Furthermore, the workforce within the

tourism sector is hard to rationalize without compromising service quality,

which means that growth within the sector has led to growth in

employment opportunities in the areas in question (Lundmark 2006).

The globalization of personal mobility has also led to large increases in

international migration streams, but also to shifts between various kinds

of migration motives. In the past 30 years, for example, Sweden has expe-

rienced large increases in especially asylum immigration while labor

immigration has declined (until 2006, when it began increasing again).

While the largest part of immigration is directed towards urban areas, rural

areas have also seen a substantial increase in immigration. By contributing

to demographic changes, the large-scale immigration is an important

phenomenon in rural areas as well (Hedberg and Haandrikman 2014).

27

Settlement and employment

development in Sweden

In this section, the settlement and population development in Sweden are

described in order to place the empirical studies in this thesis in their

historical context. However, before moving on to the historical description,

some theoretical considerations are necessary, regarding spatial

population redistribution and its relation to demographic and economic

changes.

The redistribution of the population can be broken down into the natural

component of birth and death rates, and the migration component of in-

migration and out-migration; however, these two components are inter-

twined. A local decrease in the birth rate may, for example, be the result of

a past out-migration of youths, and a war or famine may lead to both an

increased death rate and mass out-migration (consider for example the

potato famine on Ireland during the period 1845-1852, which increased

emigration from the country). While the natural component is important

for population redistribution, theories on population redistribution have

primarily focused on the migration component, as in many cases it is an

underlying cause of regional differences in the birth rate (Borgegård,

Håkansson and Malmberg 1995). These theories have concerned

themselves with a wide array of possible determinants of migration,

including demographic, social and economic conditions and their spatial

variation (Andersson 1987), policy interventions, and attitudes towards

different places, whereby the importance of these factors in shaping the

migration patterns is partly dependent on the perspective, the timeframe

of analysis, and on variations in the importance of these factors in time and

space (Borgegård et al. 1995).

From a long-term historical perspective, a fruitful view on the migratory

processes and the redistribution of the population is to consider them

together with changes in economic activities – a view often referred to as

the structural perspective (Andersson 1987). This perspective combines

the two spheres Hägerstrand (1970) refers to as “population system” and

“activity system”. The reason for this is that the advance and retreat of

settlements are closely correlated with economic activities, with the

exploitation of local resources having played an important part in shaping

the contemporary settlement pattern (Enequist 1960, Schön 2012). This is

28

not to say that changes in the activity system by necessity lead to changes

in the population system or vice versa, but rather that the two spheres are

bound to each other (Andersson 1987, Weber 1962). In an investigation of

the 20th-century urbanization in Sweden, Andersson (1987) emphasized

the constituent relationship between the birth of capitalism, industrializa-

tion, and the growth of wage laborers on the one hand, and urbanization

on the other. Zelinsky (1971), for his part, argued that there is an inherent

relationship between the demographic transition (or the “vital transition”

in Zelinsky’s terms) and changes in social norms, migration patterns,

infrastructure, technology and economic activities, and that these changes

together comprise an important part of the modernizing process itself.

The remaining part of this section presents a brief description of the

settlement development, the demographic development, and the economic

development in Sweden from 1750, with emphasis on the relationship

between the population system and the activity system.

In a study on the long-term trends of the population redistribution in

Sweden, Borgegård et al. (1995) identified five periods of redistribution

between 1780 and 1990. In the first period, 1780-1890, the settlement

pattern was characterized by dispersion on both the regional and local

levels, with areas previously unoccupied by the Swedish state becoming

occupied. One reason for the dispersion of the population was population

growth: the Swedish population grew an average of 0.5% per year during

1750-1800, and by 0.8% per year during 1800-1890. The population

increased by 3 million in the period from 1750 to 1890, reaching a total of

4.8 million by 1890 (Schön 2012, Statistics Sweden 2017). In the largely

agrarian Sweden of 1750, the growing population led to an increasing

demand for new land, which pushed the population both north and south

as well as west from the more densely populated regions of Mälardalen,

Östergötland and Kopparberg (Borgegard et al. 1995, Schön 2012).

In 1750 the region of Lappland was mostly inhabited by the nomadic Sami

people, but besides a few settlements in the southern parts of the region it

was largely unoccupied by the Swedish state. By 1800 many new

settlements had formed in the region, but these were mostly located in

proximity to the old settlements in the south. By 1850 settlements had

formed over the whole region, even in the more peripheral mountainous

parts of Lappland, and already by this period the outer boundaries of the

northern settlement expansion had been reached (Enequist 1960). While

29

the initial population growth in Lappland occurred through in-migration,

or external colonization, a large part of the subsequent population growth

in the region could be described as a process of internal colonization,

primarily driven by natural population growth and resettlement from

“mother settlements” (Bunte, Borgegård and Gaunitz 1982, Bylund 1956,

Bylund 1960). Until around 1850, the settlers of the northern expansion

used the forests primarily for animal keeping, hunting and farming, and

for the production of charcoal for use in the smelting of iron (Aldskogius

1960, Enequist 1960) and tar. One enabling factor behind the northern

expansion was the introduction of the potato, which was rich in nutrition

and could endure the harsher climate of the northern inland. Tax reduction

policies for those who chose to settle in the north and freedom from

military service among the men were two other important factors (Bylund

1956, Enequist 1960).

While Sweden was primarily an agrarian society in the 19th century,

employment in various forms of craftsmanship as well as industrial

employment increased, especially in the latter half of the century (Enequist

1960, Schön 2012). Timber rafting was also an expanding sector during the

19th century, but in contrast to work in the forest it was not easily combined

with farming, as it was performed during the summer months (Bunte et al.

1982). Work in both agriculture and the growing manufacturing sector was

segregated by gender. In agriculture, for example, milking was primarily

performed by women while men cared for the horses. Household tasks,

such as washing, cleaning and caring for the children, were also performed

by women (Ohlander and Strömberg 2002). In manufacturing, women

worked as spinners in the expanding textile industry. In the 18th century,

spinning had primarily been performed on the farm as a complement to

agriculture; in the latter half of the 19th century this practice was replaced

by textile factories (Ohlander and Strömberg 2002).

The next period identified by Borgegård et al. (1995), 1890-1930, was

characterized by regional dispersion and local concentration, meaning that

urban areas grew or took form in many parts of the country. The population

redistribution in this period was closely related to the industrialization of

Sweden. Around 1850 in northern Sweden, and earlier in southern

Sweden, smaller saw mills started competing with the traditional use of the

forest – partly as a result of increased demand from Britain (Bunte et al.

1982). Similar to smelters, the saw mills were located near streaming

rivers, which were used both as a source of power and for the rafting of

30

timber to the mill. Many farmers combined farming with work in the saw

mills or the forest, or producing charcoal (Enequist 1960, Lundmark and

Malmberg 1988, Schön 2012). However, there was also an abandonment

of smaller farms in favor of settlement in villages and employment at the

local saw mill or iron smelting factory (Enequist 1960). It should also be

emphasized, as Lundmark and Malmberg (1988) have shown, that the

early industrialization in Sweden occurred in both rural and urban areas.

This dispersed pattern of industry location has been explained by the

natural resource profile of the Swedish industries during this period

(Lundmark and Malmberg 1988).

The continued increase in the population, which grew by another 1.3

million during the period 1890-1930 to a total of 6.1 million by 1930, was

also an important part of the settlement development in this period

(Statistics Sweden 2017). While industrialization was a major development

trend in many parts of Europe, not least in Britain where the Industrial

Revolution started, the major increase in the Swedish population from

1750 onward was a contributing factor in shaping the Industrial Revolution

in Sweden. As more children survived into adulthood, many sons and

daughters of farmers did not inherit their parents’ land; this resulted in a

growing rural proletariat, dependent on wage labor for their survival

(Schön 2012).

In the 18th and 19th centuries the population increase resulted in a diffusion

of the population in search of new land, but also in a growing proletariat

who took work on the farm or in the mine as wage laborers (Enequist 1960,

Magnusson 2002, Schön 2012). While agriculture remained the major

sector of employment, the importance of the manufacturing sector grew

from 1850 onward. As the increased supply of wage labor coincided with

the Industrial Revolution, and thereby the increased demand for wage

labor, the increase in population was a factor behind the growth of urban

areas in this period (Andersson 1987, Borgegard et al. 1995). In 1900,

approximately 500,000 persons were employed in the manufacturing

sector while 1.3 million were employed in agriculture, whereas by 1930

manufacturing employment had increased to 750,000 while employment

in agriculture had declined to 1.1 million (Jansson and Antonson 2011).

There were some regional variations in the changes in employment, with

employment in agriculture continuing to increase until around 1930 in the

inland of northern Sweden. According to Bäcklund (1988), this was mainly

a result of the population growth exceeding the growth of employment

31

opportunities in manufacturing and services in northern Sweden, which,

in turn, forced people into agriculture for their livelihood. However, the

development in neither the agricultural nor the manufacturing sector could

absorb the larger new generations – a combination of the severe famine of

1867-1869, better opportunities elsewhere, and reduced transportation

costs led to a mass emigration of the population, primarily to America but

also to more developed countries on the European continent such as

Britain, Germany and Denmark. During the period 1860-1930

approximately 1.4 million emigrated from Sweden, while only 400,000

immigrated (Schön 2012, Statistics Sweden 2017).

From 1930 to 1970 the population redistributed towards a concentration

on both the local and the regional level, with urban centers growing at the

same time as the southern regions increased more than the northern

regions (Borgegard et al. 1995). During this period the population

increased by another 2 million, to a total of 8.1 million by 1970. A quarter

of the population increase was due to immigration, with the negative net

out-migration of earlier periods, from 1930 onward, changing into positive

yearly figures (Statistics Sweden 2017).

One reason for the population concentration at the beginning of this period

was industrial modernization, whereby small-scale industries were

replaced by large-scale, more modern, ones – this was not least the case for

the local saw mills (Borgegard et al. 1995, Gårdlund 1942, Lundmark and

Malmberg 1988). However, from around 1940, employment in

manufacturing grew more quickly in peripheral regions which, according

to Lundmark and Malmberg (1988), suggests that explanations for the

regional concentration after 1940 should be sought in other sectors than

manufacturing.

Between 1930 and 1950 both employment and yearly production in

manufacturing and related services, such as transportation, increased

extensively. In the period 1950-1970 employment growth in the

manufacturing sector slowed down, while the yearly production growth

continued at the same pace as for the period 1930-1950, at around 5% per

year (Schön 2012). Instead, the main employment growth took place in the

service sector (for males) and in both the public and service sector (for

females) (Jansson and Antonson 2011, Ohlander and Strömberg 2002,

Schön 2012).

32

The growth of the public sector during the period 1950-1970 was an

important factor for the concentration of the population. The welfare state

and the related municipal reforms of 1952 and 1971 reduced the number of

municipalities from 2,498 to 278, each with its own urban center, where

the majority of the public services were located. The growth of the public

sector thereby increased the employment opportunities in the municipal

centers (Borgegard et al. 1995). The public sector growth also coincided

with the increased participation of women on the labor market. In 1920

only around 5% of married women worked for wages; in 1960 this figure

was 25%; and by 1980 it had increased to 65% (Ohlander 1994).

The period 1930-1970 was also when employment in agriculture under-

went its major decline, and when the manufacturing sector became the

major source of employment. Between 1930 and 1970, employment in

agriculture declined from 1.1 million to a mere 300,000 persons. In the

same period, employment in the manufacturing sector increased

from 750,000 to 1.2 million, making it the largest sector in Sweden

(Jansson and Antonson 2011). Similarly, the number of self-employed

farmers declined by 78%, from approximately 430,000 in 1930 to 95,000

in 1970 (Figure 2). However, the decline of employment in agriculture was

only one aspect of a more fundamental transformation of the sector,

whereby manual labor was replaced by machines, fertilizers and other

refinement techniques were introduced on a large scale and, after 1960,

smaller production units were replaced by larger ones – thus, the period

1930-1970, and especially 1950-1970, saw the rise of productivist agri-

culture in Sweden (Flygare and Isacson 2003, Jansson and Antonson 2011,

Schön 2012, Woods 2005). Although the largest decline in agricultural

employment occurred between 1950 and 1960, the age distribution of

farmers in 1960 (Figure 3) shows that the aggregate employment figures

were delayed. That the number of farmers was largest in the age group 50-

60, and decreased linearly with ages below 50, indicates that the decline in

young adults entering the sector as farmers started around 1930.

33

Figure 2: Number of farmers, tractors and harvesters in Sweden (Board of Agriculture 2008)

Figure 3: Number of farmers by age, 1960 and 2010 (data source: ASTRID)

At the same time as employment in agriculture decreased, agricultural

production more than doubled between 1950 and 1990 (Figure 4). The

0

50,000

100,000

150,000

200,000

250,000

300,000

350,000

400,000

450,000

500,000

Farmers Harvesters Tractors

8,000 6,000 4,000 2,000 0 2,000

18

22

26

30

34

38

42

46

50

54

58

62

Number of farmers

Age

2010

1960

34

explanation for this increase in productivity was the introduction of more

advanced technology during this period (Jansson and Antonson 2011,

Schön 2012). One such technology with importance for agriculture, but far

from the only one, was motor vehicles. In 1944 there were 417 harvesters

and 25,000 tractors in Sweden, while by 1975 these numbers had increased

to a high point of 49,000 harvesters and 189,000 tractors. In the years

following 2000, there was almost one harvester for every farmer (Figure

2). An intertwined reason for the increased production was the

concentration of agriculture, which began in the 1940s (Jansson and

Antonson 2011). In 1940, 65% of agricultural land belonged to farms with

less than 30 ha of land while 10% belonged to farms with more than 100

ha of land, whereas in 2007 20% of total agricultural land belonged to

farms with less than 30 ha and 45% to farms with more than 100 ha of land

(Figure 5).

Figure 4. Production of wheat, grain, oat and beets in Sweden 1921–1990 in tons, five-year moving average (data source: Board of Agriculture 2008)

0

1,000,000

2,000,000

3,000,000

4,000,000

5,000,000

6,000,000

7,000,000

8,000,000

9,000,000

10,000,000

19

21

19

25

19

29

19

33

19

37

19

41

19

45

19

49

19

53

19

57

19

61

19

65

19

69

19

73

19

77

19

81

19

85

19

89

Ton

s

35

Figure 5. Hectares by size of farm in percentages of total agricultural land in Sweden 1927–2007 (data source: Board of Agriculture 2008)

The same pattern of rationalization and increase in production can also be

found in the forestry sector. In 1930 the number of those employed full-

time in forestry amounted to approximately 182,000, while the same figure

in 1970 was 32,000 and in 2015 only 16,000 (SOU 1938:58, Swedish Forest

Agency 2017). The major decrease in forestry employment coincided with

the implementation of new technologies in the sector. In the 1950s, the

chainsaw revolutionized the forestry sector in Sweden, with the yearly sales

figure of about 600 chainsaws in 1948 had increased to more than 15,000

by 1956. The chainsaw eased several work-intensive steps in forestry labor,

and thus increased the productivity per worked hour and reduced the labor

demand (Hjelm 1991). Similar to the development in agriculture,

production in forestry continued to increase despite the decline in

employment: from a five-year average of 55 million cubic meters for the

period 1944-1948 to 75 million cubic meters for 1970-1974 and 89 million

cubic meters for 2011-2015 (Swedish Forest Agency 2017). This

development can only be explained by the technological development that

has occurred in the forestry sector. Since the chainsaw’s arrival

technological development has only continued, and today the chainsaw has

been replaced by harvest machines and other technology, which have

vastly reduced the demand for employment (Schön 2012).

0%

10%

20%

30%

40%

50%

60%

70%

19

27

19

31

19

35

19

39

19

43

19

47

19

51

19

55

19

59

19

63

19

67

19

71

19

75

19

79

19

83

19

87

19

91

19

95

19

99

20

03

20

07

>30 ha 30,1-50,0 ha 50,1-100,0 ha Above 100,0 ha

36

The restructuring of the agricultural and forestry sectors from around

1930, and especially after 1950, whereby employment declined at the same

time as production increased, was an integral part of the restructuring from

agriculture to manufacturing. In the 1930s the agricultural and forestry

sectors had difficulty attracting new labor to farms, as the wages in the

expanding manufacturing sector were much higher than in employment as

a farmhand, maid or forestry worker (Bunte et al. 1982, Schön 2012). The

insufficient renewal of the sectors, with youths entering the labor market

joining the manufacturing sector to a larger extent, forced the sectors to

mechanize in order keep up production; but, on the other hand, the

mechanization of the sectors was only made possible through the new

technologies developed by the manufacturing sector (Schön 2012). The

combination of employment decline at both the local saw mills and within

the forestry and agricultural sectors, from the 1950s onward, hit the

frontiers of the northern expansion extra hard. However, the loss of jobs in

the municipalities of the northern inland was to some extent compensated

for by an expanding public sector in the municipal centers (Borgegard et

al. 1995).

In the following decade of 1970-1980, the concentration of the population

continued but decreased in intensity. Although the 1970s are often referred

to as a decade characterized by counterurbanization and a “green wave”,

the net result was still a concentration of the population on both the

regional and the local level (although mainly related to a larger natural

growth in urban areas compared to rural areas) (Amcoff 2000, Borgegard

et al. 1995). A major part of the counterurbanization, and the decline in the

pace of concentration, in the 1970s consisted of suburbanization due to

increased commuting (Amcoff 2000, Falk 1976). Another important part

was a slight shift in the continued expansion of the public sector, with

smaller urban areas in urban regions seeing a larger increase in the public

sector. This led to a redistribution of employment opportunities in the

public sector, from larger to smaller urban areas (Borgegard et al. 1995).

Thus, the dispersion – or rather the slowed concentration – of the

population in this decade was mainly a dispersion within urban regions

(Amcoff 2000, Borgegard et al. 1995, Falk 1976).

From the 1970s onward, the yearly increase in population growth slowed

from around 0.7% in previous decades to 0.4%. During the 1970s the

Swedish population increased by 220,000 (half the growth of the 1960s),

37

of which 125,000 reflected a surplus of the net migration to the country

(Statistics Sweden 2017).

The 1970s were also the decade when the manufacturing sector began its

decline in Sweden. Following the oil crisis of 1973 and the related structural

crisis in the latter half of the decade, many manufacturing companies went

bankrupt or had to downsize or rationalize production; this was the case

not least for companies in the shipyard and steel industries, and

employment in manufacturing declined from 1975 onward. While

employment in manufacturing declined, however, employment in the

service and public sectors continued to increase (Jansson and Antonson

2011, Schön 2012). The economic shift in the 1970s, which occurred in

many Western countries, has consequently been described as a structural

shift from an industrial economy to a service economy (Schön 2012).

However, the decline of employment in the manufacturing sector during

the 1970s was slower in peripheral regions, which increased their national

share of employment in manufacturing (Lundmark and Malmberg 1988).

A few empirical studies have highlighted the relationship between man-

ufacturing decline and subsequent migration of the workforce. In a case

study on the closing of the paper mill in the small urban locality of

Jössefors, Berger (1973) concluded that the closing had significant effects

on the local economy and commuting and migration patterns.

The last period described by Borgegård et al. (1995), 1980-1990, consisted

of regional concentration, while local concentration neither increased nor

decreased. However, when broken down into different groups a slightly

different pattern emerges whereby the net migration to larger cities was

different for different groups, with families with children moving to the

outskirts of larger cities while young adults and immigrants had a positive

net migration to larger cities (Amcoff 2000, Borgegård, Håkansson and

Müller 1998). In this decade the net migration surplus made up a larger

share of the total population growth, with the total population increasing

by 210,000, of which 147,000 reflected a surplus of the net migration to

Sweden.

In the 1990s, Westlund (2002) argued, Sweden experienced a new green

wave characterized by a higher net increase of the population in rural areas.

However, the “rural” population increase was concentrated around urban

centers, and on the regional level the population concentrated by

increasing more in the metropolitan regions (Amcoff 2003). The

38

“counterurbanization” was thus directed to areas within functional urban

regions, which indicates that suburbanization, ex-urbanization and urban

sprawl are more reasonable descriptions than a “new green wave”. Further-

more, Amcoff (2006) has shown that the net growth of the rural population

(defined as population outside “tätorter”) in the 1990s disappeared

entirely, when urban sprawl is controlled for. In elaborating on the study

by Westlund (2002), Westlund and Pichler (2013) found that the regional

concentration of the population continued into the 21st century, and argued

for understanding this trend as a manifestation of the transition towards a

“knowledge economy.”

The 1990s were also an important decade for the development of the

Swedish economy, with the first half of the decade tinged by a severe

economic crisis. In the aftermath of the crisis, employment in manufac-

turing and the public sector declined by approximately 200,000 jobs each,

and the service sector is the only one that has grown since the crisis.

However, although the service sector has grown, the employment rate has

not recovered to pre-crisis figures (Schön 2012). After the economic crisis

of the 1990s, the number of youths entering higher education increased

substantially, which spurred migration to urban areas and thereby had a

concentrating effect on the population (Lundholm 2007).

From 1990 onward the trend in the population growth that could be noted

in the 1980s continued, with the growth to a larger degree being due to a

surplus of the net migration to the country. Between 1990 and 2010 the

population increased by 825,000, to a total of 9.4 million, of which

650,000 reflected a surplus of the net migration to Sweden (Statistics

Sweden 2017). Since immigrants settled more in urban areas, especially

Stockholm, Gothenburg and Malmö, the growing immigration had a

concentrating effect on the population (Borgegard et al. 1995, Borgegård et

al. 1998, Hedberg and Haandrikman 2014).

By way of summarizing the changes in the settlement pattern and in

employment since 1750, it is evident that the settlement pattern, the

demographic changes, and the changes in economic activity are bound to

each other and may shift in importance over time. However, after the 1970s

the description becomes more stratified, with changes in the settlement

pattern being treated separately from those in the economic activities. The

reason for this is that most studies on urbanization in Sweden for the

period after 1970 have considered the population system in isolation from

39

the activity system (Amcoff 2000, Amcoff 2003, Amcoff 2006, Stenbacka

2001). However, in another strand of literature that specifically focuses on

the contemporary development of cities, population redistribution is

considered together with economic changes. Florida (2002), for example,

argues that economic growth in the new service economy is related to the

in-migration of a “creative class”. In writing on the transition to the new

service economy (emphasizing the heterogeneity of the service sector),

Castells (2010) argued that the new information technology would not lead

to the end of cities, as was hypothesized in the 1980s, but rather to their

growth. In Sweden, Nilsson (2011) has shown that cities with an industrial

profile have declined from the 1990s onward, while large cities with a more

diversified economy have grown. While not looking specifically at changes

in the types of economic activities, Holm et al. (2013) reach a similar

conclusion. This would lend support to the structural perspective of

considering the population system together with the activity system even

when investigating the recent history of rural and regional change.

40

Data and methods

Methodological perspective

The three empirical papers in this thesis investigate changes in employ-

ment and migration in rural Sweden 1960-2010 (Articles I and II), and the

geographic distribution of these factors in 2008 (Article III). In turn, as the

discussion in Section 2 argues, these factors should be understood as the

local products of broader technological, economic and social processes on

a global scale – processes that are not subject to empirical investigation in

this thesis. This research design – in which observable, local factors are

explained by less visible, global factors – lies at the center of the restruc-

turing approach. In this sense, “localities are places where wider processes

manifest themselves” (Lovering 1989, 213).5 Thus, it is not just that the

broader technological, economic and social processes are unobserved in

the empirical material of this thesis; it is also the case that the causal

connection between the global and the local, between the broad and the

specific, is less visible as an object of empirical enquiry (Urry 1987).

The underlying philosophical perspective thus lies closer to the method-

ological realism proposed by Bhaskar (1978) than to methodological

empiricism or positivism. Whereas an empiricist or positivist approach

would restrict the analysis to the data under observation, a realist approach

acknowledges that causal factors may be relevant to the explanation even

when they are not an object of empirical investigation (Lovering 1989). As

Sayer (1982) noted, assumptions of non-empirical objects have often been

a fruitful method of scientific enquiry, with for example manifestations of

disease (observed) being assumed to be caused by viruses even before

viruses had been discovered as an empirical object. In the case of

technological development, social modernization and globalization, there

is not really any doubt as to the existence of these processes; it is just that

their local impact (and to some extent their content) is elusive to measure

5 In the restructuring approach, the causal relationship between the local and the global is recognized as a two-way relationship, whereby “the locality [...] may be a site of emergent causal powers” (Lovering, 1989, 213). The approach in this thesis, however, is delimited to considering the relationship between the local and the global as a one-way relationship, with global processes causing local outcomes.

41

– especially as their impact varies between localities with different

conditions and trajectories (Urry 1987).

In following a realist line of thought, this thesis views space as a continuous

phenomenon, whereby all (spatial) categorizations are in some sense

arbitrary (Jones 2009). However, although a realist perspective acknowl-

edges that reality is continuous on an ontological level, epistemology (and

indeed terminology) demands that categorizations be made (Harvey 1996,

Kant 1998). Thus, there is no contradiction in this thesis concerning itself

with rural Sweden as a spatial category while at the same time

acknowledging that space is continuous. As argued for in Section 1.2, rural

is merely viewed as an area of interest, and not as an enclosed social system

(Hoggart 1988, Hoggart 1990).

Methods

As this thesis sets out to explore contemporary demographic and labor

market change in rural Sweden, the methodological starting point is one of

inductive description, with the purpose of describing the broad features of

these changes using longitudinal geo-referenced micro-data on the whole

Swedish population.

In Articles I and II a life-course perspective is applied, whereby the

differences between a selection of cohorts are investigated. In Paper I the

selection consists of two cohorts of male and female agricultural workers,

the first born in 1938-1940 and the second in 1958-1960, whose employ-

ment trajectories are studied until they turned 50-52 years old. In Paper II

the selection consists of three whole population cohorts, the first born in

1945, the second in 1960 and the third in 1980. The individuals in the

cohorts are defined as urban or rural based on their place of residence at

age 15 (see rural definition above), stratified by sex, and measured on their

sector of employment and type of residence at ages 20, 25 and 30.

It is argued here that comparing the difference in employment and

migration trajectories between cohorts reveals information on larger

societal changes that is harder to find using aggregate time series methods.

The reason for this is related to the life-course perspective, originally

developed by Elder (1994). From this perspective, the relationship between

life courses and historical time is central, whereby “differences in birth year

expose individuals to different historical worlds, with their constraints and

options” (Elder Jr 1994, 5). In this view, aggregate data contain individuals

42

from many different “historical worlds”, both old and new. Since individual

life courses are largely path-dependent – dependent on their past biog-

raphies – this means that aggregate data may hide, or at least delay, the

extent of societal changes. Therefore, comparing cohorts is often a better

method for capturing new trends than aggregate time series.

Another central aspect of the life-course perspective is the timing of events

during the individual life course, whereby the individual’s age at the time

of an event and the order of events are important (Elder Jr 1994). In con-

trast to a cross-sectional approach, a longitudinal approach also offers

information on what happened before and after the event. For example,

while cross-sectional data can show the number of individuals in higher

education, longitudinal data may also show what the individual did before

and after higher education – how many went on to a job, how many

migrated, etc. In turn, such information says something important about

the dynamics on the labor market that would have been missed with a

cross-sectional approach.

In combination, these two aspects of the life-course perspective – historical

time and individual time – make a strong argument when studying societal

change for comparing trajectories between cohorts from different

historical times.

In Article III the focus is on variation in space rather than change over time,

with a hierarchical cluster analysis performed in order to distinguish

between different types of rural areas (see definition above). The rural

areas were clustered based on their share of employment in different

sectors (see above), employment rate, share of population aged 65+, share

of females aged 15-45, and population change in percentage 1985-2008. It

was important that no spatial variable be included in the cluster analysis,

in order not to bias the results towards an urban-rural continuum model.

Data

All three empirical papers in this thesis are based on longitudinal micro-

data emanating from the ASTRID database, which is hosted by the

Department of Geography and Economic History at Umeå University.

ASTRID is a combination of annual registers from Statistics Sweden cov-

ering the whole (registered) population in Sweden annually from 1985

onward on a large array of variables. In addition, ASTRID also contains

census data on the entire population that covers the period 1960-1990 at

43

five-year intervals. Crucial for this thesis is also that the data are geo-

referenced with high geographic resolution (100 square meters). The

register data, from 1985 onward, contain the residence and workplace

location of the individual accurate to 100x100m, and the census data for

1960-1990 contain information on the parish of residence and whether or

not the individual lived within the boundaries of an urban locality.

Two variables are of special importance since they reoccur throughout the

articles: namely, a geographical division into rural and urban, and sector

of employment. As mentioned in the introduction, the definition of rural

varied between the articles, with Article II applying a regional definition

and Article III a local one.

The specific definitions of rural and urban in Articles II and III were:

Article II – Parishes containing a “tätort” larger than 5,000

inhabitants and a population greater than 50,000 within 25km

were classified as urban. The remaining population was classified

as rural.

Article III – In the first step, individuals registered in a 100x100m

square not located within “tätorter” larger than 1,000 inhabitants

were classified as rural. In the second step, the “rural” individuals

were aggregated to SAMS (Small Areas for Market Statistics).

The rural definition in Article II thus includes peripheral “tätorter” but

excludes areas outside “tätorter” in densely populated regions, while

Article III excludes peripheral “tätorter” but includes areas outside

“tätorter” in densely populated regions. While this may seem like a

problem, it is actually a strength of the thesis. The reason for this is that

the research questions vary between the articles, with Article II focusing on

the labor market and Article III on rural heterogeneity. Furthermore, the

results of Article III (see below) show that the large difference in

demographic trends and employment characteristics is between “rural”

areas close to urban centers and peripheral rural areas, rather than

between rural and urban areas understood as inside/outside “tätorter”. For

this reason, the results in Article II would be distorted with the rural

definition of Article III. At the same time, the typology developed in Article

III would be much less interesting if only peripheral rural areas were

included in the study. However, it is important to keep in mind the

difference between the articles when conclusions are drawn, as Article II

44

investigates the areas categorized as areas outside the urban shadow,

manufacturing periphery and resource periphery in the rural typology

developed in Article III (see below).

As was mentioned in the Introduction Article I does not include any

definition of rural, but rather looks at changes in the agricultural sector as

a “rural” sector. This does not imply that rural areas are defined by

agriculture but rather that agriculture almost exclusively takes place in

“rural areas” when a local definition of rural is used (as in Article III). While

Article I does not discriminate between rural areas the results of Article II

shows that changes in agricultural mainly took place in more peripheral

areas rather than areas close to larger urban areas.

The variable “sector of employment” reoccurred throughout the three

articles, and comprised a central part of the analysis. The variable is based

on four- to five-digit SNI codes6, which are collected at the plant level.

Individuals working at a plant are then ascribed the SNI code of that plant.

This means that the data do not reveal the occupation of the individual but

only the sectorial category of the workplace. In order to create the variable

“sector of employment”, the SNI codes were aggregated to broader

sectorial categories. However, these variables have slight variations in

definition between the articles:

Article I – agriculture, other natural resource sectors,

manufacturing, services, public

Article II – natural resource sectors, manufacturing, services,

public

Article III – agriculture, forestry, mining, manufacturing, tourism,

finance and other advanced services

The reason for this variation is differences in the research questions or

focus between the articles. Agriculture is broken out from natural

resources in Article I due to the article’s focus on the agricultural sector. In

Article III, only the sectors with a substantial geographical variation are

selected (the public sector, for example, is much more evenly distributed

6 SNI = Standard för Svensk Näringsgrensindelning

45

Article I Article II Article III

Aim To analyze changes in employment trajectories between two cohorts of male and female agricultural workers

To analyze changes in employment and migration trajectories between three cohorts of the Swedish population in rural areas

To analyze the geographic distribution of employment and population in rural Sweden

Data source ASTRID register database ASTRID register database ASTRID register database

Data structure Individual, longitudinal Individual, longitudinal Aggregated, cross-sectional

Population Two cohorts of agricultural workers, the first born in 1938-1940 and the second in 1958-1960

Three cohorts of the Swedish population, the first born in 1945, the second in 1960, and the third in 1980

The population in areas defined as rural

Time period Historical time: 1960-2010 Individual time: age 20-50 (22-52)

Historical time: 1965-2010 Individual time: age 20-30

Cross-section in 2008 (population change 1985-2008)

Methods Life-course perspective, comparing descriptive statistics between cohorts stratified by sex

Life-course perspective, comparing descriptive statistics between cohorts stratified by sex and area of residence at age of 15

Hierarchical cluster analysis

Rural definition

No formal definition of rural Regional Local

Indicators Sector of employment Sector of employment Area of residence (rural/urban)

Sector of employment (age 20-64) Employment rate (age 20-64) Education rate (age 20-64) Age structure of population Population change 1985-2008 (total population)

46

than manufacturing, mining, or tourism). A summary of the research

design of the three articles can be found in Table 1.

Regarding the SNI-codes used to define the variable “sector of

employment” in Article I and Article II it should also be mentioned that

three different versions of the codes was connected: SNI69 (1960-1990),

SNI92 (1990-2002) and SNI02 (2002-2010). This was done through using

a key created by Statistics Sweden between the different versions. Few of

the broader sectorial categories were affected by the transition between the

different versions of SNI. However, in the more fine-grained division with

27 sectorial categories (used in Article II) the codes in the category “other

services” became more detailed in SNI92 and SNI02.

Limitations and delimitations

In focusing on changes in employment, internal migration and population

distribution in rural Sweden, this thesis is naturally delimited to these

aspects. However, even when the focus is narrowed to only these aspects,

there are some additional delimitations that need to be considered and

made explicit. Furthermore, some limitations related to the data and

methods should also be considered.

One limitation is the extent to which temporal changes in the observed

sectorial distribution reflect real changes in types of employment or merely

changes in the structure of companies. In this regard, Knobblock &

Pettersson (2010) found that sub-contractors, not classified as mining

companies in the SNI code, stood for a substantial part of the employment

growth in the mining industry in northern Sweden during the period 2001-

2007. Schön (2012) also points out that parts of the decline of the

manufacturing sector and growth of the service sector since the 1970s are

a statistical artifact caused by increased outsourcing. Whereas

manufacturing companies previously kept many services internally (e.g.

cleaning and other forms of maintenance services), these are increasingly

being outsourced to sub-contractors. The implication of this is a reclassi-

fication of individuals from manufacturing to services, even though the

content of the job (e.g. cleaning) is identical. The extent to which this effect

is an important factor in the Swedish manufacturing sector has not, to the

author’s knowledge, been investigated empirically. One study on this effect

in South Africa between 2001 and 2007 (Tregenna 2010) found that the

effect is important. However, the short time period of that investigation

and the differences between Sweden and South Africa (with the latter, for

47

example, seeing an increase in employment in the manufacturing sector

during the study period, even without controls for outsourcing), should

caution against a direct transferal of the results to Swedish conditions.

While Schön (2012) acknowledges that the restructuring towards services

may to some extent be a statistical artifact, he also acknowledges that there

has been a real (as opposed to statistical) shift towards the service sector.

Thus, as far as this thesis is concerned, outsourcing may be a source of

error but it certainly does not account for the broad shift from

manufacturing towards services that is identified in Article II; nor can it

explain the decline in employment in agriculture identified in Article I.

Another limitation is the geographic distribution of sectorial change. While

Articles I and II look at the temporal change in employment in different

sectors (divided into rural and urban areas in Article II), Article III looks

at the cross-sectional geographical distribution of employment in 2008. As

argued for above, the rural typology developed in Article III can be viewed

as the spatial outcome of the processes described in Articles I and II, but

the more exact measurements of the geography, at a high spatial

resolution, of sectorial change are not accounted for. In this sense, one

could ask whether the mining areas in the rural typology developed in

Article III have had another trajectory in regard to sectorial changes than

the forestry areas or the areas outside the urban shadow.7 However, one

indication that the detailed geography of sectorial change is of limited

importance is that the initial stages of what would become Article II

divided Sweden into four geographic areas (rural sparse, rural dense,

urban sparse, and urban dense). As the changes in employment between

the cohorts were very similar between all categories except urban dense,

the division was aggregated to only two categories: urban and rural (see

Methods section in Article II for further explanation). Thus, while the

geographic distribution of sectorial change may vary on a detailed

geographic level, it does not seem to vary when broader regions are

considered. However, this is still an avenue for further studies, which

7 When the empirical analysis was conducted, in what would become Article III, the changes in sectorial distribution between 1985 and 2008 were examined. The tentative conclusion from this, admittedly brief, investigation was that there were no systematic differences in changes in sectorial distribution between the different categories of the typology. Thus, while the areas in the typology revealed similar characteristics in 2008, they had not necessarily reached this state through similar trajectories.

48

should also consider the differences between, and perhaps even within,

urban areas.

One delimitation is related to the temporary mobility of the population. As

mentioned above, the analyses of rural-urban migration in Article II and

population change in Article III are based on the parish and the 100x100m

square where the individual is registered in the ASTRID database. The

focus on the location of residence means that temporary mobility, for both

recreational and work-related purposes, is unaccounted for. The first

delimitation is that seasonal population changes related to various forms

of recreational activities are unaccounted for. However, insofar as seasonal

population changes for recreational purposes are important for the

combined change of the population system and the activity system they are,

in fact, captured in the rural typology developed in Article III. The tourism

areas (see below), where a large share of the local population is employed

in tourism, imply that there is an additional inflow of consumers to these

areas. Population changes through second homes are captured to a lesser

extent, except for the retirement areas in the typology.

When it comes to temporary mobility for work purposes, Article III

captures these movements to some extent by also considering rural-to-

urban commuting, whereby, for example, rural areas closer to larger urban

areas had much higher rural-to-urban commuting than more peripheral

rural areas. On the other hand, Article II (on the changes of employment

and rural-to-urban migration 1960-2010) does not account for changes in

the commuting patterns. However, as Article II defines rural on a regional

level, changes in the commuting pattern were of less importance compared

to Article III, with a local definition of rural (as the distance commuters are

willing to travel is limited, even if it has increased). The increasing group

of weekly commuters, however, is still a source of error (Amcoff, 2003).

While the empirical articles in thesis do not look at changes in commuting

patterns, the description of how commuting has developed since 1960 in

Section 3 and the references cited can be consulted in order to complement

the picture.

Another delimitation concerns population and employment change

through international migration, both temporary and permanent. This is

especially the case in the cohort design of Article II, whereby the interna-

tional migration to Sweden between 1996 and 2010 is unaccounted for in

49

the cohort born in 1980 (as they had to be registered at ages 15, 20, 25 and

30).8

8 Although Article I also uses a cohort approach, agriculture in Sweden employs very few international migrants.

50

Article summaries

Summary of Article I

In the post-World War II era, the Swedish agricultural sector underwent a

structural change consisting of a major decline in employment and a major

increase in productivity. This structural change, induced by new technol-

ogy and globalizing markets, resulted in increasing competition in the

agricultural sector, putting pressure on actors to rationalize and intensify.

The aim of the article is to examine how this structural change relates to

the labor market trajectories for men and women working in the Swedish

agricultural sector, focusing on the labor market dynamics for individuals

working in agriculture. The investigation consists of comparing employ-

ment trajectories between two cohorts of agricultural workers split by

gender, the first born in 1938–1940 and the second in 1958–1960. The

method of comparing the work trajectories of individuals rather than the

stock of agricultural workers enables the observation of exits from and

entries into the agricultural sector, and how this differed between men and

women and between the cohorts.

Previous research is cited as showing that men and women have held

different roles on the family farm, with men often having the most

influence over the business. Due to rationalization, it was therefore often

the women who left agriculture for other job opportunities. Based on such

research, it is reasonable to believe that the agricultural restructuring has

had different implications for men and women.

The data used in the article consist of a combination of census and register

data on the entire Swedish population between 1960 and 2010 at five-year

intervals. Based on these data, two cohort groups of agricultural workers

were selected: the first born between 1938 and 1940, and the second

between 1958 and 1960. These cohorts were followed until the age of 50-

52. In order to qualify into one of the groups, the individuals in the

specified cohorts also had to have worked in agriculture when they were 20

and 25 years old, respectively. This resulted in a cohort group born in

1938–1940 of 3,893 agricultural workers (3140 men and 753 women), and

a cohort group born in 1958–1960 of 2,184 agricultural workers (1824 men

and 360 women). In the second step, a status variable was created to show

which sector the individual worked in, or that the individual was not

51

working. The sector in which the individual worked was based on a four-

digit sectorial code aggregated into five broader sectors: agriculture, other

natural resource, manufacturing, service, and public.

The results of the study show that fewer entered agriculture in the 1960

cohort compared with the 1940 one (42% fewer men and 52% fewer

women). Furthermore, the individuals in the 1960 cohort exited agricul-

ture to a greater extent and had more turbulent employment trajectories in

terms of changing occupation and/or employment sector. While the

changes went in the same direction for both sexes they were more promi-

nent for women, which means that changes in the Swedish agricultural

labor market have had a greater effect on the women in the sector. The

results also show that men who exited agriculture mainly went into

manufacturing and the service sector, while women mainly entered the

public and service sectors, a pattern that reflects the general gender

segregation of the job market. Furthermore, the number entering the

service sector was substantially larger in the 1960 cohort while the number

entering manufacturing was slightly smaller compared with the 1940

cohort.

The article concludes that the larger exits and the higher turbulence in the

men’s 1960 cohort point towards an increased likelihood for this cohort to

continue working in agriculture compared with the 1940 cohort, most

likely due to increased competition, rationalization, and a lack of profita-

bility in the agricultural sector. The article interprets this as an indication

that the structural changes in the economy in general made it harder to

continue working in agriculture, and that the exits from agriculture into

other sectors was affected by the growth or decline of these sectors.

Summary of Article II

This article investigates long-term changes of employment in sectors and

urbanization in rural Sweden. Similar to Article I, this article utilizes a

comparison between cohorts split by gender. As opposed to Article I, the

selection consisted of three whole population cohorts measured at the ages

of 15, 20, 25, and 30. The first cohort was born in 1945, the second in 1960,

and the third in 1980. These cohorts were subsequently divided into three

rural and three urban cohorts, based on parish of residence at age 15.

Another important difference from Article I is that this article looks at the

combination of employment and rural-to-urban migration rather than just

employment.

52

The article departs from the theoretical framework of rural restructuring,

viewing the changes of employment and urbanization as the local products

of the global processes of technological development, social modernization

and globalization. The article also considers how life-course theory can be

utilized to reveal aggregate societal changes in a more nuanced way by

distinguishing between historical and individual time.

The most important findings in the empirical analysis are a substantial

decline in – especially male – manufacturing, and a decline in the female

public sector in rural areas in the 1980 rural cohort. At the same time,

rural-to-urban migration and work in the urban service sector increased in

the cohort. The article moves on to conclude that the rural change in

Sweden during the period 1965-2010 encompasses the rise and decline of

the rural public sector, the decline of manufacturing, a general transition

towards public- and service-sector employment in urban areas, and a

substantial increase in higher education. In turn, the article argues that

these local changes are connected to global processes, whereby the

observed local changes in rural areas should be understood in the context

of the general transition from a manufacturing to a service economy.

In light of these results, the article suggests that the importance of

manufacturing has been downplayed in rural geography. As the results

indicate that changes in manufacturing employment are an important

aspect of contemporary rural change, the article argues that manufacturing

change should receive more recognition in rural research.

Summary of Article III

This article argues that rural areas should be recognized as heterogeneous

localities, and goes on to develop a typology of rural Sweden, distinguishing

between different types of rural areas based on their socioeconomic

characteristics. The reason for empirically investigating the heterogeneity

of rural Sweden, it is argued, is that the literature on rural change consists

of many location-specific theories and many empirical claims based on

case studies, while it is harder to find overview studies showing the location

and magnitude of these changes. In this regard, some studies have claimed

that contemporary rural change is characterized by gentrification through

an in-migration of the urban middle class, whereas others have focused on

population decline and ageing. The argument in the article is that both

these processes could be occurring simultaneously, but most likely in

53

different types of rural areas. And, for this reason, it is an important task

to investigate the heterogeneity of rural areas.

The socioeconomic typology was developed using a hierarchical cluster

analysis on Swedish SAMS (Small Areas for Market Statistics). The socio-

economic data were aggregated from a set of cross-sectional micro-data on

the entire Swedish population in 2008. The micro-data were prepared in

three steps. First, the population living in rural areas was selected for

further analysis while the population in urban areas was discarded (see

details on the utilized definition of rural in the Methods section). Second,

the rural population was aggregated to the SAMS, and in the third step the

SAMS with a rural population of fewer than 20 people were excluded from

further analysis. This procedure resulted in a dataset with 3,983 SAMS.

Eleven socioeconomic variables were included in the cluster analysis:

employment, higher education, population change 1985-2008, two

variables reflecting demographic structure, and six variables reflecting

sectorial distribution.

From the hierarchical clustering on the SAMS, a solution was chosen that

consisted of five clusters and 16 sub-clusters. The five-cluster solution

consisted of the middle-class countryside within the urban shadow (12%

of the rural population), the working-class countryside within the urban

shadow (22% of the rural population), the countryside outside the urban

shadow (25% of the rural population), the manufacturing periphery (28%

of the rural population), and the resource periphery (12% of the rural

population). Among other things, the typology reveals that middle-class

presence in rural Sweden is limited to the near-urban vicinity, or

concentrated in small semi-remote but attractive locations (the sub-cluster

retirement areas), that areas with a large-scale tourism economy are few

and far between, and that areas characterized by depopulation and ageing

are mostly found in peripheral areas. The article concludes by pointing out

the necessity of geographic overview studies and country-specific

typologies to get a better understanding of rural heterogeneity.

54

Concluding discussion

As stated in the introductory section, this thesis aimed at analyzing demo-

graphic and employment changes during the period 1960-2010 in rural

Sweden from a structural perspective, acknowledging the interdependence

between the population system and the activity system. Based on the

empirical investigations in this thesis as well as previous research, a broad

distinction is made between two phases of change in rural Sweden. The

first phase, characterized by the decline of agriculture and the rise of

manufacturing, was connected to the Industrial Revolution and the

transformation from an agrarian to an industrial society (Flygare and

Isacson 2003, Magnusson 2002, Schön 2012). In this phase, urbanization

was extensive and many cities and smaller urban localities were formed or

increased in size. In the earlier stages of this period, urban structures grew

or were formed in many peripheral areas as well, including the inland of

northern Sweden (Borgegard et al. 1995).

Starting in the 1980s rural change entered a second phase, which, in

contrast to the first, was characterized by de-industrialization and the rise

of the urban service sector. This period was characterized by regional

rather than local urbanization. Peripheral urban and rural areas based on

industrial employment found themselves with a declining economic motor,

which meant that people had to find their source of income elsewhere

(Article II in this thesis). The migration streams in this period were thus

increasingly directed towards metropolitan or large city centers and their

rural surroundings within commuting distance (Amcoff 2000, Amcoff

2003, Amcoff 2006, Borgegard et al. 1995).

Scientifically, this thesis contributes in several ways: firstly, by applying a

structural perspective on contemporary rural change, it connects demo-

graphic and employment changes to each other and to larger processes;

secondly, by applying a life-course perspective on rural restructuring, it

departs from the individual level to show how urbanization and employ-

ment have changed on the aggregate level; and thirdly, by departing from

geo-referenced micro-data aggregated to SAMS, it shows how demo-

graphic change and employment vary across rural Sweden. More

specifically, the first two articles in the thesis describe changes employ-

ment in rural Sweden 1960-2010. This is accomplished by looking at

changes in the life course of individuals from different historical times. The

second article also looks at the combination of changes in employment and

55

urban-to-rural migration by situating these two factors on the individual

level. The third article, which also departs from micro-data, describes how

population change and employment in different sectors were distributed

geographically in 2008, and generalizes the distribution into a typology of

rural Sweden. Together, the three articles describe how employment and

settlement patterns have changed in the recent past and how they are

distributed in space. Furthermore, the perspective of rural restructuring

widens the understanding of rural change by viewing it as a result of

changes in layers of investment and the redistribution of the spatial

division of labor. These changes can in turn be linked to the global

processes of technological development, social modernization, and

globalization. In this view, demographic and employment changes in rural

Sweden are the local products of global processes, mediated by changes in

local investments.

The societal contribution of this research should be viewed in the context

of rural decline, with rural areas in Sweden facing several problems related

to economic and demographic decline, including lower tax revenues,

increased dependency ratio and the cutback of public services. In this

respect, this thesis contributes by specifying which areas are declining and

which are growing. Furthermore, by connecting these changes to the

transition from a manufacturing to a service-based economy, and to the

global processes driving this transition, this thesis could contribute to the

policy discussion on rural change by offering a view based on the

contemporary development.

In the remaining parts of this concluding discussion, the scientific

contribution of the thesis will be discussed in more detail (Sections 6.1 and

6.2), followed by a discussion on some policy implications (Section 6.3).

Finally, Section 6.4 discusses avenues for further research related to the

present findings.

Two phases of rural restructuring

A central finding in this thesis is that contemporary rural change is

characterized by a declining manufacturing sector and a growing urban

service sector. This finding challenges the common assumption in rural

geography that rural change during the past three decades was charac-

terized by a transformation of the rural economy from natural resource

sectors towards the service sector. This is the case not least in the discourse

of “post-productivism”, in which changes in agriculture are described as an

56

important component of contemporary rural change. Within this dis-

course, it is described how productivist agriculture has been replaced by

post-productivist agriculture, which is characterized by small-scale

production, often of ecological or luxury products (Evans et al. 2002,

Shucksmith 1993, Ward 1993, Wilson 2001, Woods 2005, Woods 2010).

While there is no doubt that this trend exists, it is more questionable to

what extent it is central in understanding contemporary rural change. In

their report on rural change in OECD countries, the OECD (2006, 39-44)

refers to the change in agricultural employment between 1983 and 2003,

and concludes that “agriculture is no longer the backbone of the rural

economy”. However, the figures in their report, to which the claim refers,

show that employment in agriculture was already marginal by 1983 in

many OECD countries, not least Britain, which means that the decline of

employment in agriculture must have happened in an earlier period

(Figure 6). However, this is not to say that agriculture is an unimportant

economic activity (it certainly is not), or even that it is unimportant for

employment in rural areas (employment in agriculture may still be studied

in its own right, as is done in Article I); it is simply that changes in

employment in agriculture are not a central factor behind the

contemporary restructuring of rural areas.

57

Figure 6: Agriculture as a share of total employment in OECD countries, 1983 and 2003 (OECD 2006, 40)

Based on the results in Article II and on the description in Section 3 of this

thesis, it is evident that in the past century rural areas have undergone two

different restructuring processes. The first was a transition from

agriculture to manufacturing, with, primarily, new generations of youths

starting to work in manufacturing instead of agriculture. As a result of this

process of industrialization, many urban areas were formed or increased

in size, including many peripheral urban areas (Borgegard et al. 1995).

Article II investigates the later stages of the first restructuring phase,

during which manufacturing was a major source of employment in rural

areas (see Figure 7). The first cohort of rural youths (born in 1945) entered

a labor market dominated by employment in manufacturing and the

service sector for males, and in the public and service sectors for females.

For this cohort of youths, who entered the labor market in 1965,

employment in natural resource sectors was already very small. Out-

migration to larger urban centers was also substantial for this cohort, with

a large share of the rural males migrating to work in the urban service and

58

manufacturing sectors, and a large share of the rural females migrating to

work in the urban service and public sectors.

The second cohort of rural youths, born in 1960 and entering the labor

market in 1980, entered a labor market where the manufacturing sector

had begun to decline but was still a major source of employment for males.

For this cohort the transition towards the service sector began to show, but

it did not foster an increase in urbanization. Instead urbanization declined

in this period, a fact that could be related to the expansion of the rural

public sector, which employed a large share of the females in the cohort.

When the third cohort of rural youths entered the labor market in the year

2000, both employment in the manufacturing sector and the rural public

sector had declined substantially. Instead, a larger percentage of this

cohort, compared with previous cohorts, migrated to urban areas in order

to study or work in the urban service and public sectors.

By situating the changes in employment and urbanization between these

three cohorts in their historical context outlined in Section 3, we see that

rural areas have undergone two different restructuring processes. In this

view, the period 1965-1975 comprised the later stages of the first restruc-

turing process, with manufacturing serving as the economic motor of rural

Sweden. Through technological development and globalization,

manufacturing gradually declined, and in the period 1980-1990 started to

lose its absorptive capacity in rural areas. However, the decline in manu-

facturing was compensated for by a large expansion of the rural public

sector, which in turn was a product of the welfare state, which was made

possible through Sweden’s success as an industrial country (see Sections

2.2 and 3). The period 1980-1990 can thus be seen as a transition period

between the first and second phases, in which elements related to both the

first phase (expanding public sector) and the second phase (shift from

manufacturing to services) could be observed. The period 2000-2010

marked the final break with the previous restructuring phase from

agriculture to manufacturing. Manufacturing declined substantially in this

period, together with a decline of the rural public sector and a large

increase in the urbanization rate and the urban service sector. These major

changes in employment and migration patterns should be understood as a

new restructuring process, whose main component is a transformation of

the economy from the manufacturing sector to the urban service sector (see

Figure 7 for a summary of this process).

59

Figure 7: Schematic model of how the two phases of rural restructuring relate to the three youth cohorts studied in Article II

60

As argued for in Section 2, the transformation of the Swedish economy

from manufacturing towards services can in turn be understood as a prod-

uct of the broader processes of technological development, social modern-

ization, and globalization. For example, the expanding market due to

globalization has increased competition, which in turn has facilitated both

spatial shifts in local investments due to comparative advantages, and

shifts in the content of local investments towards more automation, replac-

ing manpower with machines.

While the empirical investigation in this thesis is delimited to rural

Sweden, there is reason to believe that the two phases of rural restructuring

can be identified in other countries as well. For example, the decline of

employment in manufacturing is not a phenomenon unique to Sweden, but

can be observed in both other parts of Europe and the United States

(Dicken 2003). However, in the same manner that rural Sweden is

heterogeneous when it comes to demographic changes and employment

distribution (see below), it can also be expected that the processes of

technological development, social modernization, and globalization have

had different local outcomes in different countries.

The geography of the second restructuring phase

While the transformation of the Swedish economy from manufacturing to

services has affected the whole country, the outcome of this transformation

is still local and has had different implications for different places. In

Article III of this thesis, a typology of rural Sweden (see definition of rural,

above) based on socioeconomic characteristics was developed (see Figure

8). The typology divides rural Sweden into five broad categories and 16

sub-categories based on the local labor market, demographic

characteristics, and demographic change. It can be assumed that these

different rural areas have experienced, and will continue to experience, the

shift from manufacturing to services differently, with some areas having

grown in both demographic and employment terms while others have

declined. In this sense, the heterogeneity of rural areas is a product of both

growth and decline – of old development paths that are reaching their end

and of new development paths that will continue into the future (compare

with Figure 1).

61

Figure 8: Concentric model of the rural typology developed in Article III. The distance from the urban core is estimated based on the value range of the first and second quartiles for each cluster. The location of the sub-clusters of the resource periphery is approximated based on the distance to the urban core and their relative location.

62

Stretching the trends into the future, it is likely that the decline of manu-

facturing and the rural public sector, and the transition towards the urban

service sector, will have large effects in the manufacturing periphery and

in most areas of the resource periphery, where it will lead to continued

out-migration and further decline in employment opportunities. The areas

within the urban shadow, on the other hand, will continue to grow

alongside the growth in the urban service sector. In this sense, the

manufacturing periphery and parts of the resource periphery are

products of the old economic order and the settlement pattern it gave rise

to, while the middle-class countryside within the urban shadow and the

working-class countryside within the urban shadow are products of the

new economic order and its pressure on the settlement pattern. In

addition, the tourism areas and retirement areas sub-categories of the

resource periphery are also products of new development paths related to

the contemporary restructuring of rural areas. These areas have the

potential to grow, and the number of tourism and retirement areas may

increase in the future. However, it is important to recognize that these

areas are few in number and that only a small percentage of the rural

population lives there. Thus, even if these areas are growing in size and

number, they are unlikely to turn the general decline of the rural periphery

– for now, they should rather be understood as small islands of new

development paths in an ocean of old paths

Combatting rural decline?

The contemporary restructuring of rural Sweden is comprised of a trans-

formation of the economy from manufacturing towards the urban service

sector. At the same time, this transformation is a geographic shift in

economic power towards urban centers, with peripheral areas losing their

economic motor. As a result of this process, peripheral rural areas have

found themselves in a situation of a declining and ageing population, and

with an eroding tax base.

The response from policy institutes and local politicians in rural areas has

often been to say that the trends of economic and demographic decline

present a large problem, and that the goal of policy should be to turn this

decline into growth (Bontje 2005, Hospers and Reverda 2015, Martinez‐

Fernandez et al. 2012, Syssner 2006, Syssner 2014, Syssner 2016,

Wiechmann and Pallagst 2012). However, while it is understandable that

local politicians strive for the growth of their municipalities, it may not be

possible to turn these trends around. The combined forces of technological

development, social modernization, and globalization that drive

63

contemporary rural restructuring are not easily changed – a more realistic

view on the decline of the rural periphery is therefore that it is a structural

feature of the contemporary economic changes. In this view, the decline of

the rural periphery may be hard to counteract through policy initiatives.

Based on the notion that the decline of the rural periphery is a structural

feature of the contemporary economy, it would be wiser to question the

taken-for-granted goal that every community should grow (Syssner 2016).

From a historical perspective, settlement patterns are in constant change

and it is a futile task to hold onto a settlement pattern from a particular

point in time. In this view, the resource periphery and the manufacturing

periphery are products of the restructuring from agriculture to

manufacturing, and with the contemporary restructuring from

manufacturing to the service sector, a decline in these areas is expected.

Perhaps a more realistic goal for policy should be to minimize the social

effects of decline rather than focusing on population growth? This

suggestion is in line with a relatively new, but growing, field of policy

research on how to handle a shrinking population, in which the first step

to handling a shrinking population is to admit the fact (Sousa and Pinho

2015, Syssner 2016). From a national public finance perspective, it may be

counterproductive to increase the population in areas where local

investments and the absorptive capacity of economic activities are

declining. From this perspective, the optimal distribution of the population

of working age is one that, within commuting distance, correlates with

economic activities.

While it is hard to change the population decline in peripheral areas, the

consequences of an ageing population and a high dependency ratio in

peripheral municipalities need not include, for example, a diminishing tax

base. In Sweden there is already a system in place whereby wealthier

municipalities compensate poorer ones, and this system could be elabo-

rated on in order to compensate for the increased dependency ratio in

peripheral municipalities. On the other hand, it is hard to escape the fact

that a small and shrinking population means that public services, such as

hospitals and schools, become hard to preserve without the necessary

population base. A continued withdrawal of public services from declining

peripheral areas is therefore likely. While the effects of the withdrawal of

public services could be mitigated through various technological solutions,

such as telemedicine and distance studies, it is unlikely that technological

solutions could compensate in full. In this respect, the current policy

64

discussions regarding municipal merges and/or exceptions in the welfare

services for some municipalities suggest a continued phasing out of official

commitments in peripheral areas.

Suggestions for further research

By emphasizing the co-relationship between changes in economic struc-

tures and population, and empirically situating these factors on the indi-

vidual level, comparing changes between cohorts over a long time span and

placing these changes in their geographical and historical context, this

thesis contributes to the understanding of rural change in Sweden.

Whereas previous empirical research on contemporary rural change in

Sweden has tended towards isolating the population system from the

activity system, or limiting the investigation to only one sector or region,

this thesis looks at the population system in combination with the activity

system in the whole country. Furthermore, by looking at changes in

employment and migration for youths making the transition to adulthood

rather than the population stock, new trends are illustrated more clearly.

In light of these results, one avenue for further research is related to the

geographic context and the extent to which the findings from Sweden are

applicable to other countries. As the decline of manufacturing employment

and the growth of the urban service sector comprise an experience shared

by many Western countries (Dicken 2003), it is likely that this is an

important driver behind urbanization in other countries as well.

Investigations of the hypothesis that contemporary rural change in the

West is driven by a decline in manufacturing employment are thus

warranted. Similarly, it would be interesting to see how the typology

developed in Article III compares to other countries, and the extent to

which the generalized model in Figure 8 is applicable to other countries.

By comparing rural heterogeneity and rural change between different

countries, insight could be gained, for example, into how the local mani-

festations of modernization, globalization and technological development

vary between countries.

Another avenue for further research is related to the life-course perspective

and the method of comparing changes between cohorts to measure societal

change. While similar methods of cohort comparisons are common in

other fields, particularly modernization research (Kohli 2007), to the

author’s knowledge they have not previously been applied in rural

geography.

65

References

Abreu, M. 1996. Trade in manufactures: the outcome of the Uruguay

Round and developing country interests. In The Uruguay Round

and the developing countries, eds. W. Martin & L. Winters.

Cambridge: Cambridge University Press.

Albrow, M. 1996. Global Age. Cambridge: Polity Press.

Aldskogius, H. (1960) Changing land use and settlement development in

the Siljan region. Geografiska annaler, 42, 250-261.

Almstedt, Å., Brouder, P., Karlsson, S. & Lundmark, L. (2014) Beyond

Post-productivism: Clarifying Manifestations of Rural Change.

European Countryside, 6, 297-306.

Amcoff, J. 2000. Samtida bosättning på svensk landsbygd. Uppsala:

University of Uppsala.

Amcoff, J. 2003. Regional befolkningsomfördelning bland unga och

gamla. Arbetsrapporter 2003:13 Stockholm: Institutet för

framtidsstudier.

Amcoff, J. (2006) Rural population growth in Sweden in the 1990s:

unexpected reality or spatial–statistical chimera? Population,

Space and Place, 12, 171-185.

Anderson, J., Duncan, S. & Hudson, R. 1983. Redundant Spaces in Cities

and Regions?: Studies in Industrial Decline and Social Change.

London: Academic Press.

Andersson, R. 1987. Den svenska urbaniseringen: kontextualisering av

begrepp och processer. Geografiska Regionstudier 18. PhD thesis.

University of Uppsala.

Appadurai, A. 1996. Modernity al large: cultural dimensions of

globalization. Minneapolis: University of Minnesota Press.

Arrighi, G. 1994. The long twentieth century: Money, power, and the

origins of our times. London: Verso.

Arrighi, G. 1999. Chaos and governance in the modern world system.

Minneapolis: University of Minnesota Press.

Bartelsman, E. J., Haltiwanger, J. & Scarpetta, S. 2004. Microeconomic

evidence of creative destruction in industrial and developing

countries. Policy Research Working Paper The World Bank.

Bauman, Z. 2000. Liquid modernity. Polity Press: Cambridge.

Berger, S. 1973. Företagsnedläggning – konsekvenser för individ och

samhälle. Geografiska regionstudier 9. PhD thesis. Uppsala

universitet.

66

Berggren, H. & Trägårdh, L. 2015. Är svensken människa? Gemenskap och

oberoende i det moderna Sverige. Stockholm: Norstedts.

Berman, M. 1983. All that is solid melts into air: The experience of

modernity. London: Verso.

Bhaskar, R. 1978. A realist theory of science. Brighton: Harvester Press.

Board of Agriculture. 2008. Historisk statistik. Jordbruket i siffror 1866-

2007.

http://statistik.sjv.se/PXWeb/pxweb/sv/Jordbruksverkets%20st

atistikdatabas/?rxid=5adf4929-f548-4f27-9bc9-78e127837625

2017-01-10

Bolt, J. & Zanden, J. L. (2014) The Maddison Project: collaborative

research on historical national accounts. The Economic History

Review, 67, 627-651.

Bontje, M. (2005) Facing the challenge of shrinking cities in East Germany:

The case of Leipzig. GeoJournal, 61, 13-21.

Borgegård, L., Håkansson, J. & Malmberg, G. (1995) Population

redistribution in Sweden – long term trends and contemporary

tendencies. Geografiska annaler. Series B, Human geography,

77, 31-45.

Borgegård, L., Håkansson, J. & Müller, D. K. (1998) Concentration and

dispersion of immigrants in Sweden, 1973–1992. The Canadian

Geographer, 42, 28-39.

Boyle, P. J. & Halfacree, K. 1998. Migration into rural areas: theories and

issues. Chichester: Wiley.

Braudel, F. (1958) Histoire et sciences sociales: la longue durée. Annales.

Économies, Sociétés, Civilisations, 13, 725-753.

Braudel, F. 1982a. Civilization and Capitalism, 15th-18th Century: The

Perspective of the World. University of California Press.

Braudel, F. 1982b. Civilization and Capitalism, 15th-18th Century: the

Wheels of Commerce. University of California Press.

Braudel, F. 1995. The Mediterranean and the Mediterranean world in the

age of Philip II. University of California Press.

Broadberry, S. & O'Rourke, K. H. 2010. The Cambridge Economic History

of Modern Europe: Volume 2, 1870 to the Present. Cambridge:

Cambridge University Press.

Brouder, P., Karlsson, S. & Lundmark, L. (2015) Hyper-Production: A New

Metric of Multifunctionality. European Countryside, 7, 134-143.

Brown, D. J. & Pheasant, J. (1985) A Sharpe portfolio approach to regional

economic analysis. Journal of Regional Science, 25, 51-63.

67

Bruinsma, J. 2003. World agriculture: towards 2015/2030: an FAO

perspective. London: Earthscan.

Bunte, R., Borgegård, L.-E. & Gaunitz, S. 1982. Vindeln: en norrländsk

kommuns ekonomiska utveckling, 1800-1980: en analys av

bondesamhällets ekonomiska utveckling, anpassning och

förvandling under 200 år. Lund: Bröderna Ekströms Tryckeri.

Bylund, E. 1956. Koloniseringen av Pite Lappmark t.o.m år 1867.

Geographica 30. PhD thesis. University of Uppsala.

Bylund, E. (1960) Theoretical considerations regarding the distribution of

settlement in inner north Sweden. Geografiska Annaler, 42, 225-

231.

Bäcklund, D. 1988. I industrisamhällets utkant: småbrukets omvandling

i Lappmarken 1870-1970. Kungliga Skytteanska Samfundets

Handlingar 34. Umeå university.

Cairncross, F. 2001. The death of distance 2.0: how the communications

revolution will change our lives. London: Texere.

Castells, M. 2010. The rise of the network society: The information age:

Economy, society, and culture. Chichester: John Wiley & Sons.

Chesnais, F. (1986) Science, technology and competitiveness. STI review,

1, 86-129.

Chiang, S.-h. (2009) The Effects of Industrial Diversification on Regional

Unemployment in Taiwan: is the portfolio theory applicable? The

Annals of Regional Science, 43, 947-962.

Cloke, P. & Edwards, G. (1986) Rurality in England and Wales 1981 - a

replication of the 1971 Index. Regional Studies, 20, 289-306.

Cloke, P. J. (1977) An index of rurality for England and Wales. Regional

Studies, 11, 31-46.

Cooke, P. (2001) Regional innovation systems, clusters, and the knowledge

economy. Industrial and corporate change, 10, 945-974.

Cutter, S. L., Renwick, H. L. & Renwick, W. H. 1985. Exploitation,

conservation, preservation: A geographic perspective on natural

resource use. Totowa: Rowman and Allanheld.

Dicken, P. 2003. Global shift: transforming the world economy. London:

SAGE Publications.

Du Plessis, V., Beshiri, R., Bollman, R. & Clemenson, H. 2001. Definitions

of rural. Rural and small town Canada analysis bulletin 3 Statistics

Canada

Elder Jr, G. H. (1994) Time, human agency, and social change:

Perspectives on the life course. Social psychology quarterly, 4-15.

68

Enequist, G. (1960) Advance and retreat of rural settlement in

northwestern Sweden. Geografiska annaler, 42, 211-220.

Engels, F. 1963. The condition of the working class in England. St Albans:

Panther Books.

Eriksson, R. 2009. Labour Mobility and Plant Performance: The influence

of proximity, relatedness and agglomeration. Gerum 2009:2.

PhD thesis. Umeå University.

Evans, N., Morris, C. & Winter, M. (2002) Conceptualizing agriculture: a

critique of post-productivism as the new orthodoxy. Progress in

Human Geography, 26, 313-332.

Falk, T. 1976. Urban Sweden: Changes in the distribution of population –

the 1960s in Focus. PhD. Stockholm school of economics.

Feenstra, R. C., Inklaar, R. & Timmer, M. P. (2015) The next generation of

the Penn World Table. The American Economic Review, 105,

3150-3182.

Fielding, A. J. (1989) Migration and urbanization in Western Europe since

1950. The Geographical Journal, 155, 60-69.

Flint, C. & Taylor, P. J. 2011. Political geography: World-economy,

nation-state, and locality. Harlow: Pearson Education.

Florida, R. 2002. The rise of the creative class, and how it is transforming

work, leisure, community and everyday life. New York: Basic

Books.

Flygare, I. A. & Isacson, M. 2003. Jordbruket i välfärdssamhället: 1945-

2000. Stockholm: Natur & kultur.

Friedman, J. 2004. Globalization, the state, and violence. Lanham:

AltaMira.

Friedman, J. 2014. Did someone say globalization? The mystification of

intellectuals and the cunning of history.

http://www.focaalblog.com/2014/07/17/did-someone-say-

globalization-the-mystification-of-intellectuals-and-the-cunning-

of-history-by-jonathan-friedman/ Accessed 2017-01-12

Friedman, J. & Friedman, K. E. (2013) Globalization as a discourse of

hegemonic crisis: A global systemic analysis. American

Ethnologist, 40, 244-257.

Friedman, K. E. & Friedman, J. 2008a. Historical transformations: the

anthropology of global systems. Rowman Altamira.

Friedman, K. E. & Friedman, J. 2008b. Modernities, class, and the

contradictions of globalization: The anthropology of global

systems. Lanham: AltaMira.

69

Friedman, T. L. 2005. The world is flat: a brief history of the twenty-first

century. New York: Farrar, Straus and Giroux.

Fröbel, F., Heinrichs, J. & Kreye, O. (1978) The new international division

of labour. Social Science Information, 17, 123-142.

Ghose, R. (2004). Big sky or big sprawl? Rural gentrification and the

changing cultural landscape of Missoula, Montana. Urban

geography, 25, 528-549.

Gårdlund, T. 1942. Industrialismens samhälle. Stockholm: Tidens Förlag.

Haase, D., Haase, A., Kabisch, N., Kabisch, S. & Rink, D. (2012) Actors and

factors in land-use simulation: The challenge of urban shrinkage.

Environmental Modelling & Software, 35, 92-103.

Halfacree, K. H. (1995) Talking about rurality: social representations of the

rural as expressed by residents of six English parishes. Journal of

Rural Studies, 11, 1-20.

Harvey, D. 1996. Justice, nature and geography of difference. Blackwell:

Oxford.

Harvey, D. (1999) Time-space compression and the postmodern condition.

Modernity: Critical Concepts, 4, 98-118.

Harvey, D. (2007) Neoliberalism as creative destruction. The annals of the

American academy of political and social science, 610, 21-44.

Heckscher, E. F. & Ohlin, B. G. 1991. Heckscher-Ohlin trade theory.

Cambridge: MIT Press.

Hedberg, C. & Haandrikman, K. (2014) Repopulation of the Swedish

countryside: Globalisation by international migration. Journal of

Rural Studies, 34, 128-138.

Hendrickson, M. K. & Heffernan, W. D. (2002) Opening spaces through

relocalization: Locating potential resistance in the weaknesses of

the global food system. Sociologia Ruralis, 42, 347.

Hjelm, J. 1991. Skogsarbetarna och motorsågen: en studie av arbetsliv

och teknisk förändring. Lund: Arkiv Förlag.

Hjort, S. & Malmberg, G. (2006) The attraction of the rural: Characteristics

of rural migrants in Sweden. Scottish Geographical Journal, 122,

55-75.

Hobsbawm, E. 2010. Age of revolution: 1789-1848. Hachette UK.

Hobsbawm, E. J. 1981. The forward march of Labor halted? London:

Verso.

Hoggart, K. (1988) Not a definition of rural. Area, 35-40.

Hoggart, K. (1990) Let's do away with rural. Journal of rural studies, 6,

245-257.

70

Hoggart, K. & Paniagua, A. (2001a) The restructuring of rural Spain?

Journal of Rural Studies, 17, 63-80.

Hoggart, K. & Paniagua, A. (2001b) What rural restructuring? Journal of

Rural Studies, 17, 41-62.

Hollander, J. B. (2011) Can a city successfully shrink? Evidence from

survey data on neighborhood quality. Urban Affairs Review, 47,

129-141.

Holm, E., Karlsson, S., Strömgren, M. & Westin, K. 2013. Vad ska man ha

ett land till? Matching av bosättning, arbete och produktion för

tillväxt. VINNOVA Report

Hospers, G. J. & Reverda, N. 2015. Managing Population Decline in

Europe's Urban and Rural Areas. London: Springer.

Hägerstrand, T. 1970. Urbaniseringen: stadsutveckling och regionala

ojämlikheter. Lund: Gleerup.

Ilbery, B. 1998. The geography of rural change. Harlow: Longman.

Ilbery, B. & Bowler, I. 1998. From agricultural productivism to post-

productivism. In The geography of rural change, ed. B. Ilbery.

Harlow: Longman.

Inglehart, R. (1971) The silent revolution in Europe: Intergenerational

change in post-industrial societies. American political science

review, 65, 991-1017.

Inglehart, R. 2004. Human beliefs and values: a cross-cultural

sourcebook based on the 1999-2002 values surveys. Madrid: Siglo

XXI.

Inglehart, R. & Baker, W. E. (2000) Modernization, cultural change, and

the persistence of traditional values. American sociological

review, 19-51.

Inglehart, R. & Norris, P. 2003. Rising tide: gender equality and cultural

change around the world. Cambridge: Cambridge University

Press.

Jansson, U. & Antonson, H. 2011. Jordbruk och skogsbruk i Sverige sedan

år 1900 – studier av de areella näringarnas geografi och

historia. Stockholm: Kungliga Skogs- och lantbruksakademien.

Jones, M. (2009) Phase space: geography, relational thinking, and beyond.

Progress in Human Geography, 33, 487-506.

Kant, I. 1998. Critique of pure reason. Cambridge: Cambridge University

Press.

Klasing, M. J. & Milionis, P. (2014) Quantifying the evolution of world

trade, 1870–1949. Journal of International Economics, 92, 185-

197.

71

Knobblock, E. & Pettersson, Ö. (2010) Restructuring and risk-reduction in

mining: employment implications for northern Sweden. Fennia,

188, 61-75.

Kohli, M. (2007) The institutionalization of the life course: Looking back

to look ahead. Research in human development, 4, 253-271.

Kohli, M. & Meyer, J. W. (1986) Social structure and social construction of

life stages. Human Development, 29, 145-149.

Lavesson, N. 2017. Rural-urban interdependencies: The role of cities in

rural growth. PhD thesis. Lund University.

Layton, I. & Lindgren, U. 1992. Spatial changes in forest-based industries:

the case of the Swedish Cellulose Company (SCA) 1954-1992.

Meddelanden 77 Umeå: Kulturgeografiska institutionen, Umeå

universitet.

Lewis, J. (1992) Gender and the development of welfare regimes. Journal

of European social policy, 2, 159-173.

Lovering, J. 1989. The restructuring debate. In New Models in Geography,

eds. R. Pete & N. Thrift, 198-223. London: Hyman.

Lundholm, E. (2007) Are movers still the same? Characteristics of

interregional migrants in Sweden 1970-2001. Tijdschrift Voor

Economische En Sociale Geografie, 98, 336-348.

Lundmark, L. (2005) Economic restructuring into tourism in the Swedish

mountain range. Scandinavian Journal of Hospitality and

Tourism, 5, 23-45.

Lundmark, L. 2006. Restructuring and employment change in sparsely

populated areas: examples from northern Sweden and Finland.

GERUM 2006:2. PhD Thesis. Umeå University.

Lundmark, M. & Malmberg, A. 1988. Industrilokalisering i Sverige:

regional och strukturell förändring. Geografiska regionstudier 19.

PhD Thesis. Uppsala University.

Magnusson, L. 2002. Sveriges ekonomiska historia. Stockholm: Prisma.

Magnusson, L. 2006. Håller den svenska modellen? Arbete och välfärd i

en globaliserad värld. Stockholm: Nordstedts.

Markowitz, H. M. (1991) Foundations of portfolio theory. The journal of

finance, 46, 469-477.

Marks, R. B. 2006. The origins of the modern world: A global and

ecological narrative from the fifteenth to the twenty-first century.

Lanham: Rowman & Littlefield Publishers.

Markusen, A. (1996) Sticky places in slippery space: A typology of

industrial districts. Economic Geography, 72, 293-313.

72

Marsden, T., Lowe, P. & Whatmore, S. 1990. Rural restructuring: global

processes and their responses. London: Fulton.

Marsden, T., Murdoch, J., Philip, L., Munton, R. & Flynn, A. 1993.

Constructing the countryside. London: University College London

Press.

Martinez‐ Fernandez, C., Audirac, I., Fol, S. & Cunningham‐ Sabot, E.

(2012) Shrinking cities: Urban challenges of globalization.

International Journal of Urban and Regional Research, 36, 213-

225.

Marx, K. & Engels, F. 2002. The communist manifesto. London: Penguin

Books.

Massey, D. (1978) Regionalism: some current issues. Capital & Class, 2,

106-125.

Massey, D. (1979) In what sense a regional problem? Regional studies, 13,

233-243.

Massey, D. B. 1984. Spatial divisions of labor: Social structures and the

geography of production. New York: Methuen.

Mather, A. S., Hill, G. & Nijnik, M. (2006) Post-productivism and rural

land use: cul de sac or challenge for theorization? Journal of Rural

Studies, 22, 441-455.

McDowell, L. & Massey, D. 1984. A women´s place? In Geography

matters! A reader, eds. A. John & D. Massey, 124-147. Cambridge:

Cambridge University Press.

McGrath, B. (2001) "A problem of resources": defining rural youth

encounters in education, work & housing. Journal of Rural

Studies, 17, 481-495.

Mormont, M. 1990. Who is rural? or, how to be rural: towards a sociology

of the rural. In Rural Restructuring. Global processes and their

responses., eds. T. Marsden, P. Lowe & S. Whatmore, 21-44.

Müller, D. K. & Ulrich, P. 2007. Tourism development and the rural labour

market in Sweden, 1960-1999. In Tourism in peripheries:

Perspectives from the far north and south, eds. D. K. Müller & B.

Jansson, 85-105.

Neffke, F., Henning, M. & Boschma, R. (2011) How do regions diversify

over time? Industry relatedness and the development of new

growth paths in regions. Economic Geography, 87, 237-265.

Nilsson, L. 2011. Efter industrialismen: urbanisering och

tätortsutveckling i Sverige 1950-2005. Malmö: Elanders Sverige.

Nordin, S. 1989. Från tradition till apokalyps: historieskrivning och

civilisationskritik i det moderna Europa. Stockholm: Symposium.

73

Nordin, S. 2006. Globaliseringens idéhistoria. Lund: Studentlitteratur.

Nordin, S. 2016. Filosoferna – Den moderna världens födelse och det

västerländska tänkandet 1776-1900. Stockholm: Natur & kultur.

Nutley, S. 1998. Rural areas: the accessibility problem. In Modern

transport geography, eds. B. S. Hoyle & R. D. Knowles, 185-215.

Chichester: Wiley.

OECD. 2006. The new rural paradigm: policies and governance. Paris:

OECD.

Ohlander, A.-S. 1994. Kvinnor, barn och arbete i Sverige 1850-1993.

Rapport för FN: s konferens om befolkning och utveckling i Kairo

1994.

Ohlander, A. & Strömberg, U. 2002. Tusen svenska kvinnoår: svensk

kvinnohistoria från vikingatid till nutid. Prisma.

Orloff, A. (1996) Gender in the welfare state. Annual review of sociology,

22, 51-78.

Osborne, P. (1992) Modernity is a qualitative, not a chronological category.

New Left Review, 192, 65-84.

Page, S. 1998. Transport for recreation and tourism. In Modern transport

geography, eds. B. S. Hoyle & R. D. Knowles, 216-243. Chichester:

Wiley.

Pahl, R. E. (1966) The rural‐ urban continuum. Sociologia ruralis, 6, 299-

329.

Pettersson, Ö. 2002. Socio-Economic Dynamics in Sparse Regional

Structures. Umeå: Umeå University.

Phillips, M. (2002) The production, symbolization and socialization of

gentrification: impressions from two Berkshire villages.

Transactions of the Institute of British Geographers, 27, 282-308.

Phillips, M. (2005) Differential productions of rural gentrification:

illustrations from North and South Norfolk. Geoforum, 36, 477-

494.

Phillips, M. (2009) Counterurbanisation and rural gentrification: an

exploration of the terms. Population, Space and Place, 16, 539-

558.

Popper, K. R. 2002. The poverty of historicism. Brighton: Psychology

Press.

Popper, K. S. 2012. The open society and its enemies. New York:

Routledge.

Porter, M. E. 1986. Competition in global industries. Boston: Harvard

Business School Press.

74

Porter, M. E. (1998) Clusters and the new economics of competition.

Boston: Harvard Business Review.

Ricardo, D. 1817. Principles of political economy and taxation. London: J.

Murray

Sayer, A. (1982) Explanation in economic geography: abstraction versus

generalization. Progress in Geography, 6, 68-88.

Schumpeter, J. A. 1943. Capitalism, socialism and democracy. London:

Allen & Unwin.

Schön, L. 2012. En modern svensk ekonomisk historia: tillväxt och

omvandling under två sekel. Stockholm: SNS förlag.

Shanahan, M. J. (2000) Pathways to adulthood in changing societies:

Variability and mechanisms in life course perspective. Annual

review of sociology, 26, 667-692.

Shucksmith, M. (1993) Farm Household Behavior and the Transition to

Post-Productivism. Journal of Agricultural Economics, 44, 466-

478.

Smith, A. 1986. The wealth of nations. London: Penguin Books.

Smith, M. R. & Marx, L. 1994. Does technology drive history? The

dilemma of technological determinism. Cambridge: MIT Press.

Solow, R. M. (1956) A contribution to the theory of economic growth. The

quarterly journal of economics, 70, 65-94.

SOU. 1938:58. 1936 års skogsutredning. Statens offentliga utredningar.

Sousa, S. & Pinho, P. (2015) Planning for shrinkage: Paradox or paradigm.

European planning studies, 23, 12-32.

Statistics Sweden. 2017. Sveriges folkmängd från 1749 och fram till idag.

http://www.scb.se/hitta-statistik/sverige-i-siffror/manniskorna-

i-sverige/befolkningsutveckling/ Accessed: 2017-05-25

Stenbacka, S. 2001. Landsbygdsboende i inflyttarnas perspektiv.

Intention och handling i lokalsamhället. Uppsala: University of

Uppsala.

Swedish Competition Authority. 2007. Konkurrensen i Sverige 2007.

Konkurrensverkets rapportserie 2007:4.

Swedish Forest Agency. 2017. Statistikdatabas.

http://pxweb.skogsstyrelsen.se/pxweb/sv/Skogsstyrelsens%20st

atistikdatabas/?rxid=4f117184-361f-4e56-9632-dd9a2b706135

Accessed: 2017-02-10

Syssner, J. 2006. What Kind of Regionalism?: regionalism and region

building in Northern European peripheries. Frankfurt: Peter

Lang Publishing Group.

75

Syssner, J. 2014. Politik för kommuner som krymper. Linköping:

Linköping University Electronic Press.

Syssner, J. (2016) Planning for shrinkage? Policy implications of

demographic decline in Swedish municipalities. Ager, 7, 7-31.

Toffler, A. 1990. Future shock. New York: Bantam.

Tregenna, F. (2010) How significant is intersectoral outsourcing of

employment in South Africa? Industrial and Corporate Change,

19, 1427-1457.

Urry, J. (1987) Survey 12: Society, space, and locality. Environment and

Planning D: Society and Space, 5, 435-444.

Urry, J. (1995) A middle-class countryside. In Social change and the

middle classes, eds T. Butler & M. Savage. 205-219. London: UCL

Press

Wallerstein, I. (1974) The rise and future demise of the world capitalist

system: concepts for comparative analysis. Comparative studies in

society and history, 16, 387-415.

Wallerstein, I. 1979. The capitalist world-economy. Cambridge University

Press.

Wallerstein, I. (2010) Structural crises. New Left Review, 62, 133-42.

Ward, N. (1993) The Agricultural Treadmill and the Rural Environment in

the Post-Productivist Era. Sociologia Ruralis, 33, 348-364.

Warsh, D. 2007. Knowledge and the wealth of nations: A story of

economic discovery. WW Norton & Company.

Weber, A. 1962. Theory of the Location of Industries. University of Chicago

Press Chicago, Ill, USA.

Weber, M. 1968. Politics as a Vocation. Philadelphia: Fortress Press

Philadelphia.

Weck, S. & Beißwenger, S. (2014) Coping with peripheralization:

Governance response in two German small cities. European

Planning Studies, 22, 2156-2171.

Westholm, E. 2008. Vad menas egentligen med landsbygd. In Ska hela

Sverige leva, eds B. Johansson. 49-58. Stockholm: 08-tryck AB

Westlund, H. (2002) An unplanned green wave: settlement patterns in

Sweden during the 1990s. Environment and Planning A, 34, 1395-

1410.

Westlund, H. & Pichler, W. (2013) The Swedish countryside in the neo‐

urban knowledge economy. Regional Science Policy & Practice, 5,

225-236.

Wiechmann, T. & Pallagst, K. M. (2012) Urban shrinkage in Germany and

the USA: A comparison of transformation patterns and local

76

strategies. International Journal of Urban and Regional

Research, 36, 261-280.

Wilson, G. A. (2001) From productivism to post-productivism ... and back

again? Exploring the (un)changed natural and mental landscapes

of European agriculture. Transactions of the Institute of British

Geographers, 26, 77-102.

Wirth, L. (1938) Urbanism as a way of life. American journal of sociology,

44, 1-24.

Woods, M. 2005. Rural geography: processes, responses and experiences

in rural restructuring. London: Sage.

Woods, M. 2010. Rural. New York: Routledge.

WTO. 2017. Statistics database. http://stat.wto.org/ Accessed on 2017-

04-10

Zelinsky, W. (1971) The hypothesis of the mobility transition.

Geographical review, 61, 219-249.

Avhandlingar i kulturgeografi

Jonsson, Ingvar (1971): Jordskatt och kameral organisation i Norrland under äldre tid. Geografiska meddelanden nr 7. Häggström, Nils (1971): Norrland’s Direct Foreign Trade 1850-1914. Geographical Reports No 2. Borgegård, Lars-Erik (1973): Tjärhantering i Västerbottens län under 1800-talets senare hälft. Geografiska meddelanden nr 13. Weissglas, Gösta (1975): Studies on Service Problems in the Sparsely Populated Areas in Northern Sweden. Geographical Reports No 5. Liljenäs, Ingrid (1977): Allmänningsskogarna i Norrbottens län – deras betydelse för det enskilda jord- och skogsbruket. Geografiska meddelanden nr 23. Arell, Nils (1977): Rennomadismen i Torne lappmark – markanvändning under koloni-sationsepoken i framför allt Enontekis socken. Geografiska meddelanden nr 24. Nordberg, Perola (1977): Ljungan. Vattenbyggnader i den näringsgeografiska miljön 1550-1940. Geografiska meddelanden nr 25. Layton, Ian (1981): The Evolution of Upper Norrland’s Ports and Loading Place 1750-1976. Geographical Reports No 6. Khakee, Abdul (1983): Municipal Planning. Restrictions, Methods and Organizational Problems. Geografiska meddelanden nr 27. Wiberg, Ulf (1983): Service i glesbygd – trender och planeringsmöjligheter. GERUM rapport B:8. Holm, Einar (1984): Att lokalisera utbildning, sysselsättning och boende. GERUM rapport B:9. Lövgren, Sture (1986): Så växer tätorten – diskussion om en modell för urban utredning. GERUM rapport nr 5.

Egerbladh, Inez (1987): Agrara bebyggelseprocesser. Utvecklingen i Norrbottens kustland fram till 1900-talet. GERUM rapport nr 7. Johnsson, Rolf S (1987): Jordbrukspolitiska stödformer – en studie av SR-, A- och B-stödens lokala effekter 1961-1981. GERUM rapport nr 11. Malmberg, Gunnar (1988): Metropolitan Growth and Migration in Peru. Geographical Reports No 9. Sivertun, Åke (1993): Geographical Information Systems (GIS) as a Tool for Analysis and Communication of Complex Data. Geographical Reports No 10. Westin, Kerstin (1994): Valuation of Goods Transportation Characteristics – A Study of a Sparsely Populated Area. Geographical Reports No 12.

Jansson, Bruno (1994): Borta bra men hemma bäst – Svenskars turistresor i Sverige under sommaren. GERUM rapport nr 22. Lindgren, Urban (1997): Local Impacts of Large Investments. GERUM kulturgeografi 1997:2. Stjernström, Olof (1998): Flytta nära, långt bort. De sociala nätverkens betydelse för val av bostadsort. GERUM kulturgeografi 1998:1. Müller, Dieter (1999): German Second Home Owners in the Swedish Countryside. GERUM kulturgeografi 1999:2. Håkansson, Johan (2000): Changing Population Distribution in Sweden – Long Term Trends and Contemporary Tendencies. GERUM kulturgeografi 2000:1. Tollefsen Altamirano, Aina (2000): Seasons of Migration to the North. A Study of Biographies and Narrative Identities in US-Mexican and Swedish-Chilean Return Movements. GERUM kulturgeografi 2000:3. Nilsson, Karina (2001): Migration among young men and women in Sweden. GERUM kulturgeografi 2001:2. Appelblad, Håkan (2001): The Spawning Salmon as a Resource by Recreational Use. The case of the wild Baltic salmon and conditions for angling in north Swedish rivers. GERUM kulturgeografi 2001:3.

Alfredsson, Eva (2002): Green Consumption Energy Use and Carbon Dioxide Emissions. GERUM kulturgeografi 2002:1. Pettersson, Örjan (2002): Socio-Economic Dynamics in Sparse Regional Structures. GERUM kulturgeografi 2002:2. Abramsson, Marianne (2003): Housing Careers in a Changing Welfare State. GERUM kulturgeografi 2003:1. Strömgren, Magnus (2003): Spatial Diffusion of Telemedicine in Sweden. GERUM kulturgeografi 2003:2. Jonsson, Gunilla (2003): Rotad, rotlös, rastlös – Ung mobilitet i tid och rum. GERUM kulturgeografi 2003:3. Pettersson, Robert (2004): Sami Tourism in Northern Sweden – Supply, Demand and Interaction. GERUM kulturgeografi 2004:1. Brandt, Backa Fredrik (2005): Botniabanan – Förväntningar i tid och rum på regional utveckling och resande. GERUM kulturgeografi 2005:4. Helgesson, Linda (2006): Getting Ready for Life – Life Strategies of Town Youth in Mozambique and Tanzania. GERUM kulturgeografi 2006:1. Lundmark, Linda (2006): Restructuring and Employment Change in Sparsely Populated Areas. Examples from Northern Sweden and Finland. GERUM kulturgeografi 2006:2.

Li, Wenjuan (2007): Firms and People in Place. Driving Forces for Regional Growth. GERUM kulturgeografi 2007:1. Lundholm, Emma (2007): New Motives for Migration? On Interregional Mobility in the Nordic Context. GERUM kulturgeografi 2007:2. Zillinger, Malin (2007): Guided Tourism - the Role of Guidebooks in German Tourist Behaviour in Sweden. GERUM kulturgeografi 2007:3. Marjavaara, Roger (2008): Second Home Tourism. The Root to Displacement in Sweden? GERUM kulturgeografi 2008:1.

Sörensson, Erika (2008): Making a Living in the World of Tourism. Livelihoods in Backpacker Tourism in Urban Indonesia. GERUM kulturgeografi 2008:2.

Hjort, Susanne (2009): Socio-economic Differentiation and Selective Migration in Rural and Urban Sweden. GERUM kulturgeografi 2009:1.

Eriksson, Rikard (2009): Labour Mobility and Plant Performance. The influence of proximity, relatedness and agglomeration. GERUM kulturgeografi 2009:2.

Khouangvichit, Damdouane (2010): Socio-Economic Transformation and Gender Relations in Lao PDR. GERUM kulturgeografi 2010:1. Eriksson, Madeleine (2010): (Re)producing a periphery - popular representations of the Swedish North. GERUM kulturgeografi 2010:2. Phouxay, Kabmanivanh (2010): Patterns of Migration and Socio-Economic Change in Lao PDR. GERUM kulturgeografi 2010:3. Hjälm, Anna (2011): A family landscape. On the geographical distances between elderly parents and adult children in Sweden. GERUM kulturgeografi 2011:1. Sandow, Erika (2011): On the road. Social aspects of commuting long distances to work. GERUM kulturgeografi 2011:2. Sandberg, Linda (2011): Fear of violence and gendered power relations. Responses to threat in public space in Sweden. GERUM kulturgeografi 2011:3. Phommavong, Saithong (2011): International Tourism Development and Poverty Reduction in Lao PDR. GERUM kulturgeografi 2011:4. Haugen, Katarina (2012): The accessibility paradox – everyday geographies of proximity, distance and mobility. GERUM kulturgeografi 2012:1. Olofsson, Jenny (2012): Go West – East European Migrants in Sweden. GERUM kulturgeografi 2012:2. Ouma, Anne (2013): From Rural Gift to Urban Commodity – Traditional Medicinal Knowledge and Socio-spatial Transformation in the Eastern Lake Victoria Region. GERUM kulturgeografi 2013:1.

Åkerlund, Ulrika (2013): The Best of Both Worlds – Aspirations, Drivers and Practices of Swedish Lifestyle Movers in Malta. GERUM kulturgeografi 2013:2. Olsson, Olof (2014): Out of the wild: Studies on the forest as a recreational resource for urban residents. GERUM kulturgeografi 2014:1. Gustafsson, Cecilia (2014): ”For a better life…” A study on migration and health in Nicaragua. GERUM kulturgeografi 2014:2. Östbring, Lisa (2015): Relatedness put in place. On the effects of proximity on firm performance. GERUM kulturgeografi 2015:1. Turi, Ellen Inga (2016): State Steering and Traditional Ecological Knowledge in Reindeer-herding Governance. Cases from Western Finnmark, Norway and Yamal, Russia. GERUM kulturgeografi 2016:1. Liljenfeldt, Johanna (2017): Where the Wind Blows – The socio-political geography of wind power development in Finland, Norway and Sweden. GERUM kulturgeografi 2017:1. Hedlund, Martin (2017): Growth and decline in rural Sweden – Geographical distribution of employment and population 1960–2010. GERUM kulturgeografi 2017:3.

GERUM kulturgeografi (1997-20xx)

(ISSN 1402-5205)

1997:1 Holm, E. (ed.): Modelling Space and Networks. Progress in Theoretical and

Quantitative Geography. 1997:2 Lindgren, U.: Local Impacts of Large Investments. (Akad. avh.) 1998:1 Stjernström, O.: Flytta nära, långt bort. De sociala nätverkens betydelse för val

av bostadsort. (Akad. avh.) 1998:2 Andersson, E.; L.-E. Borgegård och S. Hjort: Boendesegregation i de nordiska

huvudstadsregionerna. 1999:1 Stjernström, O. & I. Svensson: Kommunerna och avfallet. Planering och hante-

ring av farligt avfall, exemplen Umeå och Gotland. 1999:2 Müller, D.: German Second Home Owners in the Swedish Countryside. (Akad.

avh.) 2000:1 Håkansson, J.: Changing Population Distribution in Sweden – Long Term Trends

and Contemporary Tendencies. (Akad. avh.) 2000:2 Helgesson, L.: Högutbildad, men diskvalificerad. Några invandrares röster om

den svenska arbetsmarknaden och vägen dit. 2000:3 Tollefsen Altamirano, A.: Seasons of Migrations to the North. A Study of

Biographies and Narrative Identities in US-Mexican and Swedish-Chilean Return Movements. (Akad. avh.)

2001:1 Alatalo, M.: Sportfisketurism i Västerbottens läns inlands- och fjällområde. Om

naturresursanvändning i förändring. (Lic. avh.) 2001:2 Nilsson, K.: Migration among young men and women in Sweden. (Akad. avh.) 2001:3 Appelblad, H.: The Spawning Salmon as a Resource by Recreational Use. The case

of the wild Baltic salmon and conditions for angling in north Swedish rivers. (Akad. avh.)

2001:4 Pettersson, R.: Sami Tourism - Supply and Demand. Two Essays on Indigenous

Peoples and Tourism in Sweden. (Lic. avh.) 2002:1 Alfredsson, E.: Green Consumption Energy Use and Carbon Dioxide Emissions.

(Akad. avh.) 2002:2 Pettersson, Ö.: Socio-Economic Dynamics in Sparse Regional Structures. (Akad.

avh.)

2002:3 Malmberg, G. (red.): Befolkningen spelar roll.

2002:4 Holm, E.; Holme, K.; Mäkilä, K.; Mattsson-Kauppi, M. & G. Mörtvik: The SVERIGE spatial microsimulation model. Content validation, and example applications.

2003:1 Abramsson, M.: Housing Careers in a Changing Welfare State. (Akad. avh.)

2003:2 Strömgren, M.: Spatial Diffusion of Telemedicine in Sweden. (Akad. avh.)

2003:3 Jonsson, G.: Rotad, rotlös, rastlös – Ung mobilitet i tid och rum. (Akad. avh.)

2004:1 Pettersson, R.: Sami Tourism in Northern Sweden – Supply, Demand and

Interaction. (Akad. avh.)

2005:1 Hjort, S.: Rural migration in Sweden: a new green wave or a blue ripple? (Lic. avh.)

2005:2 Malmberg, G.: Sandberg, L. & Westin, K.: Den goda platsen. Platsanknytning och

flyttningsbeslut bland unga vuxna i Sverige.

2005:3 Lundgren, A.: Microsimulation and tourism forecasts. (Lic. avh.)

2005:4 Brandt, B. F.: Botniabanan – Förväntningar i tid och rum på regional utveckling och resande. (Akad. avh.)

2006:1 Helgesson, L.: Getting Ready for Life – Life Strategies of Town Youth in Mozam-

bique and Tanzania. (Akad. avh.) 2006:2 Lundmark, L.: Restructuring and Employment Change in Sparsely Populated

Areas. Examples from Northern Sweden and Finland. (Akad. avh.)

2007:1 Li, W.: Firms and People in Place. Driving Forces for Regional Growth. (Akad. avh.)

2007:2 Lundholm, E.: New Motives for Migration? On Interregional Mobility in the

Nordic Context. (Akad. avh.) 2007:3 Zillinger, M.: Guided Tourism - the Role of Guidebooks in German Tourist

Behaviour in Sweden. (Akad. avh.) 2008:1 Marjavaara, R.: Second Home Tourism. The Root to Displacement in Sweden?

(Akad. avh.) 2008:2 Sörensson, E.: Making a Living in the World of Tourism. Livelihoods in

Backpacker Tourism in Urban Indonesia. (Akad. avh.) 2009:1 Hjort, S.: Socio-economic differentiation and selective migration in rural and

urban Sweden. (Akad. avh.) 2009:2 Eriksson, R.: Labour Mobility and Plant Performance. The influence of proximity,

relatedness and agglomeration. (Akad. avh.)

2010:1 Khouangvichit, D.: Socio-Economic Transformation and Gender Relations in Lao PDR. (Akad. avh.)

2010:2 Eriksson, M.: (Re)producing a periphery - popular representations of the Swedish

North. (Akad. avh.)

2010:3 Phouxay, K.: Patterns of Migration and Socio-Economic Change in Lao PDR.

(Akad. avh.) 2011:1 Hjälm, A.: A family landscape. On the geographical distances between elderly

parents and adult children in Sweden. (Akad. avh.) 2011:2 Sandow, E.: On the road. Social aspects of commuting long distances to work.

(Akad. avh.) 2011:3 Sandberg, L.: Fear of violence and gendered power relations. Responses to threat

in public space in Sweden. (Akad. avh.) 2011:4 Phommavong, S.: International Tourism Development and Poverty Reduction in

Lao PDR. (Akad. avh.) 2012:1 Haugen, K.: The accessibility paradox – everyday geographies of proximity,

distance and mobility. (Akad. avh.) 2012:2 Olofsson, J.: Go West – East European Migrants in Sweden. (Akad. avh.) 2013:1 Ouma, A.: From Rural Gift to Urban Commodity – Traditional Medicinal

Knowledge and Socio-spatial Transformation in the Eastern Lake Victoria Region. (Akad. avh.)

2013:2 Åkerlund, U.: The Best of Both Worlds – Aspirations, Drivers and Practices of

Swedish Lifestyle Movers in Malta. (Akad. avh.) 2014:1 Olsson, O.: Out of the Wild: Studies on the forest as a recreational resource for

urban residents. (Akad. avh.) 2014:2 Gustafsson, C.: ”For a better life…” A study on migration and health in Nicaragua.

(Akad. avh.) 2015:1 Östbring, L.: Relatedness put in place. On the effects of proximity on firm

performance. (Akad. avh.)

2016:1 Turi, E.I.: State Steering and Traditional Ecological Knowledge in Reindeer-herding Governance. Cases from Western Finnmark, Norway and Yamal, Russia. (Akad. avh.)

2017:1 Liljenfeldt, J.: Where the Wind Blows – The socio-political geography of wind

power development in Finland, Norway and Sweden. (Akad. avh.) 2017:2 Lind, T.: Schools in Sparse Spatial Structures. (Lic.avh.) 2017:3 Hedlund, M.: Growth and decline in rural Sweden – Geographical distribution of

employment and population 1960–2010. (Akad. avh.)