Det här verket har digitaliserats vid Göteborgs universitetsbibliotek.
Alla tryckta texter är OCR-tolkade till maskinläsbar text. Det betyder att du kan söka och kopiera texten från dokumentet. Vissa äldre dokument med dåligt tryck kan vara svåra att OCR-tolka korrekt vilket medför att den OCR-tolkade texten kan innehålla fel och därför bör man visuellt jämföra med verkets bilder för att avgöra vad som är riktigt.
Th is work has been digitised at Gothenburg University Library.
All printed texts have been OCR-processed and converted to machine readable text.
Th is means that you can search and copy text from the document. Some early printed books are hard to OCR-process correctly and the text may contain errors, so one should always visually compare it with the images to determine what is correct.
012345678910111213141516171819202122232425262728 CM
MEDDELANDEN FRÅN EKONOMISK-HISTORISKA INSTITUTIONEN HANDELSHÖGSKOLAN VID GÖTEBORGS UNIVERSITET
Studies in the economic life of women
in interwar Sweden
TRYING TO MAKE A LIVING
To Gwendolyn Laveme Lane, Katie Belle Smith and to the memory of Mary Lizzie Sheffield,
Lula Belle Jackson and GeorgiaAnn Taylor.
Thank you for teaching me how to conjugate the verb, ‘to work’
TRYING TO MAKE A LIVING Studies in the economic life of women
in interwar Sweden
Linda Lane
MEDDELANDEN FRÅN EKONOMISK-HISTORISKA INSTITUTIONEN HANDELSHÖGSKOLAN VlD GÖTEBORGS UNIVERSITET
Kopieringsförbud
© Linda Lane
Kopiering är förbjuden utan författarens skriftliga medgivande Gäller även för undervisningsbruk
Tryckt på Livréna, Kungälv 2004
ABSTRACT
TRYING TO MAKE A LIVING. Studies in the Economic Life of Women in Interwar Sweden. (Publications of the Department of Economic History, School of Economics and Commercial Law, Göteborg University, no 90)
ISSN 1403-2864. IBSN 91-85196-58-4. Göteborg 2004.
Author: Linda Lane
Doctoral dissertation at the Department of Economic History, Göteborg University.
Distribution: The Department of Economic History, Göteborg University, Box 720, SE-405 30 Göteborg, SWEDEN
Studies of the emerging economic roles of women have been dominated by the Anglo-American discourse. Female experiences of economic life in these countries are often interpreted as universal. This dissertation argues that institutional arrangements particular to specific societies generate different outcomes. It is argued that the specific Swedish historical context shaped the economic lives of Swedish women in ways decisively different from the Anglo-American experience.
Trying to make a living investigates the influence of demographic conditions; such as rising nuptiality and falling nativity, for generating processes that encouraged the emergence of Swedish women into full economic citizenship as articulated through increased female market participation.
Using gender as a central concept and drawing on a quantitative approach to analyse primary archival materials, the study focuses on how demographic change affected gender conditions in the contested terrain of the labour market, as well as the institutions governing it. The dissertation finds general support for the instrumental role of demography in generating processes that challenged the gender order and by extension the division of labour, however the strength of these processes varied in time and place and among specific categories of women.
KEYWORDS: collective agreements, demography, discrimination, economic growth, gender, gender order, income, institutions, interwar years, labour market participation, occupations, standard of living, Sweden, trade unions, women.
© 2004 Linda Lane ISBN 91-85196-58 4 ISSN 1403-2864
Printed in Sweden Grafikerna Livréna Kungälv 2004
TABLE OF CONTENTS
LIST OF TABLES...10
LISTS OF FIGURES... 12
ACKNOWLEDGEMENTS...13
1. ”YOUNG PEOPLE TODAY KNOW NOTHING ABOUT WORK”... 15 1.1 Introduction 15
1.2 Gender and demographics during the interwar years 21 1.3 Selecting a site for the study - Gothenburg and its
labour market 22
1.4 The local study approach 24 1.5 Dissertation outline 26
2. WHOSE STORY IS TOLD - RESEARCH DESIGN AND SOURCE MATERIAL... 29 2.1 Introduction 29
2.2 Cluster sampling method 29 2.3 Selection criteria 31
2.3.1 Characteristicsoftheselectedresidentialdistricts 32 2.3.2 Choice of sample years 33
2.3.3 Population size and sample size 35
2.3.4 Source Material -Population Registers (MantaIsIangder) 38
3. SEPARATE AND UNEQUAL - THE DISTRIBUTION AND DEPLOYMENT OF FEMALE LABOUR... ...39 3.1 Introduction 39
3.2 The distribution of employment 41 3.2.1 Agriculture 42
3.2.2 Industry 43 3.2.3 Trade 48
3.2.4 Communication 50
3.2.5 State and municipal employees 50 3.2.6 Domestic service 50
3.2.7 Other miscellaneous industries 51
3.2.8 Occupational distribution by residential district 51 3.3 The deployment of female labour 53
3.3.1 Three hypotheses 55 3.3.2 Method 58
3.3.3 Relationship between the three hypotheses and the decomposition method 60
3.3.4 Data 62
3.3.5 Decomposition results 64 3.4 Summary 70
4. OUT TO WORK - FEMALE LABOUR SUPPLY
IN THE INTERWAR YEARS... 73 4.1 Introduction 73
4.2 Determinants of female labour supply 77 4.2.1 Expansioninoccupationalopportunities 77 4.2.2 Growth in real wages 78
4.2.3 Social and economic status of the family 79
4.2.4 Influence of demographic change and population growth 82 4.2.5 Availability of childcare and transportation 88
4.3 Individual characteristics of female labour supply 97 4.3.1 CiviIStatus 98
4.3.2 Age 99
4.3.3 Degree of gender specific disadvantage 101
4.4 The effect of gender specific disadvantage on labour market participation 107
4.4.1 The model 108 4.4.2 The results 109 4.5 Summary 113
5. TRYING TO MAKE A LIVING - FEMALE OCCUPATIONS AND INCOME STRATEGIES... 115 5.1 Introduction 115
5.2 The variation and distribution of income from employment 118
5.2.1 Variations in income as a consequence of age 118 5.2.2 Variations in income by civil status 120
5.3 Explaining differences in income between occupations 129
5.3.1 Office Workers 129 5.3.2 Domestic Service 137
5.3.3 Industrial Workers and a few wildcards- female metalworkers 146
5.3.4 Differences and Similarities 154
5.4 Summary 170
6. BRINGING HOME THE BACON - FEMALE
CONTRIBUTIONS TO FAMILY INCOMES... 173 6.1 Introduction 173
6.2 Income differences between residential districts 177 6.3 Contributions to family budgets 179
6.3.1 Contributionsbycivilstatus 179
6.3.2 Differences in averageincome between occupations 185 6.3.3 The importance of small incomes 186
6.4 Summary 190
7. SUMMARY AND CONCLUSIONS... 193
APPENDIX... 206
BIBLIOGRAPHY 210
LIST OF TABLES
Table 2.1 Gothenburg Parishes by Income Clusters 1920 and 1930... 30 Table 2.2 Labour Market Participation Rate for the Excerpted Population .
... 35 Table 2.3 The Ratio of the Excerpted Population of Employed Women to Total Taxable Female Residents by Year...35 Table 2.4 Proportion of Women in the Sample Compared to Total Taxable Female Residents... 37 Table 3.1 Employment in Gothenburg by Economic Sector
1920-1940 iin percent...42 Table 3.2 Proportion of female to Total Demanded Labour in Various Branches for Selected Years in Percent...54 Table3.3 Women’s Predicted Employment Patterns... 61 Table3.4 Contributions to Employment Change Among Newly Recruited Female Labour in all Sectors... 64 Table3.5 Contributions to Employment Change Among Females in 9- Manufacturing Industries... 65 Table4.1 Labour Market Participation in Sweden by Sex 1870-1940...75 Table4.2 Male and Female Labour Market Participants in Gothenburg
1920-1940 ...76 Table 4.3 Female Labour Market Participation by Civil Status... 98 Table 4.4 Female Labour Market Participation by Age...100 Table 4.5 Female Labour Market Participation by Number of Children and Civil Status...101 Table 4.6 Labour Market Participation of Women in Incomplete Families...
Table 4 7 Married Women and Female-headed Flouseholds with Children ... 104
Table 4.8 Probit Estimates of Labour Market Participation... 110 Table 4.9 Probit Estimates of Labour Market Participation for
married women...112 Table 5.1 Average Annual Income by Age and Year in Swedish Crowns
in 1925 year prices... 1 is Table 5.2 Average Annual Income by Civil Status, Age, Full-time,
Part-time and Year in Swedish Crowns in 1925 years prices...119 Table 5.3 Average Annual Income by Civil Status, Yearand Quartile in Swedish Crowns in 1925 years prices... 123 Table 5.4 Average Annual Income of Single and Married Women in 1925 years prices...:...127 Table 5.5 Females Employed in Office Work in Sweden by Branch
and Year... 130 Table 5.6 Average Annual Income for Employed Office Workers by Age and Income Class in Percent in 1925 years prices... 131 Table 5.7 Average Annual Income of Office Workers by Year and Degree of Gender Disadvantage in 1925 years prices... 135 Table 5.8 Average Annual Income for Domestic Servants by Age and Earning Class in Percent in 1925 years prices... 139 Table 5.9 Average Annual Income of Domestic Servants by Year and Degree of Gender Disadvantage in 1925 years prices... 145 Table 5.10 Age Distribution of Female Industrial Workers by Year... 147 Table 5.11 Average Annual Income for Female Industrial Workers
by Age and Earning Class in Percent in 1925 years prices...150 Table 5.12 Average Annual Income Industrial Workers and for Metal workers by Year and Degree of Gender Disadvantage in 1925 years Prices... 151 Table 5.13 OLS Regression Coefficients for Office Workers, Domestic Servants and Industrial Workers... 165 Table 6.1 Annual Income by Source, Residential District and Year
in 1925 years prices... 177 Table 6.2 Average Incomes by Residential District, Civil Status
and Year in 1925 years prices... 180
Table 6.3 Ratio of Female to Male Average Income by Residential District Civil Status and Year in 1925 years prices...182 Table 6.4 Ratio of Female to Male Wage Income for Selected
Occupations... 184 Table 6.5 Total Family Income by Degree of Gender Disadvantage for Selected Occupations 1924, 1931 and 1937 in Percent of Average Income in 1925 years prices...187
Table A.l Change in Placements of New Recruited Female
Labourin Gothenburg 1919-1940... 206 Table A.2 Change in Female Employment i 9 Manufacturing
Industries in Gothenburg 1919-1940... 207 Table A.3 Job Applicants, Vacancies and Job Placements in
Domestic service 1919-1939 ... 208 Table A.4 Degree of Gender Specific Disadvantage... 209
LISTS OF FIGURES
Figure 3.1 Distributions of Occupations by Residential District and Year..52 Figure 4.1 Annual Incomes for Adult Female Workers in Manufacturing and Mining 1913 & 1916-1930 (1913=100)... 79 Figure 4.2 Employed Women by Degree of GenderSpedfic Disadvantage for Gamlestaden, Vasastaden and Masthugget by Year... 105 Figure 5.1 Job Applicants, Available Jobs and Job Placements in
Domestic Service 1919-1939...143 Figure 5.2 Average Annual Earnings by Age and Year for Office Workers Domestic Servants and Industrial Workers... 166
ACKNOWLEDGEMENTS
I would like to express my appreciation to my dissertation supervisors Bengt Berglund, Jan Bohlin, Sverker Jonsson and Ulf Olsson. I am particularly gratefully to Jan Bohlin, his keen eye, support and encouragement has proven invaluable in the final stages of this project.
I would also like to express my deep appreciation to Kent Olsson whose willingness to share his knowledge of archives and personal contacts among archivists made the data collection phase of this project an easier task. I have also received very helpful comments, criticisms and intellectual support from a number of colleagues, notably Susanne Fransson, Martin Fritz, Staffan Granér, Per Hallén, Ann Ighe, Hrefna Karlsdottir, Katarina Katz, Svante Larsson, Kerstin Norlander and Christer Thömqvist. I owe a special debt of gratitude to Joan Holmedal, who read the entire text with great care, and to Annci Haag for assistance with the layout.
I particularly wish to acknowledge the contributions of students who took my classes and seminars at the Department of Economic History, Department of History and the Department of Social Work. As this work progressed their frank, openhearted comments along the way provided excellent feedback and challenged me to think and re-think my theoretical and methodological approach to this dissertation. Thanks also to Rådet för yrkeshistorisk forskning vid Arbetets Museum and the International Working Lives Doctoral School, two forums where lively debate, the exchange of ideas and good conversations in interdisciplinary environments has enriched this project.
This dissertation was initially financed by a faculty research grant. I gratefully acknowledge the additional financial support from the following institutions whose generosity made it possible to attend conferences, summer schools, participate in seminars and in essence to complete this project: FAS, Helge Ax:son Johnsons Stiftelse, Filosofiska fakulteternas gemensamma donationsnämnd,
Vetenskapsrådet (previously HSFR) and Stiftelsen Fru Mary von Sydows, född Wijk, Donationsfond.
I am also indebted to my friends and family for their encouragement and unwavering support during the long period over which this work has evolved. My final debt is to my son, Brandon whose love, patience and support of my attempt to fulfil a dream while trying to make a living for the both of us, was and is a constant source of strength.
September 2004 Linda Lane
1.
”YOUNG PEOPLE TODAY KNOW NOTHINGABOUT WORK”
1.1 Introduction
During my first years in Sweden, one of my many jobs as a student was as a home caregiver. I worked part-time visiting six or seven old-age pensioners in their homes on Saturdays and Sundays. Most of them were elderly women over 80 years old, but still capable of living on their own with help from home caregivers like myself. Primarily my job entailed providing them with social contact with the outside world during the weekend when their ordinary caregivers were free.
One of the most important social rituals during these visits was coffee drinking, which I did at all six homes in nearly as many hours. With coffee came conversation and somehow the conversations always got around to work. In these conversations, the myriad of pictures of long dead husbands and parents, and of children, grandchildren and great grandchildren standing in places of honour on tables and in bookcases were brought to life by connecting them to places of work, jobs and employment opportunities both won and lost, and of dreams and aspirations. On occasion they wanted to know what I did when I didn’t work as a home caregiver but mostly they talked about how easy life was for young people entering the world of work and sooner or later their stories were interjected with, ”you young people today, you know nothing about work, when I was young...,” thereafter followed long, winding stories of their personal experiences in the labour market.
They told stories about the places they had worked and the work they performed. Stories of workmates, cruel overbearing bosses, sexual harassment from male colleagues, long hours and low pay, but also stories of pride in jobs well done, making good friends, earning their
own money, being able to contribute at home or spending it all on new hairdos, dances or saving for the future. They spoke of coming home to unemployed husbands who had ”done nothing all day,” of how children coped alone at home and how they coped with leaving them alone. While some spoke of their lives in the labour market as interesting and worthwhile others were bitter and felt they had been cheated because their dreams of life as full-time housewives were never fulfilled. Whatever the experience, the one thing they expressed in common was the economic independence they gained from paid work.
Their stories were testimonies of the experiences of women in the labour market during a period when the institutional framework surrounding female labour market participation was most prohibitive.
The women I visited entered the labour market in the 1920s and 30s, a period when statistics tell us that Swedish women made up between 20 and 30 percent of the labour force.1 Most women were employed in the agricultural sector,2 but a large proportion had made the journey from country to town where they were employed in industry3 and in the service sector primarily as domestic servants.4 Somehow, in the economic upheaval of the interwar years their experiences were
1 Per Silenstam, Arbetskraftsutbudets utveckling i Sverige 1870-1965, Stockholm 1970.
2 Silenstam 1970. See also Sten Carlsson, Den sociala omgrupperingen i Sverige efter 1866, Stockholm 1966. For a critique of methods used by both as regards female labour, see Anita Nyberg, “Vad är förvärvsarbete?”
Kvinnovetenskaplig tidskrift, 8:1, 1987, pp. 54-65.
3 Lennart Jörberg, ”Kvinnan i det svenska industrisamhället, ” Meddelande från Ekonomisk-historiska Institution No. 28, Lund 1983. Anita Göransson, Från familj till fabrik, Lund 1988. Ulla Wikander, Kvinnors och mäns arbeten:
Gustavsberg 1880-1980, Lund 1988. Kerstin Norlander, Människor kring ett företag, Göteborg 2000.
4 Kerstin Moberg, Från tjänstehjon till hembiträde, Stockholm 1987. See even, Herbert Malmback, På väg framåt, Stockholm 1944.
forgotten or lost in discussions of male unemployment,5 general labour market unrest,6 and the fight for breadwinner wages.7 As a result, women
”disappeared” from the labour market only to re-emerge after World War II to the surprise of returning soldiers and male economists. The one reminder that women’s roles in the labour market created ripples on the otherwise smooth waters of masculine dominated rhetoric was the debate over married women’s rights to work.8 Despite claims that women had accepted a gender contract which required them to relinquish their positions in the labour market to men in exchange for economic support as wives and mothers,0 the debate made clear that
5Arbetsloshetsutredningens betänkande I, SOU 1931:20, Stockholm 1931.
Arbetslöshetsutredningens betänkande I Bilagor, Band 1, SOU 1931:21, Stockholm 1931. Arbetslöshetsutredningens betänkande II, SOU 1935:6, Stockholm 1935. Christer Lundh, Spelets regler, Lund 2000. Rolf Ohlsson &
Jonas Olofsson, Arbetslöshetens dilemma, Stockholm 1998. Roger Axelsson, Karl-Gustav Löfgren & Lars-Gunnar Nilsson, Den svenska
arbetsmarknadspolitiken under 1900-talet, Stockholm 1983. Hans Wallentin, Arbetslöshet och levnadsförhållanden i Göteborg under 1920-talet, Göteborg
1978.
6 Christer Thömqvist, Arbetarna lämnar fabriken, Göteborg 1994. Tommy
Isidorsson, Striden om tiden, Göteborg 2001.
7 Christina Carlsson, Kvinnosyn och kvinnopolitik. En studie av svensk socialdemokrati 1880-1910, Lund 1986. Yvonne Hirdman, Att lägga livet till rätta, Stockholm 1989. Susan Pedersen, Dependence and the Origins of the Welfare State: Britain and France 1914-1945, Cambridge 1995. Lena Sommestad, ”Welfare state attitudes to the male breadwinning system,”
International Review of Social History. 5, 1997, pp. 153-174.
8 Kjell Östberg, Efter Rösträtten, Stockholm 1997. Renée Frangeur,
Yrkeskvinna eller makens tjänarinna, Lund 1998. Silke Neunsinger, DieArbeit der Frauen - die Krise der Männer, Uppsala 2001.
9 Yvonne Hirdman, ”Genussystemet - teoretiska reflexioner kring kvinnors sociala underordning,” Kvinnovetenskaplig tidskrift, 9:3, 1988, pp. 49-63.
Hirdman 1989. Yvonne Hirdman, Med kluven tunga. LO och genusordningen, Stockholm 1998. Ulla Wikander, ”Periodisering av kapitalism - med kvinnor,”
Arbetarhistoria:51, 1989, pp. 7-11. Klas Åmark & Gro Hagemann, ”Från
”husmorkontrakt” till ”jämställdhetskontrakt,” Häften för kritiska studier, 33:2, 2000, pp. 4-27.
the gender order and the role of women within it was still contested terrain.
Gender is a central analytical concept in this study. I use gender to refer to the ways in which social and historical processes construct and re-construct biological sex, male and female. These constructions vary in time and space and permeate societies on a multitude of levels creating hierarchical structures within which certain masculinities are more highly valued than are femininities. The valorisation of masculinities gives those who hold these attributes power and privilege over those who do not.
At the ideological level, during the interwar years conceptions of gender were manifested in a gender order and a gender division of labour. The gender order, defined as an informal social institution regulated acceptable social behaviour for women and men.10 With its underlying ideologies and social norms, the gender order permeated all aspects of societal life. It was embedded in the social institution of the family where the gender division of labour was most refined, as well as in both formal and informal economic and political institutions. The institutional framework discouraged women from entering the labour market through a multitude of formal and informal restrictions, among these were occupational segregation, wage discrimination, protective legislation, taxation policy, and when all else failed, by reminding them
10 The discussion of institutions and their importance for economic performance is taken from Douglass C. North’s institutional theory. Although North does not specifically discuss gender, it is indirectly included in his argument that ideas and ideologies shape the subjective mental constructs that individuals use to interpret the world around them. In addition, he claims that ideas and ideologies form informal institutions that are themselves important and not simply as appendages to formal rules. See Douglass C. North, Institution, Institutional Change and Economic Performance, Cambridge 1990. Ann Ighe & Linda Lane,
“Institutionell stabilitet: Om svårigheten i att omförhandla ett genuskontrakt,” in Eva Borgström & Anna Nordenstam (eds.), Kvinnovetenskapens vadan och varthän, Göteborg 1996, pp. 108-119.
that their rightful place was in the household. The implementation of these and other restrictions supported the illusion that at the ideological and institutional levels the gender problem was solved. However at the individual level, men and women in families considered the income contributions of female family members earned in the labour market as necessary and desirable and formulated strategies thereafter.11
The strategies of individual families to send females into the labour market were supported by the rise in male unemployment during the interwar period. This suggests that many women entered and remained in the labour market out of economic need to replace family income lost due to male breadwinner unemployment.12 For women to step into the void and support the family when the male breadwinners were unable to do so was acceptable, although even this behaviour was called into question at times.13 Moreover, there is empirical support for women working not only out of need, but because they wanted to contribute towards improving family standards of living.14 And what about those women, albeit few, that worked not only for wages but because they found their work rewarding, how then did they reconcile participating in the labour market when the gender order at best only permitted them
11 Åmark & Hagemann 2000. See also Klas Åmark, ”Arbetarrörelsen,
socialförsäkringssystemet och genusordningen 1932-1970,” in Klas Åmark et al (eds.), Välfärdsstat i brytningstid, Sociologisk Forskning: Supplement,
Stockholm 1999, pp. 253-285.
Yvonne Hirdman, Den socialistiska hemmafrun och andra kvinnohistorier, Stockholm 1992. Jane Lewis, Women in England 1870-1950, Sexual divisions and social change, Bloomington 1984. Alice Kessler-Harris, Out to Work: A history of wage-earning women in the United States, Oxford 1982.
13 Frangeur 1998, pp. 178-220. See also Åsa Löfström ”Ekonomisk kris och kvinnors arbete,” Umeå Economic Studies no 99, Umeå 1981.
14 Susan P. Benson, “Living on the Margin.” in Victoria de Grazia & Ellen Furlough (eds.), The Sex of Things, California 1996, pp. 212-243. Winifred D.
W. Bolin, “The Economics of Middle-Income Family Life: Working Women during the Great Depression,” Journal of American History, Vol. 65 Issue 1, 1978, pp. 60-74.
to work out of need? One answer is that they did not need to justify their actions at all. Increasing real wages for female wage earners, a process that had been taking place since the beginning of the 20th century raised the cost of housework relative to market work, making it advantageous for many women to reconsider decisions to remain in the household.15 Furthermore, Sweden was in the process of catching-up with the large industrialised countries, new technology and production methods brought increased employment opportunities - even for women, as new occupations developed in a modern mass-production society.16
One way of describing the interwar period from a gender perspective, is to see it as a period when expanding economic opportunities coincided with conflicts between ideological conceptions of women’s and men’s rightful roles in society and the need to transgress these prescribed roles. It is evident that the frustrations and anxieties created by these conflicts were not evenly distributed among all women but was more or less intense depending upon the level of confrontation. The question then is where do the experiences of the women I visited fit- in, were they unique? My attempt to answer that question is the primary concern of this dissertation. In the dissertation I try to make some sense of the experiences of the women I visited and millions like them who entered the labour market during the interwar years and for whatever reason remained there for long periods of their lives, combining work in the productive sphere with reproductive work as daughters, wives, sisters and mothers in families.
Although power relations embedded in gender relations are crucial to the tasks undertaken in this thesis, I interpret gender relations as one
15 Lars Svensson, Closing the Gender Gap, Lund 1995. Claudia Goldin, Understanding the Gender Gap: An Economic History of American Women, New York 1990.
16 Lennart Schön, En modern svensk ekonomisk historia, Stockholm 2000.
Johan Myhrman, Hur Sverige blev rikt, Stockholm 1994.
among many hierarchical structures such as ethnicity, class, age, and civil status that privilege some individuals above others.
1.2 Gender and demographics during the interwar years
Two distinct structural features of the interwar period had influential roles on the labour market participation of Swedish women. One, the general unrest precipitated by economic and technological transformations taking place in Sweden and in the world.17 These changes raised the economic cost to society of maintaining some aspects of the existing gender order by challenging the social norms that supported the existing gender division of labour.18 In this dissertation, I argue that during the interwar years this challenge was underwritten by the second distinctive feature of the interwar period, namely, the demographic structure of the period.
A country’s demographic structure is itself a product of its institutional framework. Social norms and traditions embedded in these institutions influence demographic variables such as age at first marriage and nativity, and these in the long-run influence labour supply and economic growth. In this respect, demographic conditions during the interwar years were products of historical decisions and, as such, exogenous.19 In that respect, demographic conditions during the interwar years compelled economic and social institutions to adjust to demography and not the other way around.20
17 Schön 2000. Erik Dahmén, Svensk industriell företagarv erksamhet, Band 1, Stockholm 1950. Ingvar Svennilsson, Growth and Stagnation in the European Economy, UN Economie Commission for Europe. Geneva 1954, pp. 1-58.
18 North 1990. Ighe & Lane 1996.
19 Maria Stanfors, Education, Labor Force Participation and Changing Fertility Patterns, Lund 2003, pp. 69-115. Also Bo Malmberg & Lena Sommestad, "The EIidden Pulse of History, Age Transition and Economic Change in Sweden, 1820-2000,” Scandinavian Journal of History, Vol. 25 1/ 2, 2000, pp. 131-146.
20 Richard A. Easterlin, Birth and Fortune. The Impact of Numbers on Personal Welfare. Chicago 1980, also by the same author, Population. Labor Force and Long Swings in Economic Growth: The American Experience, New York 1969.
The gender division of labour limits women’s labour market participation in ways that are not applicable to men. By designating certain activities such as housework and care responsibilities as female specific tasks, men are freed from performing them and can participate in other activities. During the interwar years the grip of the gender order on Swedish women was loosened, not because of profound changes in the division of labour between men and women, -women still had primary responsibility for housework and care-giving, but because demographic conditions in the 1920s and 1930s credited women with fewer gender specific caring responsibilities. A decline in the number of individuals requiring these services freed women to participate in other forms of work.
1.3 Selecting a site for the study - Gothenburg and its labour market
The site of the study was determined by selecting a city that filled the following criteria during the interwar period. First, the city should exhibit a diverse economic structure that provided diversity in employment opportunities for both male and female labour. Second, a relatively rigid residential delineation based on socio-economic variables such as income, which assured that a diversity of women from a variety of social backgrounds would be included in the study and third, a relatively even population distribution. Based on these criteria the choice fell between the four largest cities in Sweden during the interwar years Stockholm, Malmö, Norrköping and Gothenburg.21 In general, all four cities complied with criteria two and three; the deciding factor was the first - diversity in the labour market. Stockholm, afforded women many employment opportunities, however as the seat of the national government, the service sector dominated with work for women concentrated to the civil service, commercial office work and domestic service.22 In Norrköping, there was plenty of work for women in the 21 The city’s name in Swedish is Göteborg.
22 Staffan Högberg, Stockholms historia, Del 2. Småstaden Fabriksstaden Storstaden, Stockholm 1981.
textile industries but comparatively little elsewhere.23 Malmö provided a more diverse array of industrial work and a growing number of opportunities in office work and other service sector jobs,24 but lacked diversity in these sectors when compared to Gothenburg.25
I chose Gothenburg, Sweden’s second largest city. Gothenburg is located on the west coast of the country at the mouth of the Göta älv River. Historically, from the Viking age village of Lödöse via Nylose to the present city that received its charter from King Gustavus II Adolphus in 1618, Gothenburg has functioned as an important port for trade and cultural exchange.26 A stipulation of its town privileges was that foreigners and other strangers living in the city be given property to build homes. The purpose was to inspire Dutch traders and other businessmen to move to the city. The Dutch were followed by Germans, Scotsmen, and Englishmen, many of whom made lasting impressions on the business life of the city. In this respect, Gothenburg was multicultural long before the modem usage of the word. During the 19th century, Gothenburg served not only as a trading port but also as the gateway to the new world for many Swedes migrating to the Americas.
By the beginning of the interwar period, when this investigation begins, emigration had ceased but due to internal migration, the population grew from 130,619 in 1900 to 202,328 inhabitants in 1920.27
23 Sven Godlund, Norrköpings historia, tiden 1914-1970, Avsnitt 13-17, Norrköping 1976.
24 Oscar Bjurling et al (rcds.J, Malmö Stads historia, femte delen, Malmö 1989.
25 Kent Olsson, Från industristad till tjänstestad 1920-1995, Stockholm 1996.
26 The early economic history of Gothenburg and its importance as a trading city and the influence of emigrants on the development of the city has been
investigated by a number of economic historians. For an in depth study of the ties between trading families on continental Europe and Gothenburg, see for example Christina Dalhede, Handelsfamiljer på Stormaktstidens
Europamarknad, Vol. 1-3, Partille 2001.
27AlvarWestman, ”Göteborgs Folkmängdsutveckling 1800-1960,” Göteborg Förr och Nu. Skriftserie 111, Göteborg 1964, pp. 145-157.
1.4 The local study approach
The choice to study Gothenburg and to limit the study geographically to the administrative boundaries of the city during the interwar period characterises the dissertation as a local study.28 In discussions of local studies in economic history research, two advantages are put forward.
First, local studies allow the researcher to perform in-depth studies of the selected geographical area using both quantitative and qualitative data as source material. The degree to which one or the other takes priority depends on the research questions asked and the availability of data. Second, the amount of data needed to complete the study is reduced to manageable levels. An often discussed reason for why local studies given the advantages listed above, are not often used in economic history research is that local studies limit the generalizations that may be drawn from the results. However, even this disadvantage may be circumvented if the choice to use or reject local studies is based not on their inability to generate general solutions but whether or not they provide an opportunity to make inferences and generate knowledge that is impossible to obtain by other methods.29 Although the local study approach limits as to what can be said about the changing economic role of women in other localities and in the country as a whole, it provides me with an opportunity to generate answers to questions about the labour market strategies of individuals and groups of women, which justifies its use in this dissertation.
Research on female labour market participation has relied heavily on aggregate data collected from census reports. That is, data collected
28 As late as 1950, 80 percent of Gothenburgs regional population lived in the city’s administrative boundaries and 88 percent of workplaces were found there.
This does not exclude the possibility of workers from outside the administrative boundaries working inside them or the opposite. Concerning the definition and limiting of administrative boundaries see: Lars Nilsson, Den urbana
transitionen, tätorterna i svensk samhällsomvandling 1800-1980, Stockholm 1989.
29 Björn Tropp, Att sätta spaden i jorden, Göteborg 1999, pp. 44-47.
every ten years or after special decree, as was the case for the partial census taken in 1936. Alternatively, industry statistics have been used.30 Unfortunately, less than one-third of women engaged in the labour market worked in industry.31 Thus, the majority of women in the labour force are normally excluded from studies using this type of data. Aside from the problems inherent in the collection of aggregate data, conclusions drawn from analyses using such data do not necessarily hold true at more disaggregated levels; in fact, the norm is to expect discrepancies. In general, at the aggregate level, many of the underlying explanatory factors that generate the observed results are lost. Therefore, any attempt to explain female labour force participation during the interwar period based entirely on aggregated employment data at the national level distorts participation levels for groups of women in particular places, times and occupations. Furthermore, analyses at the aggregate level are incapable of taking into account changes in gender divisions of labour negotiated through individual and group actions at the local level.
Rejecting the aggregate approach and usage of census or industry data, some scholars has focused on specific occupations. Among these are studies of women in candle-making-,32 porcelain-,33 match-34 and textile factories,35 yet others have concentrated on investigating the division of labour, development of class identities and conflicts among
30 Silenstam 1970. Carlsson 1966. LennartJorberg 1983. Lynn Karlsson & Ulla Wikander, (eds.), ”Kvinnoarbete och könssegregering i svensk industri 1870-
1950: tre uppsatser,” Uppsala Papers in Economic History, No 9, Uppsala 1985.
31 Kvinnoarbetskommitténs betänkande, SOU 1938:47, Stockholm 1938.
32 Norlander 2000.
33 Wikander 1988.
34 Bengt Berglund, Industriarbetarklassens formering. Arbete och teknisk förändring vid tre svenska fabriker under 1800-talet, Göteborg 1982.
35 Eva Hesselgren, V? äro tusenden, Stockholm 1992.
specific groups of female employees such as office workers,36 teachers,37 dairy workers38 and nurses.39 However, until now, no study has had as its primary goal to systematically examine labour market participation of Swedish women from a variety of socio-economic backgrounds and occupations within the same study for the interwar years 1919-1939, an oversight that this study seeks to correct.
To empirically investigate the theoretical arguments, I use data excerpted from primary archival sources. The method used to select that data is described in Chapter two. Archival sources are complemented with data excerpted from published annual reports of private enterprises and public municipal organisations, public statistics and periodicals.
1.5 Dissertation outline
My aim is to study the economic life of women in Gothenburg during the interwar years 1919-1939 by investigating underlying developments that supported change in the supply of and demand for their labour in the labour market. To gain insight into the economic benefits of labour market participation for women I will investigate the degree of economic independence obtainable for women through wage work. The method used to accomplish these goals is by carrying out a number of in-depth studies of various aspects of women’s economic life.
The dissertation consists of seven chapters; in each with the exception of chapter two, I ask and seek answers to specific questions. Chapter two discusses the primary archival sources and the methods used to select individuals to the study. Considerable space is also given to arguments concerning specific choices. The following chapter, chapter
36 Mats Greift, Kontoristen: Från chefens högra hand till proletär, Lund 1992.
Birgitta Conradsson, Kontorsfolket, Stockholm 1988.
37 Christina Florin, Kampen om katedern, Stockholm 1987.
38 Lena Sommestad, Från mejerska till mejerist: en studie av mejeriyrkets maskuliniseringsprocess, Lund 1992.
39Agneta Emanuelsson, Pionjäre i vitt, Huddinge 1990.
three, presents an overview of the labour market in Gothenburg during the period studied. The structure of the labour market is discussed with emphasis on the potentials for female employment opportunities in an effort to capture the demand for female labour. Included in the chapter is an investigation of employer strategies in employing and deploying female labour. The aim of chapter four is to discuss and investigate female labour supply. The chapter discusses a number of variables of personal character that affected female labour supply - so called human capital attributes, but also includes a brief look at transportation in terms of distance, time and cost and child care facilities, two institutional arrangements that are thought to have profound impact on female labour market participation. A more rigorous examination of female labour supply is attempted using statistical methods in the later half of the chapter. In chapter five, women’s wages is the focus of attention. The aim is to study development in earnings between different categories of employed women from a comparative perspective. Of specific interest are differences that arose because of individual characteristics such as age, civil status and occupation. Chapter six focuses on the rewards from employment. I am primarily concerned with investigating levels of economic independence women could obtain through their individual effort given structural hinders such as wage and occupation segregation. Comparisons are made between groups of women, within and between occupations and between men and women. Chapter seven is given over to a summary and presentation of the findings.
2
.WHOSE STORY IS TOLD - RESEARCH DESIGN AND SOURCE MATERIAL
2.1 Introduction
The primary source of archival data used in this investigation was excerpted from Population Registers for the city of Gothenburg for 1925, 1932 and 1938. The excerpts were used to construct datasets suitable for statistical analysis at the household level. In this chapter, I will discuss the process used to select that data.
In quantitative research, a stochastic model represents any uncertainty in the phenomena under investigation. Thus, although the sample selection process plays no direct role in the model building process, the manner in which a sample is selected may limit any inference that can be drawn about the results of the model. To obtain data that would permit me to study the questions of interest in this dissertation the selection process should provide a representative sample of the female population of Gothenburg living in private households. The collected data should permit generalizations about the female population and it should be cost effective in terms of time spent in the archives. A method that fulfilled these criteria was cluster sampling.
2.2 Cluster Sampling Method
A cluster or stratified sampling method provides for the selection of sample members by breaking the population down into stratas, or clusters based on supplementary information. After dividing the population into clusters, a random sample is drawn from each cluster.40
40 Overviews of sampling methods are found in most textbooks dealing with research methods. See for example Alan Bryman, Social Research Methods,
Table 2.1 Gothenburg Parishes by Income Clusters 1920 and 1930
High income parish
1920 Middle income
parish
Low income parish
High income parish
1930 Middle income
parish
Low income parish
Domkyrko Haga Carl Johans Domkyrko Haga Carl Johans
Vasa Annedals Gamlestads Vasa Annedals Gamlestads
Kristine Masthuggs Kristine Masthuggs
Oscar Fredriks Oscar Fredriks
Source: Statistisk årsbok för Göteborg, Tab. 15. c) 1932 and Tab. 14. b) 1937.
The advantages of using cluster sampling aside from cost effectiveness, is that it is relatively easy to implement. Furthermore, it guarantees representativeness of all clusters and the final sample is representative of the population, thereby improving sampling reliability.
The cluster sampling method was used to select three residential districts or neighbourhoods. Each district should represent one socio
economic profile called ”high-income,” ”middle-income” or ”low income.”41 The selection process began by using data collected at the parish level to establish a group of statistically equivalent residential districts from which to choose.
In 1920, Gothenburg was divided into nine parishes and thirty-one residential districts. For the purpose of this study, parishes were grouped into clusters using a combination of average income and average wealth within each parish for two years 1920 and 1930 the goal was to maximize differences between the three groups.42 Table 2.1 shows the Oxford 2001, pp. 90-92. Lawrence W. Neuman, Social Research Methods:
Qualitative and Quantitative Approaches, Boston 2001, pp. 209-211. For views of the method using historical data see, Loren Haskins & Jeffery Kirk,
Understanding Quantitative History, Cambridge 1990, pp. 141-143.
41 The categorization by the terms high-income, middle-income and low- income as opposed to high, middle or low-class that are much more difficult todefine follows the example set by other historians. See for example, Bolin 1978, p. 4.
42 Lundby Parish is excluded because although the municipality was
incorporated into Gothenburg city in 1906, for ecclesiastical purposes the parish was held separate. Örgryte municipality and parish were incorporated into Gothenburg city after 1920 and were therefore excluded.
distribution of the parishes within clusters for the two years examined.
The result was unambiguous; residents in Domkyrko and Vasa Parishes had the highest incomes and the greatest wealth in both years while Carl Johans, Masthuggs and Gamlestads Parishes were equally distinguished as those with the lowest income and wealth in both years. From among the group of parishes that were in the same cluster in both 1920 and 1930, three were selected to represent the defined socio-economic criteria. The parishes selected were Vasa, Oscar Fredriks and Gamlestads. From each selected parish a residential district was randomly selected. The residential district Vasastaden represents the high-income parish Vasa. From the middle-income cluster, the residential district Masthugget represents Oscar Fredriks Parish and from the low-income cluster, Gamlestaden represents Gamlestads Parish.
The final stage in the selection process was divided into two steps.
First, the entire population of women who were gainfully employed in 1924, 1931 and 1937 with an income greater than null was excerpted from Population Registers for the three selected residential districts. The second step was the collection of a random sample of every 20th woman from the same residential districts and for the same years.
2.3 Selection criteria
The population of employed women and the sample were excerpted from Population Registers (Mantalslängder) for three residential districts. Registration in Population Registers takes place after the tax year has ended hence the information they contain reflects conditions of the previous year.43 Therefore, data collected from Population Registers established in 1925, 1932 and 1938 were excerpted to study the conditions of individuals for the sample years 1924,1931 and 1937.
Population Registers are kept for each parish. The register provides information about each individual residing in the parish and forms the
43 Gösta Lext, Studier i svensk kyrkobokföring 1600-1946. Göteborg 1984, pp. 310-316.
base for the national census. From 1923, the old administrative system based on ”rote” was modernized. The new administrative system subdivided each parish into units called ”kvarter” or quarters. A residential district “stadsdel” is composed of a certain number of quarters based on population density.44
2.3.1 Characteristics of the selected residential districts
Beyond the characteristics obtained in the sampling technique, the three residential districts had characteristics that added to their attractiveness as study objects. Gamlestaden belonged to the low-income parish Gamlestads. Among the characteristics making it interesting to this study were the facts that already in 1920, 31 percent of all women residing in the district were gainfully occupied, that the industrial structure provided a diversity of employment opportunities for women ranging from skilled industrial employment to typically unskilled secondary employment. Furthermore, during the interwar years as population increased, the district supplied a growing number of opportunities for female labour in the service sector.
Vasastaden in Vasa Parish was chosen to represent the high-income cluster. In Vasastaden, few married women were employed in the labour market, but the families they were members of were major employers of female labour. A large number of women in high-income families had the economic wherewithal and opportunity to pursue higher education, an advantage that opened the door to professional positions in the labour market. These women, mostly single, were likely to be found in middle- to high-level civil service positions in health, education and social welfare or in trade and commercial enterprises. Alongside them was an even larger group of low-income earners residing in the district as live-in domestic servants. An important characteristic of Vasastaden is that the number of female domestic servants was greater than the number of females from middle-income backgrounds employed
44 For an overview of the old and new system see, ”Stadens läge. indelning och areal,” Statistisk årsbok för Göteborg, Göteborg 1926, pp. 1-7.
in the labour market; as a result, the district is characterised by high levels of female employment but by low average annual incomes.
Furthermore, the two groups are distinguished by the fact that annual income from wages for women from middle-income backgrounds made up only a portion of their income; income from other sources such as capital was equally important. Vasastaden supported the employment of other groups of women and men through a flourishing cluster of boutiques, shops, and service establishments catering to the needs of the wealthy.
Masthugget in Oscar Fredriks parish was chosen because of its mixed social structure; middle-income professionals and civil servants along Linnégatan, and artisans mixed with skilled and unskilled labourers living in the Långgatans. Masthugget offered a variety of employment opportunities including industrial bakeries, textiles and the garment industries all of which employed large amounts of female labour. There was also a growing service sector and many opportunities for female owned small businesses and enterprises.
2.3.2 Choice of sample years
A goal for the selected sample is that it should provide ”photographs”
of women’s positions in the labour force at particular times. Thus, the choice of sample years is important if the ”photographs” are to be fair representations of female labour force participation. A supposition often used to describe female labour force participation is that, women enter the labour force in economic upswings when demands for labour exceed that which can be supplied by men and that they leave the labour force in periods of economic decline. Therefore, the choice of study years should include not only periods of prosperity and economic growth but also periods of decline. The years investigated were chosen to cover the period of interest as well as to reflect the fluctuations in the labour market caused by the economic turmoil and general unrest of the interwar period. The selection was based on labour market trends in Sweden and Gothenburg during the interwar years as outlined below.
In Sweden, an economic depression with over-capacity and unemployment followed the international inflation that was experienced
directly after World War I. By the latter half of 1923 however, increased demand for Swedish goods in the international market, rationalization in domestic production processes and increasing real wages had improved the unemployment situation. Furthermore, strong expansion in trade and commerce and services offered new employment opportunities for both men and women until cut short by the depression 1929-1933. The effects on the Swedish economy were perhaps strongest between 1930 and 1933, with falling export levels and high unemployment.45 The crash of Kreuger’s empire in 1932 added to the general upheaval. Gothenburg, a city where industry and related activities in the form of services and subcontractors were concentrated towards the export market was especially vulnerable.
By the end of 1933, the Swedish economy had begun to experience an upswing. Increased demand for Swedish exports and expansion in industrial production for the domestic market created an environment conducive for the employment of women. Among these were the textile and garment industries, and the meatpacking and canning industries.
These industries were characterised by the use of new techniques and organization methods that affected skills, wages and the gender of the employed. In both cases demand for female labour increased.46
With these economic conditions in mind, the year 1924 was selected because it was a year of relative economic stability, which should have provided employment opportunities for women, either in the new expanding services sector or in industries undergoing different forms of rationalization. A year of economic decline is represented by the selection of 1931. The expectation is that economic decline should affect both male and female employment levels. Whether the impact is in the same direction, that is decline in employment for both, or follows the same pattern within and across industries is determined by the structure of the labour market and the attitudes towards female labour within it.
The last year included in the investigation is 1937. Sweden by 1937 was
45 Dahmen 1950, pp. 28-44. Schön 2000, pp. 287-348.
46 Schön 2000, pp. 320-324.
Table 2.2 Labour Market Participation Rate for the Excerpted Population
Year Total female population 16-64 in Gothenburg
Percentoftotal female population 16-64 excerpted
Total number of taxable women in the three residential districts
Total number of employed women in the three residential districts
Percent of taxable women excerpted
1924 80134 4 8036 3270 40
1931 94614 4 8198 3330 40
1937 105858 3 8050 3550 44
Total 280606 24284 10150
Source: Population Registers 1925,1932 and 1938, GLA. Statistisk årsbok för Göteborg.
well underway towards economic recovery, again providing employment opportunities in new sectors, specifically in the areas of health, education and welfare. Thus, given the general developments within the labour market, the choice of the three years are deemed adequate for comparisons of a number of variables affecting female labour force participation under the variety of labour market conditions presented above.
2.3.3 Population size and sample size
In Population Registers, inhabitants were divided into two categories, those who were obliged to pay tax and those exempted. Those exempted from paying tax were the old-aged/handicapped and dependent children.
All individuals, including those belonging to the exempted categories were obligated to register in their parish of residence. Moreover, no individual was permitted registration in more than one parish at a time.
In the selection process, only taxable females, i.e. those women that were not exempted from paying taxes, were of interest. The excerpted population represents taxable women in each residential district with income greater than zero.
Table 2.2 relates the excerpted population of employed women in the three residential districts to the taxable population according to population registers and to the female population of Gothenburg for 1924, 1931 and 1937. Table 2.3 below presents the results of the excerpts in each residential district. In the table, the excerpted population of employed females is compared with the number of taxable women in each residential district for each year. The relative high
Table 2.3 The Ratio of the Excerpted Population of Employed Women to Total Taxable Female Residents by Year
Numberoftaxable Excerptedpopulationof Population as % of
women employed women taxable women
1924 Gamlestaden Vasastaden Masthugget Total
1562 557 36
3942 1714 43
2532 999 39
8036 3270
1931 Gamlestaden Vasastaden Masthugget Total
2197 776 35
3669 1586 43
2332 968 41
8198 3330
1937 Gamlestaden Vasastaden Masthugget Total
2359 941 40
3504 1602 48
2187 1007 46
8050 3550
Source: Population Registers 1925,1932 and 1938, GLA.
percentage of employed women selected results from the manner in which “gainfully” employed is defined.
In this study women are qualified if they were employed during some period during the year and had an income greater than zero. I have chosen to use this selection method based on the premise that family decisions to send women into the labour market were made in the interest of the family unit and should be respected. To exclude women simply because the sum earned was less than some arbitrary amount or because they worked only short periods of time would ignore the ability of some families to make decisions in their own interest. To exclude small earners in this study would be especially negligent since a primary reason for selecting a local study method is that it supports study of individual actors. In this case, given the view that family decisions are reflected through levels of female employment, to not include this group of women would be to ignore one of the basic tenets of the method.
Table 2.4 Proportion of Women in the Sample Compared to Total Taxable Female Residents
Number of taxable Excerpted sample Sample as % of taxable
women women
1924
Gamlestaden 1562 83 5
Vasastaden 3942 128 3
Masthugget 2532 106 4
1931
Gamlestaden 2197 87 4
Vasastaden 3669 139 4
Masthugget 2332 87 4
1937
Gamlestaden 2359 116 5
Vasastaden 3504 130 4
Masthugget 2187 79 5
Total 24284 955
Source: Population Registers 1925, 1932 and 1938, GLA
One of the aims of this dissertation is to study female labour supply that is, to understand how individual characteristics of female labour effected the decision to enter the labour market. A study of female labour supply required data not included in the original excerptions concerning women not engaged labour market activity. To that end, from among taxable women in the same residential districts and for the same years, a random sample of every 20th female, both employed and unemployed, over age 15 was excerpted.47 The results of the excerptions are presented in table 2.4.
2.3.4 Source Material -Population Registers (MantaIsIangder)
From the abundance of material included in the Population Registers the following data were collected for each individual woman: name of residential district, taxation year, occupation, date of birth, birth parish,
47 To ensure that the conditions for a random sample were not violated, a computer generated random number between 1 and 20 was used to determine the starting number for the selection.
civil status, number of dependent children48, annual income from wages and/or salary, income from business activities, income from capital and name of employer. Although men in general had higher incomes and were the designated head-of-household, their role in this study is of secondary interest, thus they enter the study only as spouses of married or cohabiting women. For that reason, the income from spouses of married and co-habiting women was also collected.
Population Registers included information on demographic structure and family relationships, income from a variety of sources; source of employment and taxes assessed and paid. Family relationships are defined relative to the head-of-household, normally the husband if the household contained a married couple or, if not, the senior adult or main breadwinner. A woman was declared head of household if she was single, divorced or widowed and had her own domicile. In such cases, she was even responsible for all dependent children.
48 For the purpose of this study, dependent children are defined as children sixteen years of age or younger, not engaged in labour market activity and for who parents, could claim a tax exemption.
3
.SEPARATE AND UNEQUAL
- THE DISTRIBUTION AND DEPLOYMENT OF FEMALE LABOUR
3.1 Introduction
It has not been enough that women have felt the need or desire to enter paid work, the types of employment opportunities available to them in both industry and the service sectors have had a profound impact on the emergence of women into the labour market. Employers’ decisions to utilize male and female labour in specific times and places and under a variety of economic conditions were influenced by the manner in which employment opportunities were distributed across and within specific sectors and branches. In turn, the distribution of job opportunities reflected the gender division of labour.
Explanations for the increased labour market participation of women either tend to focus on supply-side effects, which relate labour market participation to choice, or on demand-side effects, which focus on structural change. Studies of demand-side effects include changes in the number, characteristics and distribution of jobs in labour markets, changes in technology, and the level of discrimination among both employers and employees. Although the demand for female labour is central to understanding gender outcomes in the labour market, the study of demand-side effects influencing the demand for female labour has received less empirical attention than has the study of female labour supply.49
49 An attempt to correct this deficiency is found in the work of Janice F.
Madden. See Janice F. Madden, The Economics of Sex Discrimination, Toronto and London 1973. Janice F. Madden “Discrimination - A Manifestation of Male
The demand for labour is differentiated by quality, that is, employers demand a certain sort of labour. One of the characteristics that separate a desired labour sort from another is the biological sex of that labour.50 Notions of women as only temporary workers in the labour market and ideas about the kinds of work fit for them to perform whilst there were prevalent during the interwar period. These perceptions were instrumental in creating and maintaining two separate labour markets - one for occupations gender coded as female and one for those coded as male.51 This chapter emphasizes how the differentiation of the labour market by sex, affected the demand for female labour and the manner in which female labour was utilized by employers.
To demonstrate the effects of institutionalised separation of the labour market into a male and female component on the demand for female labour during the interwar years the chapter begins with a description
Market Power? In Cynthia B. Lloyd (ed.) Sex, Discrimination and the Division of Labor, New York and London, 1975, pp. 146-174. Maddens research adds a gender perspective to previously developed theories of monopsony in Joan Robinson, The Economics of Imperfect Competition Book IX, London 1933, pp.
292-304 and the study of trade union power in Martin Bronfenbrenner, “The Economics of Collective Bargaining,” Quarterly Journal of Economics, Vol. 53, August 1939, pp. 535-561.
50 Madden 1973, pp. 69-85. For discussions of how these conditions are institutionalised and maintained, see Yvonne Hirdman, Women - from possibility to problem? Arbetslivscentrum Report no. 3, Stockholm 1994. Robert M.
Solow, The Labor Market as a Social Institution, Cambridge 1990, pp. 2-23.
Alice Kessler-Harris, “The Just Price, the Free Market and the Value of Women,” Feminist Studies, no.14, 1988, pp. 235-250.
51 That what was considered women’s work was different from men's work was no secret in the labour market. These differences were historically codified in both formal and informal arrangements concerning labour market conditions, wages and occupations for men and women. Inger Jonsson, Deltidsarbete och deltidsarbetslöshet inom svensk detaljhandel, Arbetslivsinstitutet HELA Projektet 3, Stockholm 2003. Norlander 2000. Inger Humlesjö,
”Manlighetskonstruktion i arbetarhistoria och fackföreningar,” Häften för kritiska studier, 31, 1998, pp. 3-13. Sommestad 1992. Wikander 1988.
