Gender division of labour in unified Germany

WORC Report 00.00.000/0

Jan Künzler Wolfgang Walter Elisabeth Reichart Gerd Pfister

ISBN 00-00000-00-0 WORC Report 00.00.000/0 WORC EUROPEAN NETWORK ON POLICIES AND THE DIVISION OF UNPAID AND PAID WORK Tilburg University P.O. Box 90153 5000 LE TILBURG The Netherlands © 2001

II

Preface In 1995 the European Network on Policies and the Division of Unpaid and Paid Work was launched. It was the result of a meeting in Tilburg where researchers from the European Union countries decided to set up an international network. The aim of the network was to prepare and conduct an international comparative study on the impact of various policy measures on the division of paid and unpaid work between men and women. The starting activities of the network were financially supported by the European Science Foundation. No part of this text may be published in whole or part, reproduced electronically or distributed without the express written consent of the authors. Short quotations may be cited in scholarly works and reviews. During the following years the participants, representing most of the EU countries, agreed on a research programme. The overall purpose of the project was to gain insight into the conditions under which policy measures are (or are not) effective in influencing the division of paid and unpaid work between men and women. For that purpose the network has planned two projects within the programme: (1) National studies that follow one and the same design. These national studies focus on policies, on the structural and cultural context in which these policies are embedded and implemented, and on the (resulting) division of paid and unpaid work among men and women; (2) An international comparative study that integrates the results of the national studies and investigates the correspondence between the outcomes of the national studies, resulting in clear statements about effects of policy measures in their specific context. Furthermore, the international comparative study examines the correspondence between a specific mode of regulation (the balance of social, institutional and economic forces that characterize the division of work at a particular time) and the prevailing societal division of work between men and women. In this report we describe the present German policy mix in social policy, family policy, and equal opportunity policy and its history. We give an overview over the development of gender inequality in education, in the labour market, and in unpaid work with a focus on changes following unification. We report findings of a secondary analysis on changing gender role attitudes. We summarize recent theoretical developments in the explanation of the division of housework. Finally, we present results of our study on "Gender Division of Labour in Germany, 2000". In summer 2000, a national representative sample was interviewed using a questionnaire developed by the European Network on Policies and the Division of Unpaid and Paid Work. The project is a joint venture of the Wuerzburg University and the State Institute of Family Research at the University at Bamberg. The survey and the analyses were funded by the German Science Foundation (Deutsche

III

Forschungsgemeinschaft DFG, grant no. LI 350/3-1). The leader of the project is Wolfgang Lipp, chair at the department of sociology at Wuerzbug University. The field work of the study was done by the "Institute of Applied Social Sciences" (Institut für angewandte Sozialwissenschaft) INFAS, Bonn. At INFAS, Doris Hess, Janina Belz, and Stefan Schiele were in charge of the field work; the collaboration with them was very pleasant. When preparing this report, we had the opportunity to discuss our ideas with Menno Jacobs, Tilburg, who made valuable remarks. James Brice, Sandy Halliday, and Felix Blaser helped in the technical production of the report.

IV

The following papers and reports written by members of the Network have already been published by WORC: Frinking, G.A.B. & Willemsen, T.M. (1996). Travail et Famille dans les Pays de l’Union Européenne. Le Rôle des Politiques Étude Méthodologique. Tilburg: WORC paper 96.12.022/6. Willemsen, T.M. (1997). European Network on Policies and the Division of Unpaid and Paid Work. Survey Questionnaire. Tilburg: WORC paper 97.07.003/6. Barrère-Maurisson, M.-A. & Frinking G.A.B. (1997). The impact of Policies on the Division of Labour: A New Approach. Tilburg: WORC paper 97.10.013/6 Miettinen, A. (1997). Women in Europe. Data on Demographic Factors, Economic Activity, Education and Related Issues in Selected European Countries. Tilburg: WORC paper 97.10.014/6 Vossen, A.P., Frinking, G.A.B. & Willemsen, T.M. (1997). De invloed van overheidsmaatregelen op de verdeling van betaalde en onbetaalde arbeid binnen huishoudens: De visie van deskundigen. Verslag van een pilot study. Tilburg: WORC paper 97.12.016/6 Willemsen, T.M. & Frinking, G.A.B. (Eds.) (1998). The role of social partners in the redivision of paid and unpaid work, an international comparison. Tilburg: WORC report 98.05.002/6 Torres, A.C. & Vieira da Silva, F. (1999). Childcare and division of work between men and women. Tilburg: WORC paper 99.01.001/6 Jacobs, M.J.G. (1999). Effects of policies and gender role attitudes on the division of paid and unpaid work within households. Tilburg: WORC paper 99.03.004/6

V

VI

Contents PREFACE .............................................................................................................III CONTENTS............................................................................................................1 LIST OF TABLES ..................................................................................................4 LIST OF FIGURES.................................................................................................6 ZUSAMMENFASSUNG .........................................................................................1 ABSTRACT............................................................................................................4 CHAPTER 1 ...........................................................................................................7 GERMANY’S POSITION IN THE MODERNIZATION OF GENDER RELATIONS 7 1.1

Introduction .....................................................................................................................................7

1.2

The two German states and gender relations................................................................................8

1.3

Comparative frameworks for the modernization of gender relations ...........................................................................................................................................11

CHAPTER 2 .........................................................................................................15 POLICIES AFFECTING THE GENDER DIVISION OF LABOUR ..............................................................................................................15 2.1 2.1.1 2.1.2 2.1.3 2.1.4

Paternalist policies.........................................................................................................................17 Marriage benefits........................................................................................................................17 Family benefits ...........................................................................................................................21 Lone mothers benefits ................................................................................................................23 Conclusions ................................................................................................................................25

2.2.1 2.2.2 2.2.3 2.2.4 2.2.5 2.2.6

Maternalist policies........................................................................................................................25 Maternity protection ...................................................................................................................26 Parental leave .............................................................................................................................26 Child day care.............................................................................................................................28 Elderly care ................................................................................................................................34 Work schedule policies...............................................................................................................35 Conclusions ................................................................................................................................36

2.2

1

2.3. East-West-differences and German unification..........................................................................37 2.3.1 GDR family policy .....................................................................................................................37 2.3.2 Unification and gender relations — what is the outcome? .........................................................38

CHAPTER 3 .........................................................................................................41 STRUCTURE AND DEVELOPMENT OF GENDER INEQUALITY........................................................................................................41 3.1

Gender ideology .............................................................................................................................41

3.2

Gender differences in education...................................................................................................43

3.3 3.3.1 3.3.2 3.3.3 3.3.4 3.3.5 3.3.6 3.4

Gender differences in the labour market ....................................................................................49 Overall participation...................................................................................................................50 Unemployment ...........................................................................................................................51 Overall participation by mothers ................................................................................................53 Part-time employment ................................................................................................................54 Employment without social protection.......................................................................................56 Income........................................................................................................................................57 Division of unpaid work ................................................................................................................59

CHAPTER 4 .........................................................................................................65 EXPLAINING THE DIVISION OF UNPAID LABOUR: THEORETICAL APPROACHES ....................................................................................................65 4.1

New home economics.....................................................................................................................65

4.2

Time-availability approach...........................................................................................................65

4.3

Resource theory .............................................................................................................................66

4.4

Role theory .....................................................................................................................................67

4.5

Doing gender approach .................................................................................................................68

4.6

Integrating findings .......................................................................................................................68

4.7

Regulation approach .....................................................................................................................70

CHAPTER 5 .........................................................................................................73 GENDER DIVISION OF LABOUR IN GERMANY: THE STUDY .........................................................................................................73 5.1

2

Sample ............................................................................................................................................73

5.2

Dependent variables ......................................................................................................................77

5.3

Independent variables ...................................................................................................................78

CHAPTER 6 .........................................................................................................81 STALLED MODERNIZATION IN THE WEST — BACK TO TRADITION IN THE EAST? .................................................................................................................81 6.1

Univariate findings ........................................................................................................................81

6.2

Multivariate findings.....................................................................................................................91

6.3

Discussion .....................................................................................................................................100

REFERENCES...................................................................................................103 APPENDIX .........................................................................................................111 NOTES...............................................................................................................151

3

List of Tables Table 1: Table 2: Table 3: Table 4: Table 5: Table 6: Table 7: Table 8: Table 9: Table 10: Table 11: Table 12: Table 13:

Policies affecting the gender division of labour in the FRG .............. 16 Effects of tax splitting for spouses and child tax credit...................... 19 Employment of mothers 1972, 1991, and 1996 by age of youngest child in Germany ................................................................. 53 Time use studies in Germany: Weekly time in housework, 1965-1995.............................................. 62 Design and procedure of the survey ................................................... 73 Number and average age of respondents ............................................ 74 Household types of respondents ......................................................... 75 Number of children living in respondents’ households ...................... 76 Marital status of respondents .............................................................. 77 Theories, variables and operationalization ......................................... 79 Housework and the extent of spouse’s employment in Germany, 1995-2000 ...................................................................... 85 OLS-regression models predicting time spent on housework for all respondents and couples ................................................................ 93 OLS-regression models predicting time spent on housework and child care for couples with children.................................................... 99

Appendix-Table A1: Appendix-Table A2: Appendix-Table A3: Appendix-Table A4: Appendix-Table A5: Appendix-Table A6: Appendix-Table A7: Appendix-Table A8: Appendix-Table A9: Appendix-Table A10:

4

Germany in the modernization of gender relations, 1980s-1990s .................................................................. 113 Development of child benefit/tax credit in the FRG, 1946-2000...................................................................... 116 Comparative advantages of child benefit/tax credit in the FRG, 1996-2000.................................................. 118 Incidence and duration of social assistance in Germany, 1995 .............................................................. 118 Family policy measures in the GDR ............................. 120 Development of birth subsidy in the GDR, 1950-1990...................................................................... 123 Development of child benefit in the GDR, 1950-1990...................................................................... 123 Development of the use of day care in Germany, 1990-1999...................................................................... 124 Development of day care facilities for infants under three (Krippen) in Germany, 1950-1998....................... 125 Coverage level of day care facilities for infants under three (Krippen) in the federal states of Germany, 1990-1998..................................................... 126

Appendix-Table A11: Appendix-Table A12: Appendix-Table A13: Appendix-Table A14: Appendix-Table A15: Appendix-Table A16: Appendix-Table A17: Appendix-Table A18: Appendix-Table A19:

Appendix-Table A20: Appendix-Table A21: Appendix-Table A22: Appendix-Table A23: Appendix-Table A24: Appendix-Table A25: Appendix-Table A26: Appendix-Table A27: Appendix-Table A28: Appendix-Table A29: Appendix-Table A30: Appendix-Table A31:

Day care facilities for infants under three (Krippen) in Germany.................................................................... 127 Development of day care coverage for pre-school children between three and six (Kindergarten) in Germany, 1950-1998..................................................... 129 Coverage level of day care facilities for pre-school children between three and six (Kindergarten) in the federal states of Germany, 1990-1998 .......................... 130 Opening hours of day care facilities for pre-school children between three and six (Kindergarten) in Germany, 1994 .............................................................. 131 Day care facilities for pre-school children between three and six (Kindergarten) in Germany ..................... 132 Coverage level of day care facilities for school children (Horte), in the federal states of Germany, 1990-1998...................................................................... 134 Indicators of fertility in Germany, 1950-1998 .............. 135 Long-term care in Germany, 1984-1998....................... 136 Gender role orientations in Germany, 1982-1996: conservative answers (percentages of respective group) and overall scale scores (mean and standard deviation)....................................................................... 137 CASMIN categories and corresponding educational levels in the FRG and the GDR..................................... 138 Level of general education by birth cohort in West Germany ........................................................................ 139 Level of general education by birth cohort in East Germany ........................................................................ 140 Level of vocational education by birth cohort in West Germany............................................................... 140 Level of vocational education by birth cohort in East Germany ........................................................................ 140 Time use in West Germany (hours/week)..................... 141 Time use in East Germany (hours/week) ...................... 142 Time use in West Germany (hours/week)..................... 143 Time use in East Germany (hours/week) ...................... 144 Independent variables — descriptives .......................... 145 Housework — zero-order correlations.......................... 147 Child care — zero-order correlations............................ 148

5

List of Figures Figure 1: Figure 2: Figure 3: Figure 4: Figure 5: Figure 6: Figure 7: Figure 8: Figure 9: Figure 10: Figure 11: Figure 12: Figure 13: Figure 14: Figure 15: Figure 16: Figure 17: Figure 18: Figure 19: Figure 20: Figure 21: Figure 22:

Gender ideology in Germany, 1982-1996 .......................................... 43 Female/male ratio (ln) of persons in Germany with no degree or general elementary education by birth cohort..................................... 44 Female/male ratio (ln) of persons in Germany with intermediate degree by birth cohort ......................................................................... 45 Female/male ratio (ln) of persons in Germany with maturity degree by birth cohort ......................................................................... 46 Female/male ratio (ln) of persons in Germany with completed vocational training by birth cohort...................................................... 47 Female/male ratio (ln) of persons in Germany with university degree by birth cohort ......................................................................... 48 Labour force participation in Germany, 1965-1999 ........................... 50 Unemployment rates in Germany, 1976-1999.................................... 52 Part-time employment in Germany, 1980-1998 ................................. 55 Income of full-time employees in Germany, 1983-1996.................... 57 Participation in housework in international comparison, 1965 .......... 60 Housework time in international comparison, 1965........................... 60 Ratios of women’s/men’s time in housework in Germany, 1965-1995 ...................................................................... 61 Housework time in Germany, 1965-2000........................................... 82 Housework ratios in Germany, 1965-2000......................................... 83 Housework ratios by earner type in West Germany, 1995-2000........ 84 Housework ratios by earner type in East Germany, 1995-2000......... 84 Men’s time use in West Germany, 2000............................................. 87 Men’s time use in East Germany, 2000.............................................. 87 Women’s time use in West Germany, 2000 ....................................... 89 Women’s time use in East Germany, 2000......................................... 89 Dependence in couple households and husbands’ housework. .......... 96

Appendix-figure A1:

6

Rate of income tax in Germany, 1999 .......................... 116

Zusammenfassung 1995 veranstaltete das Work and Organization Research Centre WORC an der Katholischen Universität von Brabant in Tilburg, Niederlande, eine Tagung zum Thema "Work and Family in Europe: The Role of Policies". Die dort versammelten Forscherinnen und Forscher beschlossen, längerfristig zusammenzuarbeiten, um das Defizit an vergleichbaren Daten zur Verteilung von Hausarbeit und Kinderbetreuung zu beheben und konstituierten das European Network on Policies and the Division of Unpaid and Paid Work. Das Netzwerk wurde in den folgenden Jahren (1995-2001) unter anderem von der European Science Foundation unterstützt. Ein zentrales Ziel der Zusammenarbeit war es, in den Ländern der Europäischen Union vergleichbare, repräsentative Umfragen zur Verteilung der unbezahlten Arbeit zwischen Frauen und Männern durchzuführen; in Finnland, den Niederlanden, Frankreich, Griechenland, Portugal und in Deutschland sind die Erhebungen mittlerweile abgeschlossen. Das deutsche Projekt wurde von Jan Künzler, Wolfgang Lipp und Wolfgang Walter vorbereitet und von der Deutschen Forschungsgemeinschaft DFG mit einer Sachbeihilfe (LI 350/3-1) unterstützt. Im vorliegenden Bericht berichten wir über den Stand der Entwicklung der Geschlechterverhältnisse in Deutschland: Die deutsche Vereinigung hat 1990 zwei Staaten zusammengeführt, wie sie auch in Ausprägung und politischer Bearbeitung geschlechtsspezifischer Ungleichheit nicht unterschiedlicher hätten sein können. Im internationalen Vergleich nahmen BRD und DDR bei fast allen Indikatoren extrem entgegengesetzte Positionen ein - die DDR gehörte regelmäßig mit den skandinavischen Ländern zu den modernisierten Ländern, die BRD gehörte regelmäßig mit den südeuropäischen Ländern zu den traditionalen Ländern. Kapitel 1 gibt einen Überblick über den Stand der Modernisierung der Geschlechterverhältnisse in der BRD und in der DDR im internationalen Vergleich vor der Vereinigung. Kapitel 2 beschreibt die politischen Maßnahmen, die geeignet sind, geschlechtsspezifische Ungleichheiten abzubauen oder zu reproduzieren. Dabei wird zwischen paternalistischen und maternalistischen Maßnahmen unterschieden. Paternalistische Maßnahmen unterstützen die traditionale Ernährerrolle des Mannes. Maternalistische Maßnahmen entlasten die Frauen von ihrer traditionalen Zuständigkeit für die Betreuung von Familienangehörigen oder verpflichten sie darauf. In der westdeutschen Familienpolitik überwogen die Elemente mit traditionalisierender Wirkung bei weitem. Im Zuge der Vereinigung wurden die politischen Rahmenbedingungen des westdeutschen Geschlechterverhältnisses auf die DDR übertragen. Kapitel 3 liefert einen Überblick, wie Arbeitsmärkte, wie Frauen und Männer, wie die Geschlechterverhältnisse in Ostdeutschland auf die politische Anpassung an Westdeutschland nach 1989 reagiert haben. Die Befunde sind gemischt. Eigene Längsschnittanalysen der Allgemeinen Bevölkerungsumfrage in den

1

Sozialwissenschaften ALLBUS zeigen, dass die Welle des Antifeminismus, die die anderen Transformationsländer Osteuropas in den neunziger Jahren erfasst hat, in Ostdeutschland ausgeblieben ist. Im Gegenteil: Die Geschlechtsrollenorientierung der Westdeutschen ist vergleichweise konservativ geblieben, die der Ostdeutschen ist liberaler geworden, zwischen beiden Teilen Deutschlands hat sich eine wachsende Kluft aufgetan. Die Bildungsungleichheit zwischen Frauen und Männern ist in der DDR früher abgebaut worden als in der BRD, die jedoch nachgezogen hat, so dass bei den Bildungsabschlüssen mittlerweile in beiden Teilen Deutschlands kaum noch eine Benachteiligung der Frauen zu finden ist. Trotz des Abbaus der Bildungsungleichheit blieb die geschlechtsspezifische Ungleichheit auf dem Arbeitsmarkt in Westdeutschland allerdings erstaunlich stabil. Die hohe Erwerbsbeteiligung der Frauen in der DDR ist nach der Vereinigung deutlich zurückgegangen, liegt aber immer noch über der der Frauen in Westdeutschland. In Ostdeutschland sind die Frauen überproportional von der andauernd hohen Arbeitslosigkeit betroffen. Andererseits sind die Lohnunterschiede zwischen Frauen und Männern im Osten immer noch kleiner als im Westen. Die wenigen Studien zur Verteilung der Hausarbeit zeigen, dass der temporäre Modernisierungsvorsprung der DDR in den neunziger Jahren wieder verschwunden ist. In beiden Teilen Deutschlands verbringen Frauen noch ungefähr doppelt so viel Zeit mit Hausarbeit wie Männer. In Kapitel 4 werden mit New home economics, Time-availability approach, Ressourcentheorie, Rollentheorie und Doing gender approach die wichtigsten theoretischen Ansätze zur Erklärung unterschiedlicher Beiträge zur Hausarbeit bei Frauen und Männern dargestellt. Der Einfluss politischer Maßnahmen auf die familiale Arbeitsteilung soll im Rahmen eines neu zu entwickelnden Regulationsansatzes modelliert werden. In Kapitel 5 wird die Studie "Geschlechtsspezifische Arbeitsteilung im vereinten Deutschland 2000" (Gender division of labour in unified Germany, 2000) beschrieben. Im Sommer 2000 wurden in einer repräsentativen Stichprobe (mehrstufige Zufallsauswahl) 2019 Haushalte in Westdeutschland und 982 Haushalte in Ostdeutschland ausgewählt, in denen telefonische, standardisierte Interviews (CATI) mit Personen durchgeführt wurden, die zwischen 20 und 50 Jahre alt und deutscher Nationalität sein mussten. Gegenstand des Interviews waren unter anderem Zeitverwendung, Berufs- und Familienbiographie, Haushaltszusammensetzung, Einstellungen und Beziehungsqualität. Bei Personen, die mit einer Partnerin oder einem Partner zusammenlebten, wurde auch die Partnerin bzw. der Partner um ein Interview gebeten. In 1031 Haushalten konnten beide Partner interviewt werden. Für die hier vorgestellten Analysen wurden allerdings nur die Hauptinterviews benutzt. In Kapitel 6 wird der Zeitaufwand, den Frauen und Männer im Haushalt betreiben, für eine Reihe von Haushaltstypen berechnet und zum Teil mit älteren Daten verglichen. Zwischen 1995 und 2000 hat sich die Verteilung der Hausarbeit in Westdeutschland kaum geändert, in Ostdeutschland ist die Verteilung dagegen

2

traditioneller geworden. Mit multiplen linearen Regressionsanalysen wird für unterschiedliche Teilstichproben (alle Befragte, Paare, Paare mit Kindern) untersucht, welche Faktoren den Zeitaufwand für die Hausarbeit und für direkte Aktivitäten mit Kindern beeinflussen. Time-availability approach bzw. New home economics werden bestätigt: Der wichtigste Prädiktor ist der Umfang der Erwerbstätigkeit der Befragten. Aber auch Hypothesen der Rollentheorie und der Ressourcentheorie werden zumindest zum Teil bestätigt. Politische Maßnahmen (familienergänzende Kinderbetreuung und finanzielle Transfers wie Kindergeld und Erziehungsgeld) haben nicht nur einen indirekten Einfluss (via Erwerbsbeteiligung), sondern auch einen direkten Einfluss auf Hausarbeit und Kinderbetreuung. Die multivariaten Analysen bestätigen ebenfalls die Traditionalisierung der Verteilung der unbezahlten Arbeit im Osten. Männer in Ostdeutschland machen — im Vergleich zu westdeutschen Männern — deutlich weniger Hausarbeit, als aufgrund ihrer liberaleren Orientierungen und der größeren finanziellen Unabhängigkeit ihrer Partnerinnen zu erwarten wäre.

3

Abstract In 1995, the Work and Organization Research Centre at the Catholic University of Brabant in Tilburg, the Netherlands, organized a conference on "Work and Family in Europe: The Role of Policies". The researchers participating in the meeting soon agreed that there was a severe lack of data on the gender division of housework and child care; they decided to start an ongoing collaboration on the topic and established the European Network on Policies and the Division of Unpaid and Paid Work. From 1995 to 2001, the Network was supported by a grant of the European and the Dutch Science Foundation. One of the network’s prominent aims was to conduct representative and comparable surveys on the gender division of unpaid work in the countries of the European Union. In the meantime, surveys have been conducted in Finland, Greece, Portugal, France, the Netherlands, and Germany. The German study was prepared by Jan Künzler, Wolfgang Lipp (University of Wuerzburg), and Wolfgang Walter (State Institute for Family Research at the University of Bamberg) and supported by a grant of the German Science Foundation DFG (No. LI 350/3-1). In the present paper we report on the state of development of gender relations in Germany: German unification in 1989 brought together two countries which extremely differed with regard to both the extent of gender inequalities and to the policies aiming at a modernization of gender relations. In international comparison, the FRG and the GDR regularly occupied opposing ranks in all indicators of gender inequality. Together with the scandinavian countries, the GDR formed a group of modern countries with low inequality. Together with the mediterranean countries, the FRG formed a group of traditional countries with high inequality. Chapter 1 gives an overview of the state of the modernization of gender relations in preunification Germany in international comparison. Chapter 2 describes the political measures which are supposed to reduce or to reproduce gender inequalities. We differentiate between paternalist and maternalist measures. Paternalist policies support men’s traditional provider role. Maternalist policies either enforce or reduce women’s traditional obligation to care for dependent family members or relatives. In West German family policy, the traditional elements have been dominant all the time. German unification transfered the West German traditional policy mix to East Germany. Chapter 3 reports on the development following East Germany’s adoption of the West German policy regime. The findings are mixed. According to a longitudinal analysis of the German General Social Survey ALLBUS, in East Germany there was no wave of antifeminism as in the other postsocialist countries in Eastern Europe. On the contrary: East Germans’ liberal attitudes became even more liberal in the 1990s while West Germans’ traditional gender role attitudes persisted with a growing gap between East and West Germany. Inequalities in education between women and men have been reduced relatively early in East Germany but West

4

Germany followed in the meantime. At present, disadvantages of women totally vanished at least in general schooling. In spite of the reduction of inequalities in education, gender inequalities in labour market participation have not changed much in West Germany. The high employment rate of women in East Germany decreased in the post-unification period but until now it is still higher than the employment rate of women in West Germany. Also, gender differences in wage rates are lower in East Germany. The few studies on the division of housework show a temporary modernization lead of the GDR which disappeared until the early 1990s. In the 1990s, both in East and in West Germany women spend twice as much time doing housework as men. Chapter 4 presents five approaches currently used in quantitative research aiming at an explanation of differences in women’s and men’s contributions to housework: new home economics, time-availability approach, resource theory, role theory, and doing gender approach. We propose to include a regulation approach and hypotheses on the effects of policies in future research. In order to test hypotheses derived from the approaches mentioned above we analyze data of the network study "Gender division of labour in unified Germany, 2000". The study is described in Chapter 5. In summer 2000 in a representative sample (multistage random sampling) of n = 2,019 households in West Germany and 982 households in East Germany, persons between 20 and 50 years of age and of german nationality were interviewed by telephone (CATI) on various areas as time use, employment history, family history, household composition, attitudes, and marital quality. In case that a respondent was living together with a spouse, his or her partner was asked for an interview, too. We got interviews by n = 1,031 couples. In this paper we exclusively rely on the interview of the first respondent. Chapter 6 presents the findings of the study. We calculate weekly hours spent in housework and child care in different types of living arrangements and households. Compared to 1995, there were only minor changes in the division of housework in West Germany but a marked traditionalization in East Germany. We search for factors influencing the time spent in housework and child care using OLSregression analysis. Analyses confirm the hypotheses of time-availability approach and new home economics: Time in housework and child care is reduced by time in employment which is by far the most important predictor. Resource theory and role theory are partially supported, too. Housework and child care time are not only indirectly influenced by policies, i.e., additional child care and financial transfers (e.g., child benefits): There are direct influences, too. Multivariate analyses confirm the bivariate finding of a traditionalization in East Germany: Men in East Germany spend less time in housework compared to men in West Germany taking into account that their attitudes are less traditional and that their partners contribute a higher share to household income.

5

6

Chapter 1 Germany’s position in the modernization of gender relations 1.1

Introduction

This report on the division of paid and unpaid labour in contemporary Germany is placed in a comparative and theoretical framework that aims at the explanation of the state and the change of gender relations in European societies. This framework is meant to provide answers to three interrelated questions. First: What is the specific position of German society in the process of modernizing gender relations? Or: What is the German path to more modern gender relations? Modernization has led to a more equal sharing of life opportunities and risks between the sexes in most developed countries, but from different starting points and at a different pace. We chiefly discuss this question in this chapter. Special attention is drawn to the post-war existence of two German states and their subsequent unification. The post-war division of Germany led to two distinct paths in the modernization of gender relations that are now in the process of gradual transformation. Although the political, economic, and partly even cultural order of West Germany is dominant in this transformation, politically-unified Germany still bears the marks of its social division (see Section 1.2). The description of paths of modernization in Germany is subsequently put into comparative perspective by the use of socio-political indicators which relate them to the range of alternative models within the realm of the Western European and OECD countries. As a consequence, two conceptual frameworks are reviewed that include classifications of the paths of modernization (Section 1.3). The conclusion derived from this theoretical and methodological discussion is that the framework for assessing the division of labour in particular should take into account the interrelations of different dimensions of gender inequality, namely the relevant policies and the general pattern of social structures affecting the two genders. This leads to the next query. The second question is partly descriptive and partly heuristic in nature: What are the general factors and dimensions that affect and encompass the gender division of labour? In the second chapter (pp. 15ff.), we focus on the relevant policies. There are two reasons for doing so. First, the frameworks of comparative study of Western societies discussed in Section 1.3 suggest that policies have a predictive and/or symptomatic role for gender inequalities. And, second, this study is a product of the European Network on Policies and the Division of Unpaid and Unpaid Work (hereinafter called Network), whose members have dedicated their efforts to this very theme – a stance which, of course, has been influenced by the

7

views brought forward in the above-mentioned theoretical frameworks. In the third chapter (pp. 41ff.) we discuss several dimensions of gender inequality, their state, development and variation in terms of regional disparities (East-West) or other social influences. In part, this already provides us with data on the division of labour between the sexes as, e.g., in the section on employment; and, in part, the conditions and intermediary factors affecting the division of labour, such as, e.g., educational attainment, are reviewed more thoroughly. Up to this point, the whole analysis is more or less on the macro-level. We will leave this level in the following chapters and attempt to answer a third question: What are the predictors on the micro- or individual level that account for the division of unpaid labour? Again, there are different influences in the course of our research project that have led to this perspective. First, our research group belongs to several European countries in which, in the course of the co-operation within the Network, an extensive survey study has been undertaken. On the basis of a common questionnaire, the main aspects of the gender division of labour are being studied in this survey. The second influence which has led to the study of conditions on the micro-level has also given it its specific direction, namely the targeting of the division of unpaid labour. There is an extensive literature and a group of theories devoted to explanations of the division of housework. In our research project, we rely heavily on theoretical perspectives that will be reviewed in chapter 4 (pp. 65ff.). Partly for pragmatic reasons and partly in order to make a substantial contribution to the theoretical frameworks discussed in chapter 4, we restrict our perspective to the division of unpaid labour. The fifth chapter (pp. 73ff.) provides information on data and methods, while chapter 6 presents descriptive and multivariate findings. In short, the division of housework (and as we will show: the division of child care in the family) is — at least partly and presumably — a combined result of the division of paid labour, the demands of unpaid labour in the household, gender role orientations, the command of resources, the incentives/disincentives set by policies and the attempt to display a certain gender image. Therefore, the considerations in the chapters on policies (pp. 15ff.) and gender inequalities (pp. 41ff.) are put in a different perspective. They mutate from being an aspect of the gender division of labour on the macro-level to predictors on the micro-level. Moreover, the first question on Germany’s position in the process of modernization is viewed from a different angle. In discussing the results of our study, we claim to show that the different developments and multivariate relations in East and West Germany are part of a double process: belated traditionalization (in the East) and restricted modernization (in the West).

1.2

The two German states and gender relations

Due to the unification of East and West Germany in 1990, contemporary Germany has become a very peculiar case for the study of modernization. The two regimes

8

that existed in the former GDR and the former FRG were opposed to each other in scope as well as in depth: Institutionally, politically, economically, culturally, etc. they became paradigms of the Western (democratic, capitalist) and the Eastern (one-party politics, planned economy) paths to modernization. For a long time, gender relations (and the differences between the two systems) offered a means of distinguishing the two regimes. West Germany, on the one hand, has been — for a long time — a paradigm of the Continental European approach to family and gender relations under the auspices of Catholicism, Conservatism, and corporatism. Although the reasons for and causes of the Gestalt of gender inequality are seen differently in theories of gender, comparative social policy and history, it is obvious that the unequal division of paid and unpaid labour has its roots in cultural traditions, political measures and the unequal distribution of resources between the sexes. These conditions have contributed to the FRG’s position as a laggard in the modernization of gender relations. East Germany, or the former GDR, on the other hand, could be seen as an example of enforced modernization under the auspices of Socialism, egalitarianism, and centralism. The unquestioned GDR female model was the full-time employed mother — a model that was supported not only by ideology and propaganda, but also by employment policy, child care services, work schedule regulations, and financial incentives. Consequently, the two Germanys occupy rather opposite positions in the "benchmarking" of gender relations. By classifying two criteria (the starting point and the pace of change) for the assessment of the modernization of gender relations into four classes (traditional, take-off, modernized, modern) and applying this scheme to areas of gender inequality for which comparative empirical data are available (education, gender role attitudes, labour force participation, division of household labour), Künzler (1999a, pp. 215ff.) concludes that West Germany is mainly a take-off country (with the exception of labour force participation, which is traditional) and East Germany a modern country (with the exception of the "modernized" state of inequality in educational attainment). As is well known, German unification did not take place on equal terms. The West German polity, legal, economic, and civil society systems were transferred to (or imposed upon) East Germany. Although there is disagreement on whether the political style has changed due to unification (from the "Bonn Republic" to the "Berlin Republic"), there is no point in disputing that the structural characteristics of the new FRG are — with only minor exceptions — identical with those of the old one.1 It can be argued that this historical constellation has led to an assimilation of the East German life course and habitus to the West German counterpart, which implies its becoming more traditional — a hypothesis that will be partly supported by the data presented in this report. Because the West German gender regime seems to be the prevailing one in the unequal competition that ended with unification, it deserves attention in terms of its

9

Gestalt and underlying conditions. First: What is the system of inequalities with regard to gender? In feminist scholarship, several criteria have been discussed. Besides the conventional criteria of social structure (educational attainment, income, labour force participation, etc.), standards that are seen as specific for the situation of women have been put forward, such as, e.g., the possibility to maintain an autonomous household, anti-discrimination provisions or protection against marginalization (Orloff, 1993; Ostner, 1998). Not all of these dimensions are represented by indicators in comparative research. And: Not all indicators correspond to unequivocal data sources and standardized measures. Therefore, a compilation of empirical data in different areas is presented that gives an overview of the position of West Germany in the process of modernizing gender relations. For comparative reasons, data on these indicators for East Germany, or the former GDR, are given, as well as the relative position of the two countries in the respective set of industrialized and post-industrialized countries, with a low rank indicating a rather traditional, a high rank, a rather modern position. Appendix-Table A1 compiles several indicators of gender inequality, mainly in public policy (such as the availability of child care services) as well as outcomes of these conditions or indicators of gender inequality per se (such as differences in educational attainment, employment, etc.). The social policy variables can be differentiated as financial support and support via services. For the first type of public assistance, Germany2 displays relatively high ranks and values for the support of non-earning wives, i.e., a relatively high difference in net income between a one-earner couple and a single person (see indicators 6.1a-c)3 as well as for the support of families with children (see indicators 6.2a-b, 3.1), although the level of "economic intervention" into family functions is below average with respect to the proportion of public spending to financially support families (in per cent of the gross domestic product GDP, see indicator 9).4 The counterpart of these benefits are disincentives that take effect if the wife becomes employed. In this case, the German tax and social assistance system produces relatively high values in extra-marginal tax (see indicator 3.2), benefit loss to unemployed husbands (see indicator 3.3) and difference between gross and net income (indicator 6.3). The second type of public support, the availability of social services, has the effect of facilitating or hindering female labour force participation (and all the outcomes thereof, such as income, self-sufficiency, self-esteem, or independence) by indirectly assigning them the tasks of child care and care for other dependents. Most notably, day care coverage as a percentage of the youngest age-group (< 3) expresses this feature of the German welfare state with ca. 3% over different sources and reference dates (see indicators 1.2, 8.1, 10). For pre-school children (in the 1980s day care coverage was 60% of three to six year olds, see indicator 8.2), recent changes in policy (see below) are not included in this indicator.5 Another policy in this area is care for other dependents, measured by the provision of

10

home-help services for the aged, in which Germany ranks below average (see indicators 1.3, 8.4), and of care institutions, in which it has at least an average position (see indicators 8.3). The final set of indicators that could be seen as predictors for the division of labour are gender role attitudes.6 As indicator 12.1 shows, the level of modernity in this respect in the younger cohorts of West Germany leans to the traditional side and is below average, whereas East Germany leads the OECD countries.7 The outcomes of the modernization of gender relations, which can be seen as dependent variables, display a rather clear-cut picture. West Germany is located in the lower half of the ranking for educational attainment, labour force participation, and the division of household labour, whereas East Germany is clearly in the upper half.8 West Germany is in the lower half of the ranking for educational attainment (see indicators 4.5, 11.1), labour force participation (indicator 13.1), the multiplied score for female work desirability (indicator 4.6) and — notabene — the division of household labour (indicators 14.1, 2.3).

1.3 Comparative frameworks for the modernization of gender relations The second part of the question on the position of Germany in the modernization of gender relations is — to put it simply: Why? The question is tricky, though, as not only the causes or predictors have to be found, but also an assessment must be made of its relative position. At best, the degree of modernization is only part of this assessment. The concept of modernization suggests a seductively simple answer, because it implies dichotomous classification, one-directional progress and unequivocal evaluation. In order to give a more differentiated picture, we must use various typologies that have been developed in the comparative analysis of gender inequalities. We will discuss two of them in order to clarify the aspects that have to be taken into account in explaining gender relations comparatively. Esping-Andersen (1990, 1996, 1999) sees West Germany as a model of the "conservative welfare-state regime"9 with its étatist-corporatist approach based on employee social insurance ("Bismarck-model") intended to contribute to the well-being of different classes which are regarded as part of a given societal order. These classes or — as they were formerly understood: — "Estates" (Stände) are constituted by their legal position (wage earner, salaried employee, civil servant, or self-employed person) and/or their income level. Consequently, the welfare state encourages the traditional "bourgeois" family model, which has been maintained by offering only a low level of child care services and an level of financial support that is — comparatively speaking — at most moderate to generous. In recent times, the deficit in "state de-familialization", i.e., the lack of care services provided for children and the elderly, has — in his view — blocked further modernization. Continental Europe has welfare states where there is no work and few children,

11

because it is not creating enough jobs in the service economy and because women (especially the highly-qualified) are opting out of family formation in order to pursue their careers. Although this view has met with substantial criticism in feminist scholarship, Esping-Andersen (esp. in 1999) claims that he can also explain the situation of the "femina oeconomica" with his approach. However, his results are — at best — ambiguous. On the one hand, Esping-Andersen provides us with an ambitious explanatory claim in which causes (such as, e.g., paternalist policies adopted by conservative elites or the structure of welfare state arrangements vis-à-vis market and family) are clearly distinguished from outcomes — the social and financial situation of the (male and female) population that they are meant to account for. On the other hand, he cannot avoid the pitfalls of a one-directional concept of modernity, in which all he can explain is almost exclusively: the difference between the Scandinavian countries and the others.10 An alternative to the regime-typology is the earner model concept propounded by Lewis and Ostner (Lewis & Ostner, 1994; Ostner, 1995a, 1995b). They have discussed the German case (along with, e.g., Great Britain, Ireland or the Netherlands) as an example of the "strong wage-earner model" which expresses the norm of the traditional division of labour between the sexes, which is reinforced by several social policy provisions such as income taxation, child care supply or parental leave. The typology also includes the effects of this normative arrangement. Consequently, labour force participation or the social status of women (or subgroups as lone mothers) are indicators for the earner model. On the one hand, Lewis and Ostner provide us with a classification that is directly applicable to questions of gender inequality and therefore combines the relevant dimensions in a more pronounced way. On the other hand, it implies a homogeneity within the indicators that cannot be attributed to, e.g., the relation between child care policies and gender role attitudes that both contribute to the gendered division of labour (see Künzler, 1999a; Künzler, Schulze, & van Hekken, 1999). In other words, although the earner model concept allows for the co-existence of different modes of gender relations in modern societies, it is, at least in this respect, also one-directional. There are at least two dimensions with which these approaches can be compared. The first is the degree to which independent and dependent variables are distinguished. One could call this aspect "explanatory ambition". Esping-Andersen scores high on this dimension — and falls short of the expectations raised by his theory. Nevertheless, both formally and substantively, he does differentiate between independent variables, historical ones such as class coalitions, as well as institutional ones such as public policy profiles, and dependent variables, i.e., de-commodification, restructuring of social inequality and women’s participation in the labour force and the family. A theory of this format has two methodological consequences. First, independent variables on the macro-level have to be found that represent general characteristics of the societies to be compared, such as, e.g., the

12

public policy profile. Second, a theory has to be formulated that expresses the effects of these macro-variables on behaviour at the micro-level.11 The second dimension for comparing the approaches is found in their appropriateness for explaining the gender division of labour. Highly adequate for this task is Lewis’ and Ostner’s approach, which involves assembling comprehensive information on the most important aspects of gender inequality. Methodologically, this requires the compilation of typological indicators that cover both the conditions of gender inequality, e.g., the provision of child care, and gender inequality per se, e.g., labour force participation. Reflecting on the strengths and weaknesses of the two approaches, one can see that the empirical analysis of the gender division of labour has to integrate them. In our effort to locate the German case in a comparative framework, we will use the overall framework of policies, as well as other gender indicators (educational attainment, gender role attitudes, etc.) as hypothetically related to it. Additionally, we aim at a preliminary explanatory model to reveal the most powerful predictors of the gender division of labour on the micro-level.

13

14

Chapter 2 Policies affecting the gender division of labour In this chapter, we will focus on the social provisions of the West German welfare state (see for an exception the section on child day care, pp. 28ff.) which have been extended to East Germany since unification. These provisions currently have an impact on the gender division of labour. However, there are two reasons to also discuss the relevant measures of the former GDR. First, they might have an indirect effect on the recent situation insofar as they influenced the life courses, attitudes, resources of the population in East Germany. Second, a comparison of policies in West and East can help us understand the specific West German family policy profile. Therefore we conclude the two parts on paternalist and maternalist policies with a section on differences between East and West (see pp. 37ff.). Generally speaking, the West German welfare state as a system of objectives, institutions, policies and incentives favours a traditional division of labour between the sexes. The concept of the strong male wage-earner model (Lewis, 1992, Lewis & Ostner, 1994; Ostner, 1995a, 1995b) is a good starting point for a description and analysis of the way social policies influence the life chances and life courses of women. Role models for men and women, namely to be/become wage-earners and homemakers, respectively, are embedded in the social provisions of the West German welfare state and operate as normative preconditions for the gender division of labour. After the elimination of most types of legal discrimination against women,12 the bulk of policies affecting the gender division of labour are economic and ecological in nature, i.e., incentives/disincentives and provision/non-provision of social services. With respect to the addressees, the policies that help to enforce gender role models can be termed "paternalist" and "maternalist".13 Paternalist policies strengthen the economic position of men who provide for their wives and children (compared to single men and women in the labour force). They belong to the core of the typically German "social policy" (Sozialpolitik) from Bismarck’s time onward. The different forms of social insurance14 have been justified by the provider role of the male wage-earner that is to be supported, supplemented and substituted for by public policy. Therefore, they include family supplements in benefits (unemployment insurance), benefits without payments (for non-employed wives and minor children in the health insurance system, or parental years in the pension system) or derived benefits (widows’ pensions) (cf., e.g., Alber, 1986). The financial support of families that forms the core of the explicit family policy agenda (Gerlach, 1996; Münch, 1990) is also defined as a compensation for the family burdens (Familienlastenausgleich) of the chief wage-earners.

15

The counterparts to paternalist policies are maternalist policies. Maternalist policies have come to the attention of feminist historians in recent years (Joosten, 1990; Moeller, 1993). They consist only in part of maternity protection for health reasons (Halbach, Paland, Schwedes, & Wlotzke, 1994, pp. 274ff.).15 Mainly, they support the traditional division of labour in care work (for children and other dependents). In this respect, they are only in part an element of the explicit family policy agenda; partly they are consequences of "non-decisions". There are three elements of German family policy which should be mentioned: parental leave legislation, day care services for children and regulations for the care of frail elderly persons. As will be shown below, each of these political measures contributes to the traditional division of labour in several ways: by strengthening the social and/or financial position of (full-time) mothers (compared to childless women and women in the labour force), setting symbolic and practical preferences for certain care arrangements, re-defining opportunity costs for home-makers, creating transaction costs for pursuing alternative strategies, impeding change by dispersing jurisdiction to different arenas, avoiding incentives for change. To put it more simply: They work as an institutionalized arrangement to maintain the traditional division of care work. To sum up: It is a more or less implicit normative assumption in the traditional family model that women should be wives, and wives should be mothers, and mothers should not be employed,16 which means that, in a strict sense, no woman should be in the labour force, and unpaid work and caring activities are — more or less — solely women’s tasks. Although the German system does not constitute an ideal type, we will show that the German welfare state comes very near to being one. Table 1 gives an overview of the structure of the presentation of policies. Table 1: Policy Function 1. Paternalist Policies

Policies affecting the gender division of labour in the FRG Policy Area

Policies

Marriage benefits Taxation

Family benefits

Other marriage benefits Child benefit and tax credit Other family benefits

Lone mother benefits

Advance payment of child maintenance Assistance to unwed mothers Social assistance

16

Effect Opportunity costs of home-makers Strengthening of the earner’s role Opportunity costs of home-makers Strengthening the earner’s role Opportunity costs of home-makers Strengthening the earner’s role Opportunity costs of home-makers Strengthening the earner’s role Opportunity costs of lone motherhood Opportunity costs of lone motherhood Opportunity costs of lone motherhood Capacity to maintain an autonomous household

Policy Function 2. Maternalist Policies

Policy Area Maternity protection Parental leave

Policies Pregnancy protection Pregnancy leave Maternity leave Maternity-leave benefit Child-raising leave period Child-raising benefit Child-raising period in old-age insurance Sickness leave

Child day care Elderly care Work Schedule

Sources:

2.1

Provision of day care Long-term care insurance Working-time reduction Other work-related areas

Effect

Combining work and family Strengthening the care giver’s role Sequential compatibility Opportunity costs of home-makers Opportunity costs of home-makers Strengthening the care giver’s role Combining work and family Strengthening the care giver’s role Opportunity costs of home-makers Strengthening the care giver’s role Opportunity costs of home-makers Combining work and family Combining work and family

See below.

Paternalist policies

By definition, paternalist policies strengthen the position of the "pater familias". The core of these policies in Germany consists of benefits intended to relieve the financial burden on families of the direct and indirect costs of having children. As an indirect child cost, the reduction or giving up of employment for the sake of child-raising is targeted by these measures, too. Thus, family benefits indirectly work as a disincentive for female employment. Besides family benefits, two policy areas are considered in this section. Marriage benefits are an analogous measure aiming to support the traditional wage-earner/home-maker marriage by lowering its opportunity costs. Lone mother benefits are the counterpart of paternalist policies. Whereas "positive" paternalism institutionalizes women’s dependence on their husbands, lone mother benefits provide an opportunity for women to maintain an autonomous household despite the father’s absence from the home. Although the aims of the two policies seem to be directly opposed to each other, they share the same ideological background, as we explain below.

2.1.1 Marriage benefits There are two categories of marriage benefits. While taxation has effects on the whole population, some minor benefits, such as those in social insurance or supplements to wages and salaries in the public sector, are of a more restricted scope.

17

Taxation. The traditional one-earner marriage is favoured by the combined effects of three elements in the tax code. Joint taxation aggregates the incomes of spouses and taxes it together. Combined with the progressive tax rate, this leads to disadvantages for two-earner couples, especially for low- and middle-income families, because there is a large difference between the marginal tax rate for two separate incomes and that for a single income. The splitting system favours the traditional division of labour even more. The aggregated income of the spouses is divided by two. The tax rate valid for this halved income, in most cases resulting from the progressive tax rate, especially for middle and high-income families, is lower than when combined income is used to determine the total tax amount to be paid. There are two effects resulting from this regulation. The splitting advantage is greater the higher the aggregate income of the spouses and/or the greater the difference between the two incomes — and vice versa. The extreme case with the greatest advantage is — of course — the traditional marriage with one spouse, usually the wife, not employed in the labour market. In 1999, taxation started at incomes above 13,060 German Marks (DM) annually for singles, 26,130 for married couples, and 38,010 for married couples with two children (see Appendix-figure A1). The progressive rate started with a marginal tax rate of 23.9% and climbed to 51% for incomes over 120,000 / 240,000 DM (single/married). The curve of the tax rate 1999 (for singles) is given in Figure 1.1. With a yearly income of 60,800 DM17, a single person has a marginal tax rate of 35.6% and an overall tax rate of 24.1%, so that he/she has to pay 14,627 DM in income tax. With the same income, a married couple has a marginal tax rate of 29.4% and a tax rate of 15.7% (the tax rate can be calculated from the figure by dividing income by two and applying the tax rate of the halved income to the full one), reducing their income tax to less than two thirds (9,504 DM). From the viewpoint of economic rationality, this means that it is easier for married couples to reach a higher net income on a given gross income or a preferred net income on a lower gross income if they choose the traditional division of labour. Therefore, the division of labour can be partly accounted for by politically regulated incentives. To illustrate the point, Table 2 shows the effects for three income levels: the income at which taxation for a married couple with two children starts, the average household income of a salaried employee and that of a self-employed person (see note 17).

18

Table 2:

Effects of tax splitting for spouses and child tax credit

Category

Family status

Start of taxation

Single, no children Couple, no children Couple, 2 children

Gross income

Tax rate in % 38,010 19.2 38,010 9.6 38,010

0

Marginal tax rate in % 30.7 27.8 24.1

Net income

Net Relation gain absolute in % / in % 30,715 Base 34,349 11.83 3,661 38,010 23.75 100.74% 50,642 Base

Single, no children 67,900 25.4 37.0 Married couple, no Middle level 67,900 17.9 29.6 55,744 10.07 children income Married couple, two 4,022 67,900 15.0 29.6 59,766 18.02 children 78.83% Single, no children 172,800 39.8 52.8 104,058 Base Married couple, no Higher 172,800 28.6 42.6 123,472 18.66 children income level Married couple, two 5,772 172,800 27.4 40.7 129,244 24.20 children 29.73% Note: Income figures in DM, 1999 tax code, calculation does not consider individual deductions, start of taxation rounded off to 10 Marks. Net gain is the difference between the net income of the respective family status and the respective base net income (bold) as a percentage of the latter. E.g.: The last-reported net gain of 24.2% is (129,244-104,058)/104,058*100). Absolute Relation is the difference in DM between net income of a couple with two children and a childless couple. Relation in % (italics) is the difference between the net gain due to children and the net gain due to marriage (in the respective income category) as a percentage of the latter. Source: Own calculation of net income using www.lexware.de/cool/esttarif.asp.

There are two changes accompanying increasing income. First, the net gain in net income (as a percentage of the income of a single person without children) for a married couple increases with income. In other words: The wife of a high-income husband is more valuable in the tax code.18 The net gain for two children also increases with total income. But, second, the net gain due to children is smaller than that due to marriage19 and, moreover, the relation between the two diminishes with rising income. This means that under the tax code children are less valuable than wives and that — despite progressive taxation, which is intended to raise the net gain due to the child tax credit — the actual added benefit due to children, which rises less steeply than that due to marriage, is rather small. In other words: By reducing the marginal tax rate, splitting partly "consumes" the net gain due to The tax credit for children (see Münch, 1990, p. 83). To sum up: As the examples show, the German tax system has developed strong incentives for a traditional division of labour that, moreover, increase with rising income. Therefore, the middle and upper classes may be influenced in their labour force participation and division of household labour by the political creation of opportunity costs.

19

As comparative research on the married-female labour supply has shown, this combination of joint taxation, progressive income taxes and splitting works as a disincentive for female employment (see, e.g., Gustafsson & Bruyn-Hundt, 1991). Moreover, the proportion of this measure for all programs that support families is a good indicator of the priority assigned to upholding the traditional division of labour. In 1996, marriage taxation accounted for a 25.3% share (in absolute figures: 41.1 billion DM) of the public tax policies for marriage and families (Lampert, 1998, p. 349). Because many childless couples benefit from split taxation, it has frequently been criticized as contrary to the declared goals of German family policy (e.g., Münch, 1990, pp. 82f.; Pfaff & Kerschreiter, 1982). Despite this criticism and some minor changes, the measure has been upheld because of the strong ideological stance on the traditional marriage taken by the Christian-Democratic parties. There are many observers who claim that the splitting system is even protected by the German constitution, in which marriage and the family as institutions are placed under the guardianship of the state. There is one minor detail that is significant for the deeply-rooted traditionalism of the German tax code. If both spouses work, the general tax credit, which includes a subsistence minimum, can be divided between husband and wife. Legislators assumed that there would normally be a difference in income, so that — besides the 50/50 split of the tax credit — another split is possible on a 60/40 basis. Although this does not influence the taxation of aggregate income, it gives the monthly pay-check of the husband, who usually takes the 60% share, a higher net value. Other marriage benefits. There are many other benefits which, although they are of a minor nature and have direct or indirect effects, nevertheless taken as a whole contribute to substantially supporting the traditional division of labour in marriages. To name a few: ƒEvery employed person in Germany (except civil servants, self-employed people, and employed persons earning more than — currently — 6,450 DM) is enrolled in a mandatory health insurance program. Spouses (usually wives) who are not gainfully employed are covered by this insurance at no extra charge. Every civil servant is partly privately insured and receives a health subsidy. The spouses of civil servants are entitled to a health subsidy if they do not work outside the home or earn less than — currently — 30,000 DM a year. ƒFor civil service employment, there are wage supplements that take into account the family situation of the employee (i.e., wage earner, salaried employee or civil servant) which leads to a modest increase due to marriage. ƒAlthough the regulation of the financial consequences of divorce is rather complicated, the general principle of "continuing solidarity" results in financial support of the spouse if (and to the degree that) he or she has been engaged in housekeeping and/or child-raising.20

20

ƒIf a husband or wife dies, the surviving spouse receives a widows/widowers pension. In all these cases, the benefits result in a subsidy for the traditional marriage. Not only the current situation in a marriage, but also the expectations of future benefits from marriage reflect the advantages extended to the wage-earner/home-maker type of marriage.

2.1.2 Family benefits Child benefit and tax credit. Child benefits, introduced in 1955, have gone through several changes over the past 45 years. Child benefits were originally abolished by the Allied Occupation Forces after the defeat of the National Socialist regime in 1945, because they were regarded as an instrument of fascist population policy. From the beginning, there has been an ideological debate on the purpose of this measure. As natalist arguments were not permitted as a justification for the benefit, two competing views have determined the debate. In the conception of the political left, child benefits should contribute to reducing inequalities between social classes. Thus, they would have to decrease with increasing income (or be phased out for families in higher income brackets). In the conception of the political right, i.e., the Christian-Democratic parties, the measure should reduce the inequality between families and households without children. Consequently, it should increase with income, because needs are different in every class. Although the current regulation is a sort of compromise between the two, the basic principle of the latter conception prevailed in the form of a tax credit for children, from which the child benefit is only derived. Given a progressive income tax, a child is — literally — more valuable in a high-income family than in a low-income one. Originally, after the third child the benefit was paid by the employer. In 1961, a child benefit for the second child, financed by the Federal Government, was introduced. From 1964 on, the child benefit/tax credit was drawn from this budget. The Social-Democratic government suspended the tax credit in 1975, while increasing the child benefit and introducing it for the first child. The Christian-Democratic government restored the tax credit in 1983. The so-called "dual model", i.e., a combination of tax credit and benefit, once again came into effect. There have been two major driving forces behind the increases documented in Appendix-Table A2. The first has been politics. In every election year (or one year before or after), increases have taken effect — with the exception of 1969 — accounting for 43 of the 73 increases.21 The second has been the Constitutional Court. Especially the decisions of May 29, 1990 and of June 12, 1990, which stipulated that the minimum subsistence level of adults and children cannot be taxed, and of November 10, 1998, which stated that all families are entitled to a "child care tax credit" (from 2000 on) and a "child-raising tax credit" (from 2002 on), were of special importance in this respect. These combined influences have led

21

to a substantial, but — for many observers — inadequate increase in the child benefit/tax credit (see Appendix-Table A2). Since 1996, under the so-called "option model", most families receive a child benefit on a monthly basis, while high-income families can choose a tax credit as an option. In any case, the annual income-tax statement determines whether a family can claim the tax credit (and, thus, higher benefits). Due to changes in legislation on child benefits and tax credits, the range of families which can take advantage of the tax credit has changed in the last few years (Appendix-Table A3). While, due to the increase in the child benefit, the relative advantage of high-income families has been gradually diminishing in the past few years (from 39% in 1996/7 to 22% in 1998/9) and the number of entitled families (additionally through changes in the tax rate) has also been declining, the 2000 increase in the tax credit has made it more advantageous for more families. Although this measure is not intended to further a traditional division of labour and although it only partially fulfils its main function of relieving families (or their wage earners) of their financial burden, it has the side effect of helping families make ends meet without a second earner providing for the subsistence of the children. For a typical family with two children, the 540 DM child benefit is only slightly lower than the net income that could be earned in a part-time job (to which — in most cases — mothers’ employment is limited, given the low level of institutional child care provision) held by a low or moderately skilled woman. The picture is different for high-income families. Although they might not have financial problems and although there should be a substantially higher net income, relatively speaking, they can raise their allowance via tax credit up to 844 DM for two children. Whether families with different income levels are influenced by these incentives is a matter of dispute. Two other tax credits relevant in this respect have been generalized recently. In both cases, the special tax credits were formerly available only to single or unmarried parents, but have now become — through Constitutional Court rulings — an entitlement of all families. The rationales for the decision were that besides the subsistence minimum, certain costs of raising children have to be considered as non-taxable income and that families with married parents should not be discriminated against. The first is the "child care tax credit", already included in the tax credit for children (see Appendix-Table A2). It covers demand for child care, whether provided in the form of institutional child care, family care or parental home care. The second is the "child-raising tax credit", which relates to the costs of goods and services necessary for raising children (such as toys, music lessons, books, etc.). Legislation is pending within the coming year, because the Constitutional Court has ruled that it must be provided to all parents until January 1, 2002. The general tax credit, the two specific credits and their joint development display an important aspect of the financial support of families in Germany. Families are regarded as having specific financial needs that must be partly covered by state

22

provisions. The rationale for this political and legal strategy of "compensation for family burdens" (Familienlastenausgleich) is that the specific "contribution" of families to society is child-raising and creating the human capital that is necessary for the continued functioning of society. The official political guideline is that families should be supported in this task, especially by reimbursing costs incurred in child-raising. The policy strategy legitimated in this way has led to a reduction in the opportunity costs of women engaged in homemaking. Other family benefits. Other family benefits follow the scheme of the marriage benefits (see pp. 17ff.). Some measures partially unburden male wage-earners: ƒMinor children who are not gainfully employed are covered by the statutory insurance at no extra charge. Every civil servant receives a health subsidy for his or her children, if they are not fully insured or for medical services that are not or cannot be insured by statutory or voluntary insurance. The proportional coverage of this subsidy increases with the number of children. ƒThere are wage supplements for civil service employees with children. Other regulations partially compensate for the loss of one family wage-earner, such as alimony regulation under divorce law or orphans’ pensions. While these instruments provide aid to minor children, there are additional benefits that target the parents of older children. ƒThere is a tax credit for vocational training (Lampert, 1998, p. 346) paid to parents with minor children in training outside their parents’ household (1,800 DM per year), as well as for children older than 17 years living in or outside their parents’ household (2,400 DM / 4,200 DM, respectively). ƒPupils in upper secondary classes of grammar schools and university students are supported by a means-tested grant of up to 1,030 DM (East Germany: 1,020 DM). In 1998 about 45% of all university students received support under the Federal Training Assistance Act which was less than in 1993 (Statistisches Bundesamt, 2000g, p. 82); fewer students received this assistance in West Germany than in the East (26.8% compared to 54.8% in 1994). Although the grant is modest and has to be partly reimbursed, this support extends material relief for the wage-earner to families of older children, as do free schooling, the child benefit and the educational tax credit for children.

2.1.3 Lone mothers benefits Benefits for lone mothers represent the counterpart of breadwinner policies. The ability to found and maintain an independent household (Orloff, 1993) and the social position of lone mothers (Hobson, 1994) are prominent indicators in feminist comparative welfare state research. As is to be expected, policies providing guarantees for the financial situation and social well-being of lone mothers are scarce and stigmatized.

23

In 1996, 4% of all family households (including married couples without children) in Germany were headed by lone parents; the rate in East Germany was higher (8%), than in West Germany (3%). The majority of lone parents (86%) were female (Engstler, 1998, pp. 43ff.). In the former GDR, early family formation and high divorce rates led to a higher incidence of lone parents. Given the economic independence of most mothers in the former GDR, their financial situation was better compared to their West German counterparts. Moreover, stigmatization of lone motherhood was not as prevalent as in West Germany. Advance payment of child maintenance. This measure is intended to secure the financial support of children for whom the maintenance payment by the non-cohabiting parent, i.e., in most cases: the father, is not being paid (cf. Bundesministerium für Arbeit und Sozialordnung, 1998, p. 616; Gerlach, 1996, pp. 208-210). Since 1980 children under 12 in this situation are entitled to advance payment for up to three years, since 1994 up to 6 years; that amounts to 189 DM / 257 DM (East) and 220 DM / 296 DM (West) for children under six and between six and twelve, respectively. This means, that the amount is only as high as the child benefit or even lower. Assistance to unwed mothers. In order to help pregnant women in special emergency and conflict situations that tend to lead them to consider having an abortion, a Federal foundation, Mother and Child for the Prevention of Unborn Children was established (1984) and subsequently extended to East Germany (1994) (Gerlach, 1996, p. 209). Additionally, most federal states established their own foundations with the same purpose. However, there is no entitlement to this support and it is, in most cases, limited to a lump sum payment of several hundred DM. The majority of affected women are unwed. It has been criticized that the decision-making of the foundations is arbitrary and degrades women to the status of paupers. Social assistance. Many lone mothers are unable to work or to be employed sufficiently to reach the subsistence level because of having to care for their child/children without adequate support, especially by child care facilities. Therefore, a high percentage of lone mothers live below the poverty line. Besides support by their kin, for most of these mothers social assistance is the last resort. Social assistance is a means-tested benefit that is to guarantee the socio-cultural minimum of subsistence. In practice, maintenance obligations take priority over social assistance. Unwed mothers have no legal claim against the father; only their children have. Because of the life-long maintenance obligation between first-grade descendants, the parents of the mother have responsibility, even over against their adult daughter. In practice, the parents of the recipient of social assistance have to disclose their financial situation and to reimburse part of or the whole amount of social assistance, depending on their solvency. Although poor mothers are granted

24

social assistance and, thus, are able to maintain a minimum standard of subsistence, this policy humiliates and stigmatizes the recipients, as they are torn between dependence on their parents or the state, which acts as a surrogate father. In this way, social assistance reinforces the breadwinner model by not providing a decent alternative for lone mothers. Social assistance is also a good indicator for the risk of poverty and the outcomes of the paternalist benefits described in Section 2.1. As Appendix-Table A4 shows, being a lone mother poses a clear risk of falling below the poverty line as defined by the Social Assistance Act. Households with children have a higher poverty risk than households without children. The more children there are in a household, the higher the poverty risk. Lone mothers face a much higher risk of becoming poor. And there is a higher incidence of poverty in West Germany than in the East.22 All these trends point in the same direction if we look at the duration of support. These patterns combined, those with the highest risk of being poor are West German lone mothers with three or more children, of whom almost one half receive social assistance for almost three years without interruption on average. In terms of opportunity costs, this means that it is economically rational for a German woman to choose marriage, the (in principle) life-long financial support of her husband, which is subsidized by the state and — as a side effect — the corresponding inequitable division of labour.

2.1.4 Conclusions The official and often cited rationale for the policies described in this section under the heading "paternalist" is to secure financial support and to grant equity to families vis-à-vis persons and households without children. Although some progress has been made, esp. in the last decade, in furthering this process, the data presented above (Appendix-Table A4) give a sobering picture of the outcomes. Despite all efforts to subsidize families according to their needs, policies of financial support have not succeeded in even altering poverty rates. Again, this is — despite the best intentions associated with these policies — a good indicator that their main effect does not lie in the realm of subsistence, but in the contribution to a traditional division of labour.

2.2

Maternalist policies

Because there is no longer a clear-cut role ascription to women, wives and mothers in the law (see note 12), the term "maternalist policies" needs some consideration. It has a historical meaning referring to times in which discrimination against women was openly expressed in laws and policies. Today — and this our rationale for using it — it refers to the fact that the caregiver role which is the object of these policies is still mainly performed by women. We discuss six areas: maternity protection, parental leave, work schedule, child care and elderly care. The tax credit for paid household help, a very ambiguous measure that is difficult to classify, is considered in a concluding section.

25

2.2.1 Maternity protection The rationale of the Maternity Protection Act "is to protect a mother in paid employment and her unborn child from dangers, excessive demands, and damage to health, as well as from financial set-backs and from loss of her job" (Halbach et al., 1994, p. 231; for the following ibid., pp. 231ff.). For the protection of health before and after birth, there are regulations on the design of the work place and on suitable forms of work, a ban on additional hours, night work and Sunday work, a ban on specific forms of work (heavy physical work), an absolute ban on work six weeks prior to delivery (pregnancy leave) and 8 (to 12, in case of premature or multiple births) weeks after (maternity leave, see also Gerlach 1996, p. 205; Lampert, 1998, pp. 90f.). Wage protection is accomplished by a maternity wage for mothers who have to give up work because of a specific ban, a maternity leave benefit provided by social health insurance plus a supplement from the employer for the absolute ban on work before and after birth, which amount to the net income of the expectant mother. Finally, there is employment sheltering qua protection against dismissal during pregnancy and up to the end of four months after birth. Although these measures are aimed at mothers in paid employment, they are a safeguard for the caregiver role. The intention is to enable working women who are expecting a child to fulfil their duties as mothers. It has often been suspected that these regulations, along with others (parental leave, etc.), contribute to discrimination against women in the labour market, despite or even because of their anti-discriminatory character. Because employers may foresee possible disadvantages of hiring women, they might not give them an equal chance of being hired. Although there is no unequivocal evidence, the possibility that these measures contribute to the traditional division of labour cannot be dismissed. Concerning the discriminatory nature of the legislation intended to protect women, it should, however, be noted that most bans on employment that were in effect in former times have been (and had to be) reduced to insignificance following the tendency to enact equal opportunity in the law (see note 12).

2.2.2 Parental leave In this section, parental leave in statutory law is discussed. Other forms of leave (by collective agreements, in specific branches, etc.) will be summarized in the section on work schedule policies. Child-raising leave period. The core of the parental leave legislation is the Act on the Payment of Child-raising benefits and Child-raising leave of 1986. It established three complementary measures: child-raising leave, child-raising benefit and child-raising period in the pension insurance. The rationale for this legislation is to give an incentive for parents to provide care for their small children. Every parent of a child under 3 who is in paid employment is entitled to child-raising leave. This period has been gradually extended from an initial (1986) ten months to three years; the primary caregiver who takes the leave can be

26

changed three times. During this period, the employee cannot be given notice. The contract of employment is merely in suspension, and the employee has the right to return to a job comparable to the one held before taking the leave. Additionally, employees in the statutory health insurance program (about 90% of all employees) and their children are freed from the obligation to pay contributions during in the leave period. Employees on leave may work up to 19 hours per week. While this provision is gender-neutral, only about two per cent of all parents who take the leave are fathers (Vaskovics & Rost, 1999). Mainly, the reasons given for this lie in the financial situation, because most husbands earn more money than their wives, in gender role orientations and the fear of discrimination in the workplace. Moreover, parental leave serves as a way of selectively excluding women from the labour force. In the first years after the law took effect, about half of the mothers taking the leave did not return to their job (Landenberger, 1991). With the extension of the leave period to up to three years and the legal right for all children between 3 and 6 to a place in a day care facility (since 1996), parents should not face problems in finding a day care facility after the leave period. Actually, a co-ordination problem remains, first because of lack of places in certain regions or at the exact date when the parent wants to resume employment, second because of the regular day care facility schedule, which merely allows children to attend in the morning. Even part-time employment may be difficult to maintain under these circumstances, as the transportation time involved in bringing children to day care and picking them up again and the trip to and from the workplace have to be added. The gender inequality in leave taking is not only a reinforcement of the gender difference in paid work, but also contributes to the deepening inequality in the share of unpaid work. As has been shown in a panel survey of young couples, the traditionalization of the division of labour that starts with the formation of the family (birth of the first child) continues in later phases even if the wife eventually resumes employment (Rost & Schneider, 1995). The new Red-Green governing coalition at the national level has recently enacted a revision of the Child-raising Benefit Act that applies to parents of children born after December 31, 2000. These parents can take the renamed "parents’ time" while working up to 30 hours per week (previously: 19 hours). They have a right to claim part-time employment, and they can change the primary caregiver five times (before: three times). The third year of leave can be taken up to or before the child’s eighth birthday. Part of the amendment’s intention is to create incentives for fathers to take the leave. Child-raising benefit. A mother or father who is the primary care-giver and who is not employed or works no more than 19 hours per week is entitled to a means-tested flat-rate benefit until the child is 24 months of age. Again, the rationale is the furthering of primary care-giving for young children. This regulation is part of the (Federal) Child-raising benefit act and in some federal

27

states is supplemented by a federal state Child-raising benefit for the period after the child’s second birthday. Although the maximum benefit is modest (up to 600 DM), it can mean a considerable amount of money to a family in which one partner, usually the wife, has temporarily given up employment. However, the income limits after the first six months are very low; many recipients lose the benefit or have to face a reduction. Child-raising period in old-age insurance. Along with the Child-raising benefit act of 1986, a child-raising period was introduced. The primary carer of an infant receives three child-raising years in old-age insurance (Lampert, 1998, p. 342). In the pension, a supplement is taken into account valued as though the person had paid contributions for a 75%, since 1992 a full average income during the three years. The benefit has been extended to parents born after December 31, 1920 (see for the history of the measure Münch, 1990, pp. 90ff.). In practice, the mother automatically collects this benefit, unless the parents declare that the father shall be the recipient. Sickness leave. There is an entitlement to sickness leave for every employee in statutory insurance up to 10 days per year (20 days for a single parent) for the care of a sick child (no older than eleven years). Statutory insurance pays the health insurance benefit (Lampert, 1998, pp. 237f.).

2.2.3 Child day care Child day care is of central importance for the division of labour between the sexes. While labour-supply theory focuses on the costs of child care as an important predictor of women’s labour force participation, the West German child care regime (cf. Gustafsson, 1994) provides its major disincentive for female employment in the scarce supply of state-provided day care facilities (Kreyenfeld & Hank, 1999). In accordance with the above-mentioned basic features of the West German approach to family and gender, two principles govern the field of child care in the FRG. First, as with parental leave, the combined ideas of family-centred care for infants up to three and of the sequential compatibility of motherhood and employment lead to the ascription of child care tasks to the mother, depending on the age of the child. Consequently, the official political philosophy, sedimented in the supply structure, sees day care for infants up to three mostly as an emergency measure or very rare exception, for pre-school children between three and six as supplementary and for school children as mostly superfluous. Second, the so-called "subsidiarity principle" ascribes the task of providing day care for children to the local municipalities and non-profit organizations (NPO), most of them affiliated with one of the two major churches. For several reasons not to be discussed here (cf. Kreyenfeld & Hank, 1999, pp. 9f.), this leads to a "supply-side" bias, which means that not the (local) demand by parents but the political philosophy and the

28

readiness of the local governments and the NPOs determine how much child day care is provided. The demand for child care is determined by two sets of influences. As women are the persons to whom the task of child-raising is primarily ascribed, their preferences on time use are of foremost importance, esp. participation in education and the labour force. In the following chapters it is shown that the younger cohorts currently have intensified their participation rates in both respects. Therefore, more day care is needed for the growing number of women engaged in training, employment and careers. The second set of influences are subsumed under the heading of fertility. The number of children born in a society shapes the need for day care. As Appendix-Table A17 shows, fertility has been governed by different trends. First, there is a secular trend of fertility decline that was increased by a further drop since the mid-1960s (only partly shown in the table). Second, the fertility indicators for East Germany were higher than those of West Germany throughout the 1970s and 1980s. Although this may be a contested assertion, one can attribute this difference to the development and implementation of pro-natalist family policy measures in the former GDR (see page 37ff.). Third: While fertility indicators in the West stayed at a low level since the end of the 1980s, they dropped markedly in East Germany after unification and have been recently climbing again without reaching the old level. These developments can be interpreted as rational behaviour under changing circumstances. As the opportunity costs of homemaking had risen (the value of employment and the risk of divorce have been rising), it has become sensible for West German women to restrict the number of births, since they regularly lead to the discontinuation of employment. Given the socialist family policy of the GDR, the opportunity costs moved in the opposite direction. With the end of this regime, subsequent insecurity caused a birth decline. This also means that there is an inherent relationship between the birth-rate and day care, because the supply of day care influences opportunity costs. Besides these systematic circular relations of employment, day care and birth-rates, it is also noteworthy that the overall demand for child care not only depends on the percentages of women who work and/or give birth, but also on the number of women who are able to give birth. The rising number of births in West Germany, which have drawn much public attention in recent years, are related to the large number of baby-boomers (born between 1955 and 1965), of whom many postponed family-formation and are only now having their first child or subsequent children. While aware of these intricate relationships, we rely in the following paragraphs more on a crude ratio of use of day care or places in facilities, on the one hand, and the number of children in different age-groups, on the other. In this section we will, first, give an overview of recent developments in West and East Germany, second, describe the development of the supply of day care with respect to the three agegroups and conclude with some considerations on the gender division of labour.

29

Unification and continuing divergence. Unification had a direct effect on and an indirect outcome for the supply of day care. With the transformation following unification, the abundant supply of day care facilities in the former GDR had to be reduced. In the former GDR, most day care facilities were directly owned by the state or — to a lesser degree (16%) — by state-owned enterprises. Because child day care is assigned to towns and municipalities in the federal system of the FRG, day care became a huge financial problem at the local level. The closing down of many inefficient firms contributed to this problem. Especially the take-over of facilities for school children, which are near to non-existent in West Germany, resulted in great pressure for closing them (Wagner, Hank, & Tillmann, 1995, p. 3). As Appendix-Table A8 indicates, this did not happen immediately and was not as widespread as expected. In this table, percentages of children in different agegroups are given on the basis of the actual use of day care.23 Prior to unification (1990), East and West Germany differed markedly in day care use for infants (East: 62%, West: 6%) and school children (East: 35%, West: 2%), both in the overall level as well as in the use of full-time care (52% to 2%, 35% to 2% respectively). The difference for pre-school children was smaller for the overall level (East: 98%, West: 82%), but large for full-time use (80% to 21%). This reflects differences in the two child care regimes. While West Germany relied on family care and subsidiarity, East Germany expanded day care from the 1970s on under the auspices of pro-natalism and the socialist ideal of the life-long, full-time employed woman. In the first years after unification, there was a slight decline in overall day care use for infants in East Germany. Followed by a sharp decline in 1994, it remained around one third of all infants. The figures for full-time day care use steadily dropped and stayed around 15%. Nevertheless, East Germany has higher levels of day care use for infants than West Germany (8% overall, 1% full-time, 1998). The decline in overall use of day care for pre-school children was only modest in East Germany and approached the West German level of about 90%. Despite a steady decline in full-time day care in this age-group, the level is still higher in East Germany (51% in 1998) than in West Germany (20%). For school children, a similar pattern can be discerned: a very modest decline in use (overall and full-time use are by definition the same) and a marked difference between East and West (30% to 5% in 1998). With regard to the direct effect of unification on the decline of East German day care facilities, one can conclude that it has not been as dramatic as expected by many observers. Efforts to maintain a day care system which is seen by many East German women as an achievement of the former GDR that should be preserved and the sharp decline in birth rates (see Appendix-Table A17) after unification have contributed to this result. However, it is surprising how marked the differences are between East and West, esp. in day care for infants and school children. It can be speculated that the high level of day care outside the home has contributed to a

30

persistent lead in women’s labour force participation in East Germany compared with West Germany. The second aspect of the development of day care mainly refers to that of West Germany. Since the Unification Treaty left the East German abortion law in effect and determined to reformulate a new one for unified Germany, the regulations accompanying this amendment are an indirect outcome of unification. Of special importance is a provision of this law which created an entitlement to day care places for all pre-school children between three years and schooling from 1996 on. The rationale for this measure was to give political support to families in an area that has been neglected for a long time. Although, the measure has been watered down in implementation (Künzler, 1998, pp. 116f.), it has contributed to an increased use of day care for pre-school children from about 80% to about 90% (Appendix-Table A8). Day care for different age-groups. In this section, we discuss the day care supply for different age-groups with respect to four main criteria: availability, costs, time schedule and quality. It is our understanding that these dimensions taken together are decisive in determining whether the supply of child care can contribute to a more equal division of labour. Because of convenience and comparability, in this section we use coverage level as the indicator for availability (for methodological issues see note 23). Traditionally, day care facilities for infants up to three (the German word for these facilities: Krippen) are less developed in West Germany than day care for pre-school children (Appendix-Table A9 vs. Appendix-Table A12). The coverage level is so low that it could be considered non-existent. Prior to unification, there was a difference in scale between West and East Germany with coverage levels of 2 % and 80 %, respectively. Before unification, there were also vast regional differences in West Germany, with the city-states providing a large number of places for this age-group (Appendix-Table A10). Coverage levels in 1990 ranged from under 1% in Rhineland-Palatinate to 18% in Berlin (West), with the mean at nearly 2%. Regional differences persisted after unification. Although some West German federal states increased their supply of day care for infants, there was only a small improvement to a mean of slightly less than 3% in 1998. After unification, the high coverage level in East Germany was reduced from 54% in 1990 to 36% in 1998, but a big divergence from the West remained. Day care in facilities is a rare exception in West Germany, costs are high and quality standards are neither established nor enforced (Appendix-Table A11). The alternative is family day care by child minders outside the home who are usually called "day-mothers" (Tagesmütter). Although the children and youth welfare agencies at the local level have to help in finding family day care and there is an entitlement for some parents (e.g., lone parents who have to work) to the receipt of this form of day care, it is not widely used. Moreover, it is a very unstable form of day care (Tietze & Roßbach, 1993, pp. 147ff.).

31

In the 1950s and the 1960s day care facilities for pre-school children between three and six (Kindergarten) were viewed in West Germany as disapprovingly as the Krippen for infants, but both public perception and political attention have since changed. Although conservative groups expressed reservations in the take-off phase, they were expanded in West Germany, esp. in the 1970s and 1990s up to a level of 69% in 1990 (Appendix-Table A12). This growth began earlier in East Germany, but also peaked in the 1970s. However, in the former GDR this was accomplished under the auspices of the pro-natalist family policy that was started then. Shortly before unification the level was 95% in 1989. The introduction of an entitlement for all children between three and six to a place in a day care facility (see p. 31) from 1996 on should have raised the coverage level to 100%. But as Appendix-Table A12 indicates, the coverage level of 87 % in West Germany is below that goal, whereas in East Germany the coverage level surged to an all time high of 112 %, due to a slowed decrease in day care places and an even bigger decrease in demand by the plunging birth rate. Moreover, there are hints that the expansion necessary under the new regulation has been achieved at the expense of places for full-time day care (Künzler, 1998, p. 118) and of day care for infants (Bundesministerium für Familie, Senioren, Frauen und Jugend, 1998, p. 200; Tietze, 1998, pp. 44f.). As Appendix-Table A13 indicates, the regional differences in coverage levels of day care for pre-school children between the Western federal states in recent times have not been so marked as for the infants up to three. Nevertheless, there are some differences and some more within the federal states (not shown in the table). Whereas previously the problem for mothers who sought day care for their pre-school children was the uneven regional distribution of coverage levels, with mothers in certain federal states and in towns (vs. in the country) having comparatively better chances to find a place, the three problems now lie in the "fine print" of an allegedly extensive day care system (see Appendix-Table A15). As many mothers use the child-raising leave, the first problem arises when the child turns three. Regularly, children at the age of three are admitted to Kindergarten only at the beginning of the following school year. A mother who wants to keep the right to return to her former employment needs day care from the third birthday on, because her child-raising leave ends at that time. The second problem arises after this hurdle has eventually been taken. Unlike most of the East German Kindergarten, which still offer full-time day care (Appendix-Table A14; for variations between the federal states, see Künzler, 1998, Table 7.5), the typical West German counterpart is open only in the morning with an optional supplement in the afternoon. In 1994, only 16% of all places in West German Kindergartens were open full-time or half-days plus lunchtime, with a range from 3% in Baden-Württemberg to 59% in West-Berlin (Künzler, 1998, Table 7.5). Although efforts have been made to increase opening hours so that at least a half-day job is possible, facilities that allow for this may not be available in every neighbourhood or region. The third problem relates to all the times in which the day care demand

32

cannot be covered by the Kindergarten. Whereas when a child is ill parents can take a limited sickness leave, there is no flexible solution if the Kindergarten is closed, which is relevant during (parts of) the school holidays and certain additional closing days. Day care facilities for school children (Horte) in West Germany are rare and have not been expanded to a noticeable degree (Tietze, 1998, pp. 46ff.; see Appendix-Table A16). West Germany ranks in the lowest position in one international comparison that is available (Appendix-Table A1, indicator 7.2). Horte have to be provided by the schools under the jurisdiction of the federal states. The primary schools in which the provision of day care is most relevant are run by local municipalities. Coverage levels vary from 3% in Rhineland-Palatinate to 35% in West-Berlin (1998). As with Krippen, Horte are seen as an emergency measure, mainly for children of two-earner couples forced to work full-time, single parents, and families with social problems. In the former GDR, Horte were regularly provided by the state. Prior to unification, there were places for every child for whom day care was sought. Official GDR statistics stated a coverage level of 60% in 1989, declining from 92% in first grade to 54% in fourth (Bundesministerium für Familie, Frauen, Senioren und Jugend, 1998, pp. 216ff.), and of 49% in 1990 (see Appendix-Table A16). Although the number of places was subsequently reduced, the level in 1994 was still considerably higher in the East than in the West (48 % to 6 %). As in the West, Horte are nowadays provided by local schools. For a long time, the transition of a child from Kindergarten to primary school posed a serious obstacle for the mother’s employment (and every serious attempt to plan family life) in West Germany. Primary schools followed an erratic time schedule that bordered on absurdity. As a surrogate for the regular day care of school children, there are efforts in most federal states to establish a so-called "reliable" primary school. E.g., in North Rhine-Westfalia 80% of all primary schools in 1997 secured either lessons or care within the time slot from 8 a.m. to 1 p.m., with other federal states ranging from 3% (Saarland) to 50% (Hamburg) (Bundesministerium für Familie, Senioren, Frauen und Jugend, 1998, p. 217). Because this trend is still developing, no exact figures or comprehensive comparisons are possible (further information available from the authors). Day care and women’s labour force participation. The supply of day care is a crucial factor in women’s choices. Traditionally having the prime responsibility for household, family and the up-bringing of children in both German societies, women could choose employment under the East German welfare regime to a degree that has been impossible for West German women. Without jumping to conclusions, we might assume that the supply of day care at least facilitates mothers’ labour force participation. The decisive difference between the former GDR and the former FRG (before unification) and, to a certain extent, between East and West Germany is how secure

33

this supply is or how much a woman can rely on its availability. In the centralized system of the former GDR, day care places were guaranteed. Even if a mother could not get a needed place, there were additional provisions for her support (see Appendix-Table A5). Although the institutional basis was transformed, the provision of day care has remained at a high level, thus being, at least, more secure than in the West. There, every transition by a child from one age-group to another results in several imponderables. Overall supply, time schedule, support in critical situations (e.g., a child’s illness) are neither generous nor generally foreseeable nor flexibly adapted to the needs of mothers. There are several consequences for the typical life-courses of West German women that are discussed in the relevant literature (Cornelius & Vogel, 1994; Kirner & Schulz, 1992; Klein & Lauterbach, 1994; Lauterbach, Huinink, & Becker, 1994; Mayer, Allmendinger, & Huinink, 1991; Merkle, 1994; Notz, 1994; Nyssen, 1990). West German women tend, on the one hand, to have a more fragmented employment biography. This is partly a direct consequence of policies, as in the case with parental leave legislation (see also Landenberger, 1991), partly an indirect outcome of the scarce child care supply. On the other hand, German mothers, esp. career-oriented ones, have to organize day care for their children, which supplements or substitutes for the lack of state-provided facilities. Grandmothers, social networks, "day-mothers" and many fragile grey-market arrangements belong in this category. The whole employment/day care dilemma creates career-impediments for women and contributes to their opting out of family-formation, i.e., the growing number of women who are (or choose to be) childless (see Appendix-Table A17).

2.2.4 Elderly care An often neglected dimension of gender inequalities in the division of unpaid labour is the disparity in care for the frail elderly and ill family members, which is mostly given by women. In Germany, the newly introduced statutory long-term care insurance has since 1995 paid either care allowances for persons who provide care at home or subsidies for care (either at home or in institutions). The amount paid depends on the need for care that is determined by the medical experts of the insurance companies. The respective law declares a preference for home care, thus contributing to (West) Germany’s long tradition of supporting the caregiver role. Currently, long-term care pertains to three to four per cent of all German households (Appendix-Table A18). Whereas more than one million people receive provisions under long-term care insurance, less than half a million persons are in care institutions (ibid.). Although the two data are not directly comparable (recipients may be at home or in an institution, persons in care institutions may or may not receive a care allowance), the difference gives an impression of how much care is provided at home. Most of the recipients of long-term care insurance are elderly persons subject to different degrees of health-related restrictions.

34

The providers of long-term care are mostly women. As a study of the outcomes of long-term care insurance (Bundesministerium für Gesundheit, 2000) indicates, 80% of the main caregivers are women (West: 81%, East: 74%). The major groups are daughters (23%), wives (20%), husbands (12%) and mothers (11%). Caring is mostly done by persons in the 40 to 64 age-group (53%). Given the fact that demographic developments will increase the number of elderly people in Germany ("aging of the population"), and that the risk of becoming care-dependent grows with age, this trend will contribute to increasing the traditional division of labour.

2.2.5 Work schedule policies Whereas in the policy areas described above, there is at least a general legal framework on the federal level, the field of the work related policies is scattered on different levels and in diverse sectors. Moreover, several of these provisions have evolved in recent times. Therefore, they are often not consolidated, not wholly implemented and even less studied. Nevertheless, we will sketch some of these measures in order to give an impression on their aims, scope, and possible effects. Working time reduction. Besides a general reduction of working hours as a precondition for enabling a higher share of housework and child care (see Künzler, 1998, pp. 123ff.), there are different approaches to the reduction of working times. First, a higher percentage of jobs can be set up as part-time contract. This strategy will most likely lead to a furthering of gender inequalities. Mostly, part-time jobs are held by women (see pp. 54f). These jobs tend to be less qualified occupations and offer less social security. As of now, there are less than a few signs that part-time employment can be extended to male employees. Second, there is now an entitlement to temporary part-time employment if the employee cares for small children. The provision, which is part of the newly enacted Child-raising Leave Benefit Act (see pp. 26ff.), applies only to parents whose children are born after December 31, 2000.24 Therefore, no information on the effects and outcomes can be given at the moment. However, it can be speculated that this measure as similar provisions will be used mostly by women and in occupations that rank low in the hierarchy of jobs. Third, collective agreements of the social partners can address the issue of reconciliation of work and family. The most important collective agreement relevant in this respect is the contract that affects the salaried employees in civil service (Bundesangestelltentarifvertrag, see for others Künzler, 1998, p. 127). Similar provisions are valid for the wage earners and civil service in the public sector. Since May 1, 1994, there is an entitlement to part-time work for family reasons in the public sector. Employees with one minor child or a relative who needs permanent care can reduce their working time on demand, if there are no reasons on the side of the employer.

35

Fourth, there are many agreements on the level of the individual firm or corporation. Although some stock-taking has already been done (see Künzler, 1998, pp. 128ff.), it is difficult to assess intentions, scope and possible outcomes. At least some areas of these contracts can be identified: additional parental leave, trainings/stints in the leave period and return guarantees, part-time work, additional leave for elderly care, flexitime (see also Schwartz, Schwarz, & Vogel, 1991). Again, the main addressees of these opportunities are women. The flexibilization of working time is a particularly ambiguous area. Initially, working shifts were the main "flexible" arrangements in industry in order to use machines more extensively. In the service sector, there may be even a greater need of "flexible" arrangements because of the fluctuation of demand for the respective service. Given the problem of the unavoidable absence of an employee, the job-sharing may be a company-friendly flexitime. It cannot be discerned whether flexitime (or work at home) offered by firms and employers only serves their interests or whether it is at least compatible or — even more — supportive for the fulfilling of family duties. Again, if the latter is the case, it might not be helpful in modernizing the gender division of labour. Other areas of work-place related provisions. Other areas comprise of the provision of child care facilities by companies (cf. for the intricacies Künzler, 1998, pp. 130ff.), attempts to evaluate the "family-friendlyness" of companies or the creation of special awards for companies that engage in the reconciliation of work and family (ibid., pp. 133ff.).

2.2.6 Conclusions As women are the primary persons to whom the tasks of housework and child care are ascribed, maternalist policies are of particular importance for assessing the impact the welfare state has on the gender division of labour. The West German approach is characterized by a combination of incentives for the discontinuing of employment (e.g., child-raising leave) and disincentives for employment (e.g., the scarce supply of day care), both addressed to mothers and other care persons. The former GDR displayed the very opposite policy profile. Of special importance are the arenas in which the maternalist policies are decided upon. Whereas the centralized system of the GDR succeeded in developing a comprehensive policy framework, the jurisdictions in the FRG are scattered on different levels (federal government, federal states, local governments, NPOs, companies). Not only the priority given to a traditional mother ideal, but also this feature of the West German polity contributes to the political impact on the division of labour. As a consequence, women in East Germany enjoyed a higher economic independence than in the West. This does not mean that they had greater autonomy. Both regimes framed the female life-courses according their ideals, restricted opportunities and choices. But, the East German women were paradoxically

36

granted at least something that is seen as essential in the capitalist world: economic self-sufficiency.

2.3. East-West-differences and German unification 2.3.1 GDR family policy Within the scope of this report, it is impossible to provide a complete assessment of the GDR policies related to family and gender. However, we attempt to sketch the main differences in the family policy profile (see Appendix-Table A5), that may account for East/West differences in the division of labour. The following features of the situation of family and gender in the GDR were either stated goals or outcomes of the socialist policies. ƒHigh marriage rates: In the 1970s and the 1980s, the marriage rates in the GDR were constantly higher than in the FRG (cf. Engstler, 1998, p. 83). Of special importance is the fact that first marriages were contracted at a younger age (ibid.). Incentives for early marriage were set by the family formation credit for spouses younger than 30, and the preference married couples were given in the allocation of apartments as well by the support married students and young persons in training received. ƒHigh birth rates: As shown in Appendix-Table A17, the birth rates in the 1970s and 1960s were constantly higher than in the FRG (see also pp. 21f.). The explicitly pro-natalist policies of the socialist regime contributed to this result. First, the birth subsidy (Appendix-Table A6) was a highly visible, generous and big sum of money, which provided an incentive just-in-time. Second, incentives which progressed with the number of children were set, e.g., the child benefit (see Appendix-Table A7) and the health insurance benefit which was also the basis for the different forms of parental leave benefits as well as the deduction of the repayment of the family formation credit depending on the number of children. ƒHigh labour force participation of women: As part of the attempts to reach equality between the sexes, the employment of women and esp. of mothers was supported. The official ideal of the East German woman was the life-long, full-time employed mother (see also pp. 49ff.). Among others, the following measures are relevant factors: extensive provision of day care, generous forms of parental leave and parental leave benefits and working-time reductions. ƒHigh rates of out-of-wedlock births: As indicated in Appendix-Table A17, out-of-wedlock-births became almost a regular phenomenon in East Germany over the years. Women in the former GDR were economically more independent than in the West. Especially, the higher rate of full-time employment, which was also feasible for lone mothers, improved the self-sufficiency of the GDR women. Moreover, the special support for lone

37

mothers was systematically built in every standard measure of the East German family policy. ƒLast but not least: The main tenet of the GDR family policy was the necessity to develop the "socialist personality". The meticulous control of the education system from an early age on and the methodical replacement of the ("private") family by collective forms of child-raising are obvious signs of this creed. As described above (pp. 8ff.), the GDR was a paradigm of — depending on the theoretical perspective — the communist modification of social democratic welfare regime (cf. Esping-Andersen) or the weak male wage-earner model (cf. Lewis and Ostner). With the FRG being a paradigm of a conservative regime or the strong wage-earner model, each German state could be seen as the very antithesis to the other. However, there is at least one point in which East Germany reached a goal which is desperately sought after in the other part. In West Germany, there is a consensus that family policy should unburden the financial worries of having children. That is exactly what the socialist policies succeeded in doing. It was estimated that financial transfers in the former GDR reimbursed two-thirds of the costs of raising children (Ott et al., 1990). There is a simple reason for that. Unlike the conservative approach of the West German welfare state, the higher employment rate of East German women first and foremost unburdened the family policy of this state from a liability, namely to provide for a huge amount of home-making wives. Thus, the GDR policies could successfully target the costs of children by securing a high level of economic independence for women.

2.3.2 Unification and gender relations — what is the outcome? In the following chapter, we will present data on the question how gender relations and gender inequality in both East and West Germany have changed through unification. Mainly this process is taking place in East Germany which adopted the political, economic, social and legal framework of the West. Although it might be impossible to distinguish the different factors involved in this transformation, the German unification can be seen as a huge experiment on the societal level. Every policy measure of the former GDR was dismantled by introduction of the West German family policy via Unification Treaty of 1990. Pro-natalist birth incentives were abolished or reduced in the system of West German family benefits. Marriage benefits were transformed from a means to further early weddings to a measure for supporting the wage-earner/home-maker marriage. The unification brought tremendous changes to the system of child day care in East Germany; it became less generous and less secure, although the provision level is still high (see pp. 28). And, last but not least, the economic position of women dramatically changed. They experienced high rates of unemployment and came in contact with the inferior place the West German labour-market offers women (see also pp. 49ff.). Therefore, the German unification could be seen as a litmus test for the regulation approach (pp. 70f.) which targets

38

the role of policies in the division of labour: Has the less traditional division of labour in East Germany prevailed — based on modern gender role attitudes and the life-course experiences of East German women and men? Or: Has the complete change of the political framework succeeded in re-traditionalizing the division of labour — by force of the incentives and regulations of the conservative welfare regime? As will be clear later (see pp. 41ff. and 81ff.), we tend to the latter.

39

40

Chapter 3 Structure and development of gender inequality 3.1

Gender ideology

According to a widespread conviction, attitudes towards the appropriate roles of women and men have become much more liberal in the last decades throughout the OECD countries and in Germany as well. Again, conditions of change and paths of development differ among countries. This is the case for the two halves of Germany. According to Geißler (1996, p. 275), women’s liberation was promoted by different groups in the FRG and in the GDR. In the GDR, women’s liberation was a paternalistic top-down process initiated and controlled by an autocratic regime. The motives of the socialist state were a mixture of ideological reasoning and the poor performance of the East German state-run economy. In the FRG, the women’s movement was said to be the most important actor in the modernization of gender relations, which was allegedly a democratic down-top process. If this description is true, attitudes should be more liberal in West Germany than in East Germany. For East Germans, the ideal of gender equality should be discredited together with socialist ideals of egalitarianism. In other former state socialist countries in Eastern Europe, the transition to the Western model coincided with a wave of anti-feminism (Jalušig, 1997). Obviously, East Germany did not follow the pattern of traditionalization, and rolling-back modernization in this respect. Studies of gender role attitudes in Germany after unification in 1989 repeatedly found differences between East Germany and West Germany, with West Germans holding more traditional attitudes. Some longitudinal analyses found a growing gap between East Germany and West Germany due to traditionalization in West Germany. Research on gender role attitudes in Germany has at least two shortcomings: The few longitudinal studies analyze short-term changes. Most authors analyze changes at the level of items (Adler & Brayfield, 1996; Braun & Borg, 1997, 1998; Braun & Nowossadeck, 1992; Künzler, 1995; Kurz, 1998). Until now, a longitudinal analysis is still lacking which covers the 1980s and 1990s using a scale of gender role attitudes. The German General Social Survey (ALLBUS) offers this opportunity. The German General Social Survey is a biannual survey of a cross-sectional representative national sample. In the FRG, the first wave was in 1980. Starting in 1991, East Germany participated in the survey, too. Items on respondents’ gender ideology were included in many waves, but the same questions were asked only in 1982, 1991, 1992 and 1996. In these four waves, 9,093 respondents were interviewed in West Germany, and 3,752 respondents in East Germany. Respondents were asked to what extent they agree or disagree with the following statements: "A working mother can establish just as warm and secure a relationship

41

with her children as a mother who does not work"; "It is more important for a wife to help her husband’s career than to have one herself"; "A pre-school child is likely to suffer if his or her mother works"; "It is much better for everyone if the husband is the wage-earner and the wife takes care of her home and family"; "It is even better for a pre-school child when his or her mother works and does not concentrate on homemaking"; "A married woman should refrain from working if jobs are scarce and her husband is able to earn the family’s living". Respondents indicated their strength of agreement or disagreement on a scale ranging from 1 = strongly agree to 4 = strongly disagree. Answers to traditional statements were reversed and a summed scale was constructed ranging from 6 to 24. By another transformation, zero became the value representing the neutral midpoint of the scale. The scale ranges from -9 to +9. High scores represent liberal attitudes. Cronbach’s alpha is .789 for women in West Germany, .722 for women in East Germany, .753 for men in West Germany, and .718 for men in East Germany. Appendix-Table A19 shows the percentage of traditional answers to each item and overall score means for the different years of the survey. Between 1982 and 1996, respondents in Germany developed a more liberal gender ideology, but the trend was by no means linear and of the same magnitude for women and men and for West Germany and East Germany (Figure 1). In the early 1980s, a majority of West German respondents still held traditional gender role attitudes. Obviously, attitudes became less traditional in the 1980s, but not until 1991 did the overall mean rise to the neutral midpoint of the scale. In 1982, there was no difference between women’s attitudes and men’s attitudes, but by 1991 a gap had opened up. Compared to respondents in West Germany, both women and men had more liberal attitudes in East Germany, with women taking the lead in liberalization. In the 1990s, attitudes in East Germany became even more liberal, with the gap between women and men first widening and then closing again. In West Germany, there was a marked traditionalization of attitudes between 1991 and 1992, widening the gap between women and men. Men on average returned to traditional views of gender roles. By 1996 West Germany had returned to a path of (slow) liberalization and reached the level of 1991. In the area of gender role attitudes, there was no adoption of the West German point of view by respondents in East Germany. On the contrary, dissimilarities between East Germany and West Germany have grown since unification.

42

Figure 1:

Gender ideology in Germany, 1982-1996

4 3

Means

2 1 0 -1 -2 -3 1980

1985 Women West

1990 Men West

1995 Women East

2000 Men East

Source: ZA, 1999, own calculation.

3.2

Gender differences in education

Level of education is a major predictor of a person’s position in the labour market and hence of his or her wage prospects. For this reason, gender differences in education should be considered as a condition of gender differences in the labour market. During the last century, gender differences in educational attainment in Germany have continuously diminished. The GDR was ahead of the FRG, esp. in vocational and tertiary education. In unified Germany, younger men and women are currently almost equally qualified. In this section, we will discuss the reduction in gender inequality in education using cohort data from the ALLBUS. As educational level is an achieved status, only a cohort approach can properly inform on the chronological development of gender inequality. With regard to recent developments since unification, the cohort approach implies severe limitations. As the younger cohorts might not have received their degrees yet, cohort figures cannot account for recent developments.25 To make the different types of schools internationally comparable, we translate school degrees and vocational degrees into CASMIN26 categories (for classification criteria see Appendix-Table A20). For our assessment of gender inequality in educational attainment, we will first consider degrees from the general school system and, second, certificates from the

43

vocational training system. We use the logarithm of the female/male ratios in the following figures. The logarithm centres the ratios equidistantly around zero. If the ratio is > 0, there are relatively more women, if it is < 0, there are relatively more men (for the data and further information see Appendix-Table A21Appendix-Table A24). In the following three figures, the development of gender inequality is shown for three educational levels in the East and West German population: no degree or only general elementary education, intermediate degree, and maturity degree. Figure 2: Female/male ratio (ln) of persons in Germany with no degree or general elementary education by birth cohort 0,4

ln (f/m)

0,2 0 -0,2 -0,4 -0,6 before 1920

1920-1929 1930-1939 1940-1949 1950-1959 1960-1969 Birth cohort West (n=15,731)

Note:

East (n=2,341)

No degree or general elementary education: CASMIN 1a or 1b.

Source: Zentralarchiv für empirische Sozialforschung an der Universität Köln (ZA), 1999, own calculation.

The older cohorts of both East and West Germans show that women are over-represented at the lowest level of education. In recent cohorts, this relation is reversed. In the youngest cohort more men than women received no degree or only general elementary education. In East German cohorts, the development was even more favourable to women than in West German cohorts: A greater discrimination of women in the older cohorts changed into an over-proportion of men at the basic educational level: In the

44

youngest cohort, 72 women without a degree or with only eight or nine years of schooling corresponded to 100 men with the same educational level. Figure 3: Female/male ratio (ln) of persons in Germany with intermediate degree by birth cohort 0,4

ln (f/m)

0,2

0

-0,2

-0,4 before 1920

1920-1929 1930-1939 1940-1949 1950-1959 1960-1969 Birth cohort West (n=6,467)

Note:

East (n=2,166)

Intermediate degree: CASMIN 2b.

Source: ZA, 1999, own calculation.

At the intermediate level, the ratio seems more favourable to West German than to East German women: In West German cohorts, there is a remarkable rise in women’s shares, starting with an equal distribution and ending with a proportion of 144 women per 100 men in the youngest cohort. In the older cohorts of East Germans, the ratio develops surprisingly unsteadily, but in the two youngest cohorts, the ratio is close to equality.

45

Figure 4: Female/male ratio (ln) of persons in Germany with maturity degree by birth cohort 0.2 0

ln(f/m)

-0.2 -0.4 -0.6 -0.8 -1 -1.2 before 1920

1920-1929 1930-1939 1940-1949 1950-1959 1960-1969 Birth cohort West (n=5,371)

Note:

East (n=923)

Maturity degree: CASMIN 2c_gen.

Source: ZA, 1999, own calculation.

The maturity degree, which is a prerequisite for university study, has been obtained by more men than women in all cohorts studied here. The general trend, both among East and West Germans, moves from noticeable inequality towards gender equality. Other data show (not shown in figure), that this ideal has been reached and even surpassed recently. In the FRG in 1989 already 30.7% of all female 13-14 year-old teenagers, compared to 27.1% of their male peers, were attending grammar school (Gymnasium) (Köhler, 1992, p. 67). In 1999, 53.3% of all school-leavers with a maturity degree were female. The share of women was greater in the East (59.1%) than in West Germany (51.9%) (Statistisches Bundesamt, 2000c). In the West German age-groups born after 1929, women’s share at the maturity level has continuously increased from 50 women per 100 men in the 1930-1939 birth cohort to a proportion of 79 women per 100 men in the youngest cohort studied. In the East German age-groups, the development went faster than in West Germany but stagnated at a proportion of 85 women per 100 men in the youngest cohort. All in all, we may state that women have constantly been able to better their position in general education. In recent cohorts, the inequalities have virtually

46

vanished (see also Klammer, Klenner, Ochs, Radke, & Ziegler, 2000, p. 195). In the East German cohorts, the development towards equality went faster and included the intermediate and maturity level in a similar manner. In the West German cohorts, women came to a remarkable lead on the intermediate level (144 women per 100 men in the 1960-1969 cohort), but their "conquest" of the maturity level stood at a ratio of 79 women per 100 men in the 1960-1969 cohort. Although there are East-West differences, the overall tendencies have been similar, which also applies to the second area to be discussed here, i.e., vocational education. Figure 5 shows the development of gender differences in completed vocational training at a medium level, including degrees such as the apprenticeship certificate, the advanced technical school degree, the specialized vocational school degree or similar grades. Figure 5: Female/male ratio (ln) of persons in Germany with completed vocational training by birth cohort 0.2

ln (f/m)

0.0 -0.2 -0.4 -0.6 -0.8 before 1920

1920-1929 1930-1939 1940-1949 1950-1959 1960-1969 Birth cohort West (n=5,229)

Note:

East (n=3,027)

Completed vocational training: CASMIN 1c, 2a, or 2c_voc.

Source: ZA, 1999, own calculation.

Since the first cohort, the female/male ratio in completed vocational training substantially increased and reached parity in the youngest cohort studied. In the cohorts born before 1920, about 50 women corresponded to 100 men with this training; now, this gender gap has been closed. In this development, all East

47

German cohorts were ahead of the West German ones in the past, reaching an equal distribution already in the 1940s cohort. However, there is another aspect which should not be ignored: Over the years the gender-typed choice of occupation has not been reduced. Female teenagers tend to choose training or apprenticeship in the fields of "social issues", "education", "health", and services occupations, while male teenagers dominate technical occupational fields and more frequently choose an apprenticeship in production (Klammer et al., 2000, p. 195).

ln(f/m)

Figure 6: Female/male ratio (ln) of persons in Germany with university degree by birth cohort 0.0 -0.2 -0.4 -0.6 -0.8 -1.0 -1.2 -1.4 -1.6 -1.8 -2.0 before 1920

1920-1929 1930-1939 1940-1949 1950-1959 1960-1969 Birth cohort West (n=1,101)

Note:

East (n=592)

University degree: CASMIN 3a or 3b.

Source: ZA, 1999, own calculation.

In tertiary education, there have been the most far-reaching improvements in the position of women. Nevertheless, still in all cohorts, relatively more men than women hold a university degree (Figure 6). While in the 1930-1949 cohorts, women’s share in completed tertiary education was still equally low for East and West Germans (around 40 women correspond to 100 men in those age-groups), the reduction in inequality in the subsequent agegroups occurred faster for East than for West Germans; in the 1960-1969 cohort, 90 women in the East, compared to 81 women in the West German cohort, correspond to 100 men with a university degree.

48

The latest figures indicate that in unified Germany gender equality in tertiary education has not yet been reached: In 1999, 43.5% of university/college graduates were female (Statistisches Bundesamt, 2000b). Furthermore, gender specific differences have to be stated in relation to the choice of subjects: There is a high percentage of women studying languages and cultural subjects, while engineering studies continue to be a male domain (Klammer et al., 2000, p. 218).27 In international comparisons, it is usually the highest educational level, i.e., a university degree, that is used as an indicator for the assessment of gender inequality in education. This makes sense because of great differences in educational systems and a greater availability of information about higher education (cf. Künzler, forthcoming, p. 13). To sum up, the development of gender inequality in East and West Germany has taken usually similar, but sometimes different paths. In both parts of Germany the overall picture shows that there were enormous advances in the position of women throughout the Twentieth Century. Nevertheless, full equality has still not been reached. Up until now, more men than women have received a degree in higher education. Using this indicator, unified Germany cannot be called a fully modern country with respect to gender equality in educational attainment. The GDR was ahead of the FRG in the general modernization process, resulting in a more equal distribution of educational attainment already at an earlier point in time. But even in the GDR equality in higher education was never completely achieved. After unification, the West German educational system was transferred to East Germany. At that time, the proportions of men and women had reached a quite similar level in East and West Germany. It can be assumed that, in the future, the female/male ratios on the different educational levels in East and West Germany will develop in a similar manner. As we will see in the following section, the relatively equal position of younger men and women in education does not translate directly into a similar position in the labour market.

3.3

Gender differences in the labour market

Labour force participation is a central indicator of gender inequality in the division of paid and unpaid work. Over the last three decades, general differences between female and male labour force participation have decreased in all OECD countries (Künzler, 1998, p. 105). However, forms and structures of female employment are very different. In the following, we will address gender differences in East and West Germany concerning overall labour force participation, unemployment, the labour force participation of mothers, part-time employment, and employment without social protection. For our depiction, we use cross-sectional data, mainly official statistics provided by the Federal Statistical Office; those data only account

49

for the FRG and for unified Germany. To describe the situation in the GDR, we have to rely on the existing literature.

3.3.1 Overall participation In this field, East and West Germany differed widely, as in the GDR female employment was politically supported and taken for granted for a long period, while in West Germany the development towards more female employment occurred more slowly. Figure 7:

Labour force participation in Germany, 1965-1999

100 90 80 70

%

60 50 40 30 20 10 0 1960

1965

Women West Sources:

1970

1975

Men West

1980

1985

Women East

1990

1995

2000

Men East

Statistisches Bundesamt, 1976, 1982, 1988b, 1992b, 1996b, 2000f.

Figure 7 shows, that, from 1965 on, the difference between male and female labour force participation in the FRG and thereafter in the western part of unified Germany decreased by more than 24 per cent. While male participation dropped steadily, women’s participation rose. In 1999, around 80% of 15- to 65-year-old men and around 61% of 15- to 65-year-old women were either gainfully employed or looking for work. In the GDR, there were hardly any differences between women’s and men’s labour force participation (not shown in figure). Since the 1950s, employment rates for women rose steadily. In 1990, 92% of women aged 25-60 were working (not including students; Geißler, 1996, p. 281). Due to the transformation of the East German economy, both men’s and women’s labour force participation decreased in the first two years after unification.

50

Afterwards, East German men’s percentage rose again to the level of West German men. The labour force participation of women in East Germany remained stable in the second half of the 1990s. In 1999, women’s labour force participation in East Germany was 73%, 7 points lower than men’s (80%), but 11 points higher than the rate for West German women. Altogether, East German women are noticeably more integrated into the labour market than West German women, while men’s integration into the labour market has been similar in East and West Germany during the 1990s. Viewed from the opposite angle, the gender gap in labour force participation is much greater in West Germany. Female employment is restricted to a few economic branches. In 1996, two thirds of gainfully employed women (about one half of gainfully employed men) were working in the service sector, mainly in public and private services or commerce and the catering industry (Bäcker, Bispinck, Hofemann, & Naegele, 2000, pp. 270f.). In the GDR, there was a broader spectrum of women’s occupations, and comparatively more women than in the FRG held management positions. In spite of that, there was also a sex-segregated labour market in the GDR, although it was less marked (Klammer et al., 2000, pp. 60f.). In the light of these facts, we have to state that after unification, East German women were partly able to keep their advanced position relative to West German women in terms of overall labour force participation. But their situation is worsening, as can be seen from unemployment rates.

3.3.2 Unemployment In the 1960s and early 1970s, the West German "economic miracle" (Wirtschaftswunder) produced nearly full employment in the FRG. Since the recession following the first oil price crisis in 1974/75, unemployment oscillated in line with the economic situation, but the base line of unemployment grew continuously (Bäcker et al., 2000, p. 318). In the GDR, there was officially no unemployment. As Figure 8 shows, there was an increase in unemployment in West Germany after unification, with unemployment rates rising from 6.3% in 1991 to 11.0% in 1997. During this period, gender differences in unemployment rates in West Germany seem to have totally vanished.

51

Figure 8:

Unemployment rates in Germany, 1976-1999

30 25

%

20 15 10 5

Women West Sources:

Men West

Women East

98 19

96 19

94 19

92 19

90 19

88 19

86 19

84 19

82 19

80 19

78 19

19

76

0

Men East

Statistisches Bundesamt, 1976, 1982, 1988b, 1992b, 1996b, 2000f.

The reconstruction of the economy in East Germany after unification caused an extreme increase in unemployment rates. In 1991, the unemployment rate in East Germany was already 10.4%; it rose to 19.5% in 1998 (Statistisches Bundesamt, 1992d, 1999b). An extremely unequal development of male and female unemployment can be observed in East Germany in the early 1990s (Figure 8). Especially during the first years after unification, women were much more severely affected by unemployment than men (Beckmann & Engelbrech, 1994; Holst & Schupp, 1994, 1995; Pischner & Wagner 1995; Pfeiffer, 1996; as quoted in Künzler, 1998, p. 105). In spite of an overall reduction after the maximum level reached in 1993, women’s unemployment rates in East Germany have continued to be high: 21.6% in 1999. In the second half of the 1990s, male unemployment in East Germany has also increased substantially and reached its maximum level in 1998 (18.9%). In addition to high female unemployment, a sex-specific re-structuring of the labour market is taking place in East Germany. Women are losing their former positions and fields of activity. Since the beginning of the 1990s, an outstandingly high proportion of women has lost their highly-qualified, well-paid, or management jobs (Klammer et al., 2000, p. 63). These figures confirm widespread criticism that East German women have been the losers of unification. While there are nearly no gender differences in unemployment in West Germany, there were enormous differences in East

52

Germany at the beginning of the 1990s, which are still visible. Even though the general level of unemployment is much higher in East than in West Germany, women in East Germany have been hit hardest by unemployment.

3.3.3 Overall participation by mothers The degree and extent of maternal employment, especially the employment of mothers with very young children, depends on cultural norms, economic restrictions, and the availability of child day care. In this respect, the two German states differed widely. In the FRG, the life courses of many women were influenced by the political ideal of the so-called "three-phases-model". After school and vocational training, women should be employed; after founding a family (in earlier times this implied marriage, later the birth of the first child) they should be homemakers for some years, and after this "family phase", they should return to work. This pattern has been modified in the course of the last decades: Mothers stay at home for shorter periods after the birth of a child, and more mothers return to the labour market as their children grow older (Bäcker et al., 2000, pp. 272f.). Table 3: Employment of mothers 1972, 1991, and 1996 by age of youngest child in Germany

1972 (FRG)

1991 1996

Employment rate of mothers, % youngest child < 6 33.9 youngest child < 3 youngest child 3-5 West East West East 37.3 75.9 47.6 82.8 42.4 49.3 48 65.7

youngest child 6-14 44.2 West 59.2 62.3

East 86.6 77.9

Note: The table shows employment rates of 15- to 64-year-old mothers; unemployed mothers are not included. Especially the percentages for the 1990s for mothers of children under 3 years of age have to be interpreted with caution: In the concept of employment applied in the Mikrozensus, all mothers who had a contract of employment before childbirth are regarded as employed. As a consequence, the percentages overestimate the actual number of employed mothers of infants, and some of those "employed" mothers are actually on parental leave and not present at the workplace (see Engstler, 1998). Source: Engstler, 1998.

Table 3 provides figures illustrating this development for West Germany. There was a higher percentage of employed mothers of children under six in the 1990s than in 1972. While in 1972 only 44.2% of mothers with their youngest child aged 6-14 were employed, this percentage reached 59.2% in 1991 and 62.3% in 1996. One of the underlying factors in this development was the growing labour force participation of married women. In West Germany, maternal labour force participation generally depends highly on the number and age of children (Klammer et al., 2000, p. 74).

53

In the former GDR, continuous full-time employment of mothers was demanded and supported by the supply of child day care facilities (see Chapter 2). This model can be called "simultaneous compatibility" between family duties and employment. In 1991, the first year after unification, this model remained visible: In East Germany, a high percentage of mothers of infants was employed (75,9%, see Table 3). This percentage fell dramatically to 49.3% in 1996. The main reason is the high unemployment rate among women, especially among mothers. In spite of that, still more married mothers with children are employed, and more mothers are employed full-time in East than in West Germany (Klammer et al., 2000, p. 76). To sum up, the influence of the former models of compatibility between family and work (three-phase vs. simultaneous) is still visible, but they are being modified. Today, more East German mothers stay at home, at least when they have an infant, and more West German mothers return to work after their children have reached school age. Still in all stages of child care, the employment of mothers is clearly more common in East than in West Germany.

3.3.4 Part-time employment According to Künzler (1998), Germany belongs to a group of four OECD countries with a very high proportion of female part-time employment: In 1993, the female/male ratios of part-time employment rates were 18 to 1 in Luxembourg, 13 to 1 in Austria, 12 to 1 in Belgium, and 11 to 1 in Germany. Female and male part-time employment was less unbalanced in Finland and in the USA (about 2 to 1). In the OECD countries, the average female/male ratio of part-time employment was 6 to 1 (sd = 4.2). The rise in female labour force participation in West Germany is for the most part due to part-time employment (Bäcker et al., 2000, p. 273). Until unification, the percentage of gainfully employed West German women with a part-time job hovered around 30% (see Figure 9). In the GDR, more women worked full-time (not shown). There was a percentage of about 27% part-time working women, but this group included mainly older women at the end of their employment life course, who worked only slightly shorter shifts than regular hours (Klammer et al., 2000, p. 60). In contrast, most part-time-workers in the FRG were mothers with dependent children who worked by the hour (Klammer et al., 2000, p. 74). In 1999, 87% of all part-time employed persons in unified Germany were women (1991: 92%) (Statistisches Bundesamt, 2000a). Figure 9 informs about the rates of part-time employees among men and women in East and West Germany.

54

Figure 9:

Part-time employment in Germany, 1980-1998

45 40 35 30 %

25 20 15 10 5 0 1980 1982 1984 1986 1988 1990 1991 1992 1994 1996 1998 W om en W est (n=5,297) W om en East (n=1,538)

Men W est (n=8,962) Men East (n=1,740)

Note: The figure shows the percentage of gainfully employed people who, in the German General Social Survey, declared a "half-day" job as their main occupation. Source: ZA, 1999, own calculation.

After unification, a growing number of East Germans took part-time work. In the second half of the 1990s, the percentages were slightly falling for both East and West German women. Only a tiny group of men works part-time in both parts of Germany. The Federal Statistical Office, whose wording of the question in the 1999 Mikrozensus was different28, found higher percentages: 22% of all gainfully employed women in East Germany and 42% of all gainfully employed women in West Germany worked part-time, whereas the respective percentages for men were 4% (East) and 5% (West) (Statistisches Bundesamt, 2000a). Part-time employment is more widespread among West German women than among East German women. Asked about their personal motives for working part-time, 65% of West German women, but only 21% of East German women named personal or family reasons for their part-time employment, whereas, for 53% of East German women working part-time, the main reason was a lack of available full-time positions (West: 8%). Men in both parts of Germany chose vocational and further training as the most important motive for working part-time (30% in West and 20% in East Germany).

55

Only 13% of men in West Germany and 4% of men in East Germany mentioned personal or family obligations (Statistisches Bundesamt, 2000a). Part-time work still means substantial disadvantages for employees: Part-time work stands for less qualified, more encumbering, and poorly paid jobs. As social benefits in case of unemployment, disease, and old age, are income-based, part-time work offers insufficient social protection. As a result of part-time work, family and household tasks can more easily be assigned to women (Bäcker et al., 2000, pp. 273, 293). Hence, part-time work can be said to back up a traditional gender division of labour.

3.3.5 Employment without social protection For a long time, a work contract with tenure and statutory insurance that guarantees a life-long employment history has been the ideal and the reality for (at least most) men — the so-called Normalarbeitsverhältnis (regular employment life course). In the last decade, this concept was subjected to erosive pressures: Temporary contracts, regular and pretended self-employment, and part-time employment without below the social insurance threshold created a lot of less well-insured and less permanent jobs. In 1985, 75% of all gainfully employed persons still held a regular full-time contract for an indefinite period, whereas in 1995 this percentage decreased to about 63%. Some 78% of gainfully employed men; but only 51% of gainfully employed women (Bäcker et al., 2000, pp. 291f.) had a full-time, fully secured job. About two thirds of the part-time jobs for less than 15 hours a week (or less than 630 DM per month, with no statutory insurance; so-called "630-DM jobs") are held by women (Bäcker et al., 2000, p. 296). About one quarter of those part-time jobs are second jobs, but for about 10% of all employees in Germany a 630-DM job was the only employment in 1997. In the same year, 17% of all gainfully employed women had no employment other than such a part-time job (Klammer et al., 2000, p. 97). In the 1990s, this type of employment expanded remarkably, while the number of full-time employed women decreased (Klammer et al., 2000, p. 97). The 630-DM jobs were an opportunity for employers to lower personnel costs and to replace regular contracts. Persons with such contracts are dependent on either their partner’s (mostly husband’s) social insurance or on another form of social insurance (e.g., student insurance, etc.). Since 1999, new regulations are forcing many employers to provide social insurance for these jobs. A first study on the consequences of this new regulation shows that the number of persons who have only a 630-DM job, has virtually not decreased. There are still no data on the distribution of the sexes in 630-DM jobs since 1999, but a reduction in the share of women is improbable (ISG/Kienbaum, 1999, as quoted in Klammer et al., 2000, p. 98). Those findings indicate that women are generally worse off in terms of social security in the labour market. Fewer women than men can rely on a stable, full-time contract. In contrast, the majority of precarious jobs fall to women.

56

3.3.6 Income Women still earn less than men: Several studies in Germany have shown that a considerable difference remains between female and male earnings even if differences in work time, vocational training, and duration of employment are controlled for (Diekmann, Engelhardt, & Hartmann, 1993; Engelbrech 1996; Hübler, 1992; Künzler, 1998, p. 106; Schömann, Hannan, & Blossfeld, 1991; Wirth, 1996). This difference in earnings has to be interpreted as an effect of wage discrimination. Figure 10 gives the female/male income ratios for West and East Germany. In 1983, the average yearly female income in the FRG amounted to 70% of the average yearly male income. Full-time employed women in the GDR received 71% (1980) to 76% (1988) of male earnings (Nickel, 1993, as quoted in Geißler, 1996, p. 283); so they received slightly more than West German women. Figure 10:

Income of full-time employees in Germany, 1983-1996

100 95 Women/men, %

90 85 80 75 70 65 60 55

West

96 19

95 19

94 19

93 19

92 19

91 19

90 19

89 19

88 19

87 19

86 19

85 19

84 19

19

83

50

East

Sources: Statistisches Bundesamt, 1987, 1988a, 1989, 1990, 1991a, 1992a, 1993a, 1994, 1995a, 1996a, 1997, 1998, 1999a, 2000d, own calculation.

Until 1996, the income share of West German women rose and reached 75% of men’s earnings. East German woman earned on the average around 90% as much as full-time employed men during the first half of the 1990s. There is a remarkable difference between East and West Germany since unification: The difference in income between the sexes is much greater in West than in East Germany. However, a gender gap in income still remains.

57

All in all, in contrast to the similarities in educational developments, the position of women in the labour market differed markedly in the GDR and the FRG. Although women’s participation has risen in both parts of Germany, the structure and scope of female and maternal employment were very different. In the GDR, full-time employment of women was normal. It was demanded, necessary for economic subsistence and supported by public day care. Thus, overall female labour force participation was similar to that of men (around 80%). As a result, women and men had a quite equal position in the labour market, even if women earned on the average only about three quarters of the average male income. Unemployment such as we know today in unified Germany was a quite unknown phenomenon in the GDR. In the FRG, due to insufficient coverage with child day care and a traditional motherhood-ideology, it was much more difficult for mothers to work full-time. For many women, part-time work became the only possibility to combine child care and employment. The rise in female labour force participation to a level of about 58% in 1989 was for the most part due to a rise in part-time employment. As men rarely choose part-time work, it has been a women’s domain. Female unemployment rates were always above male unemployment rates during the 1970s and 1980s. With respect to the incomes of full-time employees, women earned about 70% of men’s wages. Women’s position in the labour market was generally worse than men’s, regarding working hours, job security, unemployment, and income. After unification, the position of East Germans and especially of East German women worsened dramatically. Due to the reconstruction of the economy, the general level of unemployment increased in East Germany, but East German women have been hardest hit by unemployment. In the recent poor economic situation in East Germany, women have been pushed out of their former positions in the intensified job competition between the sexes. A group (about one fifth) of East German women has had to accept part-time work, even though a majority would prefer a full-time contract. If they get a full-time job, East German women do, however, achieve a higher female/male ratio in income than West German women. In Germany overall, all precarious forms of employment increased during the 1990s; the percentage of employed women with a regular, stable, full-time job is currently falling. In spite of the obstacles, the orientation towards employment is still stronger among East German than among West German women. Female labour force participation and the percentage of fulltime working women are noticeably higher than in West Germany, and an even higher percentage would like to work full-time, if possible. The employment of mothers is in all stages of child care more common in East than in West Germany. Contrary to prognoses predicting an adaptation to the West German model, employment has remained a central element of female identity and life course in East Germany.

58

3.4

Division of unpaid work

Whereas statistical offices continuously gather and report various indicators of gender inequality in the labour market and in paid work, longitudinal data on the gender division of unpaid work are rare. Starting with the participation of the FRG and the GDR in the Multinational Project on Time Use in 1965 (Szalai, 1972), representative time use studies have been repeatedly conducted in the two German states. Several caveats have to be kept in mind when the findings of these studies are compared: Studies differ with regard to the instruments applied (time diary method or time estimates). The definition of unpaid work and even of housework varies from study to study (ranging from all household work including child care to routine housework which is often stereotyped as female housework). Results are reported for different subgroups (ranging from all respondents to two-parent families with specific numbers of children and specific combinations of partners’ employment). When analyzing trends in overall averages, it is impossible to differentiate between changes in population composition and changes in behaviour. Nevertheless, some clear-cut trends may be identified. Obviously, East Germany always was in the lead in the modernization of the gender division of unpaid work. In the 1960s, the division of housework was extremely traditional in West Germany. Only 35.9 per cent of all men reported doing any housework, compared to 72.8 per cent in the GDR. Men spent less than three hours a week doing housework. Women did between 13.0 times (in a city sample) and 14.5 times (in a national sample) as much housework (30.8 or 35.4 hours a week). In the GDR, men spent more than seven hours a week doing housework — women did 4.2 times as much (33 hours a week) (see Künzler, 1999a, p. 182, Table 31; Szalai, 1972). In the GDR, another time budget survey was conducted in 1974. Women’s housework time had not changed much, but within ten years the time men spent doing housework had more than doubled and risen to 18 hours a week. In 1974, women did only 1.8 times as much housework as men in the GDR (Künzler, 1999a, p. 182, Table 31; Statistisches Bundesamt, 1991b). In international comparison, the FRG was comparatively traditional and the GDR was comparatively advanced in the 1960s (Figure 11 and Figure 12).

59

Figure 11:

Participation in housework in international comparison, 1965

Participation rate (%doers)

100 90 80 70 60 50 40 30 20 10

U SS R

D

R

Z

Men

4

5

2

0

0

Men´s time

D G

C

FR

Women´s time

U SS R

10

BG

6

R

15

Z

8

PL

20

U SA

10

F

25

YU

12

H

30

G

14

Ratio

Housework time in international comparison, 1965

35

B

Hours/week

C

G Women

Figure 12:

PL

F

H

U SA

BG

YU

FR

B

G

0

Women/men

In East Germany, there was no further change in the division of housework from the 1980s onward (Figure 13 and Table 4). Until the 1990s, West Germany gained ground and almost caught up with East Germany. In the 1991 time budget study of

60

the Federal Statistical Office, in West Germany married women did 2.4 times as much housework as married men. In East Germany, between 1974 and 1991 no further change had taken place: Women still did 1.8 times as much housework as men. In an analysis of the 1995 wave of the German Socio-economic Panel Study (GSOEP), Künzler (1998, p. 108) found that in two-earner families with a wife working full-time, women did 1.6 times as much housework as men in West Germany. In East Germany women did 1.4 times as much housework as men. In two-earner families with a wife working part-time the ratios were 2.4 in West Germany and 1.7 in East Germany. Figure 13: 1965-1995

Ratios of women’s/men’s time in housework in Germany,

Ratios women´s/men´s time

16 14 12 10 8 6 4 2 0 1960

1965

1970

1975

1980

West Germany

1985

1990

1995

2000

East Germany

Sources: See Table 4

61

Table 4: 1965-1995

Time use studies in Germany: Weekly time in housework,

Study year 1965 1965 1974 1980 1981 1981 1981 1981 1981 1981 1981 1981 1981 1983 1983 1983 1985 1988 1988 1988 1988 1988 1988 1988 1990 1991 1995 1995 1995

West Germany Women Men 35.4 2.5 30.8 2.7

33.3 29.5 34.2 34.4 25.6 26.8 22.9 30.1 31.6 39.6 40.1 21.5 22.5 21.4 23.6 15.9 18.6 18.6 21.3 31.5 38.1 28,4 21,4

6.4 7.0 9.7 9.8 10.9 11.2 11.2 11.4 11.6 17.2 14.2 12.1 3.7 4.0 3.6 5.1 4.2 4.7 3.6 13.2 10.9 12,1 13,7

East Germany Ratio / Men  Women 14.5 11.5 31.0 7.4 33.0 18.0 30.2 19.1 5.2 4.2 3.5 3.5 2.4 2.4 2.1 2.6 2.7 2.3 2.8 1.8 28.9 18.7 6.0 5.3 6.5 3.1 4.4 4.0 5.9 24.0 17.6 2.4 27.3 15.5 3.5 38.0 16.1 2,4 27,5 16,0 1,6 24,2 17,5

Ratio /  4.2 1.8 1.6

1.6

1.4 1.8 2.4 1,7 1,4

Sources: Kössler, 1984; Krüsselberg, Auge, & Hilzenbecher, 1986; Künzler, 1998; Schulz, 1990; Statistisches Bundesamt, 1991b, 1995b; Szalai, 1972; see the compilation in Künzler, 1999a.

In the 1990s, differences between East Germany and West Germany have been controlled in multivariate analyses only twice. Controlling for other factors, men in East Germany still spent more time doing housework in the 1992 wave of the GSOEP (Althammer & Wenzler, 1996). In the 1995 wave of the GSOEP, the

62

difference disappeared when controlling the assumptions of the time-availability approach (Künzler, 1999b, p. 254).

63

64

Chapter 4 Explaining the division of unpaid labour: Theoretical approaches In the last decades, several theoretical approaches and middle range theories were developed in order to explain the gender division of unpaid work. Developments in theory and research were reviewed repeatedly in the 1990s (Coltrane, 2000; Godwin, 1991; Künzler, 1994; Künzler & Walter, 2001; Shelton & John, 1996). Coltrane mentions seven different approaches or families of theories: gender construction approaches, economic and exchange perspectives, conceptual approaches that focus on institutional constraints, socialist-feminist theories, morality theories, psychological and socialization theories, and middle-level hypotheses on life-course factors. Only some of these approaches got over the hurdle of operationalization for a quantitative testing of hypotheses.

4.1

New home economics

According to Gary S. Becker’s new home economics (1993), the household is not only a unit of consumption, but also a unit of production. It operates like a factory: Income is used to buy goods which still have to be made ready for consumption. The members of a household have to invest time in this processing which transforms goods into commodities. The household tries to maximize its overall utility by optimizing its members’ time allocation to the market and to household production. If there are differences in the wage rates of household members, specialization becomes a means of increasing a household’s utility. The household member with the higher wage rate specializes in paid work and devotes as much time as possible to labour-market activities. The household member with a comparatively lower wage rate will, in turn, take over a larger share of the routine, unpaid work which has to be done in the household (Althammer & Wenzler, 1996; Becker, 1993; Berk & Berk, 1983). In principle, the model is gender-neutral: The gender division of housework will diminish when differences between women’s and men’s wage rates disappear. Until recently, studies testing the assumptions of new home economics used wage rates (hourly wages) as an indicator. In order to include persons without any income from employment in the analysis, e.g., housewives, shadow prices for their time had to be calculated and imputed (Althammer & Wenzler, 1996).

4.2

Time-availability approach

Until the early 1990s, the time-availability approach and new home economics were considered to be competing approaches. The time-availability approach argues that a household member’s time spent doing housework is primarily a function of the time subject to no other obligations. The less time a person has at

65

his or her disposal, the less time he or she will be able to spend doing housework. Recently, Coltrane has proposed interpreting available time as an operationalization of the assumptions of new home economics (2000, p. 1214). New home economics models how a household decides about household members’ time allocation in paid work. The time-availability approach models household members’ time allocation in unpaid work afterwards, when decisions about employment have imposed different restrictions on household members. Coverman (1985) proposed a modified version of the time-availability approach that allows additional inferences: The demand/response-capacity approach tries to identify both the factors influencing the demand for housework in the household (household size, number and age of dependent children, size of dwelling, household furnishings) and the factors restricting household members’ possibilities to meet the demand (time spent in employment and training, health impairment), i.e., their response capacity. The time-availability approach is also gender neutral: The gender division of housework is above all an effect of differences in women’s and men’s time spent in employment.

4.3

Resource theory

Resource theory shares several assumptions with both new home economics and the time-availability approach: All approaches assume that there is an invariant demand for housework in the household. Both new home economics and resource theory are rational choice theories. In new home economics, the household is the unit of utility maximization. Household members combine their efforts and come to unanimous decisions for the sake of household utility. Of course, new home economics does not drop the assumptions of individual rational choice: Household members will stay in the household only as long as they are better off within than outside the household. In resource theory, individuals are the units of utility maximization, and household members are primarily concerned with their own utility. This individualistic orientation may result in conflicts within the household when there are competing interests. In contrast to new home economics and the time-availability approach, resource theory explicitly takes into account the individual cost of doing housework. Chores are said to be repetitive, onerous, and sometimes even disgusting (for a critique see the proponents of the morality approach; Coltrane, 2000) but it would be sufficient to consider the opportunity costs of housework: Doing chores prevents a person from spending time in more pleasurable or more rewarding activities. Anyway, people who are maximizing utility will avoid investing much time in housework. If the amount of housework that has to be done in a household is relatively invariant, when one household member avoids doing housework the workload of the other household members will increase. The members of the household will have to bargain over the division of the workload. In this bargaining process, they use their relative power as an argument. In spite of the fact that there are several categories of resources as

66

sources of power (Foa & Foa, 1980), empirical tests of the assumptions of resource theory rely solely on economic resources. Indicators usually used in empirical research are respondents’ absolute income, their relative contribution to household income, and their education, which stands for human capital, and thus for potential income. Resource theory is also gender neutral: The gender division of housework occurs because men usually have more resources, which in turn is a result of gender differences in labour-market participation.

4.4

Role theory

Another approach, normative role theory, assumes that people behave according to their normative orientation. Sometimes this approach is called the gender ideology perspective. Without much theoretical elaboration, role theory implicitly posits that gender role attitudes have only one dimension and that they vary between the extremes of traditionalism and liberalism: Supporters of the traditional model of separate spheres for women and men are firmly convinced that there are natural differences (ordained by God) between women and men. The provider role is a man’s natural vocation, women should be homemakers and rear children. Female employment takes away jobs that should be reserved for men as family-income providers. Maternal employment is detrimental to children’s development. In most gender role scales used in empirical research, there is no unique definition of the liberal position. People with liberal orientations are viewed as those who reject traditional items. Role theory searches for consistency between people’s attitudes and behaviour. It hypothesizes that men with liberal attitudes will do more housework than men with traditional attitudes, whereas women with liberal attitudes will do less housework than women with traditional attitudes. Recently, Greenstein (1996) suggested that the interaction between spouses’ attitudes should play a crucial role in the division of housework. Holding liberal attitudes is not a sufficient condition for increased male participation in housework. Wives are the gatekeepers for men’s participation in housework. Wives with traditional attitudes will be reluctant to accept men’s increased commitment to household labour. Only in couples where the partners share liberal attitudes is a more equitable division of housework likely to develop. Greenstein confirmed this hypothesis in an analysis using data from the 1987/88 wave of the National Survey on Families and Households (NSFH). The standard indicator used in the test of the assumptions of role theory are Likert-like attitude scales. Role theory assumes that orientations and attitudes are internalized by socialization processes at critical ages, i.e., in childhood and adolescence. For this reason, another label of the approach is the "socialization perspective". Adults normally do not change their attitudes: Overall changes in attitudes take place by cohort replacement. In this perspective, age, or better, birth cohort may be used as an indicator of a person’s gender role attitudes, with younger cohorts holding more liberal attitudes. Without giving any theoretical reason, sometimes education is used as an indicator of a person’s gender role

67

attitudes, suggesting that higher education correlates with more liberal attitudes. Due to differences in attitudes between persons cohabiting without being married and married persons, cohabitation is sometimes used as another indicator. Married respondents are supposed to be more traditional in their orientations. Normative role theory, too, is gender neutral: If support for the separate spheres model declines, differences in women’s and men’s time allocation to paid and unpaid work should also decline.

4.5

Doing gender approach

In the late 1980s, the mainstream approaches to the explanation of the division of housework, above all normative role theory, were challenged by an approach called doing-gender. With a background in symbolic interactionism and ethnomethodology, West, Zimmerman, & Fenstermaker have argued that doing or not doing housework plays a major part in the routine production of gender identity in everyday interaction (Berk, 1985; Fenstermaker, West, & Zimmerman, 1991; West & Fenstermaker, 1993, 1995; West & Zimmerman, 1987). Men avoid being caught doing routine housework in order to maintain gender accountability in the eyes of significant others. As evidence, West and Zimmerman mentioned Hochschild’s (1989) finding that all the unemployed men in a qualitative study, i.e., men who were unable to fulfil their provider role, did no housework at all. Brines (1984) extended these considerations, developing a compensation hypothesis open to quantitative testing. She argues that men’s reluctance to do housework grows the more their gender accountability in the area of providing is threatened. Gender accountability in the area of providing is measured by an index of economic dependence proposed by Sørensen and McLanahan (1987). According to Brines, the relation of spouses’ economic dependence on each other and the division of housework should be U-shaped: As long as men’s position as chief family providers and, consequently, their male identity is not threatened, a reduction in women’s economic dependence will result in increased bargaining power and a tendency towards a more equitable division of housework. But as soon as men become economically dependent on their wives’ incomes, they try to allay any doubts about their gender accountability by avoiding housework. There are findings supporting the compensation hypothesis (Brines, 1994; Greenstein, 2000) but most authors using this interpretation overlook that the overall influence of economic dependence on the division of housework is consistent with the assumptions of resource theory (Künzler, 1999b; Künzler & Walter, 1999).

4.6

Integrating findings

Multivariate analyses simultaneously testing the hypotheses of the different approaches have produced mixed evidence. Until now, two quantitative meta-analyses have tried to integrate the sometimes contradictory findings systematically. The results of meta-analyses depend, to a great extent, on the choice

68

of procedure. Using the box-counting method, Godwin was unable to identify a consistent pattern of results in 30 publications. Her conclusion was pessimistic: "Despite the dozens of studies and the years of effort on the part of researchers from the three disciplines (economics, household economics, and family sociology; the authors) little cumulative knowledge about the causes of family time allocation exists, and many questions remain without definite answers" (1991, p. 279). The box-counting method tends to overlook significant findings and results in an increased type II error. Meta-analytical type II errors are lower for the Stouffer-method, which takes sample size and exact probability of test statistics into account. When using the Stouffer method for integrating the results of studies using multiple regression analysis, Künzler (1994) found that the major hypotheses of all approaches are confirmed. According to the meta-analyses conducted by Godwin (1991) and by Künzler (1994), the explanatory power of multivariate models explaining the gender division of housework was moderate. Godwin analyzed 30 studies which reported squared coefficients of determination ranging from R² = .04 to R² = .54 with a median R² = .18 (calculation: J. Künzler). Künzler analyzed 58 studies which reported squared coefficients of determination ranging from R² = .01 to R² = .56 with a median R² = .17. Usually, models of women’s contributions to housework explain a greater proportion of variance than models of men’s contributions. Models of the relative division of housework lie in between, which may be attributed to the fact that the relative division of housework is a function of both women’s and men’s absolute contributions (see Künzler, 1994). Measures of the relative division of housework between spouses (e.g., men’s percentage share) only allow inferences on a couple’s degree of sharing or specialization. Inferences on their behaviour are not possible. Godwin, as well as Künzler, sampled studies published in the 1970s and 1980s. Until now, no meta-analysis of studies published in the 1990s has been reported; but obviously in spite of both theoretical and methodological efforts in the 1990s, the explanatory power of the models is still moderate in the 1990s. A major problem for any systematic comparison of the explanatory power of the competing theories and approaches is that they partly use the same indicators, or, as Coltrane put it: "...theories in this area are neither exhaustive nor mutually exclusive" (2000, p. 1213). Besides the lack of effort to integrate contradictory findings empirically there is also a lack of effort to integrate the different middle-range theories into a more comprehensive theoretical model. In such a comprehensive model, interactions between theories should be taken into account. For instance, it is highly plausible that wives will use their resources in order to bargain for a more equal sharing of housework only if their gender role attitudes legitimize these attempts. On the other hand, the compensation hypothesis in the doing-gender approach should only apply to men who are eager to maintain a traditional male identity. There are some topics which have not been covered by theoretical considerations until now, e.g., long-term effects of biographical decisions (but see Gershuny 1996; Gershuny, Bittman, & Brice 1997; Gershuny,

69

Godwin, & Jones 1994) or the possibility of reciprocal causation between the quality of a relationship and the division of housework. There are next to no studies making use of gender-schema theory (Bem, 1981, 1983, 1984; Frable & Bem, 1985; for an exception see Denmark, Shaw, & Ciali, 1985; Guntner & Guntner, 1990). Neither the proponents of normative role theory nor the proponents of the doing-gender approach discussed the question of how attitudes and the search for gender accountability interact with a persons’ basic gender identity (femininity, masculinity, androgynity).

4.7

Regulation approach

A major shortcoming is the lack of both theoretical consideration and of empirical research on the influence of policy measures on the gender division of unpaid work. In a growing number of countries, equality between men and women in all areas of society becomes a political goal incremented at the constitutional level. In addition, sexual discrimination is banned by a number of international treaties and by the directives of international organizations (e.g., ILO directives). Most countries have developed a range of equal-opportunity policies. Most of these policies try to reduce women’s disadvantages in the labour market and employment. There are few policies aiming at increasing male participation in the family and in unpaid work. This official commitment to gender equality is found in Germany, too. On the other hand, in the FRG’s actual policy mix, as shown above, a majority of family policies, taxation legislation, and social security regulations favour the traditional model of separate spheres with a male wage-earner role and a female homemaker and child care provider role. In general, the one-earner model is supported by all measures and policies which lower the opportunity costs of a household member reducing, interrupting, or even totally retreating from employment. All financial transfers to married couples and families lower the opportunity costs of a withdrawal from employment due to marriage or childbirth. As long as for various reasons women’s incomes are lower than men’s, and as long as income losses are not completely replaced by transfers, financial transfers are biased and support women’s withdrawal from employment. On the other hand, by definition the promotion of the two-earner model requires policies and measures which help women stay employed independently of their family obligations. This objective can be achieved by getting men to participate in care-giving and by externalizing other aspects of care-giving obligations. The supply of affordable and reliable high-quality child day care with adequate hours of availability is of crucial importance to relieve women from time constraints. Until now, the husband’s obligation to support his family through full-time employment often serves as an argument for low participation in household work. Whether men who work fewer hours would increase their household contribution and to what extent needs additional research. Both financial transfers and public child day care supply should primarily exert a strong direct influence on women’s employment and

70

thereby have an indirect effect on their housework time and the couple’s division of unpaid work. In addition to this indirect effect via employment, the possibility of a direct effect of these policies on the division of unpaid work should not be excluded in advance, even if the lines of influence are not clear.

71

72

Chapter 5 Gender division of labour in Germany: The study 5.1

Sample

The data for our analyses come from the German Survey of the European Network on Policies and the Division of Unpaid and Paid Work.29 In the study 3,001 respondents of German nationality in age between 20 and 50 were interviewed by telephone between mid-May and mid-August 2000. In West Germany, 2,019 (67.3%) persons were interviewed, in East Germany 982 (32.7%). The East German population was over-sampled to allow more detailed analyses. The sampling procedure was a multistage random sampling including random last digit dialling. Within households, a respondent was selected by "last-birthdaymethod". Data were gathered by computer-assisted telephone interviewing (CATI). In married couple households and cohabiting households, the respondent’s spouse or partner was also asked for an abbreviated interview. There are 1,939 couple households in the sample. In 1,031 cases (53%) both partners were interviewed (Table 5). Table 5:

Design and procedure of the survey Division of paid and unpaid work in Germany — survey design and procedure —

Population

Selection procedure

Main study Persons of German nationality aged between 20 and 50 Households: multistage random sampling, including random last digit dialling; Individuals within households by "last-birthday" method n = 31 n = 3,001 Telephone interview

Respondents of pre-test Respondents of main study Survey method Average duration of 51.2 min. interview Source: Methodological report from INFAS, 2000.

Partner study Partner living together with respondent of main study Selection of partners of target persons who were willing to participate in study n=5 n = 1,031 Telephone interview 29.0 min.

A total of 1,288 (42.9%) of the respondents were male and 1,713 were female (57.1%). The mean age of the respondents was 36.4 years in West Germany and 35.9 years in East Germany (Table 6). Besides the deliberate over-sampling of the

73

East-German population, there are minor deviations from official statistics (i.e., women and older cohorts are over-represented). Table 6:

Number and average age of respondents Average age 35.8

sd 8.0

n 876

% 29.2

Women

36.9

7.7

1,143

38.1

Total

36.4

7.9

2,019

67.3

Men

35.4

8.2

412

13.7

Women

36.3

8.4

570

19.0

Total

35.9

8.4

982

32.7

Men

35.7

8.1

1,288

42.9

Women

36.7

8.0

1,713

57.1

Total

36.3

8.0

3,001

100.0

Men West-Germany

East-Germany

Total

Source: Gender Division of Labour in Germany, 2000.

There are considerable differences between West and East Germany concerning household types (Table 7). In West Germany, 16.7% of the households were single households, 17.2% were childless-couple households, 5.4% were one parent households, 48.1% were couple-with-children households, and 12.4% were other households (i.e., adult respondents who were living together with their parents). In East Germany, 12.8% were living in single households, 12.5% were living in childless-couple households, 8.4% were one parent households, 50.5% were couple-with-children households, and 15.8% were other households. Concerning household types, there were differences not only between West and East Germany, but also between men and women. For instance, in West Germany only 1.9% of men were living in one parent households, as opposed to 8.9% of women. In East Germany, 1.2% of men were living in one parent households, but more than ten times as many women were (13.5%).

74

Table 7:

Household types of respondents West-Germany

Single Childless couple Type of household (n = 3,001)

East-Germany

n

Men 182

Women 156

Total 338

Men 76

Women 50

Total 126

%

20.8

13.6

16.7

18.4

8.8

12.8

n

165

183

348

52

71

123

18.8

16.0

17.2

12.6

12.5

12.5

%

One parent households

n

17

93

110

5

77

82

%

1.9

8.1

5.4

1.2

13.5

8.4

Couple with children

n

365

607

972

187

309

496

%

41.7

53.1

48.1

45.4

54.2

50.5

Other households

n

147

104

251

92

63

155

%

16.8

9.1

12.4

22.3

11.1

15.8

n

876

1,143

2,019

412

570

982

% 100.0

100.0

100.0

100.0

100.0

100.0

Total

Source: Gender Division of Labour in Germany, 2000.

Some 46.4% of the respondents in West Germany and 41.1% in East Germany live in a childless household30; 19.5% of the West German respondents live in a household with only one child, 25.6% with two children and 8.5% with more than two children. In East Germany, 30.8% live in a household with only one child, 24.5% in a household with two children and 3.6% of the persons interviewed are living together with three or more children. Most of the men live in childless households (56.4% in West Germany, 53.4% in East Germany). In contrast, women live chiefly in households with one or more children (61.2% in West Germany; 58.9% in East Germany) (Table 8). There was no register of more than six children living in the same household. On average, .95 (sd = 1.02) children live at all respondents’ households; with respect to respondents with children in the same household, there are 1.73 (sd = .75) children on the average.

75

Table 8:

Number of children living in respondents’ households West-Germany

0 Number of children (n = 3,001)

1 2 >2

Total

East-Germany

Men

Women

Total

Men

Women

Total

n

494

443

937

220

184

404

%

56.4

38.8

46.4

53.4

32.3

41.1

n

148

246

394

85

217

302

%

16.9

21.5

19.5

20.6

38.1

30.8

n

182

334

516

90

151

241

%

20.8

29.2

25.6

21.8

26.5

24.5

n

52

120

172

17

18

35

%

5.9

10.5

8.5

4.1

3.2

3.6

n

876

1,143

2,019

412

570

982

%

100.0

100.0

100.0

100.0

100.0

100.0

Source: Gender Division of Labour in Germany, 2000.

As Table 9 shows, most respondents were married (56.6% in West Germany; 49.9% in East Germany). Some respondents (2.6% in West Germany; 2.7% in East Germany) are married and living separately, a few respondents (.8% in West Germany; 1.1% in East Germany) are widowed, 6.5% of the people interviewed in West Germany and 9.7% in East Germany are divorced, 33.4% of the respondents in West Germany and 36.6% in East Germany were still unmarried. There are differences between men and women in regard to marital status. For example, in West Germany 61.7% of the women were married, but only 50.0% of the men. In East Germany, 52.1% of the women and 46.8% of the men were married; whereas 42.1% of the men in West Germany and 43.7% in East Germany were still unmarried, compared to 26.8% and 31.4% of the women.

76

Table 9:

Marital status of respondents West-Germany

Married Married and living apart Marital status (n = 2,997)

Widowed Divorced Still unmarried

Total

East-Germany

Men

Women

Total

Men

Women

Total

n

437

704

1,141

193

297

490

%

50.0

61.7

56.6

46.8

52.1

49.9

n

19

33

52

8

19

27

%

2.2

2.9

2.6

1.9

3.3

2.7

n

2

15

17

1

10

11

%

.2

1.3

.8

.2

1.8

1.1

n

48

83

131

30

65

95

%

5.5

7.3

6.5

7.3

11.4

9.7

n

368

306

674

180

179

359

%

42.1

26.8

33.4

43.7

31.4

36.6

n

874

1,141

2,015

412

570

982

%

100.0

100.0

100.0

100.0

100.0

100.0

Source: Gender Division of Labour in Germany, 2000.

5.2

Dependent variables

Routine housework and child care. Respondents were asked to produce estimates on their time expended in work (separately for workdays and weekend days) in four types of activities (paid work, housework, child care, and leisure/voluntary work). In each area, there was a detailed list of different activities. CATI technique was used to check for consistency in order to minimize implausible estimates. Respondents with weekly estimates in all areas of activity of less than 140 hours or more than 196 hours per week were asked to rethink their estimates. In the following analyses, we use routine housework and child care as dependent variables. Routine housework is the total amount of time respondents spent cooking and preparing meals, washing dishes, doing laundry, cleaning and shopping during the whole week in contrast to other housework (the total amount of time respondents spent gardening/repairs, repairs and maintenance of cars and bikes and managing finances) which is often stereotyped as male housework. Child care is the total amount of time respondents spent on joint activities with their children (like swimming, cinema, etc.), playing with children and reading, transporting children (e.g., to school, Kindergarten, events, doctor ...), looking after homework, feeding, clothing and bathing. Only respondents with one or more children under 18 years were interviewed as to how many hours they spent per week on child care.

77

5.3

Independent variables

Table 10 gives an overview of theories, concepts and their operationalization. The operationalization of most independent variables is self-evident, but two concepts need additional description. Respondent’s economic dependence. The earnings of a respondent’s partner are subtracted from the respondent’s earnings; the value of this subtraction is divided by the addition of the respondent’s and partner’s earnings and multiplied by 100. The potential values of this measure range from -100 (respondent’s complete economic dependence on his or her partner) to +100 (partner’s complete economic dependence on respondent). A value of 0 means that each partner is economically independent of the other (the partners have equal earnings) (Sørensen & McLanahan, 1987). Gender role orientation. Respondents were asked to state how much they agreed or disagreed with the following four statements: "A working mother can establish just as warm and secure a relationship with her children as a mother who does not work"; "A pre-school child is likely to suffer if his or her mother works"; "All in all, family life suffers when the mother has a full-time job"; "A man’s job is to earn money; a woman’s job is to look after the home and family". Possible answers ranged from 1 = strongly agree to 5 = strongly disagree. The coding of items expressing the traditional point of view was inverted. The answers to the four items were combined in an unweighted Likert-scale with scores ranging from 4 to 20, high scores representing liberal attitudes. Cronbach’s alpha is .703 for women in West Germany, .658 for men in West Germany, .675 for women in East Germany, and .615 for men in East Germany.

78

Table 10:

Theories, variables and operationalization

Theories

Variables

Children

Time-availability approach

Income-oriented work Partner employed full-time Number of household appliances Size of dwelling Owner of dwelling Garden

Years in education Resource theory

Respondent’s economic dependency

Operationalization In couple households, a set of dummy variables indicates whether there are children of different ages living in the household: - no children - children 0-6 (reference group) - children 7-12 - daughters 13+ - sons 13+ Weekly hours spent on paid work, education, and travelling to or from work or school/training centre Dummy coding: respondents with a partner employed full-time = 1 Count index summing car, second car, freezer, dishwasher, washing machine, microwave, and personal computer Measured in units of 10 square metres Dummy coding: owner of dwelling = 1 Dummy coding: Households with a garden = 1 Blossfeld’s and Timm’s (1997, p. 24) classification of the educational level in years of education: - General elementary education = 9 - Intermediate degree = 10 - General elementary education and vocational training = 11 - Intermediate degree and vocational training = 12 - Maturity degree = 13 - Maturity degree and vocational training = 15 - Degree from a university of applied sciences (Fachhochschule) = 17 - University degree = 19 (earningsrespondent - earningspartner) / (earningsrespondent + earningspartner) * 100 (Sørensen & McLanahan, 1987)

79

Theories

Variables

Operationalization

Age of respondents

Age of respondents in years Dummy coding: Unmarried respondents living with a partner = 1 Coding s. above Likert-scale of four items (see above); scale ranging from 4 to 20; high scores = liberal attitudes Squared economic dependency index

Cohabitation Role theory

Years in education Gender role orientation

Doing-gender approach

Regulation approach

Control variables

80

Economic dependency squared Cohabitation

Coding s. above Total time in hours of care on a typical working day between 6 a.m. and 22 p.m. the youngest child is in public day care (day nursery, Kindergarten, Youngest child external care school), family day care, and cared for by others (mother, father, grandparents, sibling, domestic servants, neighbours, youth centres/associations Monthly amount of public transfers from child benefit, child-raising benfit, Financial transfers supplementary child benefit, child care tax credit, and long-term care insurance benefits (in units of 100 DM) Dummy coding: East-/West Germany East Germany = 1 Total monthly household income Household income minus taxes and social insurance contributions Five different types of households were represented by a set of dummy variables: Singles One parent households Type of household Couples without children Couples with at least one child up to six years Couples with children over 6 years (reference group)

Chapter 6 Stalled modernization in the West — back to tradition in the East? 6.1

Univariate findings

Historical changes. In 2000, there are no longer any differences regarding the division of routine housework between East and West Germany at an aggregated level:31 Women in West Germany reported to do 35 hours a week in routine housework, women in East Germany reported doing 34 hours a week. Men in both parts of Germany reported to do 17 hours a week in routine housework. Compared to the early 1990s, women and men in both parts slightly have increased the time they invested in routine housework. In West Germany, increases in men’s housework time have surpassed increases in women’s housework time resulting in a more equitable division. In 1991, women had done 2.4 times as much housework as men; in 2000, they do only two times as much housework. In East Germany, increases in women’s housework time have surpassed increases in men’s housework time, the division of housework becoming more traditional again: In 1991, women had done 1.8 times as much housework as men; until 2000, the ratio has risen to 2.0 times and is the same as in West Germany. Between the 1960s and the early 1990s, the division of housework had changed enormously but obviously in the 1990s the modernization of gender inequality in the division of unpaid work was stalled (Figure 14 and Figure 15).

81

Figure 14:

Housework time in Germany, 1965-2000

40 35

35 31

32

31

Hours/week

30

34

27

25 20

16

17

13

15 10 5

17

7 3

0 1965 W omen W est

1991 W omen East

2000 Men W est

Men East

Sources: 1965: Szalai, 1972; 1991: Statistisches Bundesamt, 1995b; 2000: Gender Division of Labour in Germany, 2000.

82

Figure 15: 12

Housework ratios in Germany, 1965-2000 11.4

Ratio women/men

10 8 6 4.2 4 2.4 2

1.8

2

2

0 1965

1991 West

2000 East

Sources: 1965: Szalai, 1972; 1991: Statistisches Bundesamt, 1995b; 2000: Gender Division of Labour in Germany, 2000.

At an aggregated level, the modernization of the gender division of paid work is more advanced with a lead in East Germany. There are no differences between East German and West German men in their time in paid work and education, both reporting that they work for pay 52 hours a week. In West Germany, women’s time investments in paid work are much lower, with an average of 29 hours a week. The gender gap is smaller in East Germany, women reporting 37 hours a week in employment. In East Germany, men invest 1.4 times as much time in paid work as women, in West Germany they invest 1.8 times the amount. A traditionalization in the division of housework emerges when developments in male one-earner households, households of men working full-time and women working part-time, and two full-time earner households are compared for the years 1995 and 2000 (Figure 16, Figure 17, & Table 11). Whereas there are only minor changes in households with a working wife in West Germany, the ratios of the spouses’ housework time has changed to the disadvantage of women in all household types in East Germany. The modernization lead of two-earner households with both partners working full-time has vanished between 1995 and 2000: In 1995, women had done 1.4 times as much housework as men, compared to 1.9 times as much in 2000 in these households. In households with a part-time working wife and in one-earner households, the extent of traditionalization has been even greater, the ratios rising from 2.4 to 3.4 and from 1.7 to 2.5.

83

Figure 16:

Housework ratios by earner type in West Germany, 1995-2000

4

Ratio women/men

3.5 3 2.5 2 1.5 1 0.5 0 1995 Male earner

2000 Full-time/part-time

Full-time/full-time

Sources: See Table 10.

Figure 17:

Housework ratios by earner type in East Germany, 1995-2000

4

Ratio women/men

3.5 3 2.5 2 1.5 1 0.5 0 1995 Male earner Source: See Table 10.

84

2000 Full-time/part-time

Full-time/full-time

Table 11: 1995-2000 Household type

Housework and the extent of spouse’s employment in Germany, Study year

West Germany Women

38.1 22.1 n = 622 One-earner 44.1 2000 13.8 n = 329 28.4 1995 15.9 Two-earner: woman n = 376 with part-time 37.7 employment 2000 11.5 n = 212 21.4 1995 12.6 n = 413 Two full-time earners 28.7 2000 10.5 n = 194 Note: Means and standard deviations. 1995

Sources:

Men 10.9 9.6 n = 622 14.5 8.6 n = 212 12.1 9.8 n = 376 14.7 8.6 n = 126 13.7 10.4 n = 413 18.4 11.4 n = 135

East Germany Ratio 3.5 3.0 2.4 2.6 1.6 1.6

Women

Men

38.0 22.9 n = 264 44.4 16.3 n = 84 27.5 14.9 n = 157 36.9 11.4 n = 55 24.2 13.1 n = 515 31.1 10.9 n = 187

16.1 11.3 n = 264 12.9 8.1 n = 55 16.0 11.5 n = 157 15.1 6.4 n = 37 17.5 12.1 n = 515 16.5 8.9 n = 128

Ratio 2.4 3.4 1.7 2.5 1.4 1.9

GSOEP, Wave L, 1995; Künzler, 1998.

Overall averages mask differences between different types of living arrangements. Patterns of time use may change when a couple starts cohabitation; they change when a woman gives birth to a child, and they continue to change as children grow up. Patterns of time use are different for lone parents. The differences in time use between different types of households are shown in the following four figures. We calculate average time use in paid work (including travel and education), routine and other housework, child care, and other care (including care for elderly or sick members of the household, and voluntary work) for five different types of living arrangements, i.e., respondents living alone, one-parent families, couples without children, two-parent families with a pre-school child, and two-parent families with pre-school children. Men’s time use. Men’s time in employment does not vary much among the different types of household (Appendix-Table A25 to Appendix-Table A28); nor is there much variation between East and West German men. In West Germany (Figure 18), hours in employment are highest for fathers in two-parent families with school children ( = 52.5 hours/week) and lowest for lone fathers ( = 47.6 hours/week). In East Germany, men living alone spend the fewest hours in

85

employment ( = 47.9 hours/week). Cohabiting men without children and fathers in two-parent families with school children spend the most hours in gainful employment ( = 53.8). In most household types, men in East Germany (Figure 19) are employed longer hours compared to men in West Germany. There is more variation in men’s time spent in doing chores in the different living arrangements. On the average, men living alone spend 19 hours a week doing routine housework in West Germany and 21 hours in East Germany. Men do one hour less in West Germany and more than two hours less in East Germany when they live with a partner. In couple households with minor children, men again do less housework ( = 16 hours/week in West Germany and  = 15 hours/week in East Germany). Lone fathers spend the most time doing chores compared to other men ( = 22 hours/week in West Germany and 26.9 hours/week in East Germany). In time spent in child care, there are again next to no differences between men in East and West Germany. When there is a pre-school child to be cared for, men spend about 21 hours a week in activities with the children. When the children are older, fathers spend less time in child care (13.2 hours/week in West Germany and 11.6 hours/week in East Germany). Lone fathers are in-between, spending on an average 15 hours a week in child care.

86

Figure 18:

Men’s time use in West Germany, 2000

120

Hours/week

100 80 60 40 20 0 Single Paid work

Childless couple

Routine housework

One parent Other housework

Couple with child 6 Other care

Source: Gender Division of Labour in Germany, 2000.

Figure 19:

Men’s time use in East Germany, 2000

120

Hours/week

100 80 60 40 20 0 Single Paid work

Childless Couple

Routine housework

One parent Other housework

Couple with child 6 Other care

Source: Gender Division of Labour in Germany, 2000.

87

Women’s time use. Women’s time use varies in different types of living arrangements (Appendix-Table A25 to Appendix-Table A28). Differences between East and West Germany are also more marked. In West Germany (Figure 20), women living alone spend the same amount of time in gainful employment and do only slightly more housework compared to men who live alone (49.1 hours/week of gainful employment and 22.7 hours/week of routine housework). Women’s time spent in gainful employment is lower and their housework time is higher when they live with a partner. Living with a partner and raising a child of pre-school age results in another sharp reduction in time spent in gainful employment. At the same time, not only does the time spent doing chores increase to an average of 41 hours a week, but another 36 weekly hours are devoted to direct child care activities. When children begin to grow up and all the children in a family attend school, women halve the time they devote to child care activities and double the time spent in gainful employment. The amount of routine housework remains fairly constant. There are next to no differences between women in East and West Germany regarding the time spent in routine housework and child care activities, but the pattern of changes in employment is different for women in East Germany. In East Germany (Figure 21), more mothers raising a pre-school child in a two-parent family are employed full-time (20.9% compared to 9.3% in West Germany); when all children are attending school, only a minority (18.8% compared to 43.8% in West Germany) of mothers in two-parent families stays at home, and the majority works full-time (61.7% compared to 17.0% in West Germany). On average, women in East Germany spend 19 hours a week in employment when living with a partner and a pre-school child and 39 hours a week when living with a partner and school children. In East Germany, the patterns of time use of lone mothers are very similar to those of cohabiting mothers with school children. In West Germany, lone mothers work longer hours for pay than mothers with school children in two-parent families.

88

Figure 20:

Women’s time use in West Germany, 2000

120

Hours/week

100 80 60 40 20 0 Single Paid work

Childless couple

Routine housework

One parent Other housework

Couple with child 6 Other care

Source: Gender Division of Labour in Germany, 2000.

Figure 21:

Women’s time use in East Germany, 2000

120

Hours/week

100 80 60 40 20 0 Single Paid work

Childless couple

Routine housework

One parent Other housework

Couple with child 6 Other care

Source: Gender Division of Labour in Germany, 2000.

89

The division of child care is more equitable than the division of routine housework, both in East and in West Germany. In both parts of Germany it is more equitable in families with school children compared to families with a pre-school child. In East and West German two-parent families, mothers spend 1.7 to 1.8 times as much time in direct child care activities as fathers when there is a pre-school child in the household. Above all the sharp reduction in mothers’ time spent in child care results in the division of child care seeming more equitable in two-parent families with school children. In East Germany there is no longer any difference between mothers’ and fathers’ investments. In West Germany, mothers still spend 1.4 times as much time as men performing household work. In addition to routine housework and child care, other activities in everyday life should be considered when analysing unpaid work. Burdens like other housework, care for older or sick persons within or outside the household, and voluntary work add up to an average of 8.1 hours a week for women in West Germany, 11.6 hours a week for men in West Germany, 9.7 hours a week for women in East Germany, and 14.4 hours a week for men in East Germany. Taken together, all unpaid and paid work makes a double day both for women and men. In this respect, the weeks of both women and men have only weekdays and no days off. The lowest average total workload is that of single women in East Germany ( = 76.9). Lone fathers in East Germany have the highest total workload ( = 107.5). The total workload of parents is about 14 hours higher compared to couples without children. In general, women’s total workload is slightly lower than men’s total workload. In terms of time allocation, a state of equity seems to be present, but below this surface appearance of justice, a specialization following the model of separate spheres for women and men is alive and well. Independent variables. Means and standard deviations of the predictors used in the subsequent multivariate analyses are reported in Appendix-Table A29. East and West German respondents differ significantly in some respects: Due to the steep decline in birth rates in East Germany after unification there are fewer two-parent families with pre-school children in East Germany, but there are more childless couples in West Germany. West German households have more consumer durables; respondents in West Germany are more often the owners of the dwelling or house where they are living; their dwellings and houses are larger in terms of living space. Women are less dependent on their partner’s income in East Germany. Parents in East Germany receive fewer public transfers but have more child care at their disposal. In some respects, male and female respondents differ irrespectively of the part of Germany where they are living: Women report living with a higher number of children than men, which is a result of well-known differences in the mean age at the birth of the first child between women and men. Both in East Germany and in West Germany men more often live alone. With regard to gender role attitudes, our survey replicates the differences found in 1996

90

(see above, Section 3.1) There is only a small and non-significant difference between women’s and men’s liberal attitudes in East Germany. Women in West Germany are significantly more traditional in their orientations, but West German men are by far the most traditional group. Appendix-Table A30 and Appendix-Table A31 report Pearson’s zero-order correlation of respondents housework and child care time with independent variables and control variables. Nearly all relationships are in the expected direction; the vast majority of coefficients is significant.

6.2

Multivariate findings

We test major hypotheses of most of the approaches mentioned above using women’s and men’s weekly hours spent doing routine housework and child care as dependent variables applying ordinary least squares multiple regression analysis. We calculate different models for women and men. Testing additional hypotheses results in a successive reduction of the sample. A first set of regression analyses uses samples of all respondents (n = 1,397 women and n = 1,067 men). Respondents who are still living at their parental home are excluded (n = 406). Another 131 cases were lost due to missing values. In order to analyze the effects of economic dependence on partner income and of the partner’s available time, all respondents not living in a partnership are excluded. The second set of regression analyses uses samples of respondents living together with a spouse (n = 921 women and n = 653 men; 366 cases are lost due to missing values). In order to test the effect of political regulations, i.e., the effects of the supply of child care outside the family and of child benefits, all couples without children have to be excluded. The remaining sample consists of respondents living together with a spouse and raising minor children in the household (n = 534 women and n = 378 men; 557 cases are lost due to missing values). All respondents. In the first two regression models (Table 12Fehler! Verweisquelle konnte nicht gefunden werden.), the different types of living arrangements are represented by a set of dummy variables. Two-parent families with school-aged children are the reference category. At the level of descriptive findings, women’s but not men’s contributions in housework varies greatly by type of living arrangement. The regression models give a different picture. Controlling for other factors, there are no more any differences between women with children of different ages in two-parent families and lone mothers. There is a tendency for women in childless-couple households to do two hours less of housework each week. Women living alone spend significantly less time doing housework than the reference group (7.4 hours/week, p .0001). In the descriptive analysis, men’s contributions in doing housework seemed to be the same in the different types of living arrangements. Again, the regression models show a different picture. Men’s time doing household chores does not vary with the age of the youngest child. In

91

two-parent families, there is no difference between fathers of pre-school children and fathers of school children. All other men do more housework. Single men and men in childless-couple households both spend more than two hours more a week compared to the reference group. Lone fathers spend six hours more a week doing chores than fathers of school-aged children in two-parent families. An inspection of standardized regression coefficients shows that for both women and men, the amount of time spent in employment is the most important factor influencing contributions to housework. In practical terms, each additional hour in paid work results in a reduction of housework by 19 minutes for women and 12 minutes for men. The time women spend doing housework is also influenced by age, household income, gender role attitudes, and education. Young women do less housework compared to older women. In the youngest cohort, born in 1980, women do 9 hours less housework a week compared to the oldest cohort, born in 1950. Due to the cross-sectional nature of our data, it is not possible to decide whether this is a cohort effect or an age effect. The higher the income of the household, the less time women spend doing housework. Women with traditional gender role attitudes do more housework than women with liberal orientations. The difference between the most traditional and the most liberal women is 5.1 hours a week. For women, additional education results in decreases in housework time. There is a difference of 3.7 hours a week between women with a university degree and women with no degree or a degree in lower secondary education and no additional vocational training. The time men spend doing housework is also influenced by their gender role attitudes. Men with liberal attitudes do more housework than men with traditional attitudes but the difference between the most traditional and the most liberal men is only 3.5 hours a week. Neither age (birth cohort) nor household income nor education influence the time men spend doing housework. Neither men’s nor women’s housework time is influenced by indicators representing the demand side of housework, with one exception: Men who own their dwelling do two hours a week less housework than men living in rented accommodations. Finally, neither for women nor for men is there a difference between East Germany and West Germany. The explanatory power of the regression model of women’s housework time is much greater than that of men’s housework time. As adjusted coefficients of determination show, 35.7% of the variation in women’s housework time, but only 17.1% of the variation in men’s housework time is explained by the model.

92

Table 12: OLS-regression models predicting time spent on housework for all respondents and couples Independent variables East Germany Age Household income (ln) Type of household: Couple all children > 6 Single Couple without children One parent households Couple with children (at least one < 7)

All respondents Women Men 1.14 -.31 .04 -.02 .30*** .03 .16 .02 -1.96*** -.62 -.09 -.04 reference -7.44*** -.16 -1.99+ -.05 -1.55 -.03 .76 .02

-1.70 -.05 reference

Children aged 7-12 Girls aged 13 and older Boys aged 13 and older -.32*** -.45

-.20*** -.32

Partner employed full-time

Owner of dwelling Garden

.02 .00 .10 .04 -.15 -.01 -.05 .00

Men -1.77* -.09 -.02 -.02 -.21 -.01

2.10* .09 2.25** .09 5.99** .08 -.57 -.02

Children aged 0-6

Number of household appliances Dwelling (in 10 square meter)

Women 1.19 .04 .24*** .11 -1.99** -.08

reference

Children: no children

Income oriented work (hours/week)

Couples

.01 .00 -.01 -.01 -1.94** -.10 -1.09 -.05

.54 .02 -.19 -.01 1.04 .03 -.31*** -.46 1.67 .04 -.28 -.03 .22* .08 -.82 -.03 -.44 -.01

1.10 .05 reference -.45 -.02 -1.82+ -.08 1.93* .08 -.20*** -.33 .76 .04 -.07 -.01 -.04 -.02 -2.2** -.12 -.53 -.02

93

Couples Women Men -.29* .38*** Gender role orientation -.06 .13 -1.23 .93 Cohabitation -.03 .04 -.37** -.11 -.52*** .01 Years in education -.07 -.04 -.11 .00 Respondent’s economic -.01 -.06*** dependency -.03 -.23 Economic dependency .00 .0003+ squared -.04 .12 Constant 45.72*** 26.79*** 49.19*** 26.86*** Adj. R² .357 .171 .310 .220 F 56.26*** 16.67*** 24.01*** 11.19*** n 1397 1067 921 653 Note: Unstandardized regression coefficients and standardized regression coefficients. Independent variables

All respondents Women Men -.32** .22* -.07 .07

+p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001. Source: Gender Division of Labour in Germany, 2000.

All couples. Central hypotheses of the time-availability approach, resource theory, and the doing-gender approach apply only to couple households. In order to test these hypotheses, in the next two regression models respondents living alone and lone parents are excluded from the sample. In the models, the additional housework caused by children of various ages is controlled for by a set of dummy variables. Couple households with children of pre-school age are the reference group. Repeatedly, studies found that especially teenage girls do more routine housework than teenage boys. Teenage girls may be a support in household work, but there is still no test of whether this relief is shared equally by the parents or whether one parent profits more. In order to analyze these hypotheses, two dummy variables are included, indicating whether there are girls or boys aged 13 or older living in the household. In couple households without children, women do less housework compared to mothers of a pre-school child, and men do more housework compared to fathers of a pre-school child, but the differences are far from significant. Neither women’s nor men’s time spent doing housework is changed significantly by the presence of children of primary-school age. Teenage girls’ support in housework does not result in relief for their mothers, but their fathers profit by a reduction of their housework time by almost two hours a week. Surprisingly, fathers of teenage boys increase their time in housework by almost two hours a week. Controlling for economic dependence and household income, time spent in employment and a partner’s full-time employment are true and exclusive indicators of the time-availability approach. In the models for couples, too, respondents’ own time spent in employment is the most important predictor for their housework time.

94

Neither absolute nor relative effects differ from the first two models. Women employed full-time do 6.2 hours a week less housework compared to women employed part-time.Women with a partner working full-time do as much housework as the small group of women with a partner who works part-time or does not work for pay at all. The time men spend doing chores does not vary with the extent of their wives employment (full-time versus less than full-time), but only with the extent of their own employment. A reduction of men’s time spent in employment of one hour results in another 12 minutes spent doing household chores. Again, men who are owners of their dwelling do two hours a week less housework than men living in rented accommodations. For women, but not for men in couple households, time spent in housework increases with the size of the dwelling, but in terms of practical importance, the effect is rather small: Ten additional square meters result in another 13 minutes of housework each week. Other hypotheses of the time-availability approach are not confirmed by the data, e.g., the reduction of housework time by labour-saving household appliances. In couple households, too, respondents behaviour accords with their attitudes: Both women and men with traditional gender role attitudes do less housework than their liberal counterparts. In couple households, the effect is stronger for men than for women, both in absolute and in relative terms. The most liberal men spend additional 6.1 hours doing routine housework compared to the most traditional men. Spouses who cohabitate without being married do not differ in their housework time from married couples. Again, younger birth cohorts of women invest less time in housework than older ones, but there is no relation between men’s age or birth cohort and their contributions to housework. For women, education and birth cohort are in second place with regard to the relative importance in predicting housework time. The influence of education on women’s housework time increases, both in absolute and relative terms: Each additional year in education reduces women’s weekly housework time by more than half an hour. The difference between the most and the least educated women is 5.2 hours housework a week. Household income (logged) comes in third place. In wealthier households women (but not men) do less housework. For men, the predictor in second place is respondents’ economic dependence. In accordance with the assumptions of resource theory, men significantly increase their contributions to housework when their wives’ contributions to household income increase. On the other hand, the coefficient of the squared dependence term only slightly fails to be significant (p = .062). At first glance, the doing-gender approach is confirmed by a tendency for a curvilinear relation between economic dependence and housework time. In fact, the relation is not U-shaped, as hypothesized by the doing-gender approach: There is an overall positive relation between men’s economic dependence and their housework time, but changes in housework time are different at different degrees of dependence. Contrary to the assumptions of doing-gender’s compensation hypothesis, the marginal utility of changes in wives’ income contributions increases the more husbands become economically dependent on

95

their wives (Figure 22). The predicted housework time for men who are the sole earner is 13.6 hours a week. In a two-earner household with equal contributions to household income, the predicted housework time is 16.6 hours a week. In contrast to this increase by three hours, men’s housework time increases by another 9 hours a week when they become totally dependent on their wives’ income.32 Figure 22:

Dependence in couple households and husbands’ housework.

Predicted values (hours/week)

28 26 24 22 20 18 16 14

0 10

80

60

40

20

0

0 -2

0 -4

0 -6

0 -8

-1

00

12

Economic Independence Source: Gender Division of Labour in Germany, 2000.

In the couple sample, a significant difference is found between East and West German men. East German men spend 1.8 fewer hours doing housework each week compared to West German men when other factors (e.g., differences in gender role attitudes or in economic dependence) are controlled. Compared to the first two regression models including all respondents, the explanatory power of the models for couples decreases for women and increases for men. The models explain 31.1% of the variation in women’s housework time and 22.0% of the variation in men’s housework time. Couples with children. In Germany, many family policies are explicitly targeted towards households with dependent children. In order to analyze the effect of these policies, the sample has to be restricted to couples with children. For this subsample, in addition to housework time, time in child care activities is used as a dependent variable (Table 13). There are only a few changes in the set of predictors explaining the time spent on housework compared to the model using a subsample

96

of all couples. Women in wealthier households no longer spend less time doing routine housework. In the subsample of two-parent families, men tend to reduce their housework time with increasing household income. There is no longer a significant relief of fathers by their teenage daughters, but fathers of teenage sons now spend almost four additional hours doing routine housework compared to fathers of pre-school children. For women, the size of the dwelling does not matter any longer. In spite of the fact that the coefficent of the squared term of economic dependence does not change in magnitude, the probability is now far from significant (p = .188). For most significant predictors in the models with a sample of all couples, failing to become significant in the new model probably results from the reduction in sample size. Two new predictors are added to the model: the hours a week the youngest child spends in external child day care indicates whether and to what extent child care needs are met. External care may be public day care, but also care by relatives, licensed or unlicensed family day care, or care by au pairs. At present, external child day care is a political measure only in part. Especially for many parents in West Germany, it is an onerous day-to-day task to organize additional care for their children. Due to low public subsidies paid to day care centres for children below the age of three and for school children, especially centre day care is costly. Nevertheless the variable indicates the potential effects of a family-policy measure. It gives a hint as to what would happen if the supply of affordable and reliable public child day care were expanded. Somewhat surprisingly, fathers but not mothers reduce their time in routine housework when their children are cared for outside the family for longer hours. A possible cause could be that men use the time when they are monitoring their children to do chores. They invest the time saved by reductions in monitoring time in other activities (employment or leisure) thus also reducing housework time. Women, on the other hand, do not display this systematic pattern of behaviour. The other new predictor entered into the model, is the amount of public financial transfers the household receives (e.g., child benefits and paid parental leave). In a multivariate OLS-regression model (results not shown) of time in paid work and education on various controls in a sample of couples with children, there is a positive relation between the time the youngest child spends in external child day care and mothers’ time in employment (b = 1.07, p  .001, to be precise). One additional hour in child care at an ordinary workday results in more than an additional hour a week in employment. On the other hand, reducing the opportunity costs of reductions in women’s employment with financial transfers actually decreases women’s employment by almost one hour for every 100 Marks (51.13 Euro) (b = .91, p  .001). External child care is the most important predictor in relative terms ( = .234). Financial transfers come in third place ( = .169). A traditionalization of the gender division of unpaid work by financial transfers is not only the indirect result of a reduction of wives’ employment. There is also a direct effect of traditionalization in the division of housework. Public transfers influence mothers’

97

but not fathers’ time in routine housework. Each additional 100 Marks received by a household results in women spending another half an hour a week doing chores (controlling for their time spent in gainful employment). Again, respondents’ time spent in employment is the most influential factor in explaining their housework time. In absolute terms, the trade-off between employment time and housework time is greater for women than for men. Bringing women’s and men’s time in employment into line by reducing men’s time and by increasing women’s time by ten hours a week would result in women doing 3.1 hours less and men doing one-and-a-half hours more housework. Controlling for other factors, the gap between West German and East German men becomes even larger in the sample of two-parent families: East German men are much more traditional in their actual behaviour, taking into account that they are more liberal in their attitudes and that their wives are economically less dependent. If there were no other differences between East and West Germany, East German men would have to do additional three hours of housework to catch up with West German men. On the other hand, there are no differences in the time spent in child care activities between men in East and West Germany. Parents’ time in child care activities is influenced above all by the time they spend in employment. In the case of child care activities, the trade-off with time in employment is the same for mothers and for fathers. For both, a reduction of one hour in employment results in an additional quarter of an hour spent in activities with their children. For women, the employment elasticity of housework is greater than the employment elasticity of activities with children. For men, it is the other way round. Depending on the point of departure, both high and low elasticity may be interpreted as beneficial. Lower elasticity means that child care activities are not used as a resource to buffer increases in workload in employment; on the other side, only a smaller part of decreases in time in employment are invested in activities with children. Neither women’s time nor men’s time in child care activities varies with the time they spend doing chores. For women, there is a tendency to reduce child care activities when doing more housework, but in terms of practical importance the trade-off of four minutes for every hour in housework may be neglected. Next to time in employment, the age of children in the household is the most influential factor. Especially the raising of children of pre-school age is a time-consuming task. For older children, the time spent in direct activities with children is reduced by more than five hours for women and by about three hours for men. Men who are owners of the household dwelling reduce not only their housework time, but also their child care time (by two hours a week). On the other hand, women’s time in child care activities is reduced by more than two hours a week when there is a garden. Whereas women’s time doing chores depends on birth cohort and education, child care time does not. As with their time doing chores, fathers’ time spent in child care activities is influenced by the degree of their wives’ economic dependence. Confirming resource theory, men’s activities with their children increase when their wives contributions to household income increase. Financial transfers and external

98

child day care influence mothers’ but not fathers’ time spent in direct child care activities. In families receiving more public benefits, mothers spend more hours in child care activities. An increase by 100 DM increases their child care time by three quarters of an hour. On the other hand, and less surprising, the supply of external day care reduces women’s child care time. Again, there is no one-to-one trade-off: Every hour that the youngest child is cared for outside the family at an ordinary workday reduces the mother’s child care activities by 23 minutes a week. For both women and men in two-parent families, the model explains a greater proportion of variance in child care time than in housework time. 44.6% of variation in women’s child care time is explained by the model compared to 35.2% of their housework time. For men, 31.6% of the variation in child care time are explained, but only 21.1% of the variation in housework time. Table 13: OLS-regression models predicting time spent on housework and child care for couples with children

Independent variables East Germany Age Household income (ln) Children aged 0-6 Children aged 7-12 Girls aged 13 and older Boys aged 13 and older Income oriented work (hours/week) Partner employed full-time

Couples with children Housework Child care Women Men Women Men 1.15 -3.01** .52 -1.50 .04 -.17 .02 -.08 .31* -.005 -.06 -.09 .13 -.004 -.03 -.06 -1.40 -1.97+ -.73 -.69 -.06 -.11 -.03 -.03 reference .30 .01 -1.48 -.05 .99 .03 -.31*** -.41 1.85 .04

reference -.03 .00 -1.24 -.06 3.79*** .18 -.15*** -.27 1.23 .06

Housework (hours/week) number of household appliances Dwelling (in 10 square meter) Owner of dwelling

-.36 -.03 .004 .00 -.80 -.03

-.28 -.04 -.03 -.02 -1.90* -.11

reference -5.47*** -.20 -5.07*** -.15 -5.18*** -.16 -.25*** -.33 -.04 .00 -.07+ -.07 .25 .02 .00 .00 -1.71 -.06

reference -3.79*** -.02 -2.55* .10 -3.75** -.16 -.25*** -.40 -1.33 -.06 -1.27 -.01 .43 .06 .04 .03 -2.03* -.11

99

Couples with children Housework Child care Independent variables Women Men Women Men -.63 -.09 -2.20+ -1.23 Garden -.02 .00 -.07 -.05 -.16 .360** .06 .19 Gender role orientation -.04 .14 .01 .06 -1.07 -1.32 .23 1.50 Cohabitation -.02 -.05 .005 .05 -.72*** -.004 .31+ .05 Years in education -.16 .00 .07 .02 Respondent’s economic .002 -.06** .03 -.05* dependency .00 -.27 .10 -.21 Economic dependency -.0002 .0003 .00 .00 squared -.04 .13 .19 .19 Youngest child external -.06 -.24* -.39** -.07 care (hrs/week) -.02 -.12 -.12 -.03 Financial transfers (100 .53*** -.003 .78*** .09 DM) .14 .00 .20 .03 Constant 47.75*** 28.65*** 37.39*** 36.90*** Adj. R² .352 .211 .446 .316 F 9.83*** 6.61*** 20.68*** 9.71*** n 534 378 533 377 Note: Unstandardized regression coefficients and standardized regression coefficients. +p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001. Source: Gender Division of Labour in Germany, 2000.

6.3

Discussion

Theory testing. We test hypotheses derived from time-availability approach (new home economics), resource theory, role theory, doing-gender approach, and regulation approach. Hypotheses of all approaches are partly confirmed in different subsamples. Most relations are in the predicted direction but a majority of hypotheses is not confirmed by the analyses: 99 out of 142 tests (69.7%) of coefficients in multiple regression fail to become significant. Respondents’ time in paid work (including education) is the most important factor (both in absolute and in relative terms) in predicting time spent in housework and child care, both for women and for men; it is the only predictor which is significant in all models. This supports the core assumption of time-availability approach but most of the other assumptions of the approach are not confirmed. Especially children of different ages do not influence the amount of housework as much as common sense would assume. Other approaches are predictive only for women’s housework time or only for men’s housework time or only for some of the subsamples. Gender role attitudes are most predictive for the housework time of men in couple households:

100

Finding a partner is a prerequisite for traditional men trying to avoid routine housework. Another correlate of gender role attitudes, i.e., birth cohort, influences women’s but not men’s housework time: For men, cohort replacement will not result in the gender division of unpaid work becoming more equitable. Support of the compensation hypotheses of doing gender approach is weak at best. Marginal utility of women’s escape from their dependence on a male provider differs according to the degree of their dependence, but contrary to the assumptions of the approach and concordant with resource theory, marginal utility is higher when their economic dependence is higher. The general trend of the relation anyway supports the assumptions of resource theory but is restricted to men’s housework time: Men but not women are bargaining and buy off their housework obligations. The effect of policies. Policies influence the gender division of unpaid work in families with children both directly and indirectly. There is a strong and positive relation between supply of external child day care for the youngest child and the time women spend in employment. Independently of external child day care, financial transfers reduce women’s time in employment. Time in employment in turn is the most important predictor for women’s (and men’s) time in housework and child care. Besides the indirect effects of external child care supply and monetary transfers, especially monetary transfers have a direct effect, too, increasing women’s time in housework and child care even when controlling for time in employment. The importance of the time spent in employment gives a hint which policies might be most efficient in changing the traditional gender division of labour: Bringing women’s and men’s employment time into line by increasing women’s time and reducing men’s time will result in a more equitable division of housework and child care. Besides the direct effect, indirect effects are also to be expected. A levelling out of gender differences in employment time brings women’s incomes into line with men’s income. According to our findings, decreases in women’s economic dependence motivate husbands to invest additional time in housework. East-West differences. A traditionalization in East Germany, both in relative and in absolute terms, can be observed when the findings of our study are compared with the findings of other studies conducted in the 1990s. In East Germany, the ratio of women’s housework time to men’s housework time had been at a historical low level in two-earner families with both partners working full-time in 1995. Until 2000, it has risen from 1.4 to 1.9 and is now higher than the spouses’ housework ratio in this type of household in West Germany. In male one-earner households and in two-earner households with women working part-time, traditionalization is even more pronounced. Controlling for other factors in multivariate analyses, differences between men’s housework time in East Germany and West Germany are reversed between 1992 and 2000. In 1992, East German men had done more housework than West German men; in 1995, there was no longer any difference; in

101

2000, West German men do more housework (but not more child care) compared to East German men, taking into account that East German men have more liberal attitudes and live with less dependent partners. The pattern of development in the division of unpaid work differs from the development in other areas of gender inequality. In the labour market and in paid work, the East German modernization lead had decreased in the decade following unification but is still marked in the late 1990s. In gender role attitudes, there is even a continuous and pronounced modernization in East Germany. The standstill of attitude change in West Germany in this area produces an increasing gap between East and West. Comparing findings of cross-sectional studies, we do not know for certain whether the division of unpaid work is actually more traditional in East Germany or whether East Germans’ behaviour simply is not keeping pace with their modernizing attitudes.

102

References Adler, M. A., & Brayfield, A. (1996). East-West differences in attitudes about employment and family in Germany. The Sociological Quarterly, 37, 245-260. Althammer, J., & Wenzler, S. (1996). Intrafamiliale Zeitallokation, Haushaltsproduktion und Frauenerwerbstätigkeit. Ein mikroökonometrisches Modell zur empirischen Bestimmung des wirtschaftlichen Werts der Hausarbeit. Jahrbuch für Nationalökonomie und Statistik, 215, 398-418. Alber, J. (1986). Germany. In P. Flora (Ed.), Growth to limits. The western european welfare states since World War II, Vol. 2: Germany, United Kingdom, Ireland, Italy (pp. 4-154). Berlin: de Gruyter. Anttonen, A., & Sipilã, J. (1996). European social care services: Is it possible to identify models? Journal of European Social Policy, 6, 87-100. Bäcker, G., Bispinck, R., Hofemann, K., & Naegele, G. (2000). Sozialpolitik und soziale Lage in Deutschland. Band 1: Ökonomische Grundlagen, Einkommen, Arbeit und Arbeitsmarkt, Arbeit und Gesundheitsschutz. Wiesbaden: Westdeutscher Verlag. Becker, G. S. (1993). A treatise on the family. Enlarged edition. Cambridge: Harvard University Press. Beckmann, P., & Engelbrech, G. (1994). Frauenerwerbstätigkeit in den neuen Bundesländern — Erwerbsneigung, Erwerbsbeteiligung und Arbeitslosigkeit. In P. Beckmann & G. Engelbrech (Eds.), Arbeitsmarkt für Frauen 2000 — Ein Schritt vor oder ein Schritt zurück? Beiträge aus der Arbeitsmarkt- und Berufsforschung BeitrAB 179 (pp. 423-445). Nürnberg: IAB. Bem, S. L. (1981). Gender schema theory: A cognitive account of sex typing. Psychological Review, 88, 354-364. Bem, S. L. (1983). Gender schema theory and its implications for child development: Raising gender-aschematic children in a gender-schematic society. Signs, 8, 598-616. Bem, S. L. (1984). Androgyny and gender schema theory: A conceptual and empirical integration. Nebraska Symposium on Motivation, 32, 179-226. Berghahn, S. (1993). Ehe und Familie in der Verfassungsdiskussion — vom institutionellen zum sozialen Grundrechtsverständnis. Kritische Justiz, 26, 397-418. Berk, R. A., & Fenstermaker Berk, S. F. (1983). Supply-side sociology of the family: The challenge of the new home economics. Annual Review of Sociology, 9, 375-395. Berk, S. F. (1985). The gender facory. The apportionment of work in American households. New York, London: Plenum Press. Blossfeld, H. P., & Timm, A. (1987). Das Bildungssystem als Heiratsmarkt. Eine Längsschnittsanalyse der Wahl von Heiratspartnern im Lebenslauf. Sonderforschungsbereich Statuspassagen und Risikolagen im Lebensverlauf — Institutionelle Steuerung und Individuelle Handlungsstrategien. Bremen: Arbeitspapiere, 43. Braun, M., & Borg, I. (1997). Einstellungen zur Erwerbstätigkeit der Frau in Ost- und Westdeutschland: Trends, Strukturen und ihre Beziehung zu wirtschaftlichem Pessimismus. ZUMA-Nachrichten, 21, 21-35. Braun, M., & Borg, I. (1998). Die Rolle der Frau 1991-1996. In H. Meulemann (Ed.), Werte und nationale Identität im vereinten Deutschland (pp. 213-226). Opladen: Leske + Budrich. Braun, M., & Müller, W., (1997). Measurement of education in comparative research. Comparative Social Research, 16, 163-201. Braun, M., & Nowossadeck, S. (1992). Einstellungen zur Familie und zur Rolle der Frau. In P. P. Mohler & W. Bandilla (Eds.), Blickpunkt Gesellschaft, 2. Einstellungen und Verhaltensweisen der Bundesbürger in Ost und West (pp. 127-140). Opladen: Westdeutscher Verlag. Brauns, H., & Steinmann, S. (1999). Educational reform in France, West-Germany and the United Kingdom: Updating the CASMIN educational classification. ZUMA Nachrichten, Mai 1999, 7-44. Brines, J. (1994). Economic dependency, gender, and the division of labor at home. American Journal of Sociology, 100, 652-688. Bundesministerium der Finanzen (Ed.) (2000). Materialien zur Steuerpolitik. http://www.bundesfinanzministerium.de/f_triplesteu.htm. Bundesministerium für Arbeit und Sozialordnung (Ed.) (1998). Übersicht über das Sozialrecht. Bonn: Bundesministerium für Arbeit und Sozialordnung.

103

Bundesministerium für Familie, Senioren, Frauen und Jugend (Ed.) (1998). Zehnter Kinder- und Jugendbericht. Bericht über die Lebenssituation von Kindern und die Leistungen der Kinderhilfen in Deutschland. Bonn: Bundesministerium für Familie, Senioren, Frauen und Jugend. Bundesministerium für Gesundheit (Ed.) (2000). Wirkungen der Pflegeversicherung. Baden-Baden: Nomos. Coltrane, S. (2000). Research on household labor: Modeling and measuring the social embeddedness of routine family work. Journal of Marriage and the Family, 62, 1208-1233. Cornelius, I., & Vogel, C. (1994). Innerfamiliale Arbeitsteilung und Netzwerkhilfe in Familien mit Kindern. In Ministerium für Familie (Ed.), Familie heute — ausgewählte Aufsätze zur Situation der Familien in Baden-Württemberg (pp. 182-196). Stuttgart. Coverman, S. (1985). Explaining husbands' participation in domestic labor. Sociological Quarterly, 26, 81-97. Denmark, F. L., Shaw, J. S., & Ciali, S. D. (1985). The relationship among sex roles, living arrangements, and the division of household responsibilities. Sex Roles, 12, 617-625. Diekmann, A., Engelhardt, H., & Hartmann, P. (1993). Einkommensungleichheit in der Bundesrepublik Deutschland: Diskriminierung von Frauen und Ausländern? Mitteilungen aus der Arbeits- und Berufsforschung, 26, 386-398. Dorbritz, J. & Gärtner, K. (1998). Bericht 1998 über die demographische Lage in Deutschland. Zeitschrift für Bevölkerungswissenschaft, 23, 373-412. Engelbrech, G. (1996). Die Beharrlichkeit geschlechtsspezifischer beruflicher Segregation: Betriebliche Berufsausbildung und geschlechtsspezifische Einkommensentwicklung beim Berufseinstieg in den 80er Jahren. In S. Liesering & A. Rauch (Eds.), Hürden im Erwerbsleben. Aspekte beruflicher Segregation nach Geschlecht. Beiträge zur Arbeits- und Berufsforschung, BeitrAB 1983 (pp. 65-103). Nürnberg: IAB Engstler, H. (1998). Die Familie im Spiegel der amtlichen Statistik. Lebensformen, Familienstrukturen, wirtschaftliche Situation der Familien und demographische Entwicklung in Deutschland. Bonn: Bundesministerium für Familie, Senioren, Frauen und Jugend. Esping-Andersen, G. (1990). The three worlds of welfare capitalism. Cambridge: Polity Press. Esping-Andersen, G. (1996). Welfare states without work: The impasse of labour shedding and familialism in continental European Social Policy. In G. Esping-Andersen (Ed.), Welfare states in transition. National adaptions in global economics (pp. 66-87). London: Sage. Esping-Andersen, G. (1999). Social foundations of postindustrial economies. Oxford: Oxford Univ. Press. Fenstermaker, S., West, C., & Zimmerman, D. H. (1991). Gender inequality: New conceptual terrain. In R. L. Blumberg (Ed.), Gender, family, and economy: The triple overlap (pp. 289-307). Newbury Park, CA: Sage. Foa, E. B., & Foa, U. G. (1980). Resource theory. In K. J. Gergen, M. S. Greenberg & R. H. Willis (Eds.), Social exchange. Advances in theory and research (pp. 77-101). New York, London: Plenum. Frable, D. E., & Bem, S. L. (1985). If you are gender schematic, all members of the opposite sex look alike. Journal of Personality and Social Psychology, 49, 459-468. Fthenakis, W. E., & Textor, M. R. (1998). Qualität von Kinderbetreuung: Konzepte, Forschungsergebnisse, internationaler Vergleich. Weinheim, Basel: Beltz. Geißler, R. (1996). Die Sozialstruktur Deutschlands. Zur gesellschaftlichen Entwicklung mit einer Zwischenbildanz zur Vereinigung. Opladen: Westdeutscher Verlag. Gerlach, I. (1996). Familie und staatliches Handeln. Ideologie und politische Praxis. Opladen: Leske + Budrich. Gershuny, J. (1996). Veränderungen bei der Arbeitsteilung im Haushalt: Mikro-soziologische Analysen. In W. Zapf, J. Schupp & R. Habich (Eds.), Lebenslagen im Wandel: Sozialberichterstattung im Längsschnitt (pp. 97-124). Frankfurt a.M., New York: Campus. Gershuny, J., Bittman, M., & Brice, J. (1997). Exit, voice and suffering: Do couples adapt to changing employment patterns? Working Papers of the ESCR Research Centre on Micro-Social Change, 97-8. Colchester: University of Essex. Gershuny, J., Godwin, M., & Jones, S. (1994). The domestic labour revolution: A process of lagged adaption? In M. Anderson, F. Bechhofer & J. Gershuny (Eds.), The social and political economy of the household (pp. 151-197). Oxford: Oxford University Press. Godwin, D. D. (1991). Spouses' time allocation to household work: A review and critique. Lifestyles, 12, 253-294.

104

Gornick, J. C., Meyers, M. K., & Ross, K. (1996). Supporting the employment of mothers: Policy variations across fourteen welfare states. Working Paper no. 139. New York: Luxembourg Income Study. Greenstein, T. (1996): Husbands’ participation in domestic labour: Interactive effects of wives’ and husbands’ gender ideologies. Journal of marriage and the family, 58, 585-595. Greenstein, T. (2000). Economic dependence, gender, and the division of labor in the home: A replication and extension. Journal of Marriage and the Family, 62, 322-335. Guntner, N. C., & Guntner, B. G. (1990). Domestic Division of Labor among Working Couples: Does Androgyny Make a Difference? Psychology of Women Quarterly, 14, 355-370. Gustafsson, S. (1994). Childcare and types of welfare states. In D. Sainsbury (Ed.), Gendering welfare states (pp. 45-61). London: Sage. Gustafsson, S., & Bruyn-Hundt, M. (1991). Incentives for women to work: A comparison between the Netherlands, Sweden and West Germany. Journal of Economic Studies, 18, 30-65. Halbach, G., Paland, N., Schwedes, R., & Wlotzke, G. (1994). Labour law in Germany: An Overview. Bonn: Federal Ministry of Labour and Social Affairs. Hank, K., Tillmann, K., & Wagner, G. G. (2001). Außerhäusliche Kinderbetreuung in Ostdeutschland vor und nach der Wiedervereinigung. Ein Vergleich mit Westdeutschland in den Jahren 1990-1999. MPIDR Working Paper WP 2000-003. Rostock: MPIDR. Helwig, G. (1987). Frau und Familie. Bundesrepublik Deutschland, DDR. Köln: Wissenschaft und Politik. Hobson, B. (1994). Solo mothers, social policy regimes, and the logics of gender. In D. Sainsbury (Ed.), Gendering welfare states (pp. 170-187). London: Sage. Hochschild, A., & Machung, A. (1989). The second shift: Working parents and the revolution at home. New York: Viking. Holst, E., & Schupp, J. (1994). Perspektiven der Erwerbsbeteiligung von Frauen im vereinten Deutschland. In P. Beckmann & G. Engelbrech (Eds.), Arbeitsmarkt für Frauen 2000 — Ein Schritt vor oder ein Schritt zurück? Beiträge aus der Arbeitsmarkt- und Berufsforschung, BeitrAB 179 (pp. 140-174). Nürnberg: IAB. Holst, E., & Schupp, J. (1995). Zur Erwerbsorientierung von Frauen nach der deutschen Vereinigung — Umverteilung von bezahlter und unbezahlter Arbeit zwischen den Geschlechtern notwendig. Vierteljahreshefte zur Wirtschaftsforschung, 64, 52-71. Hübler, O. (1992). Dynamik des Arbeitseinkommens und geschlechtsspezifische Diskriminierung. In R. Hujer, H. Schneider, & W. Zapf (Eds.), Herausforderungen an den Wohlfahrtsstaat im strukturellen Wandel (pp. 435-465). Frankfurt a.M.: Campus. Institut für angewandte Sozialwissenschaft (2000): Familiale Arbeitsteilung in den Ländern der EU. Methodenbericht. Ergebnisse einer Untersuchung im Auftrag der Universität Würzburg und des Staatsinstituts für Familienforschung, Bamberg. Bonn Jalušic, V. (1997). Die Geschlechterfrage und die Transformation in Ostmitteleuropa: Kann das Geschlechterparadigma zur Transformation des Politischen beitragen? In E. Kreisky & B. Sauer (Eds.), Geschlechterverhältnisse im Kontext politischer Transformation. Sonderheft 28 der Politischen Vierteljahresschrift (pp. 450-474). Opladen: Westdeutscher Verlag. Joosten, A. (1990). Die Frau, das 'segenspendende Herz der Familie'. Familienpolitik als Frauenpolitik in der 'Ära Adenauer'. Forum Frauengeschichte, 6. Pfaffenweiler: Centaurus. Kirner, E., & Schulz, E. (1992). Das 'Drei-Phasen-Modell' der Erwerbsbeteiligung von Frauen — Begründung, Norm und empirische Relevanz. In N. Ott & G. Wagner (Eds.), Familie und Erwerbstätigkeit im Umbruch. Referate der Herbsttagung des Arbeitskreises 'Bevölkerungsökonomie' der Deutschen Gesellschaft für Bevölkerungswissenschaft, Sonderheft des Deutschen Instituts für Wirtschaftsforschung, 148 (pp. 17-55). Berlin: Duncker & Humblot. Klammer, U., Klenner, C., Ochs, C., Radke, P., & Ziegler, A. (2000). WSI-Frauen Daten Report (Forschung aus der Hans-Böckler-Stiftung, 26), edited by the Hans-Böckler-Stiftung. Berlin: edition sigma. Klein, Th., & Lauterbach, W. (1994). Bildungseinflüsse auf Heirat, die Geburt des ersten Kindes und die Erwerbsunterbrechung von Frauen. Eine empirische Analyse familienökonomischer Erklärungsmuster. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 46, 278-298. Köhler, H. (1992). Bildungsbeteiligung und Sozialstruktur in der Bundesrepublik. Zu Stabilität und Wandel der Ungleichheit von Bildungschancen (Studien und Berichte 53), edited by the Max-Planck-Institut für Bildungsforschng. Berlin: edition sigma.

105

Kössler, R. (1984). Arbeitszeitbudgets ausgewählter privater Haushalte in Baden-Württemberg. Stuttgart: Statistisches Landesamt Baden-Württemberg. Kreyenfeld, M. & Hank, K. (1999). The availability of child care and mothers’ employment in West Germany. DIW-Discussion Paper, No. 191. Berlin: DIW. Krüsselberg, H.-G., Auge, M., & Hilzenbecher, M. (1986). Verhaltenshypothesen und Familienzeitbudgets: Die Ansatzpunkte der "Neuen Haushaltsökonomik" für Familienpolitik. Stuttgart: Kohlhammer. Künzler, J. (1994). Familiale Arbeitsteilung. Die Beteiligung von Männern an der Hausarbeit. Bielefeld: Kleine. Künzler, J. (1995). History of equal opportunities policy: The case of Germany. In T. Willemsen, G. Frinking & R. Vogels (Eds.), Work and family in Europe: The role of policies (pp. 65-87). Tilburg: Tilburg University Press. Künzler, J. (1998), The case of Germany. In G. Frinking & T. Willemsen (Eds.), The role of social partners in the redivision of paid and unpaid work. An international comparison, WORC Report 98.05.002/6 (pp. 105-156). Tilburg: Work and Organization Research Centre. Künzler, J. (1999a). Wandel der Geschlechterverhältnisse im internationalen Vergleich. Habilitationsschrift. Würzburg: unpublished manuscript. Künzler, J. (1999b). Arbeitsteilung in Ehen und Nichtehelichen Lebensgemeinschaften. In T. Klein & W. Lauterbach (Eds.), Nichteheliche Lebensgemeinschaften (pp. 235-268). Opladen: Leske + Budrich. Künzler, J. (Forthcoming). Paths towards a modernization of gender relations, policies, and family building. Paper to be published in F.-X. Kaufmann, A. C. Kuijsten, H.-J. Schulze, & P. Klaus (Eds.), Family life and family policies in Europe, 2, Problems in comparative perspective. Oxford: Clarendon. Künzler, J., Schulze, H.-J., & Hekken, S. v. (1999). Welfare states and normative orientations towards women’s employment. In A. Leira (Ed.), Family policies. Yearbook of comparative social research, 18 (pp. 197-225). Greenwich, CT: JAI. Künzler, J., & Walter, W. (1999). Familiale Arbeitsteilung im europäischen Vergleich. In H. Schwengel (Ed.), Tagungsband 2 des Soziologie-Kongreß 1998 in Freiburg "Grenzenlose Gesellschaft?". Pfaffenweiler: Centaurus. Künzler, J. & Walter, W. (2001). Arbeitsteilung in Partnerschaften. Theoretische Ansätze und empirische Befunde. In J. Huinink, K. P. Strohmeier & M. Wagner (Eds.), Solidarität in Partnerschaft und Familie. Zum Stand familiensoziologischer Theorienbildung. Würzburg: Ergon. Kurz, P. (1998). Hausfrau oder Berufsfrau? Einstellungen zur Rolle der Frau in Ost- und Westdeutschland. In M. Braun & P. P. Mohler (Eds.), Blickpunkt Gesellschaft, 4. Soziale Ungleichheit in Deutschland (pp. 173-220). Opladen: Westdeutscher Verlag. Lampert, H. (1991). Familienpolitik in Deutschland. Ein Beitrag zu einer familienpolitischen Konzeption im vereinten Deutschland. In Kleinhenz, G. (Ed.), Sozialpolitik im vereinten Deutschland I, Schriften des Vereins für Socialpolitik, N.F. 208/I (pp. 116-139). Berlin: Duncker & Humblot. Lampert, H. (1998). Lehrbuch der Sozialpolitik. Berlin: Springer. Landenberger, M. (1991). Erziehungsurlaub: Arbeitsmarktpolitisches Instrument zur selektiven Ausgliederung und Wiedereingliederung von Frauen. In K. U. Mayer, J. Allmendinger, & J. Huinink (Eds.), Vom Regen in die Traufe: Frauen zwischen Beruf und Familie (pp. 262-287). Frankfurt a.M.: Campus. Lauterbach, W., Huinink, J., & Becker, R. (1994). Erwerbsbeteiligung und Berufschancen von Frauen. Theoretische Ansätze, methodische Verfahren und empirische Ergebnisse aus der Lebensverlaufsperspektive. In P. Beckmann & G. Engelbrech (Eds.), Arbeitsmarkt für Frauen 2000 — Ein Schritt vor oder ein Schritt zurück? Kompendium zur Erwerbstätigkeit von Frauen, Beiträge zur Arbeitsmarkt- und Berufsforschung, 179 (pp. 175-208). Nürnberg: IAB. Lautmann, R. (1990). Die Gleichheit der Geschlechter und die Wirklichkeit des Rechts. Opladen: Westdeutscher Verlag. Lewis, J. (1992). Gender and the development of welfare regimes. Journal of European Social Policy, 2, 159-173. Lewis, J. & Ostner, I. (1994). Gender and the evolution of European social policies. Zentrum für Sozialpolitik Working Paper, 94-04. Bremen. Liegle, L. (1990). Vorschulerziehung. In Bundesministerium für innerdeutsche Beziehungen (Ed.), Vergleich von Bildung und Erziehung in der Bundesrepublik Deutschland und in der Deutschen Demokratischen Republik (pp. 157-170). Köln: Wissenschaft u. Politik.

106

Limbach, J. (1988). Die Entwicklung des Familienrechts seit 1949. In R. Nave-Herz (Ed.), Wandel und Kontinuität der Familie in der Bundesrepublik Deutschland, Der Mensch als soziales und personales Wesen, 8 (pp. 11-35). Stuttgart: Enke. Lipp W. (1999). The family: Biological foundations and early cultural development. A plead for an integrated evolutionary perspective. Evolution and Cognition, 1, 33-48. Lipp W., Künzler J., & Walter, W. (1999). Antrag auf Gewährung einer Sachbeihilfe für das Forschungsvorhaben "Familiale Arbeitsteilung in den Ländern der Europäischen Union — Länderstudie Deutschland" zu Händen der Deutschen Forschungsgemeinschaft. Würzburg, Bamberg: unpublished manuscript. Mayer, K. U., Allmendinger, J., & Huinink, J. (Eds.) (1991). Vom Regen in die Traufe: Frauen zwischen Beruf und Familie. Frankfurt a.M.: Campus. Merkle, L. E. (1994). Frauenerwerbstätigkeit und Kinderbetreuung. Eine theoretische und empirische Analyse für die Bundesrepublik Deutschland. Heidelberg: Physica. Moeller, R. G. (1993). Protecting motherhood. Women and the family in the politics of postwar West Germany. Berkeley, CA: Univ. of California Press. Münch, U. (1990). Familienpolitik in der Bundesrepublik Deutschland. Maßnahmen, Defizite, Organisation familienpolitischer Staatstätigkeit. Freiburg: Lambertus. Notz, G. (1994). Frauen zwischen Arbeitsmarkt und Hauswirtschaft. Zum Verhältnis von Berufsorientierung, Kontinuitäten und Brüchen von Berufsverläufen von Frauen in den alten Bundesländern. In P. Beckmann & G. Engelbrech (Eds.), Arbeitsmarkt für Frauen 2000 — Ein Schritt vor oder ein Schritt zurück? Kompendium zur Erwerbstätigkeit von Frauen, Beiträge zur Arbeitsmarktund Berufsforschung, 179 (pp. 229-252). Nürnberg: IAB. Nyssen, E. (1990). Frauen zwischen Beruf und Familie. In S. Metz-Göckel & E. Nyssen (Eds.), Frauen leben Widersprüche. Zwischenbilanz der Frauenforschung (pp. 185-200). Weinheim: Beltz. OECD (1996). Employment Outlook 1996. Paris: OECD. OECD (1997). Education at a Glance. OECD Indicators 1997. Paris: OECD. Orloff, A. S. (1993). Gender and the social rights of citizenship: The comparative analysis of gender relations and welfare states. American Sociological Review, 58, 303-328. Orloff, A. S. (1996). Gender in the welfare state. Annual Review of Sociology, 22, 51-78. Ostner, I. (1995a). Arm ohne Ehemann? Sozialpolitische Regulierung von Lebenschancen für Frauen im internationalen Vergleich. Aus Politik und Zeitgeschichte, 46, 3-37. Ostner, I. (1995b). Sozialstaatsmodelle und die Situation der Frauen. In W. Fricke (Ed.), Jahrbuch Arbeit und Technik 1995 (pp. 57-67). Bonn: Dietz. Ostner, I. (1998). Frauengerechtigkeit und Familienpolitik. In A. Netzler & M. Opielka (Ed.), Neubewertung der Familienarbeit in der Sozialpolitik (pp. 35-43). Opladen: Leske + Budrich. Ott, N., Radtke, H., Thiel W., & Wagner G. (1990). Kindererziehung und Erwerbsarbeit — marktwirtschaftliche Möglichkeiten einer erziehungsfreundlichen Erwerbsarbeit in Deutschland. Wirtschaftswissenschaft (DDR), 38, 1242-1261. Pfaff, A. B., & Kerschreiter, M. (1982). Die Familie im Umverteilungsprozeß. Monetäre Leistungen für Ehegatten und Kinder. In F.X. Kaufmann (Ed.), Staatliche Sozialpolitik und Familie (pp. 131-168). München: Oldenbourg. Pfeiffer, B. (1996). Erwerbsbeteiligung und Stille Reserve ostdeutscher Frauen. Mitteilungen aus der Arbeitsmarkt- und Berufsforschung, 29, 117-137. Pischner, R., & Wagner, G. (1995). Bilanz der Erwerbschancen fünf Jahre nach der Wende in Ostdeutschland. DIW-Wochenbericht, 46. Rost, H., & Schneider, N. (1995). Differentielle Elternschaft — Auswirkungen der ersten Geburt für Männer und Frauen. In B. Nauck & C. Onnen-Isemann (Eds.), Familie im Brennpunkt von Wissenschaft und Forschung (pp. 177-194). Neuwied: Luchterhand. Rudolph, C. (1995). Einflußpotentiale und Machtbarrieren: Frauenpolitik in der Verfassungsdiskussion. Baden-Baden: Nomos. Schömann, K., Hannan, M. T., & Blossfeld, H.-P. (1991). Die Bedeutung von Bildung und Arbeitsmarktsegmenten für die Arbeitseinkommen von Frauen und Männern. In K. U. Mayer, J. Allmendinger & J. Huinink (Eds.), Vom Regen in die Traufe: Frauen zwischen Beruf und Familie (pp. 163-184). Frankfurt a.M.: Campus.

107

Schulz, R. (1990). Unterschiede in der Zeiteinteilung von erwerbstätigen Frauen und deren Entlastung durch Partner und/oder Kinder, Zeitschrift für Bevölkerungswissenschaft, 16, 207-236. Schwartz, W., Schwarz, T., & Vogel, C. (1991). Mütter und Väter zwischen Erwerbsarbeit und Familie: Probleme – Praxisbeispiele – Orientierungshilfen. Stuttgart: Ministerium für Arbeit, Gesundheit, Familie und Frauen. Baden-Württemberg. Shaver, S., & Bradshaw, J. (1995). The recognition of wively labour by welfare states. Social Policy and Administration, 29, 10-25. Shelton, B. A., & John, D. (1996). The Division of Household Labor. Annual Review of Sociology, 22, 299-322. Siaroff, A. (1994). Work, welfare and gender equality: A new typology. In D. Sainsbury (Ed.), Gendering welfare states (pp. 82-100). London: Sage. Sørensen A., & McLanahan, S. (1987). Married women’s economic dependency, 1940-1980. American Journal of Sociology, 93, 659-687. Statistisches Bundesamt (Ed.) (1976). Bevölkerung und Erwerbstätigkeit. Stand und Entwicklung der Erwerbstätigkeit 1975. Fachserie 1 Reihe 4.1.1. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1982). Bevölkerung und Erwerbstätigkeit. Stand und Entwicklung der Erwerbstätigkeit 1981. Fachserie 1 Reihe 4.1.1. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1987). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1983. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1988a). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1984. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1988b). Bevölkerung und Erwerbstätigkeit. Stand und Entwicklung der Erwerbstätigkeit 1987. Fachserie 1 Reihe 4.1.1. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1989). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1985. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1990). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1986. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1991a). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1987. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1991b). Zeitverwendung der Personen in Arbeiter- und Angestelltenhaushalten im Gebiet der ehemaligen DDR 1974, 1980, 1985 und 1990. Wiesbaden: Statistisches Bundesamt. Statistisches Bundesamt (Ed.) (1992a). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1988. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1992b). Bevölkerung und Erwerbstätigkeit. Stand und Entwicklung der Erwerbstätigkeit 1991. Fachserie 1 Reihe 4.1.1. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1992c). Sozialleistungen. Einrichtungen und tätige Personen in der Jugendhilfe 1990. Fachserie 13 Reihe 6.3. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1992d). Statistisches Jahrbuch 1992 für die Bundesrepublik Deutschland. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1993a). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1989. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1993b). Sozialleistungen. Einrichtungen und tätige Personen in der Jugendhilfe 1991 in den neuen Ländern und Berlin (Ost). Fachserie 13 Reihe 6.3.1. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1994). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1990. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel.

108

Statistisches Bundesamt (Ed.) (1995a). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1991. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1995b). Die Zeitverwendung der Bevölkerung. Ergebnisse der Zeitbudgeterhebung 1991/92. Allgemeiner Überblick. Tabellenband II. Wiesbaden: Statistisches Bundesamt. Statistisches Bundesamt (Ed.) (1996a). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1992. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1996b). Bevölkerung und Erwerbstätigkeit. Stand und Entwicklung der Erwerbstätigkeit 1995. Fachserie 1 Reihe 4.1.1. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1996c). Sozialleistungen. Tageseinrichtungen für Kinder 1994. Fachserie 13 Reihe 6.3.1. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1997). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1993. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1998). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1994. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1999a). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1995. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (1999b). Statistisches Jahrbuch 1999 für die Bundesrepublik Deutschland. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (2000a). "87% der Teilzeitbeschäftigten sind Frauen". Press release by March 21, 2000. http://www.statistik-bund.de/presse/deutsch/pm2000/p1040031.htm. Statistisches Bundesamt (Ed.) (2000b). "1999 bestanden 222.000 Hochschulabsolventen ihre Abschlussprüfung – Weiterhin Rückgang bei den angehenden Ingenieuren, Informatikern, Physikern und Chemikern –". Press release by September 25, 2000. http://www.statistik-bund.de/presse/deutsch/pm2000/p3420071.htm. Statistisches Bundesamt (Ed.) (2000c). "Am Ende des Schuljahres 1998/99 höchste Zahl von Abiturienten seit der Wiedervereinigung". Press release by April 7, 2000. http://www.statistik-bund.de/presse/deutsch/pm2000/p1310071.htm. Statistisches Bundesamt (Ed.) (2000d). Bevölkerung und Erwerbstätigkeit. Entgelte und Beschäftigungsdauer der sozialversicherungspflichtig Beschäftigten 1996. Fachserie 1 Reihe 4.2.2. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (2000e). Bevölkerung und Erwerbstätigkeit. Haushalte und Familien 1999 (Ergebnisse des Mikrozensus) — Vorbericht. Wiesbaden: Statistisches Bundesamt. Statistisches Bundesamt (Ed.) (2000f). Bevölkerung und Erwerbstätigkeit. Stand und Entwicklung der Erwerbstätigkeit 1999. Fachserie 1 Reihe 4.1.1. Stuttgart: Metzler-Poeschel. Statistisches Bundesamt (Ed.) (2000g). Datenreport 1999. Zahlen und Fakten über die Bundesrepublik Deutschland. Bonn: Bundeszentrale für politische Bildung. Statistisches Bundesamt (2001). Personal communication, March 1, 2001. Szalai, A. (1972). The use of time. Daily activities of urban and suburban populations in twelve countries. In Cooperation with Philip E. Converse, Pierre Feldheim, Erwin K. Scheuch, and Philip J. Stone. The Hague, Paris: Mouton. The British Council. (1996). International guide to qualifications in education, 4th ed., London: Mansell Publishing Limited. Tietze, W. (1993). Institutionelle Erfahrungsfelder für Kinder im Vorschulalter. Zur Entwicklung vorschulischer Erziehung in Deutschland. In W. Tietze & H.-G. Roßbach (Eds.), Erfahrungsfelder in der frühen Kindheit: Bestandsaufnahme, Perspektiven (pp. 98-125). Freiburg: Lambertus. Tietze, W. (1998). Zur Versorgung mit Kinderbetreuungsangeboten. In Fthenakis, W. E. & Textor, M. R. (Eds.), Qualität von Kinderbetreuung: Konzepte, Forschungsergebnisse und internationaler Vergleich (pp. 37-50). Weinheim: Juventus.

109

Tietze, W. & Roßbach, H.-G. (1993). Das Früherziehungssystem in der Bundesrepublik Deutschland (alte Bundesländer). In W. Tietze & H.-G. Roßbach (Eds.), Erfahrungsfelder in der frühen Kindheit: Bestandsaufnahme, Perspektiven (pp. 126-212). Freiburg: Lambertus. Vaskovics, L., & Rost, H. (1999). Väter und Erziehungsurlaub. Stuttgart: Kohlhammer. Voegeli, W., & Willenbacher, B. (1984). Die Ausgestaltung des Gleichberechtigungssatzes im Eherecht. Zeitschrift für Rechtssoziologie, 5, 235-259. Wagner, G. G., Hank, K., & Tillmann, K. (1995). Außerhäusige Kinderbetreuung in Ostdeutschland — 1990 und 1994 im Vergleich zu Westdeutschland. Diskussionspapiere 95-18. Bochum: Ruhr-Universität Bochum. West, C., & Fenstermaker, S. (1993). Power, inequality and the accomplishment of gender: An ethnomethodological view. In P. England (Ed.), Theory on gender/feminism on theory (pp. 151-174). New York: de Gruyter. West, C., & Fenstermaker, S. (1995). Doing difference. Gender & Society, 9, 8-17. West, C., & Zimmerman, D. H. (1987). Doing gender. Gender & Society, 1, 125-151. Willemsen, T. (1997). European network on policies and the division of unpaid and paid work: Survey questionnaire. WORC Paper 97.06.003/6. Tilburg: Work and Organization Research Centre. Wirth, S. (1996). Estimating discrimination: A new approach. Paper presented at the meeting of the Econometric Association in Helsinki. Hohenheim: unpublished paper. Zentralarchiv für empirische Sozialforschung an der Universität zu Köln (ZA) (1999). ALLBUS 1980-1998. Allgemeine Bevölkerungsumfrage der Sozialwissenschaften — German General Social Survey [CDROM].

110

Appendix

111

112

Appendix-Table A1:

Germany in the modernization of gender relations, 1980s-1990s

Area

Indicator

1. State de-familialization

1.1 Public spending on family services (% GDP) 1.2 Public day care coverage (% < 3) 1.3 Home care coverage (% aged) 1.4 Net, post-transfer/tax cost of child care as a percentage of average family income 2.1 Percentage of aged living with children 2.2 Unemployed youth living with parents as a share of total 2.3 Weekly unpaid hours, women

2. De-familialization within households

3. Welfare-state incentives/disincentives for working mothers

4. Female work desirability

3.1 Child benefits as a percentage of average worker’s income

113

3.2 Percentage benefit loss to unemployed person if spouse works 3.3 Percentage extra marginal tax if wife works 4.1 Female/male percentage in unemployment 4.2 Female/male percentage in labour force participation 4.3 Female/male percentage of industrial wages 4.4 Female/male percentage in management positions

Year

West Germany

Relative position / (range)

East Germany

Relative position

13 / 18 (.04 – 2.57)

-

-

7 / 17 (1 – 48) 4 / 18 (1 – 24)

-

-

4 / 8 (39.3 – 9.4)

-

-

1992

.54

1980s 1990

3 2

1990s

19.4

1980s

14

7 / 11 (65 – 4)

-

-

1991-3

11

8 / 9 (63 – 8)

-

-

19851990

35

5 / 12 (45.8 – 24.6)

-

-

1990

4.6

5 / 17 (11.4 – .7)

-

-

1991

10

5 / 16 (100 – 0)

-

-

1992

5

5 / 15 (14 – 0)

-

-

1980s

130

10 / 23 (247 – 60)

-

-

1980s

64

12 / 23 (41 – 91)

-

-

1980s

73

14 / 23 (52 – 90)

-

-

1980s

20

13 / 21 (6 – 61)

-

-

114

Area

Indicator

Year

4.5 Female/male percentage of 1980s post-secondary students 4.6 Multiplied score of the indicators for 1980s this area 5. Family welfare orientation Overall average of different ranks33 1980s 6. Recognition of wives’ 6.1.a Support for a non-earning wife at 1992 labour half of average male earnings34 6.1.b Support for a non-earning wife 1992 equivalent to average male earnings 6.1.c Support for a non earning wife at 1992 1..5 times average male earnings 6.2.a Support for a non-earning wife with 1992 1 child35 6.2.b Support for a non-earning wife with 1992 3 children 6.3 Support for a working wife36 1992 7. Policies that support 7.1 Policies for mothers with children 1984-7 employment for mothers37 under six 7.2 Policies for mothers with school-aged 1984-7 children 8. Level of social care services 8.1 Public day care coverage (% < 3) 1985/6 8.2 Public day care coverage (% > 3 < 6) 1985/6 8.3 People over 65 in institutional care 1985/6 (% age-group) 8.4 People over 65 receiving home help 1985/86 services (% age-group) 9. Economic intervention Public spending on families (% GDP) 1995 10. Ecological intervention Public day care coverage (% < 3) around 1990

West Germany

Relative position / (range)

East Germany

Relative position

72

5 / 23 (47 – 113)

-

-

.40

10 / 23 (.25 – .65)

-

-

3.0

10 / 23 (4.75 – 1.25)

-

-

8.3

7 / 15 (20.2 – 3.1)

-

-

11.9

4 / 15 (15.2 – 3.1)

-

-

13.9

4 / 15 (25.2 – 2.2)

-

-

17.3

5 / 15 (20.0 – 4.9)

-

-

36.0

4 / 15 (57.1 – 8.2)

17.0

4 / 15 (26.1 – -14.5)

-

-

34.1

7 / 14 (17.1 – 64.9)

-

-

32.6

1 / 6 (32.6 – 57.0)

-

-

3.0 60.0

6 / 14 (.5 – 44.0) 6 / 14 (25 – 95)

-

-

6.0

10 / 14 (.5 – 10)

-

-

3.0

4 / 13 (1 – 20)

-

-

10 / 23 (2.71 – .12)

see left

1.31

38

2.97

4 / 23 (2.17 – 58.28) 58.28

23 / 23

Area

Indicator

Year

West Germany

Relative position / (range)

East Germany

Relative position

11. Educational attainment

11.1 Male/female ratio (log.) of post-secondary education in younger 1990s .32 6 / 23 (1.37 – -.18) .07 12 / 23 cohorts 11.2 Difference between the respective 1990s -.74 18/ 23 (-013 – -1.23) -1.07 22 / 23 ratio in older cohorts and the former ratio 12.1 Average gender role attitudes in 12. Gender role attitudes 1994 15.9 8 / 21 (14.3 – 19.1) 18.4 19 / 21 younger cohorts 12.2 Ratio (log.) of the former indicator 1994 .25 21 / 21 (.07 – .25) .07 1 / 21 to the respective average in older cohorts 13.1 Male/female ratio (log.) of labour 13. Labour force participation 1995 .305 9 / 23 (.832 – .0) .069 21 / 23 force participation 13.2 Difference between the former ratio 1995 / .275 6 / 23 (.103 – .908) .103 1 / 23 and the respective ratio 22 years ago 1973 14. Division of household 14.1 Average z-transformed ratios of 1982-97 .11 9 / 23 (2.13 – -1.10) -.71 18 / 23 labour division of household labour 14.2 Difference between the former ratio 1982-97 / .27 15 / 19 (-1.52 – .83) .18 13 / 19 and the respective ratio for 1961-1981 1961-81 Note: Relative position is given as the rank of West and East Germany, respectively, on a scale ranging from a traditional to a modern pole in relation to the number of countries compared. Therefore, low ranks indicate traditionalism, high ranks modernism. The range is represented with the value of the most traditional and the value of the most modern country in the respective data-set. The order in which the range is given reflects the traditional-modern poles, not in every case lower and higher values. E.g., in the section on de-familialization in households, high values indicate traditionalism and vice versa. Sources: Esping-Andersen 1999, ch. 4 (the first three areas); Siaroff, 199439 (area 4-5); Shaver and Bradshaw, 1995 (area 6); Gornick, Meyers, & Ross, 1996 (area 7); Anttonen & Sipilã, 1996 (area 8); Künzler, 1999a (area 9ff.). The countries compared are in most cases members of the EU (or their predecessors EEC, EC) or the OECD or sub-samples of them. See the references for further information.

115

Appendix-figure A1: Rate of income tax in Germany, 1999

Note: Marginal tax rate (Grenzbelastung, grenz99), tax rate (Durchschnittsbelastung, durch99) in percentages, respectively. Source: http://www.bundesfinanzministerium.de.

Appendix-Table A2: Tax credit for first from child 1946 400 1948 600 1953 600 1954 600 1955 720 1957 720 1958 900 1959 900 1961 900 1962 1,200 1964 1,200 1970 1,200 1975 abolished 1978  1979  1981 

116

Development of child benefit/tax credit in the FRG, 1946-2000

Tax credit for second child 400 600 600 600 720 1,440 1,680 1,680 1,680 1,680 1,680 1,680 abolished   

Tax credit for third child, etc. 400 600 740 840 1,680 1,680 1,800 1,800 1,800 1,800 1,800 1,800 abolished   

Benefit for first child             50 50 50 50

Benefit for second child         25 25 25 25 70 80 100 120

Benefit for third child     25 30 30 40 40 40 50 60 120 150 200 240

Benefit for fourth child     25 30 30 40 40 40 60 60 120 150 200 240

Benefit for fifth child, etc.     25 30 30 40 40 40 70 70 120 150 200 240

Tax Tax Tax Benefit Benefit Benefit Benefit Benefit credit credit credit for for for for for for for for third first second third fourth fifth first second child, child child child child child, etc. from child child etc. 1982 50 100 220 240 240    1983 432 432 432 50 100 (70) 220 (140) 240 (140) 240 (140) 1986 2,484 2,484 2,484 50 100 (70) 220 (140) 240 (140) 240 (140) 1990 3,024 3,024 3,024 50 130 (70) 220 (140) 240 (140) 240 (140) 1992 4,104 4,104 4,104 70 130 (70) 220 (140) 240 (140) 240 (140) 1994 4,104 4,104 4,104 70 130 (70) 220 (70) 240 (70) 240 (70) 1996 6,264 6,264 6,264 200 200 300 350 350 1997 6,912 6,912 6,912 220 220 300 350 350 1998 6,912 6,912 6,912 250 250 300 350 350 2000 9,936 9,936 9,936 270 270 300 350 350 Note: Figures in DM (before 1949 in Reichsmark), only years with at least one change, changes in bold, base figures in parentheses. The child benefit for the second child was means-tested between 1961 and 1974. Income limits (in DM): 600 (1961-1964), 650 (1964-8/1970), 1,100 (9/1970-1971), 1,250 (1972), 1,400 (1973), 1,530 (1974). The child benefit for the second child and additional children were partially means-tested between 1983 and 1995 and were gradually reduced to the base figure given in parentheses depending on income. The tax credit in 2000 includes a child care tax credit (see below), this tax credit is valid for children under 16 or disabled children without age-limit. For older children (16+) it is still 6,912 DM. Sources: Bundesministerium für Arbeit und Sozialordnung, 1998; Gerlach, 1996, pp. 205-210; http://www.bmfsfj.de; http://www.bundesfinanzministerium.de; Lampert, 1998, pp. 339f.; Münch, 1990, pp. 77-79.

117

Appendix-Table A3:

Comparative advantages of child benefit/tax credit in the FRG, 1996-2000

Child tax credit

Benefit for first child

Range of marginal tax in %

1996

6,264

200

25.9 – 53.0

38.3

72,000 / 144,000

77

38.5

1997

6,912

220

25.9 – 53.0

38.2

72,000 / 144,000

85

38.6

1998 1999 2000

6,912 6,912 9,936

250 250 270

25.9 – 53.0 23.9 – 53.0 22.9 – 51,0

43.4 43.4 32.6

74,490 / 146,016 91,908 / 180,468 50,922 / 96,984

55 55 152

22.0 22.0 56.3

Year

Marginal Income starting Maximum Maximum tax rate tax credit advantage advantage starting advantage in % tax credit (single/married) advantage in %

Note: Own calculation, figures in DM or per cent, figures for tax credit and income on an annual basis, for benefit and maximum advantage on a monthly basis. Tax credit advantage is the gain in net income if tax credit is applied in relation to the benefit for the first child. Maximum advantage is the difference between the allowance by tax credit on the highest marginal tax and the child benefit; maximum advantage in per cent is this difference in relation to the child benefit. Sources: For child benefit and tax credit, see Appendix-Table A2; for other aspects: http://www.bundesfinanzministerium.de.

Appendix-Table A4:

Household type

All

Incidence and Germany, 1995 Recipients Germany (%)

duration

3.4%

Recipients East Germany (%) 1.9%

Recipients West Germany (%) 3.8%

.7%

.2%

of

social

Duration of support Germany

assistance

32.0

Duration of support East Germany 13.6

Duration of support West Germany 34.1

.8%

29.7

11.5

30.9

Married couples without children Married couples with children - with 1 child

2.1%

1.1%

2.3%

20.3

11.8

21.2

1.6%

.7%

1.8%

18.3

10.3

19.1

- with 2 children

1.9%

1.0%

2.1%

18.9

11.3

19.8

- with three or more children

4.6%

3.9%

4.7%

24.5

14.1

25.6

Lone mothers with children

25.1%

12.5%

29.7%

27.2

15.2

29.0

- with 1 child

21.6%

9.7%

26.1%

25.6

13.6

27.2

- with 2 children

28.8%

15.3%

33.4%

28.7

15.9

30.7

- with three or more children 44.3%

31.2%

48.6%

31.0

19.1

33.3

118

in

Note: Figures from late 1995; children are defined as minors (under 18); duration of support: the average of the respective group in months of uninterrupted assistance. Source: Engstler, 1998, p. 172.

119

120

Appendix-Table A5: Policy function 1. Paternalist policies

Family policy measures in the GDR

Policy area Marriage benefits

Policies Description Family-formation credit From 1972 on, employed spouses who married before reaching the age of 30 years were entitled to an interest-free loan of up to 7,000 GDR-Marks (M) for homebuilding and equipment (Helwig, 1987, pp. 67f.; Lampert, 1991). Repayment was to take place within 8 to 11 years, but the amount could be reduced if a couple had children (1,000 M for the first child, 1,500 M for the second child, 2,500 M for the third child). The GDR slang term for this was "abkindern", which is difficult to translate, but means something like "down-childing" or "reducing with children". Taxation For married couples, 50 M were deducted from the monthly income tax. Other marriage benefits Spouses who were not employed were entitled to provisions of the statutory insurance. The pension was supplemented with 150 M monthly, if his or her spouse did not have a pension of his or her own. A widow/er received a pension if the deceased spouse was entitled to one. University students who were married (or had children) were entitled to support from the university. Accommodation Marriage entitled spouses to their own apartment in the scarce housing market and thereby enabled them to move out of their parents’ home. However, this was a practice, not a measure designed as family policy. Family benefits Birth subsidy From 1950 on, mothers who attended courses offered by an agency for pregnancy and motherhood counselling were entitled to the benefit (see Appendix-Table A6). Child benefit From 1950 on, parents received a child benefit (see Appendix-Table A7). Other family benefits Supplements were paid in health insurance benefits depending on the family status. Starting from the seventh week of illness, employees received the following percentages of their net income if they earned less than 600 M / 600 M or more: 70% / 50% with no or one child, 75% / 65% with two children, 80% / 75% with three children, 85% / 80% with four children and 90% / 90% with five or more children. Starting in 1987 there were grants for pupils in the 11th (110 M) and 12th grades (150 M). Students were entitled to a basic grant (200 M), which was supplemented for every child of their own (50 M). Additionally, there were special grants based

Policy function

Policy area

2. Maternalist policies

Lone mother benefits Maternity protection Parental leave

Policies

121

Description on achievement. Lone mothers studying at a university who cannot place their child in a day care facility after birth are granted an allowance of 125 M (1 child), 150 M (2 children) or 175 M (3 or more). There were several supplements and additional provisions in some of the policies described in this table (see above and below). Pregnancy leave Expecting mothers had 6 weeks pregnancy leave (Schwangerschaftsurlaub) before Maternity leave birth and 20 weeks maternity leave (Wochenurlaub) afterwards (22 weeks in case Maternity leave benefit of multiple or complicated births) for which the net-income was paid by social insurance. After-maternity leave Mother or father could take a parental leave after maternity leave (bezahlte Freistellung nach Ablauf des Wochenurlaubs). It would last until the end of the first period year for the first and second child, until 18 months after birth for the third and following children, or until the second (third) year in case of twins (triplets). After-maternity leave The parental leave benefit was paid as a mothers allowance comparable to the benefit health insurance benefit (see above: other family benefits), depending on family status between 65% and 90% of net income, at least 250 M for one, 300 M for two or 350 M for three and more children. Child care leave If a place in a day care facility for infants (Krippen) could not be provided, the mother or other primary carer who had previously been employed, could take a leave lasting up until the child’s third birthday. There was no payment in this period. Lone mothers received the parental leave benefit (see above). Mothers who gave birth to another child within this period received an allowance of 200 M monthly. Sickness leave Married mothers with at least two children or one disabled child, as well as lone mothers with a child up to 14 years were entitled to a sickness leave of four weeks annually with one child, six weeks with two children, eight weeks with three, ten weeks with four and 13 weeks with five or more children. Sickness leave benefit The allowance was 90% of net income for the first two days, and 65%-90% (health insurance benefit, see above: other family benefits) for the following days. Care of disabled child For the necessary care of a disabled child, the primary care giver who had to benefit terminate his or her regular employment received 200 M monthly as long as the care was provided.

122

Policy function

Policy area Child day care

Policies Provision of day care

Work schedule

Reduction of working time Housework day Vacation supplement

Old-Age Insurance

Care periods under old-age insurance

Accommodation 3. Socialist policies

Egalitarianism

Description For every child for whom day care was sought, a place in a day care facility was available from birth until the completion of primary school (see Section 2.2.3; Appendix-Table A9ff.). Mothers (and lone fathers) who were employed full-time and had at least two children or one disabled child had to work only 40 hours a week (instead of 43 ¾ hours) without loss of income. Women who were married and/or had minor children and/or were at least 40-years-old, as well as lone fathers and husbands whose wives needed care were entitled to a day off every month if they worked full-time. Mothers who worked shifts, as well as lone fathers who had two or more children or one disabled child, respectively, were entitled to from two to five additional vacation days (base vacation: 22 days). The time in which a mother received the motherhood allowance (see above: parental leave benefit) was equal to the time of employment with statutory insurance. For every child, the mother received one year of old-age insurance. A pension supplement of 45 M was paid for every child. There were other provisions for mothers with more than two / five children, widows and orphans (see for details Lampert, 1991, p. 126). A rental allowance was paid to families with four or more children and lone parents with three or more children. Products and food were subsidized.

Subsidies and low indirect taxes Surveillance Meticulous control of The ideal of the socialist personality was enforced. the up-bringing of children and youth Notes: Own compilation. The family policy measures of the former GDR are categorized in a similar way as for the FRG. This may be disputed because the underlying gender ideology is rather different. The combined women and family ideal in the GDR was comprised of a combination of motherhood and full-time employment. Therefore many measures were to enable mothers’ labour force participation and/or had a natalist intention. All figures are monthly and in the currency of the former GDR, the Mark. Usually, the regulations that were in effect shortly before unification are given. Sources:

Gerlach, 1996; Helwig, 1987; Lampert, 1991.

Appendix-Table A6: Year 1950 1958 1972 Note:

Development of birth subsidy in the GDR, 1950-1990

First child Second child Third child 100   500 600 700 1,000 1,000 1,000 Figures in M, only years with changes in the measure.

Fourth child 250 850 1,000

Fifth child, etc. 500 1,000 1,000

Source: Gerlach, 1996, pp. 267-271.

Appendix-Table A7:

Development of child benefit in the GDR, 1950-1990

First Second Third Fourth Fifth child child child child child, etc. 1950 Up to 14 years 20 25    1978 Up to 14 years 20 20 50 60 70 1981 Up to 14 years 20 20 100 100 100 1987 Until end of secondary school 50 100 150 150 150 1990 Up to 12 years 95 145 195 195 195 Between 12 and under 16 years 115 165 215 215 215 Note: Monthly figures in M, only years with changes in the measure. Supplements were paid for children of university students and of parents in vocational training (until 1989: 60 M, 1990: 110 to 160 M) and for lone parents (25 M). Year

Sources:

Age of child

Gerlach, 1996; Helwig, 1987; Lampert, 1991.

123

Appendix-Table A8:

Year

Development of the use of day care in Germany, 1990-1999

Infants Pre-school-children School children West East West East West East West East West East West East Germany Germany Germany Germany Germany Germany Germany Germany Germany Germany Germany Germany full-time full-time full-time full-time full-time full-time

6.0 62.0 2.0 52.0 82.0 98.0 21.0 80.0 2.0 35.0 2.0 35.0 1990 63.0 43.0 94.0 70.0 31.0 31.0 1991       57.0 46.0 93.0 72.0 1992         50.0 38.0 91.0 74.0 1993         30.0 22.0 90.0 68.0 1994         6.0 21.0 1.0 14.0 80.0 92.0 15.0 60.0 3.0 27.0 3.0 27.0 1995 23.0 12.0 88.0 47.0 30.0 30.0 1996       33.0 15.0 87.0 45.0 27.0 27.0 1997       8.0 27.0 1.0 14.0 91.0 85.0 20.0 51.0 5.0 30.0 5.0 30.0 1998 34.0 15.0 89.0 56.0 24.0 24.0 1999       Notes: Use of day care on the basis of the question of whether children are in day care or with a child-minder (German Socio-Economic Panel, GSOEP); proportion of children in day care to the number of children in the age-group in per cent. Infants: children 0-3, pre-school children: 4-6, school children: 7-11. German population. For methodological issues see note 23. Missing values: not reported in the sources, not included in the data-sets, or not interpretable because of low numbers.

Source: Hank, Tillmann, & Wagner, 2001.

124

Appendix-Table A9:

Development of day care facilities for infants under three (Krippen) in Germany, 1950-1998

Year

West Germany West Germany East Germany East Germany places coverage level (%) places coverage level (%) 1950 7,491 .4 8,542 1.3 1955 16,043 .7 67,106 9.1 1960 18,351 .7 104,781 14.3 1965 18,108 .6 142,242 18.7 1970 17,457 .7 183,412 29.1 1975 24,251 1.3 242,553 50.1 1980 26,104 1.5 289,550 61.2 1985 28,353 1.6 343,787 72.7 1989 / / 535,203 80.2 1990 38,153 1.8 / / 1991 / / 255,280 54.2 1994 47,064 2.2 103,689 41.3 1998 58,475 2.8 108,452 36.3 Note: Coverage level is the ratio of number of places to the number of children in the age-group in percentages. For methodological issues see note 23. Sources: West Germany: Bundesministerium für Familie, Senioren, Frauen und Jugend, 1998, p. 200; Statistisches Bundesamt, 1992c, 1996c; Statistisches Bundesamt, personal communication, March 1, 2001; Tietze, 1993, p. 114; East Germany: 1950-1985: Gerlach, 1996, p. 249; Statistisches Bundesamt, 1993b; 1994: Statistisches Bundesamt, 1996c; 1998: Statistisches Bundesamt, personal communication, March 1, 2001.

125

Appendix-Table A10: Coverage level of day care facilities for infants under three (Krippen) in the federal states of Germany, 1990-1998 Change 1990/1 1994 1998 % % % 1998-1990/1 Baden-Württemberg 1.1 1.2 1.3 115 The free state of Bavaria .9 1.0 1.4 154 Berlin (West) 17.9 19.1 23.4 120 The free hanseatic city of Bremen 2.0 6.4 6.8 331 The free hanseatic city of Hamburg 9.8 11.9 11.7 120 Hesse 1.8 2.1 2.6 144 Lower-Saxony 1.6 1.5 1.8 115 North Rhine-Westphalia .9 1.5 2.5 272 Rhineland-Palatinate .5 .9 1.4 248 Saarland .8 1.7 2.5 276 Schleswig-Holstein .7 1.4 2.3 312 West Germany 1.8 2.2 2.8 153 Berlin (East) 70.1 54.4 52.4 48 Brandenburg 64.6 54.1 51.9 53 Mecklenburg-Western Pomerania 50.0 39.0 30.8 36 The free state of Saxony 51.9 32.8 24.1 30 Saxony-Anhalt 43.4 42.9 47.2 66 The free state of Thuringia 54.0 36.4 25.9 31 East Germany 54.2 41.3 36.3 42 Note: Coverage level is the ratio of number of places to the number of children in the age-group in percentages. For methodological issues see note 23. The change indicator is calculated by setting the number for 1990/1 = 100.

East Germany

West Germany

Federal state

Sources: Künzler, 1998, table 7.4 (see for further reference); for 1998: Statistisches Bundesamt, personal communication, March 1, 2001.

126

Appendix-Table A11: Day care facilities for infants under three (Krippen) in Germany Dimension Availability

Overall situation in West Germany Krippen are provided by the local municipalities and the NPOs (Tietze & Roßbach, 1993, pp. 142f.). The coverage level is with ca. 2-3% very low; West Germany ranks in the lower half in most international comparisons (Appendix-Table A1, indicators 1.2, 8.1, 10; Appendix-Table A9).

Costs



Differences in supply In West Germany, there is great regional variation in the coverage level (Appendix-Table A10). The city-states (Berlin, Hamburg, Bremen) have the highest level; in the other federal states, this pattern of concentration of institutional day care for infants under three in larger cities is also confirmed (Kreyenfeld & Hank, 1999, p. 6; Liegle, 1990, p. 163; Tietze, 1993, pp. 113f.). Krippen are seen as a welfare measure. Priority is given to the children of lone mothers and those from low-income families. There are still great differences in supply between West and East Germany (Appendix-Table A1). It has been suggested (Bundesministerium für Familie, Senioren, Frauen und Jugend, 1998, p. 200; Tietze, 1998, pp. 44f.) that day care for infants was reduced because suppliers tried to comply with the entitlement to day care for pre-school children (see p. 31). In West Germany, there are great variations in the costs parents have to pay. In the late 1980s, they were between 80 DM and 375 DM per child and month in most cases, with possible maximums up to 900 DM. Some federal states subsidize the Krippen, some do not (Tietze & Roßbach, 1993, p. 144). Recent information on the differences between East and West are unavailable.

Overall situation in East Germany Prior to unification, there were places for every child for whom day care was sought. Official statistics of the GDR stated a coverage level of 80% in 1989 (Wagner et al., 1995; see Appendix-Table A9). Although the supply has been reduced in the process of transformation, the level is still high.

In the GDR, costs were negligible. Parents had to pay 1.40 Marks for food services (Wagner et al., 1995). Recent information is unavailable.

127

128

Dimension Overall situation in West Germany Time schedule In the late 1980s, 87% of the Krippen were open at least 8 hours, ca. three quarters of all Krippen opened between 6 and 7 a.m. and closed between 5 and 6 p.m. (Tietze & Roßbach, 1993, p. 146). Usually, the Krippen are (at least partly) closed during school holidays. Quality The task of child-minding for this age-group is ascribed to the family or — more precisely: — to the mother. Therefore, almost no quality standards, pedagogical concepts, or legal regulations exist for these institutions. Sources: See in Table (above).

Differences in supply 

Overall situation in East Germany In the GDR, Krippen were usually open from 6 a.m. to 6 p.m. (Wagner et al., 1995).

In West Germany, children-staff-ratios vary between 6 to 1 and 4 to 1, the size of the groups between 8 to 14 children (Liegle, 1990, p. 161; Tietze & Roßbach, 1993, p. 145). Recent information on the development of quality standards are unavailable.

After unification, day care for infants in the former GDR was widely criticized because of unacceptable conditions and their function as a means of transmission of socialist ideology.

Appendix-Table A12: Development of day care coverage for pre-school children between three and six (Kindergarten) in Germany, 1950-1998 Year West Germany (%) East Germany (%) 1950 29.1  1955 29.4 34.5 1960 28.1 46.1 1965 28.0 52.8 1970 32.9 64.5 1975 56.1 84.6 1980 67.5 92.2 1985 67.7 [89.9] 94.0 1989 / 95.1 1990 69.0 / 1994 73.0 96.2 1998 86.8 111.8 Note: West Germany: Coverage level is the ratio of number of places to number of children in the age-group three to six and a half in percentages. East Germany: Coverage level is the ratio of number of places to the number of children in the age-group three to under six plus ¾ of the sixyear-olds (for 1955-[1985]), plus seven twelfths of six-year-olds (1985-1989) in percentages, since 1994 analogous to West Germany. For methodological issues see note 23. Sources: West Germany: Bundesministerium für Familie, Senioren, Frauen und Jugend, 1998, p. 200; Statistisches Bundesamt 1996c; Statistisches Bundesamt, personal communication, March 1, 2001; Tietze & Roßbach, 1993, pp. 132f.; East Germany: Gerlach, 1996, p. 250; Statistisches Bundesamt, 1996c; Statistisches Bundesamt, personal communication, March 1, 2001.

129

Appendix-Table A13: Coverage level of day care facilities for pre-school children between three and six (Kindergarten) in the federal states of Germany, 1990-1998 1990/1 1994 1998 Change % % % 1998–1990/1 Baden-Württemberg 104 108 125 128 The free state of Bavaria 72 88 97 144 Berlin (West) 65 60 91 120 The free hanseatic city of Bremen 75 76 96 131 The free hanseatic city of Hamburg 52 59 76 156 Hesse 90 91 108 129 Lower-Saxony 66 75 90 153 North Rhine-Westphalia 75 74 96 134 Rhineland-Palatinate 98 106 120 130 Saarland 95 98 115 115 Schleswig-Holstein 64 76 90 157 West Germany 80 85 102 136 Berlin (East) 117 118 91 41 Brandenburg 124 118 127 42 Mecklenburg-Western Pomerania 105 109 126 42 The free state of Saxony 98 118 135 49 Saxony-Anhalt 98 112 129 54 The free state of Thuringia 125 124 153 49 East Germany 114 117 132 47 Note: Coverage level is the ratio of the number of places to the number of children in the age-group three to six years old in percentages. For methodological issues see note 23. The difference in the means for West and East Germany compared to the respective figures given in Appendix-Table A12, stems from the difference in the number of children taken into account (three to six years old in this table vs. three to six ½ in the other one), for East Germany partly due to the difference in reference year (1991 in this table, 1989 in the other). The change indicator is calculated by setting the number for 1990/1 = 100.

East Germany

West Germany

Federal state

Sources: Künzler, 1998, table 7.5 (see for further reference); Statistisches Bundesamt, 1992c; Statistisches Bundesamt, 1993b; Statistisches Bundesamt, personal communication, March 1, 2001.

130

Appendix-Table A14: Opening hours of day care facilities for pre-school children between three and six (Kindergarten) in Germany, 1994 Category Germany West Germany East Germany Full-time with lunch 31.6 14.4 113.2 Morning and afternoon without lunch 41.4 50.0 0.2 Morning without lunch 13.7 16.3 1.0 Morning with lunch 1.3 1.1 2.4 Afternoon without lunch 2.7 3.3 .0 Afternoon with lunch .1 .1 .0 Note: Figures are a proportion of the respective category in percentages of all places in day care for this age-group in 1994. Source: Statistisches Bundesamt, 1996c.

131

132

Appendix-Table A15: Day care facilities for pre-school children between three and six (Kindergarten) in Germany Dimension Availability

Costs

Overall situation in West Germany Kindergartens are provided by the local municipalities and the NPOs. Recent data on coverage levels are unavailable, because the official statistics have not been released. The comparatively low levels found until the early 1990s (Appendix-Table A1, indicator 8.2; Appendix-Table A12) should have been improved (see also Appendix-Table A8), since an entitlement to a Kindergarten place for all children between three and six (see p. 31) has been in effect (1996). Kindergarten are subsidized by the federal states, the municipalities and the NPOs. The subsidies differ widely throughout the federal states (Künzler, 1998, pp. 119f.). A NPO regularly received 90% of the fixed costs for maintaining a facility.

Differences in supply In West Germany, there was regional variation in the coverage level (AppendixTable A13) with the typical town-country slope. (Tietze & Roßbach, 1993, p. 135). This should have decreased since 1996.

Overall situation in East Germany Prior to unification, there were places for every child for whom day care was sought. Official statistics of the GDR stated a coverage level of 95% in 1990 (see Appendix-Table A12). The supply of overall day care is now at a level comparable to the one in the West. As in the West, Kindergarten are nowadays provided by the local municipalities and the NPOs.

The costs to parents depend on how highly the Kindergarten is subsidized as well as on their income and the number of children. At the end of the 1980s, they were between 0 DM and 400 DM per child and month for full-time day care in North Rhine-Westphalia (Tietze & Roßbach 1993, pp. 150ff.). Recent information on the differences between East and West are unavailable.

In the GDR, costs were negligible. Parents had to pay for food supply. Recent information on the situation in East Germany is unavailable.

Dimension Overall situation in West Germany Time schedule Most West German Kindergartens are open in the morning, close at noon and provide an optional period in the afternoon (see Appendix-Table A14) Usually, the Kindergarten are (at least partly) closed during school holidays.

Overall situation in East Germany In the GDR, the regular Kindergarten was open full-time from 6 a.m. to 7 p.m. (Liegle, 1990, p. 162). In 1994, the East German Kindergarten were overwhelmingly full-time (Appendix-Table A14). The data in Appendix-Table A8 indicate that this standard was maintained for some time, but full-time places have been reduced in recent years.

Quality

Critique of the insufficient equipment and qualification of staff in the former GDR was similar to the assessment of the Krippen, although standards have been raised since the rapid expansion in the 1970s (Tietze, 1993, p. 116). Nowadays, West German standards apply to the East German day care system.

Sources:

Differences in supply Currently, there are efforts in West Germany to provide day care that allows the mother at least a half-day job. Recent data are unavailable. Despite the reductions of full-time places in the East, there is still a big East-West-difference (Appendix-Table A8). It has been suggested that the overall coverage level has been increased at the expense of full-time day care places (Künzler, 1998, p. 118). There are standards set by the federal Recent information on the implementation states. of quality standards are unavailable. In West Germany, children-staff-ratios were about 14 to 1 in 1990, the size of the groups between 20 to 30 children (Tietze & Roßbach, 1993, p. 152).

See in Table (above).

133

Appendix-Table A16: Coverage level of day care facilities for school children (Horte), in the federal states of Germany, 1990-1998 Change 1998-1990/1 Baden-Württemberg 133 The free state of Bavaria 149 Berlin (West) 125 The free hanseatic city of Bremen 120 The free hanseatic city of Hamburg 138 Hesse 134 Lower-Saxony 136 North Rhine-Westphalia 147 Rhineland-Palatinate 168 Saarland 196 Schleswig-Holstein 150 West Germany 139 Berlin (East)  Brandenburg 90 Mecklenburg-Western Pomerania 93 The free state of Saxony 103 Saxony-Anhalt  The free state of Thuringia  East Germany  Note: Coverage level is the proportion of places to the number of children in the age-group six to under ten in per cent. For methodological issues see note 23.

East Germany

West Germany

Federal state

1990/1 % 2.9 4.5 29.3 15.1 19.6 7.2 3.1 3.6 2.4 1.9 3.9 5.0  61.9 42.4 44.5   49.1

1994 % 2.8 4.8 27.3 16.3 21.2 7.2 3.0 3.9 2.8 2.2 4.2 5.1 22.2 65.6 45.3 52.9 7.0 1.6 34.1

1998 % 3.2% 5.7% 34.8% 16.4% 23.0% 8.3% 3.5% 4.5% 3.3% 3.5% 4.8% 5.9% 26.4% 82.3% 63.2% 69.2% 16.6% 5.0% 47.7%

Sources: Künzler, 1998, table 7.5 (see for further reference); Statistisches Bundesamt, personal communication, March 1, 2001.

134

Appendix-Table A17: Indicators of fertility in Germany, 1950-1998 Total fertility rate Net reproduction rate Out-of-wedlock births West East Germany West East Germany West East Germany Year Germany Germany Germany Germany Germany Germany .93 1.13 9.70 12.80 1950 2.09 2.37    1960 2.36 2.33 1.10 1.07 6.30 11.60    5.50 13.30 1970 2.01 2.19 .95 1.04    1975 1.45 1.54 .68 .73 6.10 16.10    1980 1.44 1.94 .68 .93 7.60 22.80    1981 1.43 1.85 .67 .89 7.90 25.60   1982 1.41 1.86 .66 .89 8.50 29.30    1983 1.33 1.79 .63 .85 8.80 32.00    1984 1.29 1.74 .61 .83 9.10 33.60    1985 1.28 1.73 .60 .83 9.40 33.80    1986 1.34 1.70 .65 .82 9.60 34.40    1987 1.37 1.74 .65 .83 9.70 32.80    1988 1.41 1.67 .80 10.00 33.40    1989 1.39 1.56 .67 .75 10.20 33.60    1990 1.45 1.52 1.45 .69 .73 .70 10.50 35.00 15.30 1991 1.42 .98 1.33 .68 .47 .64 11.10 41.70 15.10 1992 1.40 .83 1.29 .67 .40 .62 11.60 41.80 14.90 1993 1.39 .77 1.28 .67 .37 .61 11.90 41.10 14.80 1994 1.35 .77 1.28 .64 .37 .59 12.40 41.40 15.40 1995 1.34 .84 1.25 .64 .40 .60 12.90 41.80 16.10 1996 1.39 .93 1.29 .67 .45 .63 13.70 42.40 17.00 1997 1.44 1.02 14.30 44.10 17.90     1998 15.90 47.6        Note: Total fertility is the estimate of the final birth rate for all cohorts of women; it projects the actual birth rate into the future by taking into account that women in some of the cohorts are still able to bear children. The net reproduction rate indicates to what degree the next generation of women will be reproduced. Births out-of-wedlock are represented as a percentage of all births in a given year. Sources: 1950-1997: Dorbritz & Gärtner, 1998; 1998: http://www.statistisches-bundesamt.de.

135

136

Appendix-Table A18: Long-term care in Germany, 1984-1998 Incidence in Recipients of Persons in households in % long-term care insurance care institutions 1984 3.40   1985 3.45   1986 3.43   1987 3.37   1988 3.28   1989 2.93   1990 3.29   1991 4.04   1992 4.04   1993 4.02   1994 3.99   1995 4.07 1,056,100  1996 3.89 1,190,800 360,800 1997 3.80 1,254,000 425,700 1998 3.82 1,275,100 452,800 Note: Incidence in households is measured by the question "Is anyone in your household receiving care because of old age or health reasons?" in the German Socio-Economic Panel GSOEP; the percentage of households in which there is a case in relation to all households. Figures 1984-1990 for West Germany, from 1991 onwards for Germany. The numbers of recipients of provisions of statutory care insurance who are the persons who need care and of persons in care institutions are derived from official statistics. Year

Sources: Incidence: GSOEP, own calculation; recipients of long-term care insurance and persons in care institutions: Bundesministerium für Gesundheit, 2000.

Appendix-Table A19: Gender role orientations in Germany, 1982-1996: conservative answers (percentages of respective group) and overall scale scores (mean and standard deviation)

A working mother can establish just as warm and secure a Women relationship with her children as a mother who does not Men work. It is more important for a wife to help her husband’s career Women than to have one herself. Men A pre-school child is likely to suffer if his or her mother Women works. Men It is much better for everyone if the father is the achiever Women and the mother takes care of her home and family Men It is better for a pre-school child when his or her mother Women works and does not concentrate on homemaking. Men A married woman should refrain from working if working Women places are scarce and her husband is able to earn the Men family’s living.

1982 29.7

West Germany 1991 1992 21.0 20.9

1996 19.7

37.8

33.7

31.1

26.4

10.8

7.3

8.9

54.0

33.6

41.5

35.4

31.0

32.0

26.8

49.7

30.9

41.6

35.9

29.7

33.3

23.5

87.2

72.9

72.5

72.8

56.5

51.1

49.0

88.1

79.0

79.4

79.5

59.0

53.8

48.7

70.0

48.9

52.6

47.3

30.0

27.9

25.6

70.7

51.3

59.4

54.8

35.2

36.3

26.6

72.4

59.6

58.3

55.3

35.3

33.3

27.2

79.0

71.8

73.4

72.1

43.7

44.2

38.0

67.9

51.5

49.3

45.2

40.6

35.9

32.9

69.5

53.5

56.3

48.3

47.3

40.8

34.0

3.92 1,438

.49 3.95 663

.23 4.32 1,068

.66 4.28 1,004

2.37 3.71 729

2.89 3.77 517

3.20 3.47 515

-2.22 3.78 1,173

-.34 3.79 621

-.89 3.97 949

-.35 3.85 1,021

1.69 3.61 613

1.97 3.61 431

2.89 3.71 481

Women -1.96 n Gender role orientations

Men n

Source:

East Germany 1991 1992 1996 9.5 7.6 6.7

ZA, 1999, own calculation.

137

Appendix-Table A20: CASMIN categories and corresponding educational levels in the FRG and the GDR CASMIN 1a: Inadequately completed general education 1b: General elementary education 1c: Basic vocational qualification/general elementary education and vocational qualification 2a: Intermediate vocational qualification /intermediate general and vocational qualification 2b: Intermediate general qualification

2c_gen: General maturity certificate

2c_voc: Vocational maturity certificate/general maturity certificate and vocational qualification

138

FRG No degree

GDR No degree

Elementary school leaving certificate or degree of a secondary general school (final exams after 9 years of school): "Volksschule", "Hauptschule" General elementary education + completed vocational training

Leaving certificate of the 8th or 9th grade of a general polytechnical school General elementary education + completed vocational training

Intermediate general qualification + completed vocational training

Intermediate general qualification + completed vocational training

Degree of an intermediate school (more academic secondary school than "Hauptschule") with final exams after altogether 10 years of schooling ("Realschule", "Wirtschaftsschule") or corresponding level at comprehensive school ("Gesamtschule") Degree from a grammar school ("Gymnasium") ("certificate of general higher education entrance qualification") or corresponding level at a comprehensive school ("Gesamtschule") General maturity certificate + completed vocational training

Degree from a general polytechnical school ("Polytechnische Oberschule" POS)

Degree from an extended general polytechnical school ("Eweiterte Oberschule" EOS) General maturity certificate + completed vocational training or degree of an advanced technical school ("Fachschule")

CASMIN 3a: Lower tertiary education

GDR Degree from a technical university or engineering college ("Ingnieurhochschule") 3b: Higher tertiary Degree from a university ("Universität", Degree from a university education "Hochschule") ("Universität", "Hochschule") Notes: Especially the West German system is very differentiated and under the responsibility of the individual federal state, so for our purpose only school levels in generalized terms can be described.40 General elementary education marks the end of the period of compulsory schooling in the FRG, usually after 9 years. Pupils can obtain certificates on higher levels after altogether 10 or 13 years. In the GDR, the standard schooling period lasted 10 years with a limited number of places for pupils to go on for two more years; only few pupils left school after 8 or 9 years, which we assign to the general elementary level according to ZA. Sources:

FRG Degree from a university of applied sciences ("Fachochschule")

Brauns & Steinmann, 1999; The British Council, 1996.

Appendix-Table A21: Level of general education by birth cohort in West Germany Birth cohort before 1920 1920-1929 1930-1939 1940-1949 1950-1959 1960-1969       n 2,325 1,467 2,229 1,865 2,258 2,294 2,516 2,466 2,873 2,537 2,393 2,346 No degree or 77.8 69.9 73.9 69.5 71.7 65.4 58.6 54.5 45.4 44.2 31.6 35.5 gen. elem. Ed. (in %) Intermediate 16.7 16.4 16.4 14.5 19.9 17.8 28.1 23.7 31.3 22.9 39.0 27.2 degree (in %) Maturity 5.5 13.7 9.7 15.9 8.4 16.8 13.3 21.8 23.3 32.9 29.4 37.3 degree (in %) Note: For the characteristics of the cohort approach and its limitations see Section 3.2. Source:

ZA, 1999, own calculation.

139

Appendix-Table A22: Level of general education by birth cohort in East Germany Birth cohort

before 1920 1920-1929 1930-1939 1940-1949 1950-1959 1960-1969       188 110 392 279 600 525 500 446 661 626 661 532

n No degree or gen. 86.7 79.1 80.1 73.1 80.5 68.2 43.4 41.7 15.9 17.6 15.9 12.6 elem. Ed. (in %) Intermediate 8.0 10.0 13.3 9.7 9.7 13.3 39.4 30.9 64.6 61.2 64.6 67.7 degree (in %) Maturity 5.3 10.9 6.6 17.2 9.8 18.5 17.2 27.4 19.5 21.2 19.5 19.7 degree (in %) Note: For the characteristics of the cohort approach and its limitations see Section 3.2. Source:

ZA, 1999, own calculation.

Appendix-Table A23: Level of vocational education by birth cohort in West Germany Birth cohort

before 1920 1920-1929 1930-1939 1940-1949 1950-1959 1960-1969       359 175 609 486 713 783 753 747 867 876 1,016 1,010

n Compl.voc. 35.1 67.4 41.1 70.0 49.6 70.0 61.4 70.4 64.5 66.8 65.9 68.4 training (in %) Universitiy 3.9 16.6 3.8 13.8 4.9 14.3 8.8 20.2 15.5 22.7 12.0 14.8 degree (in %) Note: For the characteristics of the cohort approach and its limitations see Section 3.2. Source: ZA, 1999, own calculation.

Appendix-Table A24: Level of vocational education by birth cohort in East Germany Birth cohort

before 1920 1920-1929 1930-1939 1940-1949 1950-1959 1960-1969       141 74 301 191 450 394 374 320 464 467 436 399

n Completed voc. 42.6 82.4 57.5 80.6 67.1 79.2 78.3 77.2 81.5 78.8 81.0 81.7 training (in %) University 2.1 13.5 3.3 18.8 6.9 19.0 11.8 28.4 18.3 21.0 12.4 13.8 degree (in %) Note: For the characteristics of the cohort approach and its limitations see Section 3.2. Source:

140

ZA, 1999, own calculation.

Appendix-Table A25:

Time use in West Germany (hours/week) All Men n = 730 51.9 16.2

Single Women Men n = 155 n = 182 49.1 51.7 12.6 16.0

Childless couple Women Men n = 183 n = 165 41.9 52.0 14.5 16.3

One-parent family Women Men n = 93 n = 17 34.7 47.6 18.0 15.9

35.3 14.4

17.3 10.3

22.7 9.5

18.8 9.7

29.4 11.5

18.3 10.9

33.4 13.1

22.3 11.6

5.5 4.8

8.5 7.9

4.0 5.0

5.5 6.9

4.7 4.3

8.6 9.2

4.9 5.2

7.1 3.5

16.1 17.2

9.1 11.4

17.0 15.0

15.2 11.4

2.6 4.1

3.1 4.9

3.3 4.6

4.1 5.4

2.7 3.8

2.6 4.4

2.3 3.2

2.7 3.8

Total workload

87.9 18.3

89.8 18.5

79.1 15.3

80.1 19.3

78.7 14.9

81.5 16.6

92.2 16.9

94.8 16.7

Leisure

16.3 10.7

17.7 11.5

22.6 12.2

24.8 13.7

22.0 11.4

22.4 10.0

13.5 9.2

11.9 5.1

8.3 4.4

7.6 4.1

8.7 4.7

7.2 4.3

8.6 4.5

8.3 4.6

7.7 3.5

8.0 4.0

Sleep

50.8 7.6

48.9 7.5

50.4 8.4

48.8 8.2

51.7 6.6

49.8 7.5

50.0 6.2

49.4 10.9

Note:

Means and standard deviations.

Paid work, education Routine housework Other housework Child care Care for sick or old household members, voluntary work

Meals, personal care

Women n = 1,038 28.5 20.8

Source: Gender Division of Labour in Germany, 2000.

141

142

Appendix-Table A26: Time use in East Germany (hours/week) All Women n = 507 37.3 20.0

Men n = 320 52.0 16.8

Single Women Men n = 50 n = 76 42.5 47.9 19.0 20.7

34.1 14.2

17.2 9.2

25.4 10.6

21.0 9.9

29.1 12.0

18.5 10.4

34.4 20.4

26.9 11.2

6.6 5.8

11.2 8.8

5.8 6.6

7.0 7.0

5.6 6.0

10.0 7.7

6.5 7.6

8.9 6.6

13.0 15.0

8.3 10.4

16.5 13.8

15.3 14.1

3.1 6.1

3.2 4.4

3.4 3.9

3.9 4.7

3.0 4.7

3.9 5.2

2.9 6.6

2.7 1.9

Total workload

94.1 19.8

91.8 18.5

76.9 20.6

79.8 20.4

82.7 18.9

86.1 17.5

98.5 20.7

107.5 16.3

Leisure

14.4 9.9

15.8 10.4

24.5 13.1

23.6 12.0

19.7 11.2

20.0 11.0

13.2 10.0

6.3 2.2

8.4 4.7

7.6 4.4

9.5 4.6

7.9 5.9

9.2 4.6

7.5 3.5

7.9 4.5

7.5 7.0

50.9 7.1

49.9 7.5

52.9 8.2

51.2 8.1

52.0 6.5

50.3 9.5

48.6 6.8

48.8 6.9

Paid work, education Routine housework Other housework Child care Care for sick or old household members, voluntary work

Meals, personal care Sleep Note:

Means and standard deviations.

Source: Gender Division of Labour in Germany, 2000.

Childless couple Women Men n = 71 n = 52 45.1 53.8 17.2 18.6

One-parent family Women Men n = 77 n=5 38.5 53.7 19.2 15.7

Appendix-Table A27: Time use in West Germany (hours/week)

Women n = 607 18.3 17.7

Men n = 366 52.1 16.3

Two-parent family With a preschool child Women Men n = 237 n = 180 11.3 51.6 15.3 15.4

40.6 13.6

15.8 10.0

41.1 13.6

16.1 10.1

40.3 13.5

15.5 9.9

6.2 4.7

9.9 7.4

5.7 4.2

9.1 6.9

6.5 4.9

10.7 7.9

24.9 15.7

17.4 10.3

36.0 14.1

21.7 10.2

17.8 12.2

13.2 8.6

2.5 4.1

2.9 4.9

1.6 2.7

2.5 4.8

3.1 4.7

3.3 4.9

Total workload

92.3 18.3

98.0 14.6

95.5 18.2

100.9 13.4

90.3 18.1

95.3 15.2

Leisure

13.4 8.8

12.4 7.7

11.9 8.3

11.6 6.9

14.4 9.0

13.1 8.3

8.1 4.3

7.5 3.8

7.7 3.9

7.4 3.9

8.4 4.6

7.6 3.8

50.8 7.9

48.5 7.0

50.5 8.2

47.9 7.3

51.0 7.7

49.0 6.7

All

Paid work, education Routine housework Other housework Child care Care for sick or old household members, voluntary work

Meals, personal care Sleep Note:

Means and standard deviations.

Source: Gender Division of Labour in Germany, 2000.

With school children Women Men n = 370 N = 186 22.7 52.5 17.6 17.1

143

144

Appendix-Table A28: Time use in East Germany (hours/week) Two-parent family All Women n = 309 34.4 20.4

Men n = 187 53.2 14.2

With a preschool child Women Men n = 67 n = 47 18.6 51.3 22.4 15.1

36.6 13.9

15.0 7.7

40.4 13.6

14.6 7.4

35.6 13.9

15.1 7.8

7.0 5.0

13.3 9.1

6.0 4.2

9.9 8.6

7.3 5.2

14.5 9.0

17.2 15.5

13.8 10.2

35.9 16.6

20.4 11.4

12.1 10.5

11.6 8.7

3.1 6.6

2.7 4.0

2.8 9.7

1.5 1.9

3.2 5.5

3.1 4.5

Leisure

11.9 7.1

11.8 6.7

9.8 6.3

11.0 5.8

12.4 7.2

12.0 7.0

Total workload

98.3 17.0

97.9 14.8

103.6 18.1

97.7 15.9

96.8 16.4

98.0 14.4

8.2 4.8

7.4 3.8

7.9 3.8

7.1 3.4

8.3 5.0

7.5 4.0

50.9 6.9

49.3 6.6

50.5 6.9

47.9 6.7

51.0 6.9

49.7 6.6

Paid work, education Routine housework Other housework Child care Care for sick or old household members, voluntary work

Meals, personal care Sleep Note:

Means and standard deviations.

Source: Gender Division of Labour in Germany, 2000.

With school children Women Men n = 242 n = 140 38.8 53.8 17.5 13.9

Appendix-Table A29: Independent variables — descriptives

Independent variables Age Household income (in 1000 DM) Number of children

All respondents West Germany East Germany All Women Men All Women Men n = 1,639 n = 919 n = 720 n = 825 n = 478 n = 347 36.95 37.31 36.48 36.33 36.60 35.98 7.56 7.39 7.77 8.14 8.30 7.92 5.03 5.69 5.66 5.03 5.06 5.00 7.62 8.64 6.02 8.60 9.59 7.01 1.03 1.18 .83 .97 1.06 .84 1.08 1.09 1.02 .90 .85 .94

Household type Single Lone parent Couple without children Couple with children (at least one < 7) Couple with children all children > 6 Income oriented work (hours/week) Number of household appliances Dwelling (in 10 square metre) Owner of dwelling Garden Gender role orientation Years in education

.17 .38 .06 .24 .18 .38 .22 .41 .29 .45

.14 .34 .09 .28 .16 .37 .22 .47 .33 .47

.21 .40 .02 .15 .20 .40 .21 .40 .23 .42

.13 .34 .09 .29 .12 .32 .13 .34 .40 .49

.09 .28 .15 .36 .12 .32 .13 .34 .43 .50

.19 .40 .01 .11 .12 .32 .13 .34 .37 .48

39.14 21.41 6.03 1.68 11.85 6.61 .51 .50 .71 .46 13.86 3.15 13.23 3.01

29.51 20.37 6.02 1.70 11.74 6.48 .51 .50 .68 .47 14.31 3.16 12.99 2.89

51.43 15.64 6.05 1.66 12.00 6.77 .51 .50 .74 .44 13.29 3.04 13.54 3.13

43.68 20.04 5.41 1.69 9.93 5.29 .40 .49 .68 .47 15.22 2.84 13.37 3.01

38.01 20.12 5.42 1.62 9.61 4.22 .41 .49 .66 .48 15.44 2.85 13.41 2.85

51.48 17.12 5.40 1.79 10.36 6.47 .40 .49 .70 .46 14.93 2.81 13.41 3.00

145

Independent variables Cohabitation Children: no children Children aged 0-6 Children aged 7-12 Girls aged 13 and older Boys aged 13 and older Partner employed full-time Respondent’s economic dependency

Youngest child external care (hrs/week) Financial transfers (100 DM)

Couples West Germany East Germany All Women Men All Women Men n = 1,063 n = 616 n = 447 n = 511 n = 305 n = 206 .14 .10 .18 .21 .21 .20 .34 .31 .38 .41 .41 .40 .26 .22 .32 .18 .18 .19 .44 .42 .47 .39 .38 .39 .31 .30 .32 .20 .20 .20 .46 .46 .47 .40 .40 .40 .35 .37 .33 .32 .29 .36 .48 .48 .47 .47 .45 .48 .19 .23 .13 .28 .27 .29 .39 .42 .34 .45 .44 .45 .20 .23 .16 .30 .31 .27 .40 .42 .36 .46 .46 .45 .66 .93 .29 .74 .85 .58 .47 .25 .45 .44 .36 .49 48.23 -5.70 52.93 -1.16 -18.57 24.62 64.18 39.91 40.71 35.03 34.41 40.60 Couples with children West Germany East Germany All Women Men All Women Men n = 639 n = 380 n = 259 n = 273 n = 154 n = 119 6.21 6.47 5.84 7.49 7.05 8.07 4.23 4.34 4.04 3.68 3.72 3.55 6.39 6.43 6.32 5.41 5.63 5.13 3.50 3.64 3.30 2.82 2.92 2.67

Notes: Means and standard deviations. +p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001. Household types do not sum up to 100%: Differences to 100% are other households. e.g., adult respondents still living in the parental home.

146

Appendix-Table A30: Housework — zero-order correlations Independent variables

West Germany Women Men n = 919 n = 720

East Germany Women Men n = 478 n = 347

Income oriented work (hours/week)

-.543***

-.370***

-.563***

-.312***

Age

.184***

.004

.165***

-.067

Household income

-.003

-.139***

-.084+

.017

Number of children

.454***

-.176***

.328***

-.210***

Single

-.333***

-.116***

-.166***

.186***

Couple without children

-.158***

.085*

-.177**

.048

Lone parent

-.028

.074*

.045

.106*

Couple with children (at least one < 7)

-.245***

-.089*

.204***

-.101+

Couple with children (all > 6)

.239***

-.122***

.125**

-.159**

Girls aged 13 and older

.120***

-.107**

.110*

-.136**

Boys aged 13 and older

.171***

.019

.142**

-.037

Partner employed full-time

.048

.201***

.044

.130+

Number of household appliances

.200***

-.143***

.045

.239***

Dwelling (in 10 square metres)

.186***

-.159***

.005

-.143***

Owner of dwelling

.140***

-.166***

.060

-.224***

Garden

.112***

-.120***

.057

-.201***

Gender role orientation

-.234***

.103**

-.228***

.043

Cohabitation

-.084*

.031

-.045

.001

Years in education

-.184*** -.350*** n = 616 -.104* n = 457 .176*** n = 562

-.077* -.272*** n = 447 -.007 n = 279 -.102+ n = 336

-.161*** -.239*** n = 305 -.204** n = 210 .062 n = 277

-.056 -.262*** n = 206 -.130 n = 129 -.022 n = 170

Respondent’s economic dependency Youngest child external care (hours/workday) Monetary public transfers (100 DM)

Note: +p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001.

147

Appendix-Table A31: Child care — zero-order correlations Independent variables

West Germany Women Men n = 539 n = 321

East Germany Women Men n = 478 n = 158

Income oriented work (hours/week)

-.520***

-.355***

-.493***

-.415***

Age

-.352***

-.122*

-.374***

-.158*

Household income

.045

-.129*

.070

.010

Number of children

.084

-.045

.001

.065

Lone parent

-.121**

.011

-.081

-.042

Couple with children (at least one < 7)

.547***

.397***

.557***

.327***

Couple with children (all > 6)

-.392***

-.262***

-.360***

-.216**

Girls aged 13 and older

-.245***

-.209***

-.334***

-.031

Boys aged 13 and older

-.317***

-.238***

-.223***

-.202*

Partner employed full-time

.006

-.131*

-.084

-.033

Number of household appliances

-.031

.063

-.036

-.126

Dwelling (in 10 square metres)

-.076+

-.173**

-.077

-.141+

Owner of dwelling

-.120**

-.157**

-.109+

-.293***

Garden

-.013

-.137*

-.153*

-.170*

Gender role orientation

-.144***

.025

.006

.026

Cohabitation

-.010

.072

.070

.115

-.034 -.060 n = 284 -.216*** n = 279 -.015 n = 304

.027 -.189** n = 196 -.443*** n = 210 .378*** n = 266

-.049 -.032 n = 146 -.281*** n = 129 .139 n = 146

Years in education

.033 -.289*** Respondent’s economic dependency n = 436 Youngest child external care -.376*** (hours/workday) n = 457 .296*** Public financial transfers (100 DM) n = 523 Note: +p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001.

148

149

Notes 1

In the area of family-related law, there is one exception, the regulation of abortion. As a transitional provision, the Unification Treaty gave East-German women the right to continue their more liberal abortion practice until the revision of the law by the newly-elected parliament of the unified Germany.

2

In this section, we will use "Germany" for the FRG before and after unification, i.e., West Germany until 1990, and unified Germany from October 3, 1990, onwards. Only in some of the following comparisons to the former GDR or East Germany will we be more specific in our terminology.

3

That the effect of financial support on the traditional homemaker/wage-earner couple increases with income is another peculiarity of German family policy. It is based on joint taxation and a progressive income tax whose effects are ameliorated by the splitting of the income of spouses (see below). This means that the higher a couple’s income is, the higher the marginal tax, and consequently the higher the benefits for non-earning wives.

4

It should be noted that indicators of public spending on and financial support of the family have been classified as positively associated with traditionalism. The higher family benefits or workers benefits (as a form of "paternalist" policy, see below) are, the higher the opportunity costs of female employment and the incentives for a traditional division of labour – ceteris paribus. However, this point may be open to debate, because some Scandinavian countries score high on financial support to families, which leads to their ambiguous classification if we do not take into account the different political intentions, i.e., — in their case: — supporting individual family members, and the whole policy profile.

5

It should be noted that for public spending on family services (see indicator 1.1) at least the ranking is higher and the private costs for child care are below average (see indicator 1.4), which most probably means that supplying child care is rather expensive, while the contributions of families are kept relatively low.

6

It should be noted that gender role attitudes themselves could be seen as — at least partially — determined by political influences (and vice versa). The traditionalism or modernism of a country, the extent to which its policies provide for educational attainment and employment (esp. for women) plays a role in the formation of attitudes.

7

As indicator 12.2 shows, the rank positions are reversed. This means that West Germany has undergone a substantial change from older cohorts, which are very traditional, to younger cohorts that are moderately traditional, whereas the older cohorts in the former GDR already held more modern attitudes, making the change only a slight one. There are analogous relations between the indicators of the state of modernization (11.1, 12.1, 14.1) and of the process of modernization (11.2, 12.2, 14.2) in West Germany, with the former in the lower half and the latter in the upper. A notable exception is labour force participation (13.), whose two indicators are more on the traditional side, which means that less development has taken place in this area in the last twenty years than in other areas.

8

In the four areas for which indicators of the East German situation are available, East Germany is ahead and at the top of the modernity scale in all of the indicators for the state of modernization except educational attainment (i.e., gender role attitudes, labour force participation, division of household labour), in which — nevertheless — a rather steep increase in cohort change has taken place (11.2), whereas East Germany was already modernized in other areas some years ago (12.2, 13.2). This means that East Germany shows a pattern (early modernization, small change in recent cohorts) of gender role attitudes and labour force participation which is the opposite of that found in West Germany (see note 7).

151

9

This type includes nearly the whole of (West) Continental Europe. The other two types are the "SocialDemocratic welfare state regime" of the Scandinavian countries and the "Liberal welfare state regime" of the Anglo-Saxon countries.

10

Whereas the situation is — due to Esping-Andersen’s rather erratic classification methods — more complicated in the case of the indicators he uses in his 1990 book, namely de-commodification and restructuring of stratification by the welfare state, Künzler shows in a cluster analysis of EspingAndersens "de-familialization"-indicators (see Appendix-Table A1, p. 113) that, in the best solution, they differentiate between the "Social-Democratic" and all other countries (see the comprehensive overview of the discussion devoted to Esping-Andersen’s typology in Künzler (1999a, pp. 60ff).).

11

Again, we do not claim that Esping-Andersen has succeeded in formulating a clear-cut and empiricallyconfirmed theory of this format, nor that his specific framework is the one best suited for the task at hand. However, he has set standards that should be considered in further research.

12

The legal situation is more complicated. There were several provisions in German civil law that directly or indirectly contributed to the traditional division of labour, e.g., the ideal of the wageearner/home-maker type marriage in family law until 1977. On the one hand, the abolishing of measures like these furthered the equal opportunities of the sexes (for a general overview: Lautmann, 1990; Limbach, 1988; Voegeli & Willenbacher, 1984). On the other hand, despite some progress in constitutional law following German unification, there is no anti-discrimination law that would give women (or men who are engaged in family tasks) rights enforceable in legal proceedings (for recent developments: Berghahn, 1993; Rudolph, 1995). The amendment of the German constitution, called the ‘Basic Law’ (Grundgesetz), that followed unification resulted in the reformulation of the paragraph on equal opportunities which now explicitly determines the responsibility of the state to further the actual implementation of equal rights. However, no concrete measures have been put into effect thus far.

13

The terminology is widely used but somewhat unclear (see Orloff, 1996, p. 57, for a brief discussion).

14

In Germany, the term "Social Insurance" is widely used to denote the dominant form of statutory insurance, esp. old-age, unemployment and health insurance. In a nutshell, all wage and salary earners (with some exceptions) are required by law to contribute to the statutory insurance program. There are some additional regulations. Old-age insurance: All wage and salary earners are required by law to join the scheme; employer and employee each pay half of the respective contribution. There is additional tax-funding for the statutory pension system, which is justified by the additional benefits provided by the social pension scheme, one of them being the child-raising period (cf. p. 28). Civil servants (Beamte) receive a tax-financed state pension. Self-employed persons can join voluntarily. Health insurance: As in old-age insurance, wage and salary earners are required by law to join the scheme, except at higher income levels; starting at a monthly salary of 6,450 DM on, employees can choose to join the social insurance program voluntarily or purchase private insurance. Employer and employee each pay half of the contribution. Statutory health insurance includes marriage and family benefits such as coverage of non-employed spouses and children at no extra charge. Like high-income employees, civil servants can choose to join the social insurance program voluntarily or to obtain private insurance. Self-employed persons can join voluntarily. Since they receive a health subsidy, they do not have to be privately insured with full coverage. Again, in the regulation of health subsidies, there are marriage and family benefits comparable to those in statutory health insurance. Unemployment insurance: All wage and salary earners are required by law to be in the scheme; employer and employee each pay half of the contribution. Civil servants (Beamte) need not, self-employed persons cannot join the unemployment insurance program. Unemployment insurance contains a family benefit based on a higher percentage of income which is granted to unemployed persons with minor children.

15

Some of the regulations that applied to all women (like the ban on night work) had to be dropped because they conflicted with the European Union’s equal opportunity legislation (Halbach et al., 1994,

152

pp. 32f.). Even in the details of the regulations, one can discern the latent function of maintaining the traditional division of labour. E.g.: "The general ban on work performed on Sundays or public holidays does not apply to expectant mothers who are employed in the family household, performing housekeeping activities." (Halbach et al., 1994, p. 278) 16

Therefore, the economic and social conditions of lone mothers can be seen as a litmus test for how the traditional family ideal has been put in practice politically (Hobson, 1994). Lone parent families are a special case, because there is an inherent contradiction in the traditional family ideal; in the "paternalist" view they should not be supported, because support should be channelled through husbands, while in the "maternalist" view they should receive assistance. It also follows that programs such as the recent "welfare reform" in the U.S. (and, in a similar fashion, in the UK), which are intended to raise the employability and labour force participation of lone mothers in accord with the norm of family self-sufficiency, are virtually impossible in the German gender regime.

17

This was the overall average household income in 1996. The German tax code is too complicated to present the average net effects of single measures, because taxable income depends on many individual aspects and regulations. Therefore, we only illustrate the magnitude of effects. Please note that the reported average incomes are for available income, which in most cases is net income. Nevertheless, this average enters the paradigmatic calculation as gross income. The figures are for 1996; see Statistisches Bundesamt, 2000a, p. 114.

18

The net gain figure for the subsistence level or the income at which taxation starts between the single and the couple w/o children (11,8%) is only a partial contradiction. There are two effects that are combined. Besides the splitting, there is a subsistence minimum for every person that cannot be taxed. The effect of this element diminishes with increasing income, but is rather strong in the low-income zone.

19

The exception in the lowest income category is explained in note 18.

20

The example shows — although it is, besides alimony, the only one — that "paternalist" policies do not mean exclusively relieving husbands and fathers of their financial burden. It could also mean enforcing the obligations entailed by family roles (for further considerations about the development of the father's role in the family see Lipp, 1999, pp. 39ff.).

21

Every change in the three areas [tax credit, child benefit, income limit] is counted separately starting from 1953. Tax credit changes for the first three children are counted until 1975, afterwards only for the first child [because only one figure applies for all children]. Child benefit changes are counted until the fourth child, but the change in the figure for the fifth child in 1964 [the only deviation from the figure for the fourth child in the reported period and a change before an election year] is counted as an exception. Election-year changes are changes occurring in a year before, after or of an election; changes that result in reduced benefits or tax credits are not counted as election-year changes.

22

This difference can be explained by the employment policies for East Germany that were still in effect at the time of data collection (Engstler, 1998, p. 139).

23

In this section, we discuss the use of day-care as opposed to the coverage level. Use of day-care is seen from the perspective of the children: How many children in a particular age-group are actually in daycare outside the home? Coverage level takes the perspective of the facilities: How many places are supplied for children in a particular age-group? Both indicators are usually given as percentages of the respective age-group. But, for several reasons, they could not be directly compared. The second indicator regularly leads to higher rates, because counting without actual use is based on estimates on the maximum number of children that a facility might be able to accept (on the basis of staff, rooms, etc.), there may be places not used or simply because of over-supply or false methods of official counting. This over-counting was, e.g., detected for the official statistics of day-care supply in the

153

former GDR (Hank et al., 2001). Despite these disadvantages, coverage level is a useful indicator, because it is usually provided by official statistics of the facilities. Consequently, comparative and time-series data are more readily available. The use-indicator can only be obtained by surveys. Because panel data for the recent development in Germany fortunately do exist, they can be discussed in this section. Data on the real use of day-care may be more accurate. E.g., they can also encompass other forms of day-care outside the home than in facilities (child-minder) as is the case with the data used in this section. One methodological problem pertains for both indicators. There is no perfect fit between the particular facilities and the age-groups. E.g., not every child enters Kindergarten at the age of three and leaves it at the age of six. In other words, the relation of use or places to the respective age-group is only an approximation to the supply-demand-relation. Although the approximation is optimized in the data used in this section by setting age-limits different from those used in the statistics of coverage level (see Appendix-Table A8, p. 124), thus increasing dissimilarity between the indicators, it is still far from perfect. 24

Employees in firms under a minimum size are excluded and so are any cases in which severe reasons on the side of the company or employer prevent that part-time work cannot be granted.

25

There are two more restrictions. The first is the data structure of the ALLBUS. Respondents are classified as East or West German according to their place of residence at the respective time of the survey. As a consequence of (selective) migration, persons who earned their degree in either the GDR or the FRG may now live in the other part of Germany. The second restriction concerns the relationships of the cohorts and their development and history. Not all cohorts earned their degree (partly or completely) in either the GDR or the FRG. The older age-groups reached their highest educational level before the end of WW II. We can assume that the 1930-39 cohort in part and the 1940-1959 cohorts almost exclusively experienced only one of the educational systems of the two separate German states. The youngest viewed cohort includes persons who were fully educated in the FRG or the GDR, and persons, who might have started their education in the GDR, but who now have to complete it in the West German system. So the differences between East and West that are discussed in this section can only in part be attributed to the different educational systems in the GDR or the FRG, respectively.

26

Comparative Analysis of Social Mobility in Industrial Nations. We do not apply the internationally accepted ISCED (International Standard Classification of Education), because CASMIN differentiates between general and vocational-oriented education, which is crucial for our purpose. For further discussion of the two classification approaches for the German case see Braun & Müller, 1997, p. 198, footnote 5.

27

In 1997/98, women’s percentage in language and cultural studies was 72.0%, and men’s percentage in engineering studies in the same time period was 79.3%. There are also subjects with nearly equal percentages of men and women (e.g., business management and law).

28

The results are based on a self-assessment: "Is your employment full-time or part-time employment?"

29

The survey questionnaire of the European Network on Policies and the Division of Unpaid and Paid Work (Willemsen, 1997) was adapted to the German situation (Lipp, Künzler, & Walter, 1999; INFAS 2000).

30

Children include children by birth, step-children, adopted children, foster children, and children of the partner.

31

Differences have been tested for significance using the ANOVA and Scheffé-tests (findings not shown).

32

Predicted values are calculated using the unconditioned mean of men’s weekly housework time.

154

33

The indicators are ranks in social-security spending, family-policy spending, child care, maternal/parental leave.

34

The percentage difference in net income (after tax and social security contributions) of a single person and a one-earner couple set at half of the average, the average and 1.5 times average male earnings.

35

The percentage difference in net income (after tax and social security contributions) of a single person and a one-earner couple set at average male earnings, one and three children of school age.

36

Difference between the net income of a wife with .66 of average female earnings and a husband with average male earnings and the respective gross income as a percentage of net income.

37

Calculated index scores with weighted combinations of relevant indicators (see for details: Gornick et al., 1997). The ranks given in the source have been inverted in order to fit them to the traditionalismmodernism-scale.

38

The value is for unified Germany as a whole.

39

Siaroff (1994) classifies West Germany as one of the Continental European "advanced ChristianDemocratic welfare states" that support families financially and set restrictions on women’s labour force participation and careers, thereby discouraging women from employment in two ways, indirectly by lowering the opportunity costs for the decision not to work and directly by creating a societal climate of low female work desirability.

40

For a brief description of the two school systems see The British Council, 1996, pp. 324ff.

155