Tài liệu Families and Households in Post-apartheid South Africa docx

Edited by Acheampong Yaw Amoateng & Tim B Heaton

Families and households in

post-apartheid South Africa:

Socio-demographic perspectives

Free download from www.hsrcpress.ac.za

Published by HSRC Press

Private Bag X9182, Cape Town, 8000, South Africa

www.hsrcpress.ac.za

First published 2007

ISBN 978-0-7969-2190-1

© 2007 Human Sciences Research Council

Copy-edited by Vaun Cornell

Typeset by Robin Taylor

Cover design by comPress

Print management by comPress

Distributed in Africa by Blue Weaver

Tel: +27 (0) 21 701 4477; Fax: +27 (0) 21 701 7302

www.oneworldbooks.com

Distributed in Europe and the United Kingdom by Eurospan Distribution Services (EDS)

Tel: +44 (0) 20 7240 0856; Fax: +44 (0) 20 7379 0609

www.eurospangroup.com/bookstore

Distributed in North America by Independent Publishers Group (IPG)

Call toll-free: (800) 888 4741; Fax: +1 (312) 337 5985

www.ipgbook.com

Free download from www.hsrcpress.ac.za

Tables and figures iv

Preface vii

Acronyms and abbreviations ix

Chapter฀1

Social and economic context of families and households in South Africa 1

Acheampong Yaw Amoateng & Linda M Richter

Chapter฀2

Towards a conceptual framework for families and households 27

Acheampong Yaw Amoateng

Chapter฀3

Living arrangements in South Africa 43

Acheampong Yaw Amoateng, Tim B Heaton & Ishmael Kalule-Sabiti

Chapter฀4

The economic well-being of the family: Households’ access to resources in

South Africa, 1995–2003 61

Daniela Casale & Chris Desmond

Chapter฀5

Family formation and dissolution patterns 89

Ishmael Kalule-Sabiti, Martin Palamuleni, Monde Makiwane &

Acheampong Yaw Amoateng

Chapter฀6

Fertility and childbearing in South Africa 113

Martin Palamuleni, Ishmael Kalule-Sabiti & Monde Makiwane

Chapter฀7

Children’s household work as a contribution to the well-being of the family

and household 135

Sharmla Rama & Linda M Richter

Chapter฀8

The family context for racial differences in child mortality in South Africa 171

Tim B Heaton & Acheampong Yaw Amoateng

Contributors฀ 188

ContentS

Free download from www.hsrcpress.ac.za

iv

©HSRC 2007

Tables

Table 3.1 Distribution of household types by race of head in rural and urban areas 52

Table 3.2 Percentage distribution of husbands living apart from their spouses by

race 53

Table 4.1 Households’ main source of income (percentage of households), 2002 63

Table 4.2 Household income from employment, 1995–2003 66

Table 4.3 Percentage of households with a member receiving a private pension 69

Table 4.4 Percentage of households with a member receiving a welfare grant 69

Table 4.5 Percentage of households receiving remittances and average remittance

value 70

Table 4.6 Proportion of households in nominal total monthly expenditure

categories 72

Table 4.7 Percentage of households with access to and making use of various services,

1995 76

Table 4.8 Percentage of households whose main dwelling is an informal structure 77

Table 4.9 Percentage of households using mains electricity for lighting 79

Table 4.10 Percentage of households using mains electricity for cooking 80

Table 4.11 Percentage of households with access to a tap in the household or yard,

or a public tap, as their main source of water 82

Table 4.12 The distribution of access to different toilet types 84

Table 5.1 Singulate mean age at first marriage by province and race, 1996 and

2001 96

Table 5.2 Age at first marriage by selected background variables, all women 1998 98

Table 5.3 Percentage distribution of timing of first birth in relation to first marriage by

race of respondent, all women 1998 99

Table 5.4 Percentage distribution of the population by current age and marital status,

South Africa 2001 101

Table 5.5a Percentage single males by age group and province, South Africa 2001 102

Table 5.5b Percentage single females by age group and province, South Africa

2001 102

Table 5.6 Proportions married within each five-year age group by race 105

Table 5.7 Logistic regression analysis of marriage patterns 106

Table 6.1 Use-effectiveness of different contraceptive methods 120

Table 6.2 Mean number of children ever born to women by age and selected

socio-economic factors, South Africa, 1998 122

Table 6.3 Indices of proximate determinants of fertility by population group,

South Africa 1998 124

Table 7.1 Derived household income by households with children, and children by

derived income of household 139

LiSt฀oF฀tabLeS฀and฀FigureS

Free download from www.hsrcpress.ac.za

v

©HSRC 2007

Table 7.2 Distribution of children by background characteristics (weighted and

unweighted) 148

Table 7.3 Diary of main activities for two girls aged 10, residing in an area categorised

as other rural 150

Table 7.4 Total and mean time children spend on cooking-related activities, by age

and gender (unweighted) 153

Table 7.5 Total and mean time children spend on the cleaning and upkeep of the

dwelling, by age and gender (unweighted) 153

Table 7.6 Total and mean time children spend on the care of textiles, by age and

gender (unweighted) 154

Table 7.7 Total and mean time children spend on the combined household activities,

by age and gender (unweighted) 155

Table 7.8 Total and mean time children spend on the collection of fuel sources for

the household, by age and gender (unweighted) 156

Table 7.9 Total and mean time children spend on the collection of fuel sources for

the household, by gender and locale (unweighted) 157

Table 7.10 Total and mean time children spend on the collection of water for the

household, by gender and age (unweighted) 158

Table 7.11 Total and mean time children spend on the collection of water for the

household, by locale (unweighted) 158

Table 7.12 Total and mean time children spend on chopping wood, lighting fires and

heating water for the household, by gender and age (unweighted and

unweighted) 160

Table 7.13 Total and mean time spent by children on caring for household and

non-household members (unweighted and weighted) 161

Table 7.14 Total and mean time children spend on shopping for the household,

by gender and age (unweighted) 162

Table 7.15 Activities engaged in by one 13-year-old girl for the diary day,

Tuesday 163

Table 8.1 Cox regression models predicting child mortality: demographic and

socio-economic conditions 178

Table 8.2 Cox regression models predicting child mortality in South Africa:

reproduction and health 181

Table 8.3 Cox regression models of child mortality: summary model 182

Figures

Figure 3.1 Distribution of household types in South Africa, 1996 and 2001 48

Figure 3.2a Distribution of household type by race of head 49

Figure 3.2b Distribution of household type by race of head 49

Figure 3.3 Education and complex household living by race 50

Figure 3.4 Education and household type 54

Free download from www.hsrcpress.ac.za

vi

©HSRC 2007

Figure 3.5 Rural/urban residence and household type 55

Figure 5.1 Percentage of males never married by population group, South Africa

2001 103

Figure 5.2 Percentage of females never married by population group, South Africa

2001 103

Figure 5.3 Percentage of males married, by population group, South Africa 2001 104

Figure 5.4 Percentage of females married, by population group, South Africa 2001 104

Figure 5.5 Race differences in marriage 106

Figure 6.1 Linkages between fertility and the socio-economic and cultural system

through biosocial and proximate determinates 115

Figure 6.2 Impact of proximate determinants on fertility 121

Figure 6.3 Proximate determinants of fertility in South Africa by population group 125

Figure 8.1 Child survival by race group 174

Figure 8.2 Maternal education by race/ethnicity 183

Figure 8.3 Contraceptive use by race/ethnicity 183

Figure 8.4 Utilisation of pre- and post-natal healthcare by race/ethnicity 184

Free download from www.hsrcpress.ac.za

vii

©HSRC 2007

preFaCe

Because of the devastating effect apartheid-induced policies such as migratory labour,

influx control, the Immorality Act and so on, had on families and communities before

the democratic transition in 1994, concerns about families and their well-being have

come to occupy centre stage in the post-transition period both by policy-makers and

the general public. One indication of this increasing concern about families and their

social and economic circumstances is the rapid rate at which social and economic

data on families and the households they occupy are becoming available for the

purpose of planning to meet their needs.

The idea for the present publication originated in 2002 when I joined the Human

Sciences Research Council (HSRC) from the University of the Western Cape. In this

new position the Executive Director, Professor Linda Richter put me in charge of an

in-house project called the Strengthening Families Project. Essentially, this project

involved secondary and descriptive analyses of the various survey and census data

that had proliferated in the country in the immediate post-transition period. Even

though before 1994 sociologists and other social scientists had documented the

nature of changes in families and households in the country, limitations of such

studies in terms of coverage and scope had made works like the present monograph

imperative. In other words, the idea was to take advantage of the myriad large-scale,

quantitative socio-economic data sets that were increasingly becoming available to

the South African public to describe the changes that families and households were

experiencing as a result of the political, economic, and social transformations that

were engulfing the broader society. Moreover, because of the multifaceted nature of

domestic organisation, such a study was to be multidisciplinary.

The idea to write the monograph was communicated to social science colleagues

both in and outside the HSRC, many of whom readily welcomed the challenge and

agreed to attend a workshop in the Pretoria offices of the HSRC in November 2003,

to discuss issues such as chapter outlines, data sources and timelines.

At the workshop, consensus was reached on important issues. Firstly, we agreed

to use secondary data sources in the form of the two censuses and sample surveys

(the October Household series, the South African Demographic and Health Survey,

the General Household Survey series and so on). Secondly, we agreed that the

analyses for the respective chapters would be essentially descriptive to render the

study accessible to both undergraduate and postgraduate students in the family field,

academic researchers, policy-makers and the lay public at large.

The present publication has been a protracted and combined effort of patient and

diligent authors, critical readers, and a supportive and wise publisher. Thus, it is

expected that some of the information in the study may be out of date, especially

given the rapidity with which quantitative socio-economic data are being generated in

the country. Even though alteration of established patterns of social interaction takes

time, if the need to update the information contained in this study serves as a basis

for further works of this nature, then our initial purpose in producing the monograph

would have been served. The development and completion of this publication was

Free download from www.hsrcpress.ac.za

viii

©HSRC 2007

due to the indefatigable efforts of friends and colleagues. First, we would like to

thank the Executive Director of the Child, Youth, Family and Social Development

Research Programme (CYFSD) of the HSRC, Professor Linda Richter, who gave me

carte blanche in my research and the support for this work in particular. The authors

of this monograph deserve a very special thank you for their thoughtful and well￾written contributions. But for their enthusiastic timely revisions, it would have been

impossible to complete the project. Over the years, we have been blessed with

various interns and research assistants in the Cape Town office of CYFSD who all

contributed enormously to the development of this publication: Ms Thandika Gana,

Ms Mihloti Mushwana, and Mr Anthony Burns. Finally, we would like to thank the

staff at the HSRC Press, for their diligence.

Acheampong Yaw Amoateng

Human Sciences Research Council, Cape Town

Free download from www.hsrcpress.ac.za

ix

©HSRC 2007

CPR contraceptive prevalence rate

CRC United Nations Convention on the Rights of the Child

DHS Demographic and Health Survey

EA enumerated area

GHS General Household Survey

IES Income and Expenditure Survey

IMR infant mortality rate

LFS Labour Force Survey

OHS October Household Survey

PSU primary sampling unit

SADHS South African Demographic and Health Survey

SAYP Survey of Activities of Young People

SMAM singulate mean ages at marriage

SNA System of National Accounts

Stats SA Statistics South Africa

TMFR total marital fertility rate

TN total natural fertility rate

TF total fecundity rate

TFR total fertility rate

TUS Time Use Survey

VIP ventilated pit latrine

Note: Names of South African population groups

During the apartheid regime, legislation divided the South African populace into

four distinct population groups based on racial classification. Although the notion of

population groups is now legal history, it is not always possible to gauge the effects

of past discriminatory practices, and the progress of policies designed to eradicate

them, without reference to it. For this reason, the HSRC continues to use the terms

black/African, coloured, white or Indian/Asian people where it is pertinent to the

analysis of data.

aCronymS฀and฀abbreviationS

Free download from www.hsrcpress.ac.za

Free download from www.hsrcpress.ac.za

1

©HSRC 2007

Chapter฀1

Social and economic context

of families and households in

South Africa

Acheampong Yaw Amoateng & Linda M Richter

Introduction

This study uses some of the most recent quantitative datasets generated in the country

to look at families, and their residential dimension, households. It essentially uses a

socio-demographic perspective to examine aspects of family life in South Africa in

light of the transformation in the society’s social structures since the democratic

transition in 1994. We begin this task by examining the social structure of the country

both before and after the transition to serve as the broad context for the substantive

analyses of families and households. Even though we make the implicit assumption

in this study that the prevailing political, social and economic conditions before the

democratic transition were not conducive to a more objective analysis of family and

household structures, to the extent that we focus on developments with regard to

changes in family and household structures in the post-apartheid era, the present

work is duly informed by family scholarship prior to this period.

Family and household structures in pre-transition society

The institution of the family is essentially multidimensional in nature in that it affects

and is affected by the various social, economic, cultural and political institutions

which together form the social structure of any society. Thus, changes in the structure

and functions of the family are fundamentally occasioned by changes in other

institutions in the family’s environment. More broadly, social change is a function of

two main sets of factors, namely, endogenous and exogenous factors. Writing about

the heterogeneity of pre-colonial African social organisation, Adegboyega (1994), for

example, has traced the source of the variation in family forms to the variations in

environmental conditions.

Specifically, Adegboyega has argued that ecological factors seemed to have played

a major role in determining the form the family assumed in different parts across

the continent. For example, he has observed that among pastoralists like the

Masai of eastern Africa, the family tended to be nuclear in form compared to the

extended family form found among the more sedentary horticulturalists like the

Akan of western Africa. Moreover, the mere fact of the existence of different rules of

primogeniture observed amongst indigenous African peoples is a clear indication of

the diversity observed in the pre-colonial African family. But, aside from these internal

Free download from www.hsrcpress.ac.za

Families฀and฀households฀in฀post-apartheid฀South฀africa

2

©HSRC 2007

strains and stresses, there are other exogenous factors that engender family change.

For instance, there appears to be consensus among writers of the African social

experience that the incorporation of African societies into the international capitalist

economy through the colonial project has been one of the major causes of family

change on the continent (see, for example, Mazrui 1986; Russell 2002).

Some of the family patterns often cited as evidence of the immense economic,

demographic, political, legal and religious innovation that occurred consequent to

confrontation with the international capitalist order, are not only changes in the

rules of kinship, which were essentially the political backbone of society itself, but

also changes in relationships between husbands and wives, parents and children,

and between members of the conjugal family and their kin. These changes were

facilitated by such mechanisms as formal education, wage employment and adoption

of Western belief systems, the direct transposition of the Western nuclear family

system through the European settlers.1

Several Western scholars concur with this

general observation about changes in social institutions as a result of increased

interactions of different cultures. For example, they have observed that throughout

the period of modernisation and especially in the early stages of globalisation, major

changes in social institutions have taken place across the world (for example, see

Giddens 2000; Turner 2002).

Among the social institutions that have received considerable attention are the family,

the economy, polity, and educational and religious institutions. Anthony Giddens,

for instance, has argued that globalising forces are impacting on the family in ways

such as the emergence of more egalitarian relationships between men and women,

the increasing participation of women in work outside of the home and in public

life, the separation of sexuality from reproduction, and the growing tendency for

family relations to be based on the sentiments of love rather than economic or social

concerns, with the intimate couple being the primary family unit. In South Africa,

colonisation and its natural extension in the form of apartheid were institutionalised

through such mechanisms as land expropriation, political disenfranchisement of the

majority indigenous populations, and the institutionalisation of wage labour through

industrial development.

These developments resulted in significant alterations in the social structures of

the society. For instance, large numbers of people – resident Africans, imported

labourers and settlers – left family homesteads and migrated to earn cash income to

meet imposed taxes and supplement declining agricultural resources and to support

their relocation. The massive movement of people from the countryside to urban

centres following the development of industries led to the rapid urbanisation of

South Africa. These patterns of socio-economic development were exacerbated by the

institutionalisation of racism through the apartheid policy of separate development.

Through a series of legislation, the life chances of the non-white groups in the

1 In South Africa, the importation of Indian and Malay indentured slaves by the white settlers to work in Natal and the

Western Cape respectively brought in other family systems, while the interracial marriages that led to the creation of the

coloured population group added yet another dimension to this diversity of family systems in the society.

Free download from www.hsrcpress.ac.za

Social฀and฀economic฀context฀of฀families฀and฀households฀in฀South฀africa

3

©HSRC 2007

society became severely restricted, while whites were given advantages in such

critical domains as agriculture, education, employment, housing, and healthcare.

For instance, as a result of the 1913 and 1936 Land Acts, African ownership of land

was restricted to only 13 per cent of South Africa’s land area (Wilson & Ramphele

1989), thus considerably limiting opportunities for African farming.2

In the area of

education, the country’s education system was racialised with the ultimate aim of

providing inferior education to the non-white groups, especially the African majority

under the Bantu Education Act (Fedderke et al. 2000; Naicker 2000). In their study of

the country’s different education systems based on data from 1910 to 1993, Fedderke

et al. (2000) observe that white people’s educational opportunity was consistently and

considerably better than black people’s educational opportunity.

On the specific issue of pupil–teacher ratios, Fedderke et al. found that while the

white public school pupil–teacher ratio never rose above the mid-20 level, the best

black pupil–teacher ratio provided by the private schooling system in 1941 was 31:1;

the pupil–teacher ratio for black public schooling remained in the range from 50 :1 to

70 :1 for the period from 1957 to 1993. Moreover, under this unequal education of the

races, Fedderke et al. found that the real expenditure on the schooling systems for

white people was far larger than the absolute level of expenditure on any other race

group until the mid 1980s; white per pupil expenditure remained at least seven times

the level of that of black pupils between 1972 and 1992. Besides these indicators,

black schools had inferior facilities, teachers and textbooks; although 96 per cent of

all teachers in white schools had teaching certificates, only 15 per cent of teachers in

black schools were certified.

The same pattern of unequal distribution of resources to the various race groups

prevailed in the area of housing where, in terms of both quantity and quality of

housing, white people had the greatest advantage. For example, research conducted

by Real Estate Surveys between January and May 1992 showed that of the 11 500

formal houses built in the country, the distribution was as follows: Africans 23 per

cent; coloureds 17 per cent; Indians 6 per cent; and whites 54 per cent. In terms of

the average cost of the houses, the research revealed the following racial variations:

Africans R36 290; coloureds R33 661; Indians R83 882; and whites R132 613 (The Natal

Mercury 9 February 1993; Business Day 10 February 1993). Moreover, the Central

Statistical Services (1992) reported that between January and June 1992, there was a

45 per cent decrease in the number of formal houses built for Africans, while there

was a 31 per cent increase in those built for white people during the same period.

Thus colonialism-induced processes such as urbanisation, industrialisation and

subsequent apartheid-imposed restrictions affected family and household formation

patterns in the society, especially among Africans, who bore the brunt of such

policies. Among Africans, the limitations on geographical mobility reinforced dual

2 Other notable pieces of legislation include the Group Areas Act of 1950, the Immorality Act of 1949, the Population

Registration Act of 1950, the Bantu Education Act of 1953 and the Industrial Conciliation Act of 1956 (Mbeki 2001;

Sampson 1987).

Free download from www.hsrcpress.ac.za

Families฀and฀households฀in฀post-apartheid฀South฀africa

4

©HSRC 2007

urban–rural homesteads and circular migration as organisational mechanisms

of economic and social adaptation (Okoth-Ogendo 1989). With limitations on

geographical mobility came limitations on social mobility in areas such as education

and employment which, in turn, had profound implications for family life. For

instance, urban-bound migration, which resulted from the lack of opportunities for

farming in the country, impacted the family life of Africans in a number of ways.

First, in rural areas the absence of males, who predominated in the migratory labour

system, meant either the postponement or complete avoidance of marriage among

Africans.3

Second, in cases where marriages were contracted, economic necessity

meant that the husband/father left his wife and children behind to participate in the

migratory labour system, a situation that led to such family patterns as female-headed

households, out-of-wedlock births, and unstable household composition, especially

among Africans in the rural areas (see, for example, Pasha & Lodhi 1994; Oberai

1991; Pick & Cooper 1997; Seager 1994; Simkins & Dlamini 1992).

Similar economic rationale underlay other family patterns observed among the

various groups in the society. One example is the formation of complex households,

which is usually associated with Africans and often attributed to the communalist

ethos found in many African cultures; that is, the wealthier the patriarch, the more

complex the household tended to be in the pre-colonial systems. A wealthy powerful

man would tend to have more wives, more children and other dependants than a

poor man.

In fact, empirical studies in certain African societies have found that households of

the elite in African towns and cities tend to be complex due to poverty and perhaps

the high incidence of fosterage in these cultures (Oppong 1974).4

However, empirical

evidence based on studies conducted both in Africa and among other African

populations outside the continent appeared to suggest that the formation of complex

households was as much a function of poverty as it was of culture among Africans

(Stack 1974). Specifically, it was argued that as formal education spreads, especially

among Africans, the nuclear family replaced the complex or extended family as the

modal family type among this educated elite. In fact, in the context of apartheid

South Africa, political factors interacted with economic ones to prevent the formation

of extended family households among Africans. Specifically, Section 10 of the Urban

Areas Act of 1945 and a housing policy that facilitated single family units of three or

four rooms tended to compel nuclear families.

Conversely, the establishment of independent households among white people upon

marriage is as much a function of economics as it is of the Western cultural values of

independence and privacy. Needless to say, under pre-colonial and apartheid South

African conditions, white people’s relative access to societal resources in the form of

3 Apart from the sheer physical separation of the sexes as a result of male migration, the meagre wages paid to African

workers and the widespread unemployment among them in the face of increasing commercialisation of the lobolo

(bride wealth) ensure that for a large number of African males marriage is simply unavailable.

4 In the South African case, apartheid-era restrictions on African housing may have contributed significantly to the

formation of complex households, especially in the urban areas.

Free download from www.hsrcpress.ac.za

Social฀and฀economic฀context฀of฀families฀and฀households฀in฀South฀africa

5

©HSRC 2007

education, income and occupational status meant that this type of living arrangement

was available to them.

Because of the institutional racism that prevailed during the apartheid era, it is often

tempting to think that the political, social and economic conditions engendered by

the state only affected African family life. Contrary to this belief, these same processes

affected the family lives of other racial groups in the society, given the fact that

apartheid was a zero-sum game. For instance, the deliberate strategy of drawing

white people to towns and cities, and hence to the modern sectors of the economy,

ensured that they possessed the requisite social and economic resources for a viable

family life. Maconachie (1989), for example, has noted that a central constraint on

white married women’s employment was their responsibility for the care of their

children, especially younger children. However, because of white women’s ability

to access domestic help (provided mostly by African and coloured women), their

labour force participation rates were generally higher than both African and coloured

women. It is important to note that job reservation as legislated by the apartheid state

meant that very few jobs other than domestic work were open to African women

in the so-called ‘Coloured Labour Preference Area’ in particular, and in towns in

general. Even though the employment of African women was equally constrained by

childcare responsibilities, in their case this constraint was removed largely by African

mothers’ ready accessibility to childcare through kin, including politically subordinate,

unpaid extended kin.5

Against this broad background of the political economy of

South African society during the colonial and apartheid eras, how were families and

households depicted by family scholarship?

Pre-1994 family scholarship in South Africa

In the 1950s and 1960s, modernisation theory became popular in the explanation of

the evolution of families and households, especially among family sociologists (UN

1995). According to this interpretation, before industrialisation the size of the family

was relatively large, usually extended by the presence of several relatives. However,

as society developed, such an extended family gave way to the nuclear family, a

process that naturally reduced the household size (Giddens 1987).

Using this theoretical perspective several early empirical studies of black and white

domestic organisations in South Africa concluded that the family patterns of the

two groups were converging in the direction of the nuclear family system (see

for example, Clark & van Heerden 1992; Nzimande 1987; Steyn 1993). In a study

examining the relationship between exposure to urban life and patterns of domestic

organisation in an African township near Johannesburg, Marwick (1978) found that

the patterns of domestic organisation change to resemble those normally found in

industrial societies. Specifically, he found that 48 per cent of the households in his

sample were of the nuclear family type.

5 In personal communication, Jeremy Seekings of the University of Cape Town’s Departments of Sociology and

Economics drew our attention to this point about black and white labour force profiles.

Free download from www.hsrcpress.ac.za