committee for the comprehensive reform of social … report.pdf · iv transforming the present Œ...

Consolidated Report

iTransforming the Present � Protecting the Future

Transformingthe Present �Protectingthe Future

ii Transforming the Present � Protecting the Future

Consolidated Report

All photographs courtesy of The National Business Initiative Collection housed at The Mvula Trust.

ISBN: 0-621-32392-6

First published March 2002

Number: RP/53/2002

This publication may be reproduced free ofcharge for education, training, advocacy and otheractivities. Please acknowledge source.

Design & Typsetting by Crazy Cat Dseigns

Copy editing by Sheila Swan

Printed and bound by

Consolidated Report

iiiTransforming the Present � Protecting the Future

Transformingthe Present �Protecting the FutureConsolidated Report

Report of the Committee of Inquiryinto a Comprehensive System of Social

Security for South Africa

March 2002

iv Transforming the Present � Protecting the Future

Consolidated Report

Dr ZST Skweyiya, MP

Minister for Social Development

South Africa

Minister

Report of the Committee of Inquiry into a Comprehensive System of SocialSecurity for South Africa

We have the honour to present to you the Report setting out our findings, conclusionsand recommendations.

A comprehensive Report of the Committee of Inquiry is presented here, drawn fromthe detailed research chapters developed by the Committee.

Careful consideration has been given to meeting the Terms of Reference of theCommittee as approved by Cabinet. The Committee has been mindful of thesignificance of social protection as an instrument to redress the legacy of apartheidand promote social and economic development that prioritises the mostdisadvantaged.

Professor Viviene Taylor

Chairperson

Signed: March 2002

Consolidated Report

vTransforming the Present � Protecting the Future

Committee MembersThe Cabinet appointed the following as Committee Members.

�Professor Viviene Taylor (Chairperson): Special Advisor, Minister for Social Development

�Mr Fezile Makiwane, Chief Director: Social Security, Department of Social Development

�Professor Pieter le Roux, Director: Institute for Social Development. University of WesternCape

�Mr Elias Masilela, Chief Director: Macro-economic Policy Department of Finance

�Mr Alex Van den Heever, Advisor, Council for Medical Schemes

�Dr Charles Meth, Senior Lecturer: Department of Economics. University of Natal

�Mr Shadrack Mkhonto, Commissioner: Unemployment Insurance Fund. Department ofLabour

�Professor Guy Mhone, Director: Graduate School of Public and Development Management,University of Witwatersrand

�Professor Anthony Asher, Department of Statistics and Actuarial Science, Wits University

�Mr John Kruger, Chief Director: Social Services. Department of Finance

�Professor Marius Olivier, Director: Centre for International and Comparative Labour andSocial Security Law (CICLASS), Rand Afrikaans University (RAU)

�Mr Ravi Naidoo, Director: National Labour and Economic Development Institute(NALEDI)

�Mr James Chiumya, Chief Director: Department of Transport

�Mr Patrick Masobe, CEO: Council for Medical Schemes

�Ms Nobayeni Dladla, Chief Director: The Presidency

�Ms Doris Sikhosana, Executive Director: Community Internship and DevelopmentProgramme. University of Natal

�Adv. Michael Masutha. Member of Parliament

�Ms Brenda Khunoane, Director: Department of Health.

Professor Francie Lund was unable to take up her membership on the Committee because of otherwork pressures and Ms Doris Sikhosana was also unable to fully participate in the work of theCommittee. However, both Professor Lund and Ms Sikhosana made useful contributions to thework of the Committee.

vi Transforming the Present � Protecting the Future

Consolidated Report

ContributorsSouth African

Many academics, research institutions, non-governmental agencies and universities made valuableresearch and other contributions. Their role in the Inquiry was essential to the work of theCommittee. The various policy research institutes and others that assisted the Commission include:

�Economic Policy Research Institute (EPRI) especially Dr Michael Samson who with hiscolleagues, Mr Gilbert, Mr Kenneth Mac Quene, and Ms Ingrid van Niekerk, providedsignificant insights on various scenarios and micro-simulations commissioned by theCommittee

�National Labour and Economic Development Institute (NALEDI) especially EbrahimHassen, Mandy Moussouris, David Jarvis, Eric Watkinson and Claire Horton

�National Institute of Economic Policy (NIEP), especially Dr Asghar Adelzedah

�Dr Claudia Haarmann and Dr Dirk Haarmann from the Institute for Social Development atthe University of the Western Cape for sharing their research with us

�Professor Nicoli Nattrass of the University of Cape Town and Marlese von Broembsen forcontributing extremely useful research papers and analysis

�Professor Sandra Liebenberg of the Community Law Centre of the University of Western Cape

�Centre for International and Comparative Labour and Social Security Law (CICLASS),RAU � staff members, researchers and associates: Ms Adriette Dekker, Mr George Mpedi, MsGail Kuppan, Ms Elsabe Klinck, Prof. Linda Jansen van Rensburg, and Mr Francisco Khoza

�South African Development and Policy Research Unit (SADEP) and the Department ofSocial Development of tha University of Cape Town

�Health Economics Unit especially Prof. Di McIntyre

�The Child Health Policy Unit of the University of Cape Town

�Industrial Health Research Group of the University of Cape Town

�Ms Debbie Budlender of the Community Agency for Social Enquiry (CASE)

�The South African Federal Council on Disability.

In addition views and insights were received from bodies such as:

�The Commission on Gender Equality

�The Human Rights Commission

�The Youth Commission

�The Finance and Fiscal Commission

�The Office of the Public Protector

�The Financial Services Board

�The Competition Board.

Major stakeholders who contributed through consultations and submissions include:

�Business South Africa

�Cosatu

�Nedlac

�Sangoco

�Sanco

�SACC

�Catholic Bishops Conference.

Consolidated Report

viiTransforming the Present � Protecting the Future

International ExpertsThe Committee benefited from the contributions made by leading international experts andorganisations. Particular mention must be made of the role of Mr John Langmore, Director of theBureau for Economic and Social Policy in the Division of Economic and Social Affairs of theUnited Nations and Dr Sergei Zelenev, Senior Policy Analyst who facilitated the participation ofan expert group of international participants in the South African process.

In addition Dr Guy Standing of the ILO and Dr Robert Holzman from the World Bank togetherwith other leading international experts shared valuable experience.

United Nations Expert Group on Social ProtectionParticularly for their participation, as the United Nations Expert Group on Social Protection, in a number ofworkshops and discussions.

Jozef van Langendonck: Institute of Social Law, Catholic University of Leuven, Leuven, Belgium

Mohammed A Maait: Researcher, City University London, London, UK

Valda Frederica Henry: Researcher, Warwick Business School, University of Warwick Coventry,Coventry, UK

AD Kiwara: Director of International Development Studies, Muhimbili University College ofHealth Sciences, University of Dar es Salaam, Dar es Salaam, Tanzania

Osvaldo Larrañaga: Director, Department of Economics, University of Chile, Santiago, Chile

Maria Cruz-Saco: Associate Professor of Economics, Universidad del Pacífico, Lima, Peru

S Mahendra Dev: Director, Centre for Economic and Social Studies, Nizamiah ObservatoryCampus, Begumpet, Hyderabad, India

Mikko Mantysaari: Head of Research and Development, STAKES, National Research andDevelopment Centre for Welfare and Health, Helsinki, Finland

Nana Apt: Centre for Social Policy Studies, University of Ghana, Ghana

Edwell Kaseke: Principal, School of Social Work, Harare, Zimbabwe

Syed Mansoob Murshed: Research Fellow/Project Director, UNU/WIDER, Helsinki, Finland

Guy Standing: Director, InFocus Programme on Social Security, International LabourOrganization, Geneva, Switzerland

Robert Holzmann: Director, Social Protection, World Bank, Washington DC, USA

Metsi Makhetha: National Programmes Manager, United Nations Development Programme,Pretoria. South Africa

Sergei Zelenev: Chief: Social Analysis and Policy Unit, Division for Social Policy andDevelopment, Department of Economic and Social Affairs, United Nations, New York, USA

Peter Lehmann Nielsen: Social Affairs Officer, Division for Social Policy and Development,Department of Economic and Social Affairs, United Nations, New York. USA

Faith Innerarity: Chairperson of the Commission for Social Development and Director, SocialSecurity, Ministry of Labour, Social Security and Sport, Kingston, Jamaica.

United KingdomThe role of Mike Noble and Robert van Niekerk of Oxford University facilitated the sharing ofexperiences from the United Kingdom in research debates on �Alleviating Poverty and BuildingCitizenship in South Africa through Social Security: What lessons from the UK Experience�. The RhodesTrust and DFID provided financial assistance for the UK debates. The contributors are listed below:

Sir Anthony B Atkinson: Warden, Nuffield College, Oxford University. He served as memberof Royal Commission on the Distribution of Income and Wealth, Pension Law Review Committee,and Social Justice Commission

viii Transforming the Present � Protecting the Future

Consolidated Report

Fran Bennet: Research Fellow, Department of Social Policy and Social, previously worked as OxfamPolicy Adviser on EU/UK Poverty Issues

Martin Evans: Research Fellow at the Centre for the Analysis of Social Exclusion at the LondonSchool of Economics

John Hills, CBE: Director of the Centre for Analysis of Social Exclusion and Professor of SocialPolicy at London School of Economics

Marilyn Howard: Independent Social Policy Analyst related to sick and disabled people

Jane Lewis: Barnett Professor of Social Policy in the Department of Social Policy and Social Work I(Oxford)

Ruth Lister, CBE: Professor of Social Policy and Acting Head of Department of LoughboroughUniversity

Jane Millar, OBE: Professor of Social Policy Department of Social and Policy Sciences Directorof the Centre for the Analysis of Social Policy, University of Bath

Michael Noble: Head of Department, Department of Social Policy and Social Work, OxfordUniversity and Reader of Green College

David Piachaud: Professor of Social Policy at the London School of Economics. Professor Piachaudhas written extensively on child poverty, social security and social exclusion

Katherine Rake: Lecturer in Social Policy at the London School of Economics

Adrian Sinfield: Professor Emeritus of Social Policy of The University of Edinburgh

George Smith: University Research Lecturer in the Department of Social Policy and Social Work,Oxford University and is Co-director of the Social Disadvantage Research Centre

Gavin Williams: Fellow of St Peter�s College, Oxford University.

Other International ExpertsProf. Bernd von Maydell: Director, Max Planck Institute for International and ComparativeSocial Security Law.

Prof. Jos Berghman: Catholic University of Leuven, President: European Institute of SocialSecurity

Ms Pauline Barret-Reid: International Labour Office, Team for Eastern European and CentralAsia, Moscow, previously: Social Security Specialist, ILO Multi-disciplinary Team, Harare,Zimbabwe

Mr Wouter van Ginneken: Co-ordinator, Extension of Social Protection: Research and PolicyDevelopment, ILO, Geneva. Switzerland.

Consolidated Report

ixTransforming the Present � Protecting the Future

AcknowledgementsThe Committee has benefited from a large number of individuals and organisations from all sectorsof our society. We are grateful to all of them, and especially to the following:

�Ministers, Deputy Ministers, MECs, Directors General, senior officials and many othersfrom all levels of Government who assisted with, or participated in the process in one way oranother. Special thanks are due to the Department of Social Development for institutionalsupport provided for the work of the Committee.

�Consultations held with the Directors General of Social Development, Finance, Health,Labour, Public Service and Administration, Transport as well as feedback duringpresentations made to the Social Sector Cluster of Directors General was useful inidentifying current Government measures under way and the issues related to policymonitoring.

�Members of the Social Transformation Committee of the ANC also made useful commentson the work in progress of the Committee.

�International agencies, including the Frederich Ebert Stiftung, the UN Expert Group onSocial Protection, The Rhodes Trust and The UK Department for InternationalDevelopment (DfID) for support to the process.

�We also want to acknowledge all those individuals and organisations who participated in thepublic hearings, those who made written submissions, and those international agencies,NGOs, private sector, unions, universities and research institutions which participated in anumber of very useful workshops. A full list of these together with the consultation processis made available separately.

The work of the Committee and of the Chair was supported during various stages in the Inquiry bya core team of individuals whose effort, commitment and energy in the process was invaluable. MrFezile Makiwane and his staff, Mr Ravi Naidoo and his team of researchers, Dr Charles Meth, MrAlex van den Heever, Prof. Pieter le Roux and other Committee members who chaired task teamsand subcommittees such as Prof. Marius Olivier and his team of researchers, Prof. Anthony Asher,Prof. Guy Mhone must be acknowledged.

The staff in the Ministry of Social development, particularly Ms Theresa Small and Lana Peterson,provided invaluable support to the Chair. Ms Lydia Ntenga provided invaluable assistance both asa researcher and with administrative co-ordination. Ms Sandra Campbell who consistently supportedthe work of the Committee and other administrative and secretarial staff from the Social SecurityDirectorate must be acknowledged with grateful thanks.

In the process of cross checking and verifying research findings and data and sourcing the mostup to date research I was pleased to have the able and committed assistance of many individuals butMr Oliver Babson, from Princeton University, was extremely helpful and invaluable in this regard,as a research assistant to the Chair.

There were many individuals and organisations who contributed and whose support in the processwas important such as Kate Skinner and Allan Taylor.


Chairperson

x Transforming the Present � Protecting the Future

Consolidated Report

Foreword by the ChairThe findings and recommendations contained in the Report of the Committee of Inquiry intoComprehensive Social Security represent a wide range of views, empirical research and policyanalysis based on the most recent data available. It is in keeping with the Terms of Reference as laidout by the South African Cabinet. It also includes additional areas that the Committee was requestedto investigate such as Disability and a necessary expansion of the brief based on certain problemsidentified that required urgent attention in the short term.

The findings in the report indicate that South Africa has made significant advances in addressingsome of the most fundamental problems in social development. However, given the historic backlogsas a result of apartheid and the contradictory impacts of this phase of economic globalisation, thechallenges of addressing structural poverty and inequality are many. Government�s emphasis onthe need to redesign the existing piecemeal and fragmented system in response to these challengesis timely and significant.

Throughout the Inquiry, the Committee was mindful of the complex nature of the issues thatwere before it and also of the many diverse and competing claims and interests that had to beconsidered. Despite this, the overwhelming experience of the Committee was the commitmentshown by all sectors across South Africa to address persistent and pervasive poverty.

This report provides a comprehensive attempt to bring together the different elements of afragmented social security system to address, in a coherent and phased way, the constitutional anddemocratic imperatives as well as the socio-economic challenges facing South Africa.

While this is not the first time issues of social security have been engaged, this Report is significantfor a number of reasons. First, it is one of the most comprehensive inquiries into both public andprivate forms of social security in South Africa and seeks to identify those who fall through thesystem and are without any social protection. Second, it identifies the need to ensure that there arecomplementary and considered relationships between economic and social objectives. Third, itprovides recommendations that could ultimately lead to comprehensive social protection withinan integrated public-private benefit system. Fourth it prioritises the needs of the poorest and excludedin line with Government�s policy framework. Fifth, it provides a baseline of social policy informationand analysis that could contribute in a small way to building knowledge and capacity within andoutside of Government to make informed decisions on available options to transform the lives ofthose who are in desperate need and protect the future of all.

Navigating and negotiating a process of social security reform, especially within a comprehensivesocial protection framework, as suggested by the Committee, is an ongoing and complex process.The Committee is pleased to be able to contribute to this process. The Committee recognises thatwhile a long-term vision and policy framework is essential to ensure a balanced response to economicand social objectives, there is also an urgent need to respond to those whose everyday survival isbeing compromised.


Chairperson

Consolidated Report

1Transforming the Present � Protecting the Future

Contents

1 Introduction ........................................................ 91.1 Purpose of the Committee of Inquiry (the Committee) ............................................................... 9

1.1.1 Government commitments ..................................................................................................... 9

1.1.2 Findings of the inter-departmental task team reporting in 1999 ........................................... 9

1.2 Terms of reference .......................................................................................................................... 10

1.2.1 Broad terms of reference ......................................................................................................... 10

1.2.2 Interpretation of terms of reference ....................................................................................... 11

1.3 Process followed by the Committee .............................................................................................. 12

2 The Socio-economic Context and the Need forComprehensive Social Protection ............................. 15

2.1 Introduction ................................................................................................................................... 15

2.2 Definition of socio-economic concepts ...................................................................................... 15

2.2.1 Poverty .................................................................................................................................... 15

2.2.2 Inequality ................................................................................................................................ 16

2.2.3 Unemployment ...................................................................................................................... 17

2.2.4 Social exclusion ...................................................................................................................... 17

2.2.5 Vulnerability ........................................................................................................................... 18

2.3 South Africa�s demographic challenge ........................................................................................ 18

2.4 The structure of the labour force and the employment challenge .............................................. 19

2.4.1 The extent of the problem ...................................................................................................... 19

2.5 The apartheid labour-welfare nexus ............................................................................................. 21

2.5.1 Income distribution ................................................................................................................ 21

2.5.2 Inequality at the end of the apartheid era .............................................................................. 22

2.5.3 The importance of public transfers ....................................................................................... 24

2.5.4 Access to wage income ........................................................................................................... 25

2.5.5 Wage inequality ...................................................................................................................... 26

2.6 Changes in inequality in the 1990s .............................................................................................. 26

2.6.1 Intra-racial inequalities .......................................................................................................... 27

2.6.2 The growth/inequality debate ............................................................................................... 28

2.7 Poverty in South Africa .................................................................................................................. 28

2.7.1 Recent evidence on poverty in South Africa ........................................................................ 28

2.7.2 What is the level of poverty in South Africa? ....................................................................... 28

2.8 Addressing the challenges of social security reform in a democratic South Africa ................... 29

2.8.1 Inequalities in the distribution of wealth .............................................................................. 29

2.8.2 Shortcomings in the South African social security system ................................................. 30

2.9 The challenges ............................................................................................................................... 32

2 Transforming the Present � Protecting the Future

Consolidated Report

3 Approach to a Comprehensive SocialSecurity Provisioning .............................................35

3.1 Introduction .................................................................................................................................... 35

3.2 International trends in social security provisioning .................................................................... 36

3.2.1 Western Europe ...................................................................................................................... 36

3.2.2 The United States ................................................................................................................... 37

3.2.3 South East Asia ....................................................................................................................... 37

3.2.4 Developing countries ............................................................................................................. 37

3.3 Implications for South African context ......................................................................................... 37

3.3.1 Focus primary social security interventions on formally unemployed? ............................ 38

3.3.2 Workfare? ................................................................................................................................ 39

3.4 Defining an appropriate concept of comprehensive social security for South Africa ................ 40

3.4.1 A comprehensive social protection �package� ...................................................................... 41

3.4.2 Determining the �minimum� requirements for the CSP package .................................... 43

3.5 Economic, institutional and social mobilisation implications ofcomprehensive social protection ................................................................................................... 44

3.5.1 Social protection and the economy ....................................................................................... 44

3.5.2 Social protection and institutional arrangements ................................................................ 45

3.5.3 Incorporating social mobilisation into social protection ..................................................... 45

3.6 Social protection and private social insurance ............................................................................. 46

3.6.1 Mandatory cover .................................................................................................................... 46

3.6.2 Lower earnings limits for mandatory social insurance ........................................................ 46

3.6.3 Articulation and means tests .................................................................................................. 47

4 Constitutional Frameworkfor Social Security in South Africa .............................49

4.1 Introduction .................................................................................................................................... 49

4.2 Status and impact of the Constitution .......................................................................................... 49

4.3 General principles .......................................................................................................................... 49

4.4 A human rights approach .............................................................................................................. 49

4.5 The meaning of social security from a constitutional perspective ............................................. 50

4.6 Limiting, interpreting and enforcing social security rights ......................................................... 51

4.7 Concluding remarks ...................................................................................................................... 52

5 Poverty, Social AssistanceGrants and the Basic Income Grant ...........................55

5.1 Introduction .................................................................................................................................... 55

5.2 Government programmes addressing poverty: findings ............................................................. 55

5.2.1 Poverty programmes to address capability and asset poverty ............................................... 55

5.2.2 Poverty measures: requirement to target social policy ......................................................... 56

5.2.3 Government service delivery programmes .......................................................................... 56

5.3 South Africa�s social assistance system ......................................................................................... 57

5.3.1 An overview of the social assistance system .......................................................................... 58

Consolidated Report


5.4 The feasibility of a Basic Income Grant for South Africa ........................................................... 60

5.4.1 What is a Basic Income Grant? ............................................................................................. 61

5.4.2 Will the Basic Income Grant create dependency? ............................................................... 61

5.4.3 Will the Basic Income Grant target the poorest? .................................................................. 61

5.4.4 Can the Basic Income Grant be administered efficiently .................................................... 61

5.5 The impact of a Basic Income Grant ............................................................................................ 62

5.6 Recommendations ......................................................................................................................... 62

5.6.1 Poverty measures and indicators ........................................................................................... 62

5.6.2 Basic Income Grant ................................................................................................................ 62

5.6.3 A comprehensive and integrated medium- to long-term framework for income support ...... 63

5.7 Conclusions .................................................................................................................................... 65

6 Employment and Unemployment ..............................696.1 Introduction .................................................................................................................................... 69

6.1.1 Key issues ................................................................................................................................ 69

6.1.2 Implications for poverty ......................................................................................................... 70


6.2.1 The appropriate form of social security ................................................................................. 71

6.2.2 Social insurance: The UIF .................................................................................................... 71

6.2.3 Active labour market and job creation policies .................................................................... 73

6.2.4 Public work programmes ....................................................................................................... 73

6.2.5 Informal social security .......................................................................................................... 74

6.2.6 Institutional arrangements ..................................................................................................... 74

6.2.7 Policy evaluation .................................................................................................................... 75

7 Protecting the Children..........................................777.1 Introduction .................................................................................................................................... 77

7.2 The Constitution ........................................................................................................................... 77

7.2.1 Constitutional obligations ....................................................................................................... 77

7.2.2 International instruments ....................................................................................................... 78

7.3 Evaluation of current policies towards children .......................................................................... 78

7.3.1 Social assistance ...................................................................................................................... 78

7.4 Policy recommendations to address problems, gaps and inconsistencies of the current socialsecurity paradigm for children .............................................................................................................. 81

7.4.1 Dealing with the special needs of children ........................................................................... 81

7.5 Conclusion ..................................................................................................................................... 83

8 Health ..............................................................858.1 Introduction .................................................................................................................................... 85

8.2 Findings .......................................................................................................................................... 85

8.2.1 Problems identified with the existing strategic framework .................................................. 85

8.2.2 Role and scope of Government involvement ....................................................................... 86


8.3.1 Reform strategy ...................................................................................................................... 86


Consolidated Report

8.4 Strategic financial framework ....................................................................................................... 90

8.5 Coverage ......................................................................................................................................... 90

8.6 Concluding remarks ...................................................................................................................... 91

9 Retirement and Insurance ......................................939.1 Introduction ..................................................................................................................................... 93

9.1.1 Coverage ................................................................................................................................. 93

9.1.2 Proposed strategic framework ................................................................................................ 93

9.2 Findings and recommendations ..................................................................................................... 93

9.2.1 Compulsion and preservation ............................................................................................... 94

9.2.2 Administration ........................................................................................................................ 95

9.2.3 Retirement fund investment ................................................................................................. 95

9.2.4 Taxation of retirement funds .................................................................................................. 96

9.2.5 Defined benefit vs defined contribution ............................................................................... 96

9.2.6 Consumer protection issues .................................................................................................. 97

9.2.7 Missing beneficiaries .............................................................................................................. 98

9.2.8 State Old Age Pension ............................................................................................................ 98

10 Disability ......................................................... 10110.1 Introduction ................................................................................................................................ 101

10.2 The demographics of disability ................................................................................................. 101

10.2.1 The imperative to act .......................................................................................................... 102

10.3 Findings ....................................................................................................................................... 102

10.3.1 Conceptualisation and definition of disability and implications for national policy..... 102

10.3.2 Consequences of disability ................................................................................................. 102

10.3.3 Progress and gaps in the current framework ..................................................................... 103

10.4. Integrating disability measures into social security .................................................................. 103

10.4.1 Fundamental principles and legislative framework........................................................... 103

10.4.2 Concept of social security for disabled people .................................................................... 104

10.4.3 Definition of disability ......................................................................................................... 104

10.4.4 Purposes and eligibility criteria ........................................................................................... 104

10.4.5 Assessment procedures ........................................................................................................ 104

10.4.6 Targeting ............................................................................................................................... 104

10.4.7 Benefits ................................................................................................................................. 105

10.5 Short-term measures .................................................................................................................... 105

10.6 Recommendations ....................................................................................................................... 105

11 Road Accident Fund (RAF) .................................... 10911.1 Background ................................................................................................................................. 109

11.2 Conclusions ................................................................................................................................ 109

12 Coverage Against Employment Injuries and Diseases ... 11312.1 Background ................................................................................................................................. 113

12.2 Fragmented statutory framework .............................................................................................. 113

Consolidated Report


12.3 Employee protection and interests ............................................................................................ 114

12.4 International standards ............................................................................................................... 114

12.5 Prevention ................................................................................................................................... 114

12.6 Reintegration .............................................................................................................................. 114

12.7 Benefits ........................................................................................................................................ 114

12.8 Exclusion ..................................................................................................................................... 114

12.9 Commuting injuries ................................................................................................................ 115

12.10 Defining accidents and diseases ............................................................................................... 115

12.11 Administration .......................................................................................................................... 115

12.12 Civil suits against employers ................................................................................................... 115

12.13 Financial issues ......................................................................................................................... 116

12.14 Conclusion ................................................................................................................................ 116

13 Institutional Framework ...................................... 11913.1 Introduction ................................................................................................................................. 119

13.2 Challenges .................................................................................................................................... 120

13.2.1 Overview ............................................................................................................................ 120

13.2.2 Public sector ....................................................................................................................... 120

13.2.3 Social insurance ................................................................................................................. 121

13.2.4 Regulated private markets ................................................................................................. 121

13.2.5 Policy co � ordination ........................................................................................................ 121

13.3 Recommendations ....................................................................................................................... 122

13.3.1 Integrated institutional and organisational framework .................................................. 122

13.3.2 Social security board and agency ...................................................................................... 123

13.3.3 Governance structures for social insurance and related organisations .......................... 123

13.3.4 Social Protection Commission ......................................................................................... 123

13.3.5 Private sector regulation .................................................................................................... 123

13.3.6 Social security adjudication mechanism ......................................................................... 123

14 Financial Framework for Comprehensive Social Protection 12714.1 Introduction ................................................................................................................................. 127

14.2 Constitutional provisions affecting social security financing .................................................. 127

14.3 An overview of South African social security financing ........................................................... 128

14.3.1 Determination of the allocation to the social sector functions ......................................... 128

14.3.2 Budget prioritisation ............................................................................................................ 128

14.3.3 Mixed financing principles ................................................................................................. 129

14.3.4 Allocation of the horizontal division .................................................................................. 129

14.3.5 Medium-term Expenditure Framework ............................................................................ 129

14.3.6 Conditional grants and special projects .............................................................................. 129

14.3.7 Allocation of the vertical division ....................................................................................... 129

14.4 Key issues in social security financing ....................................................................................... 130

14.5 The existing framework for financing social security in South Africa .................................... 130


Consolidated Report

14.6 Overall allocations to social security .......................................................................................... 131

14.6.1 The budget framework and inter-governmental system................................................... 132

14.6.2 Sources of financing............................................................................................................. 133

14.6.3 Net burden of financing an income grant .......................................................................... 134

14.7 The tax system and social security benefits ................................................................................ 134

14.8 Review of the current tax dispensation ....................................................................................... 134

14.9 Review of principles underlying the classification and use of mixed financing options ........ 134

14.9.1 General taxation ................................................................................................................... 135

14.9.2 User charges ......................................................................................................................... 136

14.9.3 Consumer tariffs .................................................................................................................. 138

14.9.4 Nominal levies and sundry revenue ................................................................................... 139

14.9.5 Earmarked taxes .................................................................................................................... 139

14.10 Generic financial framework for social security ...................................................................... 141

14.10.1 Principles underpinning social security financial management .................................... 141

14.10.2 Recommendations ............................................................................................................. 142

14.11 Social budget .............................................................................................................................. 143

14.11.1 Fiscal capacity and the prioritisation of social security expenditure ............................... 143

14.12 Reform proposals and their financial implications .................................................................. 146

14.12.1 Assumptions ....................................................................................................................... 146

14.12.2 Results ................................................................................................................................. 147

15 Social Security Within a Regional Context ................ 15115.1 Background ................................................................................................................................. 151

15.2 Regional implications for South Africa .....................................................................................151

15.3 Findings ....................................................................................................................................... 152

16 Conclusion....................................................... 154

References ........................................................... 155

End Notes ............................................................ 158

TablesTable 1: Declining incomes and growing inequality (1991�1996)................................................... 17

Table 2: Unemployment in South Africa, (1996-2001) ..................................................................... 20

Table 3: Employment (1 000s) in South Africa, (1996-2001) ........................................................... 20

Table 4: Racial income and population shares (1970�1996) ............................................................. 22

Table 5: The redistributive effects of public transfers (1993) ............................................................. 24

Table 6: Matrix of means and ends12 .......................................................................................................................................................................... 41

Table 7: Comprehensive social protection package and components ............................................... 42

Table 8: Social impact of the Basic Income Grant .............................................................................. 63

Table 9: Summary of coverage by broad income category ................................................................. 91

Table 10: Summary statistics on the retirement industry .................................................................. 96

Table 11: Summary of key issues affecting the operational and policy effectiveness of different

Consolidated Report


organisational options within the social security system .................................................................. 120

Table 12: Financial estimate of the total expenditure within the South African social securitysystem system ....................................................................................................................................... 144

Table 13: Consolidated provincial and national expenditure from 2000/01 to 2003/04 (nominalprices) R� million ................................................................................................................................. 145

Table 14: Social security expenditure in 2015 if all the maximum scenarios evaluated by theCommittee are implemented .............................................................................................................. 147

Table 15: Social security systems in Southern Africa, a comparative assessment .......................... 152

FiguresFigure 1: Framework for evaluating the social security system ......................................................... 12

Figure 2: Estimate of the South African population by age and gender (millions of people) .......... 19

Figure 3: Income deciles by race, 1993 ................................................................................................ 23

Figure 4: Composition of household income, by income decile and source (%) ............................ 23

Figure 5: Participation rates and broad unemployment rates by income decile (1993) ................... 24

Figure 6: Employment per household by income decile ................................................................... 25

Figure 7: Shares of income by race (1970-1996) ................................................................................ 27

Figure 8: Growth of household income by race (1991-1996) ........................................................... 27

Figure 9: Percentage of population in poverty .................................................................................... 29

Figure 10: South Africa�s distribution of income (assuming full take-up of existing social securityprogrammes) .......................................................................................................................................... 60

Figure 11: Distribution of income with a Basic Income Grant ......................................................... 63

Figure 12: Suggested phasing in of an income support grant system ............................................... 65

Figure 13: Reform strategy and approximate timeline ....................................................................... 87

Figure 14: Institutional framework for a universal contributory system .......................................... 89

Figure 15: Strategic financial framework for the South African health system ............................... 90

Figure 16: Proposed strategic framework for retirement provision in South Africa ........................ 94

Figure 17: Recommended integrated organisational framework. ................................................... 122

Figure 18: Proposed structure of a social security board and agency .............................................. 123

Figure 19: Generic financial framework of a social security system ............................................... 142

Figure 20: Social security expenditure in South Africa (estimates for 2001) ................................. 143

Figure 21: Consolidated provincial and national expenditure from 2001/02 to 2004/05 .............. 145

Figure 22: Fiscal capacity over and above the existing MTEF with no negative macro-economicimplications .......................................................................................................................................... 146

Figure 23: Social security policy evaluation for the period 2001 to 2015 expenditure in constant2001 prices ............................................................................................................................................ 148

Figure 24: Social security policy evaluation for the period 2001 to 2015: expenditure expressedas a percentage of GDP ........................................................................................................................ 148


Consolidated Report

Consolidated Report


Chapter 1Introduction

1.1 Purpose of theCommittee of Inquiry

(the Committee)1.1.1 Government

commitmentsFollowing the 1994 elections, the Governmentcommitted itself to a number of specific goalsin the area of social policy, including:

�The elimination of poverty and theestablishment of a reasonable, and widelyacceptable, distribution of income

�The provision of a reasonable income inold age

�The provision of affordable, decent andeffective healthcare for all

�Full employment, or if this proves notpossible, an adequate mechanism to dealwith poverty.

The above are reflected in the Constitutionwhere, as stated in Chapter 2, section 27 (1)(c),everyone has the right to have access to socialsecurity, including appropriate social assistance,which is part of a publicly funded social securitysystem.

These commitments have also been takenforward in a number of policy documentsincluding the Reconstruction and DevelopmentProgramme (RDP) and in tripartite agreements.

The essence of these policy objectives is alsoincluded in the Growth, Employment andRedistribution (GEAR) framework.

1.1.2 Findings of theinter-departmental taskteam reporting in 1999

An inter-departmental task team, convened bythe Department of Social Development,reviewed the social security system andidentified crucial gaps. These include:

�The Unemployment Insurance Fund(UIF) covers less than 40 per cent of thelabour force at any given point in time,and offers benefits to less than 6 per centof the unemployed.

�The private old age pension systemprovides benefits with insufficientlyreliable replacement rates, vesting andportability problems.

�Disability provisions are notcomprehensive with overlaps existingbetween Compensation for OccupationalInjuries and Diseases Act (COIDA), theRoad Accident Fund (RAF), theDepartment of Social Development andthe private sector.

�Many people remain financiallyvulnerable in respect of healthcare.

�No child benefits are available forchildren older than 7 years, and underschool-leaving age.

�Large numbers of South Africans remainvulnerable to harsh poverty with limitedmeans of advancement.

In light of the above challenges, the task teamrecommended that South Africa should


Consolidated Report

investigate a move to a comprehensive andintegrated social security structure. The taskteam also recommended that a commonrevenue collection system be investigated.

The task team noted that changes of this kindwould require considerable planning, politicaldebate and consultations with the social partnersand all sections of the community.

The task team therefore proposed that aninter-departmental task team or committeeengage in the necessary consultations andgenerate final proposals with respect to animproved and better structured social securitysystem.

1.2 Terms of reference1.2.1 Broad terms

of referenceThe terms of reference given to the Committeerequires the review of a broad number ofelements relating to social security. The generalobjectives of this analysis include:

�Options on ultimate objectives and targetsfor the social security system: Alternativeoptions indicating an envisaged finalstructure should be provided. These shouldbe extensively motivated and viable (Termsof Reference, 2000, par. 2.1.1).

�Options for immediate practicalimplementation: Alternatives consistentwith envisaged ultimate objectives shouldbe outlined. These would need to bepractical and focused on immediate needs,the current level of South Africa�sdevelopment and affordability (Terms ofReference, 2000, par. 2.1.2).

�Viability and implications of optionsconsidered: All relevant informationconcerning the viability and significantnegative or positive implications linked toany options considered must be provided(Terms of Reference, 2000, par. 2.1.3).

The specific social security areas that must becovered are:

�National pensions system: This mustinvolve an assessment of the entireenvironment providing for post-retirement cover, as well as generalfinancial support for the aged (Terms ofReference, 2000, par. 2.2.1).

�Social assistance grants: This must involvean evaluation of the entire socialassistance mechanism including all grants,their funding mechanisms, and theefficiency with which they achieve theirgoals (Terms of Reference, 2000, par.2.2.2).

�Social insurance schemes: All socialinsurance schemes, including fundingand protections for injury on duty andcover for road accident victims, must beexamined (Terms of Reference, 2000, par.2.2.3).

�Unemployment insurance: The currentsystem of unemployment protection mustbe examined. This must include theadequacy of all forms of support for theunemployed, including specialemployment programmes (Terms ofReference, 2000, par. 2.2.4).

�Health funding and insurance: The publicand private sector environments must beexamined with a view toward ensuringuniversal access to basic healthcare (Termsof Reference, 2000, par. 2.2.5).

Each of the specific areas identified above mustinclude the following analyses (Terms ofReference, 2000, section 2.3):

�Existing processes: In many instancesthere are existing policy processesexamining specific funds and safety nets.The Committee will be expected to liaiseextensively with these initiatives in orderto inform the final recommendations.

�Core issues: Each policy area must beexamined taking account of the following:

o Adequacy of adherence to principles ofsocial solidarity

o The legislative and general regulatoryenvironment

o The social budget

o Institutional structure

o The tax environment

o Sources of finance

o Perverse incentives

o Significant gaps and the underlyingreasons

o Macro-economic environment

Consolidated Report


o Impacts on Government as anemployer

o Income distribution.

�Key recommendations on future directions:

o Long-term or ultimate objectives andtargets

o Short-term or required intermediatereforms consistent with the long-termobjectives.

�Implementation process: The Committeemust make concrete recommendations onimplementation steps and prerequisites.

In addition to the specific analyses indicatedabove, the Committee is also required todevelop a social budget for all the key socialsecurity areas (Terms of Reference, 2000,section 2.4):

�The Committee must generate a detailedsocial budget for the country, outliningpublic and private expenditure on keyareas of social policy.

�The Committee must also set up the basisfor the annual presentation of a socialbudget for the country. This will involvethe creation of the relevant capacitywithin key Government departments toensure this can be done.

The Committee is also expected to enter into afairly broad consultation process with allstakeholders (Terms of Reference, 2000,section 2.5):

�The Committee will be required toconsult with all relevant stakeholderslinked to the core issues underexamination. The nature and structure ofthis consultation will be at the discretionof the Committee.

�The Committee will be expected to takeinputs from all relevant South Africanexperts in the various policy areas underexamination.

�The Committee will be expected toconsult directly with all Governmentdepartments affected by the proposals.

�The Committee will be expected toreview all relevant material oninternational practice in bothindustrialised and developing countrysettings.

1.2.2 Interpretationof terms of reference

The broad and complex nature of the issuesraised in the terms of reference provided to theCommittee required the initial development ofa conceptual framework for defining andevaluating social security in South Africa. Theframework settled upon is reflected in figure 1and highlights the following:

�Underlying nature of society: This isdistinct from the socio-economicstructure and reflects the objective state ofaffairs resulting from the naturalincidence of a range of contingenciesendemic to human life. These refer toperiods of particular vulnerability to whichall persons are exposed such as, accident, theloss of parental support, old age, etc.

�Adopted values: The manner in which anyindividual or group of individualsresponds to the objective conditionsunderlying the nature of society isessentially based on a set of subjectivelydetermined values. A society can decide toabandon all orphans to the streets, ordecide to only assist those from wealthyfamilies. Values can be interpreted fromthe degree to which organised responsesto particular contingencies are arranged(or not arranged). Such values can beexplicitly framed in a constitution, a billof rights and/or legislation. In this sense,socio-economic structure and the relateddistribution of power in society influencevalues.

�Concept of social security: The adoptedvalues can be made explicit in the form aconcept of social security, whichelaborates the principles that willunderpin any organised response toparticular contingencies. Such a conceptwould interpret explicit and implicitvalues and seek to give them practicalform.

�Socio-economic situation: The prevailingsocio-economic environment reflects anadditional consideration that mayexacerbate particular contingencies facedby society. Certain groups may be moreseverely affected by disease, early parentalmortality, etc.


Consolidated Report

�Current institutional set-up: The currentinstitutional set-up reflects the prevailingresponse to the underlying nature andsocio-economic situation. This mayaccurately reflect the adopted values, or besignificantly out of step. If the latter,reform options need to be considered.

�Options for reform: Where the response toprevailing needs in a society isinconsistent with the adopted values,reform is required. Such reform willrequire short- medium- and long-termobjectives to be set.

1.3 Process followedby the Committee

In addition to the above, the Committee notedthe fragmented nature of social security policyformulation in a number of critical areas,notably retirement and old age, health, disability,poverty and unemployment. This required thatall major areas of social security be examinedholistically. The Committee formed a numberof sub-committees to facilitate a processconsistent with this approach. The sub-committees dealt with health, retirement andold age, unemployment, and poverty. Sub-committees were also formed to focus on cross-cutting issues related to institutionalarrangements, financing social assistance, socialinsurance and non-traditional forms of socialsecurity.

The Committee collated information onSouth Africa�s social security system, includingviews from researchers, Government andprivate sector stakeholders, and internationalexperts as rapidly as possible. This processincluded the establishment of sub-committees in key areas focused on by theTerms of Reference. In addition, hearingswere set up, research works commissioned,and there was a call for written submissions.Individual meetings were also held with keystakeholders and organisations where thisproved necessary.

Figure 1Framework for evaluating the social security system.

Consolidated Report


Consolidated Report


Chapter 2TheSocio-economicContextand the Need forComprehensive Social Protection

2.1 IntroductionSouth Africa�s social safety net has its roots in aset of apartheid labour and welfare policies thatwere racially biased and premised on full-employment. The last vestiges of state racialdiscrimination have subsequently beenremoved, but a key underlying principle of theold system remains in place, i.e. the assumptionthat those in the labour force can supportthemselves through work, and thatunemployment is a temporary condition. Inreality, those who cannot find work (and whodo not, or no longer, qualify for UIF payments)fall through a vast hole in the social safety net.

This chapter outlines an analysis of the socio-economic context within which the Committeeof Inquiry formulated its recommendationsfor a reform of the social security system. Itpoints to the changing nature of inequalityin South Africa and shows that the currentsafety net needs adapting to suit today�s labour-surplus economy.

Ideally, people should be able to earn aliving through employment rather than relyon welfare transfers. In this regard,Government�s macro-economic strategy aimsto push the economy onto a sustainablegrowth path that will generate jobs. However,given the size of the unemployment problemand the extent of the growth challenge, fullemployment is not a feasible scenario in theshort to medium term. Moreover, the highlevels of unemployment and the social deficit(including extreme poverty and inequality)pose a barrier to the attainment of asustainable growth path.

2.2 Definition of socio-economic concepts

It is important to clarify the meaning of theterms �poverty�, �inequality�, �unemploy-ment�, �social exclusion� and �vulnerability�because these terms have specific connotationsin social security. Also it is necessary to estab-lish why these concepts, in practice, matter topeople in their everyday lives.

2.2.1 PovertyPoverty is usually defined either in absolute orrelative terms. In absolute terms, povertyreflects an inability to afford an adequatestandard of consumption. In this event, onewould use a poverty line, reflecting an incomelevel sufficient to afford adequate consumption,as a point to determine who is poor and who isnot. This definition overlooks the distributionof resources in society that often underpinsabsolute poverty. Thus �relative poverty� refersto the individual�s or group�s lack of resourceswhen compared to that of other members ofthat society.

Bringing these absolute and relative conceptstogether, the Committee of Inquiry states thatfor a comprehensive social security system,poverty can be defined as the inability ofindividuals, households or entire communitiesto command sufficient resources to satisfy asocially acceptable minimum standard of living.

Aside from these technical definitions, howdo the poor themselves see poverty? Duringresearch for the Poverty and Inequality Report(PIR), developed for the then Deputy President


Consolidated Report

in 1998, the poor characterised their poverty inthe following manner:

�Alienation from the community They areisolated from the institutions of kinshipand community.

�Food insecurity The inability to providesufficient or good quality food for thefamily is seen as an outcome of poverty.

�Crowded homes The poor are perceived tolive in crowded conditions and in homes inneed of maintenance.

�Use of basic forms of energy The poor lackaccess to safe and efficient sources ofenergy. In rural communities the poor,particularly the women, walk longdistances to gather firewood or water,risking physical attack and sexual assault.

�Lack of adequately paid, secure jobs Lack ofemployment opportunities, low wages andlack of job security is a regarded as a majorcontributor to people�s poverty.

�Fragmentation of the family Many poorhouseholds are characterised by absentfathers or children living apart fromtheir parents. Households may be splitover a number of sites as a survivalstrategy.

Apart from the above, poverty matters becauseit is the best predictor of one�s future lifetrajectory. In particular, the negative impactsof poverty tend to accumulate through aperson�s lifetime. The greater vulnerabilityresulting from poverty in youth tends to resultin a continuous exposure to various risks(unemployment, ill health, and disability)effectively trapping people and theirdependants in a cycle of poverty.

Committe analysis of poverty in South Africatoday revealed the following:

�Forty five per cent of the population (18million people) live on less than $2 a day,as measured by the World Bank.

�Twenty five per cent of African children arestunted (that is, short for their age)

�Ten per cent of Africans are malnourished(that is, underweight for their age)

�Sixty per cent of the poor get no socialsecurity transfers.

The Committee of Inquiry has taken intoaccount the quantitative measures of poverty as

well as people�s actual experiences andperceptions of poverty and how these manifestin social exclusion and instability.

2.2.2 InequalityInequality refers to the unequal benefits oropportunities for individuals or groups withina society. Inequality applies both to economicand social aspects, and to conditions ofopportunity and outcome. Social class, gender,ethnicity, and locality generally influenceinequality. Reducing inequality includes:

�Increasing the relative share of the leastwell-off

�Improving relative mobility of the poorthrough reducing barriers to advancementin social and economic life, throughpromoting participation of disadvantagedgroups, and eliminating thedisproportionate advantages of the rich interms of education, access, politicalpower, etc.

Measured by Gini-coefficient, inequality inSouth Africa is ranked as the fifth highest in theworld.

Inequality between races is a striking featureof South Africa. In 1996, 61 per cent of Africanslived in poverty, compared with only 1 per centof whites. While this figure has probablychanged somewhat since then, the continuedstark contrast contributes to a perpetuation ofapartheid-style prejudice, where notions of an�underclass� are carried over into the post-apartheid era. Flowing from this there is also astark race differential in terms of who accessesprivate services (catering for the better off) andwho accesses public services (catering mainlyfor the poor).

In the period 1991-1996, while inequalitybetween races decreased, intra-racial (that is,class) inequality increased. This suggests thatthe racial divide of the apartheid era, if left to itsown devices, could become entrenched as adeep class divide in the post-1994transformation period.

In designing a comprehensive conceptualframework for social security, the Committeehas considered the implications of these factorsand the inequalities that emerge from itstrajectory into contemporary society. From asocial security perspective the key issue is toensure that those who are currently excluded are

Consolidated Report


given a stake in the present, and that those whohave benefited from past privileges promote alevel of solidarity.

2.2.3 UnemploymentThe loss of a job, or the inability to find one, hasa devastating impact on individuals and theirdependents. This goes beyond the loss of incomeand what it can buy, to questions of socialparticipation and personal identity.

There are two widely used definitions ofunemployment: a �strict� and an �expanded� one,with the latter including �discouraged workers�who have given up looking for work.

The Committee�s review has shown thatemployment statistical data and analysis isgenerally highly contested. Although significantimprovements have been made to employmentdata, some important problems remain to betackled.

Importantly, with changing forms ofemployment, and hence changing statisticaldefinitions of unemployment, the distinctionbetween �employed� and �unemployed� is alsobecoming blurred. For the purposes of socialpolicy, for example, the difference between anunemployed person and someone employed inthe informal sector at virtually no income appearsinsignificant, since such work does not provideadequate job and income security.

However, using the one available set ofcomparable employment statistics,1 formal sectoremployment has fallen significantly. Thisdecline has serious social and economicimplications. As mentioned, there well may beconcurrent informal-sector employment that isunrecorded. However, household surveyevidence demonstrates most of these jobs to beconsiderably lower-paid and less secure, and thusnot significantly compensating for the loss offormal sector jobs.

An important factor that the Committee hashad to consider is that South Africa is

characterised by a labour surplus economy thatis unlikely to change in the foreseeable future. Alabour surplus economy with high skills� deficitsat the lower end has significant implications forthe design of a comprehensive social securitysystem in the short to medium term.

2.2.4 Social exclusionSocial exclusion covers both the static condition(poverty) and the dynamic process (exclusion)through which poverty is caused. As such, theterm �social exclusion� is, conceptually, moreappropriate for integrated policy purposes than�poverty�.

Social exclusion functions through the twinmechanisms of exclusion and inclusion. It isessentially based upon the power of one group todeny access to reward and life-chances to anothergroup; this is on the basis of certain criteria thatthe former seeks to justify. These criteria could,for example, be income, education, skin colour,language, sex or religious belief.

As in South Africa�s apartheid past, socialexclusion was about mobilising statemachinery and policies to exclude others(black people) from reward and privilege.Social exclusion manifests itself in barriers toadvancement based on the economicallyarbitrary individual characteristics mentionedabove. Many of these take the form of exclusionfrom markets; others take the form of exclusioneven when markets are competitive.

Social exclusion implies a dynamic set ofprocesses with a number of aspects:

�Relativity People are excluded from aparticular society, as opposed to a focus onability to purchase an �absolute� basket ofgoods that might have been regarded asadequate at another time.

�Multi-dimensionality Income andconsumption are central, but so are otheraspects of participation such as the abilityto carry out socially valued activity (not

Table 1Declining household incomes and growing inequality (1991�1996)

African White Coloured AsianPoorest 50 -11 per cent -16 per cent 0 per cent -5 per centper cent

Richest 10 17 per cent -0,3 per cent 16 per cent 28 per cent


Consolidated Report

just paid work), political involvement andsocial interaction. In each dimensioninclusion/exclusion are matters of degree,rather than simply of sharp cut-offs.

�Agency Someone, something or someprocess is responsible for exclusion orinclusion occurring, while inability tocontrol major aspects of one�s life is animportant aspect of being excluded.

�Dynamics Such processes occur over timewith long-lasting or cumulative effects.Duration in particular states matter and sodo prospects for the future.

�Multi-layered Exclusion operates atdifferent levels � individual, household,community/neighbourhood, andinstitutions.

2.2.5 VulnerabilityVulnerability refers to the risk of a particularindividual or group falling into poverty or insituations that compromise their human well-being. Certain individuals or groups, due to theirposition in society, are more vulnerable thanothers to the negative consequences of economic,political and social trends, cyclical changes or�shocks�. Generally women, children, theunemployed, households and communities withlimited assets are least able to cope with the effectsof, say, a negative change in basic serviceprovision or access to remittances.

Therefore policy interventions that seek tomove people out of poverty, and have them stayout of poverty, need to address questions ofvulnerability. Such an approach requiresstrategies to advance vulnerable individuals orgroups more easily out of their precariousenvironments or at least enable them cope betterwith any future negative changes.

2.3 South Africa�sdemographic challenge

Demographic trends are important indetermining the direction of a country�s socialsecurity system. In countries where populationgrowth outstrips economic growth, thepressures on state centred social protection tendto increase. However, population change itselfis not the problem, but rather the manner inwhich the population changes. In manyOECD countries, for example, decliningpopulation growth with increasing numbers

concentrated in the higher age groups haveplaced pressures on health and retirementcover among others.5 Where social securitybenefit systems are designed in part on inter-generational solidarity, such trends raiseconcerns. Governments� responses to thesetrends vary with some encouraging birth ratesthrough family and children�s allowances andother measures. However, population trendsand features that characterise countries in thesouth (less industrialised countries) differ. Inthe south, growth trends are higher,populations are concentrated in the youngerage categories and l i fe expectancy i srelatively low.

Over the past five years, the South Africanpopulation has grown at an average rate of 2,2per cent. Currently South Africa�s populationfeatures by age and gender, depicted in figure 2,reveal a high concentration of the populationbetween the age categories of 0-34 with a higherproportion of women above the ages of 20.

HIV/ AIDS will affect population trends anddynamics such as the size, growth and agestructure of the population.

HIV/AIDS is having a profound impact onfertility and mortality rates in South Africa. Lifeexpectancy at birth is also declining. Availableinformation indicates that the average lifeexpectancy at birth in South Africa has alreadydeclined from about 63 in 1996 to about 55 in1999. It is expected to decline even further. Therewill also be a decline in the number of people inspecific age groups, namely 0-4 year olds and 25-34 year olds.

Simultaneously, however, the real number ofthe old age population in South Africa isincreasing rapidly, due to the improved qualityof life of people who live through the young adultyears of high risk to HIV infection.

The proportion of women in the totalpopulation will decline, as women are morevulnerable to HIV infection due to their lowersocial and economic status, and because ofphysiological reasons.

As a result of the above, it is expected that HIV/AIDS will impact upon dependency ratios inSouth Africa. The projected age structure of thepopulation shows that the proportion of thepopulation in dependent age groups, bothchildren and elderly people, will increaseconsiderably in relation to the potentially

Consolidated Report


economic active proportion of the population.This, in turn, means that the dependency ratewill increase substantially during the comingyears, and that there will be proportionally fewerpeople to care for children and elderly people.

In short, HIV/AIDS will continue tochallenge the capability of existing social securityprogrammes to address the increasedimpoverishment and vulnerability of people.

An important consideration in the design ofcomprehensive social security for South Africais the extent to which social security can promoteprevention and mitigate the impacts of HIV/AIDS and other chronic illnesses. Critical in thisis the increased vulnerability and risk experiencedby many because of poverty.

2.4 The structureof the labour force

and the employmentchallenge

2.4.1 The extentof the problem

Unemployment levels have risen almost withoutpause for the past decade. Apparentlyaccompanying this, at least until fairly recenttimes, has been a steady loss in formal sector jobs.This latter finding has given rise to an energeticdebate, one whose weight can scarcely be borneby the statistics. The September 2000 LFS resultsrecord a statistically insignificant increase in

formal sector employment over the February2000 figures � the February 2001 figures, in turn,record a statistically insignificant decline overthe September 2000 figures. It seems that at best,formal sector employment is now roughly static,with gains in those industries where employmentis growing being counterbalanced by losses inthose where it is shrinking.

In the absence of significant formal sectoremployment growth, the burden of absorbing thecountry�s expanding labour force falls on theinformal sector. It is difficult to construct acoherent time series for informal sectoremployment. Apart from anything else, thesurvey instruments used to capture the desiredinformation changed in the middle of the periodwith which we are concerned (the 1999 OctoberHousehold Survey [OHS] gave way to theFebruary 2000 LFS). As far as can be determined,once unpaid subsistence agricultural producershave been removed from the picture,employment in the informal sector was roughlyconstant at about 1,8-1,9 million in October 1999and September 2000.

Official unemployment figures show thatunemployment increased from 2,2 million (19,3per cent) in 1996 to 4,2 million (26,4 per cent)by 2001. An expanded definition ofunemployment increases the figure from 4,6million (33 per cent) in 1996 to 6,96 million (37per cent) by 2001 (table 2).

South Africa�s employment creation record ispresented in table 3. The SEE (a survey ofearnings and employment) covers a sample of

Figure 2Estimate of the South African population

by age and gender (millions of people).


Consolidated Report

Table 2Unemployment in South Africa (1996-2001)

Official unemployment Expanded unemploymentPeriod Number (1000s) Rate (%) Number (1000s) Rate (%)1996 2 224 19,3 4 566 33,0

1997 2 451 21,0 5 202 36,0

1998 3 163 25,2 5 634 37,5

1999 3 158 23,3 5 882 36,2

Feb 2000 4 333 26,7 6 553 35,5

Feb 2001 4 240 26,4 6 961 37,0

formal sector firms. The non-SEE figures aredrawn from household surveys, and arediscounted by the SEE figures to give an estimateof employment in firms that are not covered bythe SEE sample. The findings of these surveyshave been controversial.

The areas of the economy covered and inwhich jobs are reportedly being lost wereprobably those containing �good� jobs (relativelysecure and relatively well-paid). Expanding areas,by contrast, mainly in services, will be creatingsome �good� jobs (in areas such as financialservices), but will also see the growth of manyinsecure and poorly-paid jobs.

According to table 3 formal sectoremployment declined from 6,8 million in 1996to 6,7 million in 2001, a loss of around 100 000jobs. Over the same period the informal sectorgrew from 1 million to 2,7 million.

With the economically active populationgrowing by more that half-a-million each year,the outcome depicted in the tables � risingunemployment offset somewhat by risinginformal sector employment (mainly in

survivalist activities) � seems inevitable. This, atfirst sight, and as far as can be ascertained fromthe official statistics, is the reality with whichsocial security policies have to cope.

2.4.1.1 Youth unemployment

All told, there were some 2,5 millionunemployed young people in 1999, 1,4 millionwomen, and just fewer than 1,1 million men.Slightly more than 1,3 million of them are inurban areas; the remainder in non-urban areasface a reality in which economic opportunitiesare few and far between. The age category 15-19years contained only about 8 or 9 per cent of theunemployed youth � the others were dividedroughly evenly between the two age categories20-24, and 25-29 years. About 600 000 of theyoung men, and 700 000 of the women hadpreviously been employed. Those above the ageof 25 years who had never previously beenemployed, more than 860 000, would have beenstarting to move into the �difficult to place�category � almost half of them (410 000) hadalready slipped into non-searching status. As maybe expected, this tendency is more marked in thenon-urban areas.

Table 3Employment (1 000s) in South Africa (1996-2001)

Formal sectorPeriod SEE/STEE Non SEE/STEE Total Informal sector1996 5 242 1 550 6 792 996

1997 5 139 1 587 6 726 1 136

1998 4 945 1 445 6 390 1 316

1999 4 840 1 724 6 564 1 907

Feb 2000 4 754 1 924 6 678 1 821

Feb 2001 4 676 2 002 6 678 2 665

Note: The formal sector employment figures exclude agriculture.

Consolidated Report


2.4.1.2 The informal economy

Two sets of changes are driving trends inemployment totals in the sector. The first of these,the massive decline in informal agriculturalemployment (mainly unpaid), has witnessed adecline of almost 850 000. The other big change,that in employment in wholesale and retail trade,has seen 600 000 new workers joining theindustry between September 2000 and February2001. This is difficult to verify. The suddenappearance of 600 000 workers in six months,not to mention the fact that so few observationsexist at present, make it difficult to determine withany degree of certainty the trends in the sector.

The Committee notes that whereas almost onequarter of formal sector workers earnR1 000 or less per month, more than threequarters (76,3 per cent) of informal sectorworkers, and more than 90 per cent of domesticworkers are to be found in this income category.Another striking result is the figure of 18 percent for informal economy workers who receiveno income (it was 30 per cent in September2000). Their condition is relatively easilyexplained � they fall either into the category�helping without pay in a family business�, orthat of subsistence agricultural workers.

Ignoring those working in subsistenceagriculture, the earnings of domestic workersare even lower than those of informal economyworkers in general � almost two thirds ofdomestic workers (64,3 per cent) were paidR500 per month or less. With a further 27 percent being paid between R501-R1 000, that leftonly 7 per cent earning close to a living wage.

2.5 The apartheidlabour-welfare nexus2.5.1 Income distribution

Governments affect income distribution inindirect and direct ways.

�Indirect ways include labour market andeconomic policies that shape the growthpath (and hence the level and pattern ofincomes in society).

�Direct ways include taxation and the publicprovision of social services (understoodbroadly to include public education, healthand housing programmes) as well asincome support (such as old age pensionsor unemployment benefits).

Personal income is particularly affected by thecombination of labour market and welfare orsocial policies: the labour-welfare nexus.

The apartheid system discriminated alongracial lines, with poor white people benefitingespecially. Education, health and housingbenefits were biased towards whites, and jobreservation ensured that white wage earningswere protected. Labour policies were designedto protect the labour-market position of whiteworkers. The industrial conciliation machineryprovided wage protection, job reservationensured that the least competitive white workersobtained employment, and unemploymentinsurance was provided on a temporary basis.In this respect, the apartheid state was a raciallyexclusive variant of the Australian �wage-earners� welfare state, i.e. a welfare state thatsought to ensure a certain standard of living forAustralians as wage earners rather than ascitizens.12

By contrast, black South Africans weresubjected to extensive labour-marketdiscrimination and disadvantage. Inferioreducation, influx control, the Group Areas Actand a range of other instruments underminedblack incomes and consequently their socialdevelopment. Given the chronic labourshortages that plagued low-wage sectors(notably agriculture and mining) during thepost-war period, the apartheid state was averseto providing any alternative means ofsubsistence for African job seekers. Instead, itrelied on coercive labour legislation to channel(mostly unskilled) African labour where it wasneeded most.

In contrast to its limited welfare coverage ofthe working-age population, the Governmentprovided a universal (albeit raciallydiscriminatory) welfare net in the form of theold age pension. This proved to be an importantlifeline for poor African families, particularlyfrom the 1970s onwards, as unemployment roseand as the real value of the pension increased.The Committee calculates that limitedredistribution from white to black South Africansprobably occurred through patterns ofGovernment taxation and expenditure (see table4). In other words, even under apartheid, the finaldistribution of income was significantly moreegalitarian than the market distribution ofincome.

The development of the labour-welfare policynexus under apartheid reflected the changing


Consolidated Report

Table 4Racial income and population shares (1970�1996)

Share of total income Share of population1970 1980 1991 1996 1970 1980 1991 1996

African 19,8% 24,9% 29,9% 35,7% 70,7% 72,4% 75,2% 76,2%

White 71,2% 65,0% 59,5% 51,9% 17,0% 15,5% 13,5% 12,6%

Coloured 6,7% 7,2% 6,8% 7,9% 9,4% 9,3% 8,7% 8,6%

Asian 2,4% 3,0% 3,8% 4,5% 2,9% 2,8% 2,6% 2,6%

Total 100% 100% 100% 100% 100% 100% 100% 100%

class interests of powerful white constituencies.The massive investment in public education forwhite children in the 1950s and 1960s resultedin white workers securing the skills that enabledthem, in the 1970s and 1980s, to command highincomes in free labour markets. This largelyremoved their dependence on direct stateinterventions (such as job reservation throughthe �colour bar�). As labour market regulationwas de-racialised, the wage-setting machinerywas extended to the African working class.Essentially apartheid social welfare for whites wasbased on a combination of income (cash)measures through job reservation and otherforms of assistance and in kind benefits througheducation, health and housing to name a few. Thiscould be characterised as a state-driven orinstitutional approach to social policy for whites.

Despite the decline in discrimination,inequality remained stubbornly high. In the lastyear of apartheid (1993), the poorest four deciles(40 per cent) of households, equivalent to 52 percent of the population, accounted for less than10 per cent of total income. While the richestdecile (10 per cent) of households, equivalent tojust 6 per cent of the population, captured wellover 40 per cent of total income.

High and persistent inequality is one of theenduring legacies of apartheid. But itsdeterminants shifted over the apartheid period.Whereas inequality was initially driven by thegap in racial incomes, this situation changed overtime as African workers advanced up theoccupational ladder and as unemploymentincreased. By the end of apartheid, the gapbetween the incomes of the employed and theunemployed had become a significant driver ofinequality.

A range of economic factors contributed to therise in open unemployment from the mid-1970sonwards. These included:

�The slowdown in economic growth fromjust under 6 per cent per annum in the1960s, to 1 per cent per year in the 1980s

�Capital-intensive strategic investment bythe state

�Policies to replace labour with capital inagriculture

�Tax breaks for capital investment

�The coincidence of negative real interestrates and rising wages in the in the 1970sand early 1980s.

By the end of the apartheid period, the economicstructure had shifted away from absorbing largequantities of unskilled labour, and openunemployment had become a permanent featureof the socio-economic landscape.

2.5.2 Inequality at theend of the apartheid era

At the dawn of South Africa�s new democracy,there was still a clear correlation between raceand household income (see figure 3). Affirmativeaction policies are designed in part to address thislegacy of apartheid.

However, by the end of apartheid, intra-racial(class) inequality was contributing more tooverall inequality than inter-racial inequality.The contribution of �within-group� inequalityto total inequality rose from 38 per cent in 1975to 58 per cent in 1991 and to 67 per cent in 1996.

Figure 4 shows the composition of meanhousehold income for each income decile � inother words, the sources from which the

Note: Totals may not add up to 100% due to rounding.

Consolidated Report


average household in each decile received itsincome. The bottom, or poorest decile, receivedmost (48,5 per cent) of its income fromremittances (including monetary remittancesand remittances in kind, for example in theform of food). Old age pensions were veryimportant to deciles 2 through 4. For everydecile from the fourth up, wages comprised byfar the most important source of householdincome.

�The top five deciles were heavilydependent on wages from regularemployment. The top, or 10th, decilesupplemented its wage income (64,5 percent) with small but significant incomesfrom agriculture (6,8 per cent), self-employment (6,7 per cent) and income

from capital (12,1 per cent).

�Government old age pensions were ofminimal importance to the top decile (atless than 1 per cent). The lower, poorerdeciles relied more heavily on remittancesand old age pensions.

�Income from agricultural production wasof little importance, except to the top decile(which included high-income, capitalistfarmers) and the bottom decile (where theincomes were so low that even R8 permonth from smallholdings was animportant contribution to the decile�sincome).

�Old age pensions were the most importantpublic transfer payment.

Figure 3Income deciles by race, 1993.

Figure 4Composition of household income, by income decile and source (%).


Consolidated Report

Table 5The redistributive effects of public transfers (1993)

2.5.3 The importanceof public transfers

Figure 5 testifies to the continuing importanceof transfers in South Africa. The scale of privateinter-household transfers, i.e. remittances, reflectsthe continuing importance of migrancy. One-third of all African rural households in 1993 hadmembers who were migrant workers. The scaleof public transfers, in the form of theGovernment old age pension and other forms ofwelfare, reflects the expansion of the publicwelfare system since the 1980s.

As shown in figure 4, over a quarter ofhousehold income in the second and thirddeciles came from state old age pensions. Indeed,the presence of an old age pensioner in ahousehold was often the main reason for liftinghouseholds out of abject poverty. Remittanceswere a further important source of income �although, overall, they were much less importantthan Government old age pensions, contributingless than 4 per cent of total income to pensions�12 per cent (figure 4). Not all remittances cameout of wages � there were cases of pensioners

Figure 5Participation rates and broad unemployment rates by income decile (1993).

(Lower income households have less access to the labour market.)

Income Distribution Incidence of Net transferdecile of public taxation on through taxes and

transfers received (%) the poor (%) public welfare (%) Low High Low High

1 0 +7 -2 -3 +5 +4

2 +7

3 +12 +26 -2 -4 +23 +22

4 +13

5 +15 +29 -4 -5 +25 +24

6 +14

7 +12 +23 -11 -12 +12 +10

8 +11

9 +9 +15 -80 -76 -65 -61

10 +6

Total 100 100 -100 -100 0 0

Consolidated Report


sending a share of their pension to family membersliving elsewhere � but it is safe to assume that mostremittances came out of wage income.

Table 5 provides information to assess theredistributive effect of public transfers, i.e. thetransfer of resources from taxpayers to old agepensioners and the recipients of other non-contributory welfare payments (primarily disabilityand child maintenance grants). The first columnshows the distribution of public transfers, by decile.Clearly evident in this column is the fact that the lowestincome deciles, those in destitution, receive the lowestshare of public transfers. Once again, this demonstratesthe perverse effects of ineffective means testing �the exclusion of a significant number of the pooresthouseholds from public transfers.

The second column shows the incidence oftaxation. The final column shows the nettransfer of resources through taxation andpublic welfare.

2.5.4 Access to wage incomePoverty and inequality in South Africa arerooted in the labour market: in part in lowwages, and in part in very high rates ofunemployment. Whereas inequality until the1970s was determined largely by the gapbetween white and black incomes, inequalityin the 1990s is primarily driven by: (a) inequalitywithin the distribution of wages, and (b) by the factthat 30 per cent of households had no wage incomeat all. In other words, households in the lowerechelons of the income distribution are those withno access to the labour market (the very poor) orwith low-paying jobs.

Participation and broad unemployment rates bydecile in South Africa are shown in figure 4. Theparticipation rate corresponds to the proportion ofadult household members participating in thelabour force, and the unemployment ratecorresponds to the proportion of the labour forcethat is unemployed. Both rates are presented hereusing an expanded definition of unemployment,which includes people who are not actively lookingfor jobs because they believe there are noneavailable (i.e. the �discouraged� unemployed).

There is a clear and close correlation betweenunemployment rates and income. This can beseen in Figure 5. However, not only are poorhouseholds likely to have more unemployedadults than richer households, but they are alsolikely to have more adults who said they are notavailable for work (and hence are defined outsidethe labour force). Figure 5 shows that labour forceparticipation rates rise steadily up the incomedeciles. The dual correlation betweenunemployment and income, and labour forceparticipation and income, suggests that low-income households are significantlymarginalised from the labour market.

Because low-income families tend to be largerthan high-income families, the link betweenunemployment and poverty is stronger whenincome deciles are calculated on a per capita basis,as shown in the third bar in each set(unemployment rate*).

The incidence of employment is shown in adifferent way in figure 6. This shows theproportion of households in each income decileaccording to the number of household members

Figure 6Employment per household by income decile.

(The number of workers increases with household income.)


Consolidated Report

in employment. A majority of households in eachof the bottom four deciles have no members inemployment. At the opposite extreme, a majorityof households in each of the top two deciles havetwo or more members in employment.

The link between lack of employment andpoverty is particularly strong in South Africa. Inthe OECD, the proportion of households in thebottom quintile without any members inemployment is 42 per cent, with figures rangingfrom 21 per cent in Luxembourg, to 65 per centin Ireland and 74 per cent in Finland. In SouthAfrica, the corresponding figure is 83 per cent.This contrast is all the more striking when oneconsiders that most jobless households haveaccess to income support in the OECD, whereasthis is not the case in South Africa.

Although participation rates are low andunemployment rates high in the lower deciles,those deciles nonetheless include a significantnumber of low-paid workers.

�About 30 per cent of employed workers arein households in the bottom five deciles.These workers are predominantly farmworkers and, to a lesser extent, domesticworkers.

�Only 13 per cent of manufacturing workersare in households in the bottom five deciles.Fully 77 per cent are in the top four deciles.

�Mineworkers are distributed more widely,with the largest numbers in deciles 4-7.

�In terms of occupation, people inprofessional, technical, managerial andadministrative jobs are unsurprisingly inhouseholds in the top two deciles.

�Most machine operators and similar semi-skilled workers are in deciles 6-9.

�Unskilled labourers are spread acrossdeciles 4-8.

In short, access to the labour market is an importantdeterminant of inequality. Whether an individualhas a job, or what kind of job he or she is able to get,plays a crucial role in determining their position inthe income distribution. Labour-market institutions(bargaining councils and the EmploymentConditions Commission) protect the incomes ofwage-workers. Those who do not have jobs,however, enjoy no such income support.

2.5.5 Wage inequalityDuring the apartheid era, racial discriminationwas an important determinant of wage

inequality. The contribution of racialdiscrimination to wage determination declinedsignificantly between 1980 and 1993, droppingfrom 20 per cent to 12 per cent of the Africanwage. The racial wage gap is now predominantlyexplained by factors other than discrimination,such as differences in education and skill,location (urban or rural), and economic sector.African workers have the lowest educationalqualifications, live predominantly in rural areas,and have the highest concentration in low-paying sectors such as agriculture. Education isparticularly important, with an estimated half ofthe difference in racial earnings attributed todifferences in educational qualifications.

Despite the decline in racial discriminationand in the wage gap between white and Africanworkers, overall wage inequality has notdeclined. This is because within-race wageinequality rose as between-race inequalitydeclined. The increase in wage inequalityamongst Africans was in part the result ofincreased occupational mobility. There was asignificant movement of Africans up theoccupational ladder, with the proportion in thelabourer and semi-skilled categories droppingfrom 57 per cent in 1980 to 38 per cent in 1993.As the number of Africans in higher-payingoccupations increased, so the gap between high-and low-paid African workers increased, thuswidening wage inequality.

Union membership appears to benefit thoseat the bottom end of the wage distribution themost. By boosting the incomes of low-paidworkers relative to higher-paid workers, the tradeunion movement would thus have acted tonarrow the wage distribution in the unionisedsector. But whether these efforts served to narrowthe overall distribution of household income,however, is another matter (as the gap betweenunionised and non-unionised wages would havegrown, and to the extent that job sheddingoccurred, the gap between the employed and theunemployed would have widened).

2.6 Changes ininequality in the 1990s

The distribution of income appears to havebecome more unequal between 1991 and 1996.The income share of the top decile increasedfrom 52,3 per cent to 53 per cent, whereas that ofthe poorest 40 per cent dropped from 3,8 per cent

Consolidated Report


to 3,4 per cent. This resulted in the Gini-coefficient rising from 0,68 to 0,69. Howeverracial income differences narrowed between 1991and 1996. The share of total income received byAfrican people rose from 29,9 per cent to 35,7 percent, whilst the share received by white peoplefell from 59,5 per cent to 51,9 per cent. The steadydecline in the share of total income received bywhites in the face of the steady increase in thatreceived by Africans from 1970 to 1996 is depictedin the figure 7.

Average incomes per capita among Africanpeople rose by 4,1 per cent per year, whilstincomes per capita among white people fell by 0,7per cent per year. Figure 8 depicts household

income growth by economic group within race.In 1991 only 9 per cent of the top or richestincome decile were African. In 1996 theproportion had risen to 22 per cent.

2.6.1 Intra-racialinequalities

At the same time as inter-racial inequalities weredeclining, intra-racial inequalities werecontinuing to grow.

�The Gini-coefficient for incomedistribution among African people rosefrom 0,62 to 0,66 and among white peoplefrom 0,46 to 0,5.

�The incomes of the richest 10 per cent of

Figure 8Growth of household income by race (1991-1996).

Figure 7Shares of income by race (1970-1996).

(Racial shares of income.)


Consolidated Report

African households rose by 17 per cent,whilst the incomes of the poorest 40 per centof African households fell by 21 per cent.

�Overall, relatively poor African and whitehouseholds experienced absolute declinesin their income, whilst relatively richAfrican, coloured and Indian householdssaw big income gains.

The real winners in terms of rising incomes werethe better off African and Asian households. Ofthe total real increase in income between 1991and 1996, 40 per cent went to the richest 10 percent of African people, and a total of 62,5 per centwent to the richest 40 per cent of African people.The poor majority of African people barelybenefited at all.

These findings are broadly corroborated bydata from the KwaZulu-Natal IncomeDynamics Study (KIDS). KIDS entailed tracingand re-interviewing in 1998 the African andIndian households in KwaZulu-Natal that hadbeen interviewed in 1993 as part of the PSLSDsurvey. KIDS found that the proportion whowere poor had risen from 35 per cent to 42 percent, and that a greater proportion of householdswere in the lower end of the income distributionin 1998 than in 1993. At the same time, theproportion of rich people had also grown. Bothtrends contributed to a rise of 4 percentage pointsin the Gini-coefficient between 1993 and 1998.

2.6.2 The growth/inequality debate

It is axiomatic that the greater the degree ofincome inequality, the higher the growth raterequired to reduce poverty. The difficulties ofgenerating sustained and sustainable growthrates high enough to reduce poverty at anacceptable pace have started to attract increasingattention in recent times. The Committeeconducted an extensive survey of theinternational literature into this question, andconstructed a simple model to compare theimpacts of different growth strategies on theincomes of the poor. The model�s results confirmfindings reported in other studies � modest andsustained redistribution, even under conditionsof relatively slow growth, does much more forthe poorest of the poor (the bottom three decilesof the income distribution) than does fast �trickledown� (distribution neutral) growth.

This finding, coupled with the finding that

some substantial proportion of the unemployedprobably fall into the category of �difficult-to-place�, and coupled as well, with the fact thatearnings for most of those catapulted into theinformal sector are very low, prompts theCommittee to recommend a re-evaluation of thebalance of resources devoted to poverty reduction.

Even if the only target were the eliminationof destitution, redistribution can achieve thismuch faster than sustained trickle-downeconomic growth. The strong likelihood is thatgrants will do more than merely alleviate poverty.

In many households, the guarantee of aminimum consumption level will enablehousehold members to engage in the risk-takingbehaviour so necessary to the generation ofadditional income.

2.7 Poverty in South Africa2.7.1 Recent evidence onpoverty in South Africa

New evidence of changes in poverty emergesfrom the preliminary results of a study that looksat socio-economic conditions in householdscontaining no workers.

�In 1995, Statistics SA reported that: 32 per centof African households (a minority of whichare pensioner households) were �workerless�(contained no employed people).

�By 1999, that percentage had risen above38. Translated into numbers of households,the data suggest that whereas there wereabout 1,9 million African workerlesshouseholds in 1995, that number had risento 3,1 million by 1999.4 Only a few ofthese were �true� pensioner households,i.e. households in which the pensioner didnot have to share a pension with otherhousehold members.

�Of approximately 210 000 Africanhouseholds in which there was no workingage person present (many of them so-called�skip generation� households), about182 000 of them spent, on average, lessthan R800 per month.

2.7.2 What is the level ofpoverty in South Africa?5

If a single set of statistics can disclose the extentof poverty in South Africa, it may be this � of theapproximately 717 000 live births in 1999 that

Consolidated Report


can be sorted by household expenditure category,about 176 000 took place in households wheretotal monthly expenditure was between R0 andR399. A further 231 000 babies where born intohouseholds where monthly total expenditure laybetween R400 and R799. Into the next incomeclass, R800 � R1 199 per month, some 119 000babies were born. Accounting for almost three-quarters of the total, their prospects are bleak.

About 11 per cent of households with childrenunder 7 went hungry in 1999 due to lack ofmoney to buy food. Another 2,3 millionhouseholds with people aged 7 years and olderwent hungry due to an inability to purchase food.The percentage of households reporting hungerin 1999 was 21,9 per cent.

Malnutrition remains one of the biggestcontributors to child morbidity and mortalityin South Africa. According to the national FoodConsumption Survey of 1999, nearly 20 percent of children aged 1-9 are affected bystunting, which is by far the most commonnutritional disorder in South Africa. Around 23,3per cent of children 1-6 are stunted.

Depending on which poverty line is used,researchers put the number of South Africansliving in poverty at anywhere between 45 and 55per cent (figure 9). Despite existing measures toaddress the various dimensions of poverty thereality is that, depending on the poverty line used,about 20-28 million citizens are living in poverty.The incidence of poverty differs between thedifferent provinces. In all estimates the WesternCape and Gauteng have the lowest rates ofpoverty, and Mpumalanga, the Eastern Cape andthe Northern Province the highest rates.

2.8 Addressingthe challenges of social

security reform in ademocratic South Africa

The distribution, extent and characteristics ofsocial exclusion in South Africa have a definitematerial and structural basis. In short, theeconomic aspects of social exclusion are linkedto the inability to command a sufficient flow ofresources to avoid growing inequalities and toprevent deprivation � be it nutritional, medical,in terms of shelter, or a lack of full participationin society.

2.8.1 Inequalities in thedistribution of wealth

There are five broad categories of economic flowsreflecting different economic positions in aneconomy. These include income derived fromowning property, income received in terms ofsalaries and wages, economic resources mobilisedthrough subsistence and household activities,transfer payments received from private orGovernment sources and self-employment. Eachof these categories is intimately connected to aset of economic relationships that define andstructure a modern capitalist economy.Exclusion from, or marginalisation within, thesesources of economic resource flows greatlyincreases the risk of poverty.

A key factor in South Africa is the skeweddistribution of economic assets. Apartheid wascentral to this skewed distribution, driving thesocial exclusion of the majority and social

Figure 9Percentage of population in poverty.


Consolidated Report

inclusion of the minority. In particular, theapartheid regime constructed citizenship andsubject rights to determine which groups wouldhave access to what level of social protectiondepending on their functionality to the raciallyconstructed economic and social system.

The following are all examples of apartheidmeasures, which excluded the majority of people.

�The Land Act of 1913, which confined theland area that Africans could legally own orrent to 13 per cent of South Africa

�The 1913 Mines Act, which contained thefirst of many job reservations policies

�The Urban Areas Act of 1945

�Coloured Preference Policy

�The Group Areas Act of 1956, whichrestricted African access and Africaneconomic activity in the urban areas.

White South Africans were from 1924 onwardsgiven substantial protection against poverty andvulnerability, partly by measures to exclude blackSouth Africans referred to above, and partly bythe introduction of social and economic policiessimilar to those adopted in the social democraticcountries of Europe. Black South Africans weregenerally either excluded from these positivemeasures, or were protected to a much lesserextent than the white counterparts. For example,substantial maintenance grants were paid tosingle mothers with low income, but not if theywere African.

Over the last few decades of the 20th century,blacks began to gain access to work relatedbenefits and social insurance institutionsprimarily due to increasing unionisation andpolitical pressure. This formed an importantpressure that precipitated the breakdown of thecitizenship/subject relationship underpinningthe apartheid regime.

As democracy approached, the expectations ofthe excluded majority increased in regard to therole of the post-apartheid state in ensuring redress,employment and development. However, theapartheid regime, seeking to pre-empt effectivepost-apartheid intervention, sought to pushthrough various pieces of legislation looseningthe grip of Government over social andeconomic policy. In some instances they weresuccessful (such as deregulating aspects of privatehealthcare and food production). In other areas,such as broader economic policy, social resistance

manifested through mass mobilisation and theestablishment of tripartite socio-economicforums restrained them.

2.8.2 Shortcomingsin the South African social

security systemFirst, regarding non-contributory socialassistance, there are large gaps resulting in a largeproportion of the poor being excluded, and thosewho are uncovered are often not givenappropriate support.

The Social Assistance Act covers the followingcategories of people:

�Aged Women over 60 years of age, and menover 65 years of age receive a state old agegrant of R570 per month. This grant is thelargest current social security transfer inthe country, and, for those elderly personswho receive it, the grant plays a pivotalpoverty alleviation role for the entirehousehold.

�Disabled There is a disabled grant of R570per month for medically-diagnoseddisabled persons over 18 years of age. Thisgrant is a de facto poverty grant, as 77 percent of recipients are also in poverty.

�Foster care There is a grant of R470 permonth for caring for foster families caringfor children under 18 years of age.

�Care dependency grant There is a grant ofR540 per month for parents of a disabledchild (0-18) who requires care at home byanother person. Thereafter applicationmust be made for the adult disability grant.

�Child support grant There is a childsupport grant of R110 per month, paid tothe primary caregiver, for children underseven years of age. This grant has beenafflicted by a slow take-up rate, with only25 per cent of the targeted group receivingthe grant three years after implementation.

The following categories remain uncovered bysocial assistance:

�Children 75 per cent of poor childrenbelow seven years of age do not get thechild support grant. All children over sevendo not get any support. Finally, all childrenwithout primary caregivers, and child-headed households (which are in theincrease) do not get any grant.

Consolidated Report


�Disabled Those with a chronic illnessbut who do not meet the strict medicallybased criteria are excluded. Thus if thedisability is not medically complete, butdoes prevent that person from carryingout their trade, they would not qualifyfor the grant.

�Unemployed The UIF covers only 5 percent of the unemployed. Thereforeexcluding those unemployed gettingdisability and childcare related grants,about 5 million unemployed people arewithout any form of income support fromthe social security system.

�Poverty Those with incomes below thepoverty line, including working poor, arewithout any social security transfers.Currently about 60 per cent of the all thepoor, or 11 million people, are uncovered.

�Non-citizens While the Constitution inS27 (1) (c) states that �everyone� has aright to social security, current socialassistance mostly excludes non-citizens. Inthis regard, there will probably beconstitutional pressure to ensure all people(including illegal immigrants) have accessto certain basic services (such asemergency healthcare), and full access tocertain categories such as refugees.

In short, there are large gaps and inadequacies inthe social assistance system.

Second, regarding contribution-fundedsocial insurance and regulated private schemes,these, too, cover a relatively small number ofthe population. Moreover, the increasingnumbers who fall outside of the formal sectorundermine the scope of social insurance�scontribution base, further limiting the system.In addition, there are internal distributioninadequacies within social insurance thatprovides the most vulnerable workers with asmaller share of benefits. The better paidgenerally seem to secure the largest share ofbenefits. Furthermore, some of the mostvulnerable workers are often legally excludedfrom the system.

�The limited nature of the publiclyprovided benefits means that social securitycosts are, de facto, passed onto employers.This has increased non-wage costs in theeconomy. This acts as a disincentive fordirect employment, that is, the

employment of regular workers. Theincrease in indirect employment(including casualisation), and thestagnation of net permanent jobs, is partlythe result of employers trying to avoidthese non-wage costs.

�The growth of employer-linked benefitsnecessarily excludes a significant share ofthe population. The unemployed,informally employed and many temporaryworkers have no access to these benefits. Acurrent legal framework that cannot copewith the changing forms of employmentaids this. In the South African situationthis forces the non-wage costs to be higherstill, since workers require more to supportmany unemployed persons (with nobenefits) in their household. Consequently,this contributes to upward pressure onworkers� remuneration. This extremepressure on the breadwinner effectivelyundermines any worker support for labourflexibility � since loss of employmentequals loss of income.

�Private-provision schemes are contributingto escalating costs of services. Healthcare isa primary example of this. South Africaspends, including both public and privateexpenditure, about almost twice on healththan considered necessary by the WorldHealth Organisation � all this for relativelyinadequate public and private service. Thenational overspend derives primarily fromthe cost-intensive private medical schemeenvironment that has created perverseincentives for over-servicing by privateproviders (who are paid on a fee-for-servicebasis), excessive administrator profit-takingand undermined consumer protection.The relatively high levels of expenditure isdue to most health spending going towardsthe well resourced, private health sectorwhich caters for only 15 per cent of thepopulation.

�Profit-driven provision of services hasoften been accompanied by a removal ofcross-subsidies to those unable to affordservices. There is a tendency to �cherry-pick� high-income, low-risk groups thataccordingly increases the difficulty offunding publicly provided social servicesthrough contributions.


Consolidated Report

2.9 The challengesThe inadequacy of current interventions, in acontext of persistently high risk and deprivation,has contributed to several growing and,potentially, unsustainable challenges. The mostimportant of these are the following:

�The wage-income relationship is breakingdown � High unemployment, includingthe massive net loss of formal sector jobs,and growing shift towards so-called�atypical� work, has reduced the incomesof the poor. Historically, the working poorhave supported the poor and unemployedvia remittances and intra-householdtransfers. However, this relationshipbetween formal sector wages andhousehold incomes has declined due tothe considerable loss of net formal jobs,and the downward qualitative shift informal employment being created. In thisperiod, there has been a decline in theincomes of the poorest 40 per cent.

�The state is vulnerable to ConstitutionalCourt challenges � The ConstitutionalCourt, in its State v. Grootboom judgement,has increased pressure on the state to put inplace a coherent and comprehensiveprogramme for progressively realising theconstitutional obligations. This requiresdevising, formulating, funding,implementing and constantly reviewingrelevant measures. The Grootboomjudgement opens the way for furtherconstitutional challenges against the stateon the basis of not complying with theConstitution�s Bill of Rights. It shouldalso be noted that the ConstitutionalCourt has the power to enforce socio-economic rights, with direct implicationsfor budgetary matters.

�The impact of AIDS will exacerbate povertyand inequality � Research conducted forthe Department of Health indicates thatthere will be 5,5 million people infectedwith HIV/AIDS by 2004, and the impactwill be increasingly felt in society. Theimpact is apparently already evident onsocial service institutions (mainly publichealthcare). This will increase downwardpressure on households and householdincomes, and could significantlyundermine the country�s medium- tolong-term economic growth and social

development potential.

�The racially differentiated composition ofpublic service users reinforces apartheid-styleprejudice � The poor (mainly black) aredependent on strained public services. Therich (mainly white) make use of privateservices. This contributes to a lack of racialtolerance and understanding, andundermines social solidarity.

�Delivery of key services is affected byinability of poor to pay for inclusion �More people have access to important basicservice such as telecommunications, waterand sanitation, electricity, housing, andprimary healthcare. While these have beenhard fought gains, this extension of serviceshas been undermined by an inability of thepoor to afford payment.6

�Poverty-related increase in crime and socialinstability � This is potentiallyundermining to legitimacy of newdemocracy, and investment strategies.Levels of crime remain far too high, andinclude an increasing incidence ofdomestic violence. This reflects theunderlying causality of poverty and theresulting depressed aspirations among thepoorest in society.

�Social development/investment backlogs arenow widely recognised as barriers toeconomic growth and development There isgrowing recognition among internationalfinancial and credit rating institutions,national Government and domestic socialformations, that insufficient socialinvestment and social developmentbacklogs are a primary barrier to theachievement of sustainable levels ofeconomic growth and development.

The findings in this chapter are based on thepolitical economy of South Africa�s socialsecurity system and the challenges posed bysome of its key features. The demographic trendsand impacts of poverty, HIV/AIDS and otherchronic illnesses as well as unemployment in theformal sector reinforce the need for acomprehensive approach to social security.

Furthermore, existing levels of poverty havereached unsustainable levels, and left unattendedhave the potential to reverse democratic gainsachieved since 1994. The urgent need to addressdeepening social exclusion and alienation of

Consolidated Report


those households living in destitution cannot beignored. A redesigned social security system hasto respond to the constitutional, democratic andhuman development imperatives that providethe framework for transformation.


Consolidated Report

Consolidated Report


Chapter 3Approach to aComprehensive SocialSecurity Provisioning

3.1 IntroductionThe two-fold nature of the South Africaneconomy means that a comprehensive socialsecurity system has to deal with two different setsof needs. The first relates to the needs of the poor,excluded largely from the productive capacityand rewards of the formal economy. The secondrelates to the security needs of the informallyemployed. Without some protection against thecontingencies of death, disability, ill-health andretirement, even the comparatively wealthybeneficiaries of the formal economy can bereduced to destitution. The brief of theCommittee covers both areas.

Tackling poverty and deprivation, and itseffects, is thus a critical challenge facing SouthAfrica. The Reconstruction and DevelopmentProgramme (RDP) states that:

No political democracy can survive and flourishif the mass of our people remains in poverty,without land, without tangible prospects for abetter life. Attacking poverty and deprivationmust therefore be the first priority of ourdemocratic Government.

This challenge, in its general sense, is of coursenot a new one, having been with South Africathroughout its past. However, the coming intobeing of a democratic dispensation in 1994,followed by a new Constitution (with a Bill ofSocio-economic Rights) in 1996, has presentedthe nation with a unique opportunity to find apath away from this legacy.

The task of addressing, in the final instance,

the reality of poverty and deprivation is generallyregarded as a central feature of a country�s socialsecurity system. In South Africa, however, up to60 per cent of the poor are not getting any socialsecurity transfers at all. Further, the current socialsecurity system, principally for reasons ofinherited design, is archaic, lacks integration, andhas many gaps.

Economic globalisation is posing furtherchallenges through changing labour markets andtechnological challenges. These changes aredisplacing full-time regular employment andchanging the nature of work. Most new jobsbeing created are in the �informal sector�, or of apart-time, casual, temporary, or home-basednature. There is thus a growing army ofunemployed, underemployed and working poorsubsisting alongside an increasingly threatenedpermanent workforce. The socio-economicchallenge facing South Africa is made moreominous by the danger that the dynamics ofglobalisation may further fasten (at least in theshort to medium term) onto these existingrelations of vulnerability and exploitation, andexert pressure to intensify them. Theconsequences would be growing poverty,inequality, social polarisation, job insecurity, andcrime � and a fraying social fabric.

With this scenario in mind, the 1998Presidential Job Summit � comprisingGovernment, labour, business, and community� agreed to �implement an effectivecomprehensive social security system, aimedespecially at those living in poverty and theunemployed�. This agreement aligns well with


Consolidated Report

the needs articulated in the White Paper for SocialWelfare (1997) and the South AfricanConstitution�s Bill of Rights (especially S27(1)(c)).

In short, South Africa faces two sets ofimperatives. The first is the constitutional anddemocratic imperative, centred on a humanrights approach. The Constitution gives socio-economic rights exactly the same status as civiland political rights. In particular, the followingaspects are relevant:

�Ensure promotion of values of dignity,equality and freedom

�Build participation and voice of theexcluded

�Support citizenship claims throughequality of administrative justice, access toinformation, application procedures,adjudication of rights, monitoring ofcompliance and non-compliance.

The second is the socio-economic imperative,encapsulated in the RDP, to fundamentallyimprove the living standards of all people in thecountry. The socio-economic imperative stressesthe following:

�Reduction in poverty, deprivation andsocial inequality

�Increased access to adequate basic services

�Create an environment for sustainablesocial and economic advancement of allpeople, and especially the poor andunemployed.

Both these sets of imperatives are inter-relatedand mutually reinforcing. The ConstitutionalCourt, in the matter of The Government of theRepublic of South Africa et al v. Grootboom et al, stated:

There can be no doubt that human dignity,freedom and equality, the foundational values ofour society, are denied to those who have nofood, clothing or shelter. Affording socio-economic rights to all people therefore enablesthem to enjoy the other rights enshrined in [theConstitution].7

Thus there is a clear need to develop a new,comprehensive social security system thatsupports the achievement of socio-economicrights, and in so doing the overriding values ofSouth African society. In this regard, this chapterof the Committee�s report begins to outline theconceptual framework for such a system.

In developing a conceptual framework forSouth Africa, the Committee considered,amongst other things, the question of whetherthere is an international convergence of socialsecurity reform, and the potential implicationsof such for South Africa.

3.2 Internationaltrends in social

security provisioningThe term �social security� has, internationally,attracted a wide range of meanings, and needsto be clarified at this point. In developedcountries, where the term first originated, socialsecurity refers mainly to the following:

�Social assistance �This refers to stateprovided basic minimum protection torelieve poverty, essentially subject toqualifying criteria on a non-contributorybasis.

�Social insurance � This refers to amandatory contributory system of onekind or another, or regulated privatesector provision, concerned with thespreading of income over the life cycle orthe pooling of risks.

Social security, as defined by its Europeanorigins, developed as a complement to theformal employment relationship.

3.2.1 Western EuropeIn many Western European countries, welfaresystems are undergoing significant change.Three causes of the �welfare crisis� are oftenidentified.

�The first is that welfare states stifle themarket and erode the incentive to work.

�The second is the demographic challengein that long-term effects of ageing areundermining inter-generational basedsolidarity systems.

�The third is that the global economypunishes high Government socialexpenditure and uncompetitive economies.

The Committee�s analysis of some WesternEuropean social security systems indicates thatperceptions of a �welfare crisis� appear somewhatexaggerated. Indeed most of the problems facingwelfare states are exogenous � essentially drivenby increasing costs due to badly performing

Consolidated Report


economic policies and labour markets thatproduce an overload on existing socialprogrammes. Where the social security system�sinternal workings cannot respond to the new risksand needs of the socio-economic order, however,the causes of the crisis are endogenous.Additionally, important aspects of the reformdebate may reflect an ideological shift or approachto the concept of social security � for example,the desire to create a private market in socialservices in place of a state function.

3.2.2 The United StatesIn the United States (US), the main focus inrecent times has been to promote �back to work�schemes, through a combination of incentivesand disincentives. However, inequality andpolarisation have risen even with increasedlevels of job creation. While the incomes of 80per cent of working families have stagnated,incomes for the top 20 per cent have risensharply. Under the competitive pressures of theglobal economy, employers increasingly seek tolower their fixed labour costs and thus excludemany categories of workers (mainly non-unionised, atypical and women) from socialsecurity benefits. Hence the US has a very largeproportion of its population without healthcare,for example. Social polarisation and exclusionis extensive.

3.2.3 South East AsiaIn the �emerging� economies of South East Asia,the competitive wage cost advantage isevaporating (due to competition from evencheaper economies), forcing these countries topush towards new social programmes. Furtherthe recent global economic crisis, which centredon South East Asia, has led to rapid job lossesand expanding unemployment. This hasmotivated the development of unemploymentinsurance in these countries that, until recently,had achieved close to full employment.

3.2.4 Developingcountries

In developing country contexts, such as LatinAmerica and Africa, the problem is usuallypoverty, chronic inequality and exclusion fromthe informal sector, and the extent of �atypical�employment (as it is understood in Europe) thatcomprise up to 50 per cent of all jobs. Themajority of the population often stands outsideformal systems of social security, being engaged

in rural and self-employment. Therefore theEuropean social security focus on the risk offormal sector job loss is generally less relevanthere.

Further, the contributory-social insurance biasinherent in unemployment benefit schemes willhave limited effect in the context of high andpersistent levels of unemployment and growinginformal work. In such an environment, therewould be little possibility of insuring oneselfagainst the �contingency risk� of unemployment� rather the entire environment would be one ofuncertainty, in which insurance would beimpossible. Further, attempts to get the all of theworking poor and socially excluded to contributeto such systems invariably fall short.

The implication of the above analysis is thatthere is no uniform system that is generallyapplicable across countries. Rather one can inferthat a country�s social security system needs toaddress its own particular set of risks andchallenges in a manner that best reflects itssocietal values and resource base.

3.3 Implications forSouth African context

In looking at international reforms, theCommittee has considered the pressures thatprimarily motivate the reform dynamic. In thecontext of globalisation, there is increasingpressure to promote social security markets inhealthcare, retirement, education, welfare andlivelihood services. This pressure is premised onthe understanding that private markets are ableto achieve better efficiency gains than the publicsector. The Committee�s research indicates thatin such markets the tendency is to create profitsrather than address underlying social needs. As aprincipally people-centred set of concernsmotivates South Africa�s need for social securityreform, any conflicting market-centredmotivations need to be made explicit, andevaluated against their ability to supportfundamental social objectives.

Referring to the �mischievous euphemisms�behind which reform has taken place, Standingnotes that:

There has been �deregulation� that hasinvolved many new regulations, and there havebeen �safety nets� without safety, as millionsmore people are pushed into poverty and asinequalities have grown.8


Consolidated Report

Underlying the mischievous euphemism of�social safety nets� are three principal changes.These he identifies as:

�Increased selectivity of state transfers

�Multi-tierism in modes of provision ofsocial protection

�Partial privatisation of social policy.

Developing social security systems in aglobalising era characterised by insecurity hasled to significant debates on what type ofapproach or mix of interventions can bestrespond to the continuing crises of livelihoodsand human security. �Third Way� supporterssuch as Giddens9 are eloquent on the need forsocial democrats to find ways to deal with risksthat welfare states are no longer able to address.Giddens refers to �technological change, socialexclusion or the accelerating proportion of one-parent households� but says little about theincreasing risks encountered in the labourmarket. Standing gives an indication of the needfor coping mechanisms to be developed torespond to insecurity.

Labour security, previously the basis ofwelfare policy, has given way to insecurity alongthe following axes:

�Labour market insecurity has grownalmost globally, with much higherunemployment, slower rates ofemployment growth (except in the US)and higher �labour slack�.

�Employment insecurity is high and rising,with growing proportions of those in thelabour force having insecure employmentstatuses and more workers lackingemployment protection.

�Work insecurity has become greater, dueto more people being in work statuseswithout coverage by protectiveinstitutions and regulations.

�Job insecurity has worsened, with moreworkers having to switch jobs and learnnew ways of working.

�Skill reproduction insecurity is considerable,in part because skills become obsolescentmore quickly and because few workers arereceiving career skills.

�Income insecurity is greater for those inemployment, due to flexible wages and soon, and for those outside employment,due to explicit disentitlement to benefits.

�Representation insecurity is growing due tode-unionisation, erosion of �tripartite�institutions and the changing character ofcollective bargaining.

International experiences also reflect two trendsthat are central to the South African discussion.These concern the merits of a social securitystrategy focused on the unemployed, and theusefulness or otherwise of �workfare�.

3.3.1 Focus primary socialsecurity interventions on

formally unemployed?With wages being the key source of income,there is clearly a need for a range of nationalpolicies to focus on increasing employment andreducing unemployment. In terms of socialprotection, however, there is a need to considerthe condition of being unemployed within theoverall context of poverty and social exclusion.

Social security in industrialised countrieslargely developed around formal sectorunemployment. These countries traditionallysaw unemployment as their big problem, andtherefore focused on �contingency risk� involvedin the loss of that employment. However, massunemployment and long-term unemploymentlasting for a year or more is returning to thesecountries. At the same time, more flexible andinformal labour markets mean that full-time,regular and stable wage labour is no longer theoverwhelming norm.

Therefore the concept of social security, asfocused on the formal sector unemployed, isincreasingly challenged. For this reason theCommittee has had to consider whether, in thiscontext, social security reforms should primarilyfocus on the unemployed. Is the person inchronic �under-employment� not just as�deserving� of income security? Why provideincome support to someone with zero hours ofwork last week, and not to someone who did afew (two) hours?

In developing countries, where stable full-timewaged formal sector labour was never the norm,it is increasingly unlikely that it will become thenorm. Moreover unlike industrialised countries,large proportions of the formally employed arein poverty and are categorised as the �workingpoor�. In the context of a labour surplus economy,more and more people are being pushed into theinformal economy. The Committee�s research

Consolidated Report


into unemployment trends and workerlesshouseholds reveals that those involved ininformal work or in the �informal sector� alsotend to fall into the category of the working poor.

In short, there is a growing need for a platformof general social protection that supports both theunemployed and the working poor.

3.3.2 Workfare?Workfare originated in the US. Its philosophicalunderpinning is the ancient conservativedistinction between the �deserving� and the�undeserving� poor. Its theoreticalunderpinning derives primarily from theorthodox economic perspective that attributesunemployment largely to the behaviour andexpectations of workers. In a sense, the neworthodox economics regards unemployment aslargely �voluntary�, due to behavioural andinstitutional rigidities.

Therefore one response to the persistence ofhigh unemployment and the perceivedbehavioural rigidity has been to move socialprotection towards more active or regulatorysystems. This typically links entitlement tobenefits and the obligation to take a low-paying

job or labour market training place.

The overall economic context in whichworkfare emerged in the US was the attempt tokeep unemployment to a minimum (to maximiseemployment) by allowing wages at the bottomend of the labour market to be set at market-clearing levels. These lower-end wages diddecline, resulting in household incomes for thebottom 20 per cent falling from $10 000 in 1977to $8 800 in 1999.

A focal point of workfare activism in the USwas against a piece of legislation called Aid toFamilies with Dependent Children. Knownpopularly as �Mother�s Pensions�, it originatedin the period 1911-1920 as a form of socialprotection �for �worthy�, Protestant, whitewidows�. Increasingly claimed by African-Americans and other minorities such as thedivorced; the separated; the deserted; andincreasingly, the never married; the numberson welfare rolls grew between the late 1950suntil the early 1980s from 2 million to about13 million.

Allegedly due to the rising cost of theprogramme, conservatives began to advocate avariety of measures such as �� work

Common pitfalls for countries� mix of SP interventions

In many countries the mix of socialprotection (SP) interventions havesuffered several generic pitfalls. TheWorld Bank has highlighted thefollowing pitfalls10

�Trying to cure the ills caused by poor policychoices more generally (for exampleinappropriate macro structural policies).

�Lack of co-ordination of the many diversepolicies, programs, and actors involved inSP interventions.

�Having so many interventions that fewhave adequate resources to operateefficiently, much less to accomplish theirobjectives.

�Missing the possible synergies andcomplementarities between programs(which leads to duplication or to missedeconomies of scale).

�Expanding the intervention�s coverage orbenefit level without dealing with the

design or implementation issues that wouldmake the interventions more effective.

�Focusing on the groups for which there ispopular support but only a moderatecorrelation with poverty � for example, insome countries formal sector pensionsmay not reach the poor.

�Concentrating attention on the formalsector when poverty is largely in theinformal sector, or on urbanoccupations when poverty is largelylinked to agricultural activities orresidence in rural areas.

�Failing to reach groups that may be highlycorrelated with poverty but outside thereach of traditional mechanisms orsympathies � for example refugees,internally displaced persons, and ethnicminorities.

�Not taking into account long run impactwhen designing initial interventions.


Consolidated Report

requirements for all able-bodied parents over age21, and adoption for children whose parents areunable to provide support through work, familyor private charity �� The real issue, researchhas shown, was not the increasing cost (whichwas actually relatively small) but that �welfare�has operated as a code-word for tensions overrace, gender and ethnicity, focusedoverwhelmingly on young African-Americanwomen, allegedly breeding a criminal�underclass�.

The result of workfare is that while welfarerolls have declined, it has resulted in littlesustainable job creation. Analysis of the low-wage labour market into which nearly all of theworkers from these families have been divertedreflects an abundance of part-time, temporary,contract or contingent work with no benefits,and for which there is often quite stiffcompetition. Despite the rising prosperitybrought about by sustained economic growth,the proportion of the population below thepoverty line continued to rise � from 11,8 percent in 1997 to 12,8 per cent in 1998. Of the 34million poor, some 13,8 million survived onincomes less than one-half of the poverty line.

Despite this background, there have beensuggestions that workfare is a viable concept fordeveloping countries. However, workfarepolicies require the existence of jobs. In a contextof structural unemployment, as is the case inSouth Africa, such policies are unlikely to haveany positive impacts. Further the workfareexperience in the US has shown itself to be veryadministrative-intensive and expensive system,with little sustainable job creation.

Apart from these more technical drawbacks,in the South African context a workfare scheme,generated on notion of an �undeserving poor�,or past apartheid state�s manipulations of thelabour market, may counteract the democraticstate�s efforts to rid society of its race-codedprejudices.

3.4 Defining anappropriate concept ofcomprehensive social

security for South AfricaIt follows from the earlier discussion that theextent to which one can adopt the traditionalemployment-centred concept of social securityfor South Africa can be questioned.

As a result of these weaknesses in the traditionalconcept of social security, the concept of �socialprotection� has originated, largely toaccommodate the realities of developingcountries.

The United Nations (UN) Commission onSocial Development describes socialprotection as:

Social protection embodies society�s responsesto levels of either risk or deprivation � Theseinclude secure access to income, livelihood,employment, health and education services,nutrition and shelter.

Further, the UN Commission notes that:

The ultimate purpose of social protection is toincrease capabilities and opportunities and,thereby, human development. While by itsvery nature social protection aims at providingat least minimum standards of well-being topeople in dire circumstances enabling them tolive with dignity, one should not overlook thatsocial protection should not simply be seen as aresidual policy function of assuring the welfareof the poorest � but as a foundation at a societallevel for promoting social justice and socialcohesion, developing human capabilities andpromoting economic dynamism and creativity.11

Clearly a broad conceptualisation of socialprotection has many merits for South Africa.

�First, it incorporates developmentalstrategies and programmes moreappropriate to a developing country such asSouth Africa. For instance, it increasesopportunities for people doing �informal�work to gain access to social protectioncoverage.

�Second, it provides a coherent frameworkfor integrating existing and proposed socialand economic policy interventions. Thesewider functions and objectives of socialprotection are better able to address sociallyand economically embedded problems,new risks and increased vulnerabilities.

�Third, social protection could create addedpotential for integrated and linked private,public and community sectorinterventions and benefit systems.

For these reasons, the Committee of Inquiry hastaken on board the concept of social protection.

Consolidated Report


Table 6Matrix of means and ends12

MeansEnds to promote Creation of Improvements Building capacities

entitlements in terms ofexchange

Healthy Access to Grants andhealthcare, institutionalwater, sanitation reforms

Productive Redistribution Restructuring of Improving access toof assets markets and and affordability of

redistribution of education andopportunities economic services

Secure lives Tenure rights Social welfare and Community andsafety nets individual safety

However, such a system in South Africa, evenmore than suggested by the UN Commission,needs to be embedded in economic organisationand social relations enabling it to address thecountry�s underlying structural and materialbasis of social exclusion.

For this reason, the Committee of Inquiry hassettled on the term Comprehensive SocialProtection (CSP). The Committee definescomprehensive social protection thus:

Comprehensive social protection for SouthAfrica seeks to provide the basic means for allpeople living in the country to effectivelyparticipate and advance in social andeconomic life, and in turn to contribute tosocial and economic development.

Comprehensive social protection is broaderthan the traditional concept of social security,and incorporates developmental strategies andprogrammes designed to ensure, collectively, atleast a minimum acceptable living standardfor all citizens. It embraces the traditionalmeasures of social insurance, social assistanceand social services, but goes beyond that tofocus on causality through an integratedpolicy approach including many of thedevelopmental initiatives undertaken bythe State.

This definition is consistent with that of theWhite Paper for Social Welfare of 1997, whichbegan to map out the need for a broadconceptualisation of social security in SouthAfrica. At the time it gave the objective ofcomprehensive social security as the

�provision of a national social security system�with the ultimate goal of ensuring that �allSouth Africans have a minimum income,sufficient to meet basic subsistence needs, andshould not have to live below minimumacceptable standards�.

3.4.1 A comprehensivesocial protection �package�

CSP will work through a variety of mechanisms,embracing a �package� of social protectioninterventions and measures. The need for apackage derives from an understanding thatthere are certain basic requirements that shouldbe available to all, and not subject to beingtraded off against each other. For example, it isnot acceptable to ask a poor parent to choosebetween attaining a certain level of householdincome or sending their children to school,though this is not an uncommon choice inreality.

Further, a package approach enables one toachieve a degree of balance between measuresfocused on reducing income, services (capability)and asset poverty. In this way, a dependence oncash benefits, ignoring the potential for basicservice cost inflation, is avoided, or vice versa.Rather a poor person is guaranteed some cashsupport and a basic level of service delivery. Thisallows comprehensive social protection to betterdeliver on minimum acceptable living standardoutcomes.

The �capabilities� approach developed byAmartya Sen, the recent Nobel-laureat, has beenuseful in developing the content of the CSP


Consolidated Report

Application Key componentsIncome poverty Universal (a) � Basic Income Grant

� Child support grant� Maintained state Old Age grant

Capability poverty Universal/ � Free and adequate publicly-providedEligibility criteria healthcare(b) � Free primary and secondary education

� Free water and sanitation (lifeline)� Free electricity (lifeline)� Accessible and affordable public

transport� Access to affordable and adequate

housing� Access to jobs and skills training

Asset poverty Universal/ � Access to productive and income-Eligibility criteria generating assets such as land and credit(c) � Access to social assets such as

community infrastructure

Special needs Eligibility (d) � Reformed disability grant, foster carecriteria grant, child dependence grant

Social insurance Eligibility (e) � Cover for old age, survivors�, disability,unemployment, and health needs

package. Basic incomes, services, and assets emergeas central components of the �capabilities� approach.This is set out in table 6.

In identifying the practical aspects of such anapproach, and taking into account necessaryadaptations for South Africa, the Committee ofInquiry has arrived at the following:

a)Measures to address �income poverty� Thisincludes measures to ensure that peoplehave adequate incomes throughout theirlife cycle, covering childhood, workingage and old age. Income poverty can beaddressed through a range of measures.However, the CSP package shouldcomprise at least one primary incometransfer which ensures that all SouthAfricans have some income to mitigate oreradicate destitution and starvation. Abasic level of income would also haveother developmental spin-offs related toenabling that person to participate moreeffectively in the economy (for example,afford the bus fare to engage in job search).

b)Measures to address �capability poverty�This can be achieved through the provisionof certain basic services deemed crucial to

enable a person to live and function insociety. This includes the provision of basic(lifeline tariff) water and electricity, free andadequate healthcare, free education, foodsecurity, and affordable housing andtransport.

c) Measures to address �asset poverty� Thisincludes income-generating assets, such asland, and social capital such as communityinfrastructure. This addresses the keyunderlying structural basis of poverty andinequality in South Africa.

d)Measures to address �special needs� Thisincludes mainly standard measures toaddress special needs such as disability orchild support.

In the CSP package, (a) + (b) + (c) are coreelements of the comprehensive social protectionbasic platform that should be available to allSouth Africans (including certain categories ofnon-citizens). In general, these components needto be established as a universal-as-possiblepackage of income transfers, services and accessprovided in a non work-related manner andwhose availability is not primarily dependent onan ability to pay.

Table 7Comprehensive social protection package and components

Consolidated Report


To this basic floor, (d), which addressesspecial needs, and largely as it currentlyapplies, will be added. A social insurancecomponent (e), reformed for purposes ofinclusivity, equity, consumer protection andefficiency of the benefit types, completes thepackage.

Crucially, what are the key components ofsuch a CSP package? After detailed analysis ofthe social and economic cost-benefits ofpossible components, the key income transfers,services, assets and special needs measures putforward by the Committee of Inquiry are setout in table 7.

3.4.2 Determining the�minimum� requirements

for the CSP packageThe Constitution obliges the state to takepositive action to meet the needs of those livingin extreme conditions of poverty, basic services,and suffering from a lack of access toconstitutionally stipulated socio-economicrights.

The difficulty for the state, and anyoneinsisting on the state�s obligations, is that the�minimum essential level� must be describedfor each of the socio-economic rights (forexample, the right to adequate housing). As theConstitutional Court, in The Republic of SouthAfrica et al v. Grootboom et al, has observed:

It is not possible to determine the minimumthreshold for progressive realisation of the rightto access to adequate housing without firstidentifying the needs and opportunities forthe enjoyment of such a right. This will varyaccording to factors such as income,unemployment, availability of land andpoverty. The differences between city andrural communities will also determine theneeds and opportunities for enjoyment of sucha right. Variations ultimately depend on theeconomic and social history and circumstancesof the country.

While the Grootboom case has emphasised thatit is incumbent on the state to take reasonablemeasures to give effect to each one of theserights, the Committee believes that this shouldbe translated into making available a minimumlevel or measure of provision to everyone. As aresult, it may be advisable for the State to

stipulate up front its considered minimumobligations for service delivery, such as it isdoing for the free water programme, and itsintended schedule for progressively realisingthis.

Further, even while the state is rolling outthese medium- to long-term programmes, it ishas to ensure �temporary� relief for the poorwho are �particularly vulnerable�. In alllikelihood, the state will be unable to ensurethat all of its capability and asset programmesadequately have built-in measures fortemporary relief for those most vulnerable.The result is that the state is again exposed toConstitutional Court challenges, andinstances where the poor feel forced to takematters into their own hands (such as with landinvasions).

In this regard, the �income poverty� aspectof the CSP package is relevant, particularly forthree reasons:

First, income poverty measures are easier torollout in the short term than moreinfrastructural and institutional intensive�capabilities� and �asset� poverty programmes.

Second, people who are in �capability� and�asset� poverty, or for that matter having�special needs�, are invariable also facing�income poverty�.

Third, the Constitutional Court hasrecognised that if the state were providing bettersocial assistance to the poor there would be lesspressure on other socio-economic rights.

The poor are particularly vulnerable and theirneeds require special attention. It is in thiscontext that the relationship between sections26 (housing) and section 27 (social security)and other socio-economic rights is mostapparent. If under section 27 the state has inplace programmes to provide adequate socialassistance to those who are otherwise unableto support themselves and their dependants,that would be relevant to the state�s otherobligations in respect of other socio-economicrights.

In other words, the state could buy time forprogressive realisation of its other socio-economic rights if it improved income transfersto the poor in the short term.


Consolidated Report

3.5 Economic,institutional and social

mobilisationimplications ofcomprehensive

social protectionWhile the Committee deals with financial andinstitutional implications in more detail later inthis report, at this point it may be valuable insharing some of the important considerationsin this regard that were considered by theCommittee in developing its conceptualisationof comprehensive social protection.

3.5.1 Social protectionand the economy

The PIR concluded that the perpetuation ofextreme poverty in South Africa would mostlikely act as a brake on the Government�seconomic growth strategy. And where highergrowth was achieved, a noticeable reduction inpoverty and inequality may not follow. As a result,it proposed that South Africa could pursue moreredistributive policies without underminingcurrent growth objectives � and rather that suchpolicies would instead promote economic growth.

Indeed, the UN Commission for SocialDevelopment finds that:

Experiences of countries successful in economic,political and social terms show that economicdevelopment and social protection are mutuallyreinforcing � essentially they are elements of thesame paradigm. Any trade-off between publicspending items, between various economicneeds and the need for social protection mustincorporate recognition of the long-term socialpathologies � This approach (of objecting tosocial protection because it costs too much) hasproved to be short-sighted and superficial.

Further, the UN Commission states, socialprotection facilitates the process of social andeconomic change by moderating the costs ofeconomic transition and structural change. Byproviding a cushion, it can encourage thenecessary economic restructuring.

Internationally, two factors have beenimportant in shaping the limits, or otherwise, of�affordability�. These two factors are:

�A country�s level of economic developmentThe level of economic developmentbroadly determines the limits of the socialsecurity system. It is clear that, all otherthings being equal, a rich country canafford to provide a more comprehensivesystem than a poor one. South Africa, inthis regard, is defined as an upper-middleincome country.

�The relative strengths of social forces Therelative strength of social forces andinstitutions determines the distribution ofthe country�s resources. A rich countrymay be able to afford to provide foreveryone, but may instead develop asystem that caters for the wealthy.Therefore, conceptually, the�affordability� or otherwise of a socialprotection system is partially dependenton social contestation. However, politicaland institutional mechanisms can be usedto avoid zero-sum trade-offs. For example, aproductivity/ investment accord could beagreed to in the context of a newcomprehensive social protection system.

A further factor is the extent to which�affordability� is determined by policydecisions, for example one to reduce publicspending as a percentage of GDP. In this regardthe parameters of �affordability� may beartificially constrained.

Further, Committee research (see chapter 14)indicates that there has been a reprioritisationaway from social spending, without any priorexplicit policy decision. The extent to which thishas occurred represents some degree of �fiscalspace� that can be reclaimed without a new policydecision being made.

The state provides numerous tax breaks or�subsidies� to private sector providers. Several ofthese tax arrangements � running into manybillions of Rands each � are inherited from thepast, and do not appear based on any clearrational or equitable basis. Over time these couldbe reallocated on a clear, rational and equitablebasis in line with the chosen social protectionapproach.

Finally, the Committee is strongly of the viewthat there is a cost to not acting. Indeed, it is notalways economic to defer importantinterventions and preventative steps for primarilyshort term cost reasons. In this regard, the social

Consolidated Report


backlog and accumulating challenges present abarrier both to social and economic development,and intervention sooner rather than later may beeconomically and fiscally prudent. Indeed it couldbe argued that via the negative social externalitiesgenerated by lack of state action, the society, orthe affected communities bear the cost.

3.5.2 Socialprotection and institutional

arrangementsIt has been a relatively short space of time sincethe advent of democracy. Therefore much ofthe institutional framework necessary to reversethe consequences of the previous system andaddress poverty, unemployment and inequalityis weak or absent. There is thus a difficult taskahead to rapidly resolve many institutionalchallenges within a relatively short frame. Thesechallenges include, for example, the design ofnew policy and legislation; new administrativestructures at various levels of Government;attempts to put into place measures that ensurethe inclusion of the previously excluded; theestablishment of mechanisms to deliver socialgoods efficiently and equitably; and theestablishment of monitoring and evaluation.

Since 1994 national Government policyinitiatives have attempted to finely-target poorand vulnerable groups within South Africa.However, the institutional mechanisms toimplement such policies have been uneven,with crucial governance failures resulting.

Public spending cutbacks have oftencontributed to growing institutional crises. Thecurrent economic strategy has introduced atighter fiscal approach from Government, withless fiscal support for expanding socialprotection. There is thus a resulting tensionbetween increasing access to social protection (asrequired by the constitutional and democraticimperatives) and declining real per capitaspending (driven by fiscal policies). Institutionshave therefore struggled to both reduce costs andincrease access.

This declining public spending, concurrentwith increasing commercialisation of keyservices, has pushed many people into theregulated private market. In this regard, theproblems related to the health sector are relevant.To address this tension, institutional efficienciesclearly need to be improved substantially and/or

fiscal support needs to be increased. Regardinginstitutional efficiencies, efforts to devolvefunctions and create new responsibilities forprovincial and local Government have run upagainst un-funded mandates and uneveninstitutional capacity.

Further, the means test has negatively affectedthe ability of the poor to access benefits. Means-tested schemes invariably have low take-up rates,that is, only a small proportion of those entitledto assistance actually applies for or receives them.Some may argue that if people do not apply for abenefit then they must really not need the benefitvery much or do not qualify for it. However, theseare often not the reasons for non-application.More likely fear, a lack of public awareness of theschemes, an inability to afford the transport tothe welfare offices, stigma, or difficulty inherentto the administrative requirements are the keyfactors. In reality, it may be a combination of allof these.

Means testing also intensifies the problem ofthe �welfare-trap�. At its simplest, this ariseswhere you receive a benefit only if you are notearning anything else. As soon as you startearning, you lose the benefit. While in practicesome means tests allow for some income to beearned, the welfare-trap remains, if somewhatreduced. In the South African context, and thetendency for incomes to fluctuate, applying themeans test correctly becomes a very complicatedand generally impossible task within theavailable institutional capacity. Crucially, itpromotes corruption, where Governmentofficials are in a position to waive or overlookcertain requirements. In short, the Committeehas sought, wherever possible, to find alternativesto or simpler forms of means tests.

3.5.3 Incorporatingsocial mobilisation into

social protectionSocial mobilisation is important in embeddingsocial protection in economic organisation andsocial relations. Social mobilisation is animportant resource in developing countries thatcan create a positive forward energy andsupplement the modest financial resources ofthe state. South Africa, particularly because ofits history of effective social mobilisation againstapartheid, may find that such mobilisation hasmuch to offer in the post-1994 period, too.


Consolidated Report

In this regard, the Committee considered thenotion of a Youth Corp engaged incomprehensive social protection activities. Forinstance, there is a need for approximately 54 000community-based caregivers to assistcommunities by dealing with the HIV/AIDSoutcomes. Such a necessary scheme could begiven to specially trained youth, potentially aspart of a learnership programme, supported byexisting social programme funds, withcontributions from relevant job creation/skillsdevelopment funds.

Elements of social mobilisation are alsoimportant in terms of increasing the level ofparticipatory governance, institutionalaccountability and, hence, contributing towardsinstitutional effectiveness and efficiency.

Moreover, non-governmental organisations(NGOs) and community-based organisations(CBOs), with Government support, have animportant role to play in creating and supportingan environment of social mobilisation. Such arole seems preferable to NGOs/CBOs as partnerdelivery agents; such organisations generally facefunding and capacity constraints, and attemptsto use them, as is the case with for-profitinstitutions, has exposed several weaknesses inGovernment administrative and managementsystems. Those Government departments usingsuch organisations had the most prevalence ofunder spending (due to their inability to processthe funding), and also limited delivery outcomes(due to their inability to ensure contractcompliance).

3.6 Social protectionand private social

insuranceSouth Africa has a highly developed privatesocial insurance market offering life, disability,health, property and casualty cover as well as arange of retirement benefits. There are howevera number of issues that require action in theseareas, which are covered in later chapters.

3.6.1 Mandatory coverOne question they have in common is that theinsurance is either entirely voluntary or, at most,cover is a matter of the employment contract. Ofcountries at comparable levels of development,South Africa is unusual in not mandating cover.

It is frequently argued that people cannot berelied on to make adequate cover because theyare myopic (short sighted). The state, theargument goes, should therefore intervene toprotect people from themselves, and compelthem to belong to insurance and retirementschemes. Such undemocratic contempt for otherpeople should usually be rejected. Beingcompelled to contribute to a retirement schemein one�s 20s for instance, is likely to lead to areduction in welfare as the costs of a home loanwill probably exceed the rate earned by theretirement fund on the investment.

On the other hand, a major advantage ofmandatory cover is that there is no need tounderwrite members, nor discriminate againstpoor risks. This saves administrative costs andallows more people to obtain affordable cover.Costs can be further reduced if there is less needfor marketing. It can also be argued that state-supervised funds offer greater financial security.The introduction of mandatory cover alsoprevents people from becoming a financialburden on other members of the community.The Committee is persuaded of the need forsome mandatory insurance cover for allparticipants in the formal sector � and theirdependants.

3.6.2 Lower earnings limitsfor mandatory social

insuranceMandatory social insurance requiresmechanisms for insuring that contributions arecollected. It is effectively only possible for thoseemployed in the formal sector � if this is definedas those where formal records of income are keptand tax is paid. A clear division is requiredbetween those from whom contributions can becollected, and those who cannot effectively beincluded. If the mechanisms for collection arenot likely to be efficient, regulations for theintroduction of compulsory cover will not beeffective, and should not be introduced.

Mandatory social insurance also involvesadministrative costs to the regulator, the serviceprovider and the contributors. The costs rise as aproportion of contributions for lowercontributions and smaller employers. These costsmay have a negative impact on employment forsmaller employers and low-income earners. Thecontributions themselves may be regarded as an

Consolidated Report


additional tax if the contributors believe that thebenefits offer little value. This is particularlylikely when benefits are small relative to the socialassistance available to non-contributors.

The Committee recommends that anunambiguous and manageable dividing line bedeveloped between those for whom cover shouldbe compulsory and those who could voluntarilycontribute to social insurance. Such a dividingline should take into account the relative size ofadministrative costs and the likelihood that it beenforceable.

3.6.3 Articulation andmeans tests

However this line is drawn, it is likely that manyindividuals will not contribute for their wholeworking lives.

The current articulation between socialassistance and social insurance is ostensiblymanaged by the means tests mentioned in theprevious section. They, however, are notmanaged in a consistent fashion, and theCommittee has had difficulty envisaging howthey might be. Even if they were consistentlyapplied, it is held that in their current form, theyare unfair and create perverse incentives to hideincome, or avoid earning cash income.

Means tests are currently applied, in differentways, in the granting of the benefits provided bythe Department of Social Development, but alsoto those applying for access to public hospitals,and for housing benefits. In addition, exemptionsfrom municipal services and school fees aremeans tested. Taxes can also be regarded aseffectively means tested.

The Committee believes it helpful, from oneperspective, to see social assistance and the systemof taxation as a whole, rather than consider theirparts individually. This perspective is illustratedin table 5 in chapter 2. One implication of thisview is that there should be consistency betweenthe phasing out of means tested benefits and ofrates of taxation.

The current position is complex, inefficientand unfair. As monthly income increases fromR100 per month, the following benefits may, forinstance, be lost:

�Old age and disability grants reduce by50% of additional income.

�Other grants may be removed in total.

�The costs of a visit to a state hospital maytreble.

�The entitlement to a housing subsidy mayhalve.

�Exemption from municipal rates may beremoved.

�School fees may be increased by 3,3% ofthe additional income for every child.

�Tax will become payable at a rate varyingfrom 18% to 40% of additional income.

The Committee believes that the most efficient,developmentally most effective and fairest wayforward is to abolish all means tests and to recoverthe costs through increases in tax.

If means tests are to be retained however, theneed for efficiency and fairness would suggestthat there be some rationalisation inadministration. In particular, there is a need for acareful integration between the evaluation ofmeans and the collection of taxes. If it isconsidered necessary to retain means testing inorder to target benefits, then it is recommendedthat the Department of Social Development beresponsible for making such evaluations. Theinformation as to which elements of the socialsecurity package a person is entitled shouldultimately be captured on their identity cards.


Consolidated Report

Consolidated Report


Chapter 4ConstitutionalFramework forSocial Securityin South Africa

4.1 IntroductionFundamental reform of South Africa�s socialsecurity system that aims to redress past injusticeis in accordance with the provisions of theConstitution of the Republic of South Africa 108of 1996 (hereafter referred to as the constitution).For the first time in South Africa�s history, theConstitution compels the state to ensure the�progressive realisation� of social security.Section 27 of the Constitution clearly andunambiguously commits the state to develop acomprehensive social security system. It affirmsthe universal right to social security, includingappropriate social assistance for those unable tosupport themselves and their dependants,mandating the state to take reasonable legislativeand other measures, within its availableresources, to achieve the progressive realisationof each of these rights.

The critical question is how these rights canbe adjudicated. The Constitution imposedobligations on all spheres of the state to realisethe right to social security, and the ConstitutionalCourt has affirmed that many aspects of thesocio-economic rights included in our Bill ofRights are justiciable.

4.2 Status and impact ofthe Constitution

Section 2 of the Constitution expresses the role theConstitution is meant to play with regard to socialsecurity regulation, policy-making andadministrative practice. It states that theConstitution is the supreme law of the country;law or conduct inconsistent with it is invalid, andthe obligations imposed by it must be fulfilled.

The state is obliged to conform to the rightscontained in its Bill of Rights, as they are said tobind the legislature, the executive, the judiciaryand all organs of state,13 as well as, to the extentforeseen by the Constitution,14 natural andjuristic persons. In recent times, the courts havenot hesitated to enforce the supremacy of theConstitution in the area of social security incircumstances where its prerequisites have notbeen adhered.

4.3 Generalprinciples

The overarching aims of the Constitution areclosely related to social security goals: healingthe injustices of the past, ensuring social justice,improving the quality of life for all South Africancitizens (inter alia by alleviating poverty andsuffering), and freeing the potential of eachcitizen. The meaning of the constitutionalfundamental rights has to be determined andunderstood against the background of pasthuman rights abuses.

Certain constitutional values are key to theinterpretation of fundamental rights pertainingto social security: human dignity, equality andfreedom, as well as the advancement of race andgender equality.

4.4 A humanrights approach

The Constitution favours a human rightsapproach by not differentiating between social,civil and political rights. By not differentiatingbetween this apparent �categories� of rights,


Consolidated Report

emphasis is placed on the fact that these rightsare interrelated, interdependent and indivisible.The Constitutional Court has made it clear thatrealising a particular socio-economic right, suchas the right to access to housing, would requirethat other elements which do at times form thebasis of particular socio-economic rights, suchas access to land, must be in place as well.Together these rights are mutually supportiveand have a significant impact on the dignity ofpeople and their quality of life.

In the Constitution the human rights-basedapproach towards social security (fundamental)rights is strengthened by provisions which: (a)state that the duties imposed by the Constitutionmust be performed, and (b) require of the state torespect, protect, promote and fulfil the rights inthe Bill of Rights.

The right to access to social security is backedby a host of other social security relevantfundamental rights. These include the right tohave access to healthcare services, to sufficientfood and water, to adequate housing, the rightto education, as well as the right of children tobasic nutrition, shelter, basic healthcare servicesand social services. Together these rights can besaid to ensure, from a constitutional and humanrights perspective, adequate social protection.

4.5 The meaningof social security

from a constitutionalperspective

The right to (access to) social security in SouthAfrica, however, is not yet cast in concrete terms.It is incumbent on Government to set minimumstandards for defining the right to access to socialsecurity and its realisation. This, it is suggested,has to be done within the framework of theoverall goals of the social security systemenvisaged by Government. The definitionalstandards so adopted then have to be appliedand implemented in programmatic fashionaccording to an appropriate time-frame, settingout the goals to be achieved, mapping thedifferent programmes and systems,determining the priority order, and indicatingthe time targets.

In keeping with modern social securitythinking and policy-making, social security isno longer seen as merely curative (in the sense

of providing compensation), but also aspreventative and remedial in nature. The focusshould be on the causes of social insecurity (inthe form of, amongst others, social exclusion ormarginalisation), rather than on (merely dealingwith) the effects.

Further, the social security concept does notmerely cover measures of a public nature.Social, fiscal and occupational welfare measures,collectively and individually, whether public orprivate or of mixed public and private origin, mustbe taken into account when developing coherentsocial security policies. In a country such asSouth Africa such an approach may not only beadvisable, but also necessary, in order to fullyutilise limited resources. This implies that afunctional definition of social security be adopted,which includes all instruments, schemes orinstitutions representing functional alternativesfor the publicly recognised schemes � that is, allinstruments available to society for guaranteeingsocial security.

While the right to access to social security isgranted to �everyone�, it is clear that the rightsof children in this regard are exercised mainlyvia their parents and families. In these caseswhere family support is available, the role ofthe state is restricted to provide the legal andadministrative infrastructure necessary toensure that children are accorded the protectedcontemplated by the Constitution. In addition,according to the Constitutional Court in theGrootboom case, the state is required to fulfil itsconstitutional obligations to provide familieswith access to land, access to adequate housing,as well as access to healthcare, food, water andsocial security.

Important implications flow from the aboveperspective. For the right to access to socialsecurity (and the other social security-relatedrights) to fully mature and to be known anddirectly enforceable, the state should initiatelegislation to provide for the substantive rightscapable of being claimed (what actually shouldbe claimed). The procedure and mechanism forclaiming such rights (how the rights should beclaimed); and where (venue) the rights shouldbe claimed also needs to be specified. On thequestion of how and where the right shouldbe claimed, the state also has to concern itselfwith the institutions that will hear anddetermine disputes arising from claims forsocial security benefits provided for under therelevant legislation.

Consolidated Report


The Constitution places a duty on the state torespect, protect, promote, and fulfil the rights inthe Bill of Rights.15 On a primary level the dutyto respect requires negative state action and thecourts will only expect the state not to unjustlyinterfere with a person�s fundamental rights. Ona secondary level, the duties to protect, promote,and fulfil places a positive duty on the state and itis argued that this duty also requires positiveaction from the courts.

This positive obligation does not, as such,require that the state distribute money orresources to individuals, but requires setting upa framework wherein individuals can realisethese rights without undue influence from thestate. It requires in particular of the state toprotect especially vulnerable groups andencompasses protection against third (non-state) party violations of these rights.Practically this would, for example, mean thatpensions, medical insurance andunemployment insurance legislation shouldbe construed in such a manner that theysufficiently protect individuals againstdiscrimination in acquiring benefits.

4.6 Limiting,interpreting andenforcing socialsecurity rights

The constitutional rights relating to socialsecurity are subject to two forms of limitations.First, the limitation must comply with therequirements contained in the (general)limitation clause of the Constitution(sometimes referred to as external limitations).Secondly, the limitation can also be justified onthe basis of the specific qualifications containedin respect of a particular right (sometimes referredto as internal limitations).

As far as the general limitation clause isconcerned, section 36 of the Constitutionsubjects a limitation of a fundamental right to athreefold test, in terms of which the limitationmust be:

�Contained in a law of general application

�Reasonable

�Justifiable in an open and democraticsociety based on human dignity, equalityand freedom.

An example of a non-justifiable limitation wouldoccur where the classification or criteria adoptedto indicate beneficiaries of a particular stateproject (such as a social relief programme) arenot sufficiently determined, with the result thatcertain groups are, contrary to the intention ofthe programme, excluded.

The state�s duty to respect, protect, promote,and fulfil the right to access to social security isfurther qualified by the phrasing of section 27(2).Section 27(2) states that the state must takereasonable legislative and other measures, withinits available resources, to achieve the progressiverealisation of each of these rights. Thereasonableness of the measures will be evaluatedagainst criteria such as:

�The social, economic and historical contextof the deficiencies in the system themeasures aim to address

�The institutional capacity to implementthe programme adopted

�Whether the programme is balanced,flexible and open to review, and makesappropriate provision for attention to thedeficiencies in the system and to short-,medium- and long-term needs

�Whether the programme is inclusive anddoes not exclude a significant segment ofsociety

�Whether the measures ensure that basichuman needs are met and takes intoaccount the degree and extent of the denialof the right they endeavour to realise

�Whether the programme and measuresensure that a larger number of people anda wider range of people benefit from themas time progresses.

When interpreting social security-relatedfundamental rights, every court, tribunal andforum is required to promote particularconstitutional values, and to considerinternational law (both binding and non-binding), while foreign law may be considered.The power to enforce and adjudicate these rightsvests in the courts (in particular theConstitutional Court), while the Human RightsCommission is given the mandate to monitorcompliance with and support the developmentof fundamental rights.

Where parliament or the provinciallegislature failed to comply with a constitutional


Consolidated Report

obligation that requires positive state action theConstitutional or High Court may grantappropriate relief.

4.7 Concludingremarks

Committee analysis indicates that there areunequal, exclusionary and inequitable structureof the present social security system as a whole,and of particular elements of the system, whichare therefore not in conformity with theconstitutional prohibition of unfairdiscrimination.

In view of the need to urgently address thischallenge, the Committee recommends thatsocial security policies and programmes must bereasonable both in their conception and theirimplementation. Showing that the measures arecapable of achieving a statistical advance in therealisation of the right to access to social securitymay not be sufficient to meet the test ofreasonableness. Rather, the needs that are themost urgent must be clearly addressed.Particularly vulnerable communities must begiven priority by Government and their needsmust be effectively addressed.

From the Grootboom and other constitutionalcase law it appears that:

�A reasonable programme must clearlyallocate responsibilities and tasks andensure appropriate financial and otherresources

�It is not only the state that is responsible forthe provision of social security, but that theresponsibility and/or involvement offamily structures, other (non-state)providers and private provision has to befactored in, acknowledged, supported/protected, and, where necessary,regulated

�As a rule (but subject to the reasonablenesscriterion), the court will not considerwhether other more desirable or favourablemeasures could have been adopted orwhether public money could have beenbetter spent

�The socio-economic/social security right atstake must be coherently andcomprehensively addressed

�Guidelines drawn up in the wake of budgetconstraints have to be reasonable

�A minimum floor of benefits or provisionhas to be made for the most vulnerable insociety.

The Committee recommends that given SouthAfrica�s commitment to relevant internationalobligations, social security laws be amended tocomply with these.


Consolidated Report


Consolidated Report

Chapter 5Poverty, SocialAssistance Grantsand the BasicIncome Grant

5.1 IntroductionThe Committee paid significant attention to theurgent problem of poverty and its impacts onSouth African society. Moreover, through theprocess of public hearings, consultations withrepresentative groups, submissions andcommissioned research, the Committee hasconsidered existing and proposed measures toalleviate poverty.

The Committee was required by the termsof reference to examine the feasibility of a BasicIncome Grant. This section of the reporttherefore evaluates the potential impact of aBasic Income Grant, assessing its role inreducing poverty.

The Committee�s findings on the state ofpoverty have been drawn from a wide rangeof sources, including representations made bythe poor themselves. The Committee basedits analysis on more than just a quantitativeanalysis of indicators and statistical measures.Providing opportunities for poor peoplethemselves to voice their experiences was away of ensuring broad based participation inthe Inquiry as well as verifying the extent towhich existing policy and programmeinitiatives of Government are understood byand reach the poorest people. The realities ofintense poverty and inequality were alsoevident in direct representation made to theCommittee by poor people themselves.Moreover, hearings and visits to provincesindicated that the poorest communities areliving in conditions that dangerouslycompromise their human security andwell-being.

5.2 Governmentprogrammes addressing

poverty: findingsDuring the past 7 years Government haslaunched new programmes and expanded andrevised existing programmes to deal with assetand capability poverty. These measures aredesigned to provide people (especially thosewho have been excluded as a result ofapartheid) with access to certain assets andcapabilities needed to overcome their poverty.Through such measures it is expected thatopportunities will open up for people toimprove their income earning capacity and asa result to address levels of destitution andpoverty.

5.2.1 Povertyprogrammes

to address capabilityand asset poverty

The Committee finds that Governmentprogrammes to address deprivation inhealth, education, housing, land, basicservices such as access to water andsanitation, electricity and access to credit arewell conceived and potentially well targeted.The barriers to access, especially in regardto the poor, remain administrative andinstitutional.

A specific problem repeatedly raised is the useof communal land currently governed bycustomary law. Rights to use of communal landshould be secure and protected from arbitrary


Consolidated Report

seizure. Concerns have been voiced that the currentposition does not lead to the optimal use of land,and tends to promote migrant labour. A way needsto be found to allow people to retain their rights toreturn to land in retirement, but for the land to befruitfully used while they reside in the urban areas.The Committee recommends that such concernsbe considered in reviews of customary law.

Evidence put to the Committee shows that thepoor have particular difficulties in accessinghealthcare and primary education because they donot have even the most basic income for transport,food and basic clothing. It is the Committee�sconsidered view that these key Governmentprogrammes, put into place to address capabilityand asset poverty, are essential and will in the longterm result in sustained human development andeconomic growth. However, in the immediate termtheir policy efficiency, especially with regard to thosepoorest people in the rural and informal areas, isbeing compromised because of unsustainable levelsof income poverty.

Moreover the Committee finds that givenapartheid and structural unemployment, thecurrent range of poverty relief projects, while inmany cases innovative and responsive, are unableto make any significant impact on mass basedunemployment and levels of income poverty inthe immediate term. Many of these projects, arealso not cost efficient in terms of their outcomes.

It is therefore the Committee�s view that forthe long-term policy benefits of health, educationand more generally, basic services to make adevelopmental impact on the poorest,Government needs to take urgent steps to providethe basic means to enable the poorest to access thesebenefits. The Committee also recognises thatbarriers to access are not only bureaucratic(administrative) or to do with capacity constraintswithin Government itself but, based on evidenceput before it, also that many of the poorest aretrapped in income poor households. Addressingincome poverty is therefore fundamental to a socialprotection reform process.

5.2.2 Povertymeasures: requirement to

target social policy

The World Bank says:

The measurement and analysis of poverty,inequality and vulnerability are crucial for

cognitive purposes (to know what the situationis); for analytical purposes (to understand thefactors determining this situation); for policymaking purposes (to design interventions bestadapted to the issues); and for monitoring andevaluation purposes (to assess whether currentpolicies are effective; and whether the situationis changing).16

The Committee�s review of human developmentand various poverty studies in South Africaindicate the need for nationally agreed povertymeasures. Nationally agreed poverty measuresmust have a conceptual and empirical basis.Conceptually, the Committee�s framework forcomprehensive social protection acknowledgesthat poverty has many dimensions. Effectivepolicy must focus on capability poverty(deprivation in health and education), incomepoverty (lack of earnings and other sources ofincome) and asset poverty (access to resources).

Social policy intervention also needs to considerhow inequality, participation, social exclusion andvulnerability contribute to poverty and people�ssocial capacity for self-reliant development. Such abroad view of poverty requires special attention tothe measurement and analysis that informs specificpolicy options. For this reason a conceptually soundand an empirically based understanding of povertyindicators and measures of well-being is critical.

Statistics South Africa continues to buildnational capacity for collecting this important socialand economic data. The 1996 census and nationalsurveys of social indicators such as the South AfricanLabour Development Research Unit (SALDRU)/World Bank 17 survey in 1993 and the subsequentOHS have provided important evidence to helporient current policies as well as to inform many ofthe submissions to this Committee.

Much still needs to be done particularly in thearea of longitudinal data. Income data is notoriouslyunreliable. A proper understanding of the nature ofpersistent and intermittent poverty requires studiesthat follow the same people through thesefluctuations. This understanding might then allowfor effective interventions.

5.2.3 Governmentservice delivery

programmesIt is evident that the democratic Government hasmade significant strides to ensure policy and budgetreprioritisation to provide access to resources.


Consolidated Report

Reprioritisation of resources is directed at such basicingredients as access to water, sanitation, electricity,housing, knowledge/education, healthcare, landand other productive resources.

Government aims to meet the basic needs ofcommunities through the provision of basiceducation (pre-primary and primary), basic health(primary and district level services), housing, waterand sanitation, and electricity. It records thefollowing measures to meet basic needs:

�Forty six per cent of the total educationbudget is spent on basic education

�Twenty per cent of the total health budgetis spent on basic healthcare

�There is free healthcare for pregnantwomen and children under six

�A nutrition programme reaches 12 000primary schools

�Two million people have access to safe water

�Since 1994, 1 167 435 houses have beenbuilt or were in the process of being builtby the end of 2000

�The Consolidated MunicipalInfrastructure Fund subsidises the cost ofcapital investment in municipalinfrastructure

�An �equitable share� of national revenueis provided for local Government; fundsare transferred to municipalities on thebasis of need to enable them to provideservices to poor communities.

�Transport subsidies (such as for buses) andthe rural transport strategy aim to alleviatepoverty. The taxi recapitalisation process willcontribute to better access to transport for theaged, the disabled and school children.

The Committee engaged in a systematic reviewof some major current Government measuresto address poverty. Such a review was designedto provide a holistic picture of the potentialimpact of interventions related to asset andcapability poverty as well as income poverty.

After the review the Committee concludedthat while there has been considerable progressin some aspects, current measures do notadequately contribute to a minimum package ofgoods, services and benefits. Many programmesfor alleviating poverty and meeting basic needsare not targeted effectively in rural areas and tobeneficiaries. Gaps in coverage and inequalities

exist in the provision of services.

For example, the Committee found that mostrural women and disabled people still experiencedifficulties in gaining access to land. Researchshows that about 80 per cent of the total landclaims registered and settled to date by theDepartment of Land Affairs are urban, yet ruralclaims involve a far larger number of people.

A survey done on housing subsidy schemes alsoindicate that people in the rural areas are less likelyto know about subsidy schemes. In terms ofhousehold income, most of those who knew aboutthe subsidies were those earning between R2 000and R3 500 (63 per cent), households in the lowestincome category had less knowledge of the subsidyscheme (58 per cent).

The PIR (1998) indicated that nationalprogrammes for Small, Medium, and MicroEnterprises (SMME) development havefocused primarily on the needs of urban SMMEdevelopment. Results of an evaluation done byCommunity Agency for Social Enquiry(Case) and the International LabourOrganisation (ILO) highlighted that there isa generally inequitable geographic spread ofPublic Works projects.

While the Government has implemented newinitiatives with regard to sanitation and waterschemes, there are still millions of South Africanswithout adequate access to services.Representation made to the Committee indicatedthat lack of access to running water forces manypeople, especially the rural poor, to walk longdistances to fetch water. Most households in ruralareas have no toilet facilities and have access to waterfrom a river, a borehole or tank.

Several problems have also been encounteredwith regards to the provision of nutrition toschool children. In 1998/99, the Primary SchoolNutrition Programme aimed to target 17 500schools and serve a meal to 5 574 305 schoolchildren. Figures from the Department of Healthindicate that the target number of beneficiarylearner and schools has never been reached. Thisindicates that not all children are benefiting fromthe school feeding scheme.

5.3 South Africa�ssocial assistance system

South Africa�s social assistance system of grantsis important in addressing income poverty.


Consolidated Report

Moreover, social assistance can address the factthat low or non-existent incomes contributedirectly to poor access to healthcare, education,housing, and social infrastructure. This sectionevaluates the impact of Government�s incomegrant programme on the state of poverty in SouthAfrica. The evaluation is based on householdlevel micro-simulation models commissioned bythe Committee.

5.3.1 An overviewof the social

assistance systemIn April 2001 an estimated 3,5 million SouthAfricans received social assistance through someform of income grant.18 The State Old AgePension (SOAP) is the largest social assistanceprogramme with about 1,9 million beneficiaries.The important redistributive impact of thisprogramme has been recognised byGovernment, labour and academia.19 TheDisability Grant (DG) is the second largestprogramme in Rand terms, but smaller than theChild Support Grant (CSG) in terms ofbeneficiaries. DG beneficiaries numbered 643107 in April 2001. Eligibility for the grant isdetermined based on a medical diagnosisassessing the degree of disability, along with ameans test. Reform of the DG has been thesubject of a recent task team report.20

The introduction of the CSG represents animportant reform introduced by the Governmentsince the transition to democracy. In April 2001,800 476 caregivers received grants with anestimated value of R120 million. The distinctivefeature of the programme is the concept of �followthe child�, meaning that the benefit is independentof the child�s family structure. This grant wasintroduced in April 1998 and provided R100 permonth per child for children under the age of 7within a means test. The declared goal at that timewas to reach 3 million children within the next fiveyears. At the same time, the phasing-out of the StateMaintenance Grant (SMG) with about 350 000beneficiaries started. The Department of SocialDevelopment had to phase out the SMG over aperiod of 3 years. In April 2001, the CSG benefitwas raised to R110, with a commitment to adjustingit for inflation in subsequent years.21

The Committee, through research andsubmissions, has noted the following concernsregarding the CSG:

�The take-up rate has increased in the 2001/2 financial year to around 1,5 million outof, at the very least, 3 million eligiblechildren who could receive the benefit.However, there is still a concern thatchildren most in need are not targetedeffectively.

�It is widely accepted that the level of thegrant does not come close to meeting thebasic costs of childcare.

�The age limit has no real rational basis andis not consistent with the Constitution�sdefinition of a child, that is, those under 18years of age.

�In stopping the grant at school-going agesome children apparently are unable toattend school because their parents cannotafford the costs associated with schooling.

�The means test as currently appliedrepresents a barrier to many applicantsgaining access to the CSG.

�For many reasons, including those above,there is a widely held view amongbeneficiaries and potential beneficiariesthat the grant is not operating effectively.

Other programmes include the Foster Care Grant(FCG), and the Care Dependency Grant (CDG).At the age of 18, the disabled individual can applyfor a DG. In terms of numbers of beneficiaries,the SOAP, the DG and the CSG are the largestsocial assistance programmes within theGovernment funded social security system.

5.3.1.1 Household structure and socialsecurity reform

A number of household structural characteristicsare important when considering social securityreform. South Africa�s household structure wasanalysed using the best available data. The resultshave to be interpreted with care, but showed thefollowing:

�Most pensioners (84 per cent) live inhouseholds with non-pensioners, so it islikely that most old age pensions supportliving standards beyond their immediatebeneficiaries.

�Nevertheless, most adults (81 per cent) andchildren (76 per cent) live in householdswith no pensioners, so they are less likelyto benefit from any grants paid topensioners. It becomes clear that while


Consolidated Report

pension money benefits many of the poor,pensions alone are wholly inadequate attargeting them as a group.

�Households with only of working ageadults comprise 10 per cent of allhouseholds, or over 4,6 million people.The poor in these households are excludedfrom a social security system that protectschildren and pensioners.

�Most South Africans live in large households(more than six people). Since largerhouseholds tend to be poorer, a fixed grant toeach household will not be efficient intargeting the poor � larger per capita benefitswill accrue to wealthier households.

5.3.1.2 The impact of the current socialassistance system

An assessment of the social implications of thecurrent system indicates that:

�In the absence of social assistance transfers,58 per cent of South African householdswould fall below the subsistence line ofR401 per adult equivalent.

�Out of a projected 23 840 471 people in thebottom two quintiles, the study estimates that11 840 597 individuals (50 per cent) live inhouseholds that receive no social assistance.

�Existing social security programmesreduce the average poverty gap by 23 percent. The �poverty gap� gives an estimateof the extent of poverty, by adding, for eachhousehold, the amount by which incomefalls below the subsistence line.

5.3.1.3 Summary of findings

The existing social security programmes do notadequately address the problem of poverty. Halfof the poor live in households that receive no socialsecurity benefits at all, and the rest remain poor inspite of the benefits they receive. Nevertheless,South Africa�s social security grants make asignificant impact, reducing the average povertygap by approximately 23 per cent at the beginningof 2001.

This relatively low percentage belies importantvariances. The SOAP reduces the poverty gapfor pensioners by 94 per cent. Poor householdsthat include pensioners are on averagesignificantly less poor than households withoutpensioners.

�Skip generation� households (comprisingchild and grandparent), on average, have their

poverty gap closed by over 60 per cent. For three-generation households the poverty gap is closedby less than 50 per cent due to the burden of theworking age members.

For the average poor household without apension-eligible member, however, socialsecurity�s impact is almost negligible. Forhouseholds with no pensioners, the reduction isless than 10 per cent.

Clearly, South Africa�s social safety net has avery loose weave.

Committee modelling indicates that theexisting social security system, even if all benefitsare distributed to everyone entitled (that is,achieve a full-take up rate), has the capacity toclose only 37 per cent of the poverty gap.

Even that partial closing of the gap, however,is not evenly distributed across household types.Households containing only working-age adultshave on average only 11 per cent of the povertygap closed, while the entire poverty gap forhouseholds containing only adults inpensionable age would be closed. Householdscontaining only children and working-age adultshave an average of only 22 per cent of the povertygap closed, while �skip generation� householdshave an average of 80 per cent of the poverty gapclosed. Sixty point four per cent of the povertygap for three-generation households is closed.

Most of the benefit of the existing socialsecurity system with full take-up still comes fromthe SOAP � but it falls to approximately 60 percent of the per capita social assistance transfer,while the share attributable to the CSG rises to athird (from 10 per cent).

Figure 10 depicts the impact of the distributionof income if all potential beneficiaries of SouthAfrica�s social security programmes received thefull grants for which they were eligible. Thegraph is constructed with population on thevertical axis and relative income categories onthe horizontal axis. That is, an increment alongthe horizontal axis represents a 10 per centincrease in income. The dotted vertical linerepresents the subsistence line of R401 per adultequivalent.

Figure 10 documents that even with full take-upof all social security programmes, 21 955 935 people(over half the population) fall below the poverty line,while 20 768 683 are above. However, even thesefigures may be somewhat optimistic.


Consolidated Report

Figure 10South Africa�s distribution of income (assuming full take-up

of existing social security programmes).

Simulations done for the Committee indicatethat with full take-up of existing social securitybenefits an estimated 843 164 people move outof poverty. However, these benefits are unlikely tobe realised with the current structure of the socialsecurity system. Means tests, rigid eligibility criteria,and the high relative cost of applying for socialsecurity all contribute to low take-up rates.

In terms of the fiscal impact of this �loose weave�social safety net, at full take-up, South Africa wouldspend R26,5 billion on the transfer payments �R14,8 billion for the SOAP, R7,2 billion for theCSG, and R4,5 billion on the DG. This is inaddition to the administration costs.

As a result, the Committee finds that from acomprehensive social protection framework theexisting programme of social assistance grants isconsiderably high cost relative to its level of socialeffectiveness.

The Committee is of the view that one of themost effective means of reducing destitution andpoverty is to provide some minimum support inthe form of a social assistance grant.

Furthermore, the Committee recommendsthat any income support grant be set at a level

that would address destitution in the mediumterm and absolute poverty in the long term.

5.4 The feasibilityof a Basic Income Grant

for South AfricaThe coverage gaps within South Africa�s socialsecurity system combined with the structurallylow rate of take-up of the CSG underscore theneed for comprehensive reform. The nature ofstructural unemployment in the face of achanging global economy is such that itmarginalises unskilled workers and results in theneed to expand the scope of a social safety net.Not only do children, retirees and the disabledneed social protection � millions of potentialworkers are vulnerable to unemployment andresulting impoverishment.

This section evaluates the potential of theBasic Income Grant to address the severe povertycharacterising South African society. The natureof an income transfer has important implicationsfor its socio-economic benefits and fiscalfeasibility.


Consolidated Report

5.4.1 What is aBasic Income Grant?

A Basic Income Grant is provided as anentitlement and without a means test that willmore readily reach the poorest population. Byremoving the stigma that labels the recipient as�poor�, the grant is said to bolster economicsupport without draining psychologicalresources. The Committee understood the BasicIncome Grant as a social policy option, to bedefined as �a general social assistance grant forall South Africans.� Further, the Committee hadto determine what could be some of the concretecharacteristics of this option.

In practice the grant could be calculated on aper person basis. For instance, a Basic IncomeGrant of R100 would mean that a single personliving alone receives R100 per month. Ahousehold with six people (the average for theSouth African population) receives R600 amonth. The working assumption in the modelused to evaluate the feasibility of the BasicIncome Grant is that there is no overlap betweendifferent grants. The grant is meant for peoplecurrently not receiving social assistance � thosewho fall through the social safety net.

A Basic Income Grant would serve as a socialentitlement for all South Africans. Such anentitlement supports the right to appropriatesocial assistance as entrenched in the SouthAfrican constitution 27(1)(c) while furthering thevision of a comprehensive social security system asidentified in the White Paper for Social Welfare.

5.4.2 Will the BasicIncome Grant create

dependency?The Basic Income Grant has no means test andtherefore avoids many of the disincentives towork inherent in other social assistance systems.A Basic Income Grant stands in stark contrast tothe targeted unemployment grant or what isknown as a �dole system�, which employsconventional means tests to target theunemployed, the unemployable or the very poor.

The level at which a Basic Income Grant is setwill be crucial. At the very least it should addressdestitution. By providing such a minimum levelof income support people will be empowered totake the risks needed to break out of the povertycycle. Rather than serving as a disincentive toengage in higher return activities, such a

minimum (and irrevocable) grant couldencourage risk taking and self-reliance. Such anincome grant could thus become a springboardfor development.

5.4.3 Will the BasicIncome Grant target

the poorest?The targeting of the poor within the context of aBasic Income Grant depends on the tax system.The SARS is one of the most capable arms ofGovernment, reflecting a transformation processthat has supported consistent over-achievementof revenue targets over the past five years.Appropriate tax reform linked to the BasicIncome Grant can achieve very effectiveredistribution. Several financing mechanismshave been proposed. Cosatu has proposedrecuperating the amount of the grant from alllow to middle income earners whileimplementing a �solidarity tax� for higherincome earners. Other proposals have focusedon the Value Added Tax (VAT)22 as well asprogressive taxation.23

5.4.4 Can the Basic IncomeGrant be administered

efficiently?The Basic Income Grant would be paid to allSouth Africans. In the case of children, the grantwill be paid to the primary care givers for thebenefit of such children. To ensure that everySouth African has full access to this grant and toavoid any duplication of payment, a reliableidentification and verification system will haveto be established. There will thus be none of thecosts associated with a benefit targeted through ameans test.

One could, however, target the income grant byusing the tax system, so that the net benefit islarger for the poorer recipients. If the costs of theBasic Income Grant are recuperated, for example,through the VAT system, research shows that thegrants can be more effectively targeted. The lowerthe per capita expenditure of a household, thelarger the benefit per person.

A Basic Income Grant, which is calculated ona per person basis, favours larger households thaton average are poorer than smaller ones. Poolingof income leads to economic efficiencies and themore equitable intra-household distribution ofincome, could contribute to the empowermentof women and younger people in the family.


Consolidated Report

5.5 The impactof a Basic

Income GrantThe analysis of the impact of a Basic IncomeGrant is based on an income grant of R100 permonth for all South Africans. With full take-upof a Basic Income Grant, the number of poorSouth Africans excluded from the social securitysystem is reduced to zero. The dispersion amonghousehold types in the closing of the poverty gapis substantially reduced. The household typecurrently with the least reduction in the povertygap is the household with only working ageadults; with a Basic Income Grant the povertygap in these households is closed by over 56 percent, compared to less than 8 per cent with thecurrent system.

For households with children but nopensioners, the poverty gap is closed by two-thirds, and for households with children andpensioners, the gap is closed even moresuccessfully. For �skip generation� households,95 per cent of the poverty gap is closed, for �three-generation� households, 85 per cent of thepoverty gap is closed.

The number of people covered by the socialsecurity system increases more than five-fold.The Basic Income Grant will account for 63 percent of comprehensive social security coverage.

Most of the benefits (53 per cent) would bedistributed to rural households, reflecting thespatial character of South African poverty. Two-thirds of the transfers to three-generation and�skip generation� households would be to ruralrecipients, reflecting the household structure�srole in coping with rural poverty.

Figure 11 depicts the impact of the BasicIncome Grant on the distribution of income. Asin figure 10, the blue curve replicates thedistribution depicted in the baseline simulation.The red line represents the distribution ofincome with the Basic Income Grant.

Figure 11 documents important impacts. Theincidence of extreme poverty is nearlycompletely eliminated. The closing of the povertygap improves to 74 per cent. On a headcountbasis, approximately 6,3 million are moved outof poverty. The number of destitute individuals(measured using half the poverty line) falls by10,2 million people. Most of the remaining poorindividuals are clustered fairly close to the

poverty line, so that broad-based growth woulddemonstrate substantial success in movingadditional numbers of people out of poverty.

In sum, Table 8 compares the social impact ofthe BIG with the current situation and a situationwhere full take-up of all existing grants isachieved.

5.6 Recommendations5.6.1 Poverty

measures and indicatorsIt is necessary for the Government to decide on aminimum poverty line. It is recommended thatthis should be an absolute poverty line, e.g. R400per person per month in 1999 prices, and not arelative poverty line, e.g. the income per personof the household in the 40th percentile. Adestitution line should be set as a first realisticbenchmark that could be used by Governmentto monitor progress in alleviating poverty. Adestitution line could be set at half the poverty line.

Implementing concrete policy actionsrequires that Government develop mechanismsand procedures for gathering and evaluating dataon poverty and human development. TheCommittee recommends that Governmentensure that such data and research is gatheredand developed on a regular and consistent basisto support measures aimed at alleviating andreducing poverty.

5.6.2 BasicIncome Grant

Analysis indicates that the Basic Income Granthas the potential, more than any other possiblesocial protection intervention, to reduce povertyand promote human development andsustainable livelihoods. A universal BasicIncome Grant has the potential to fortify theability of the poor to manage risk thuscontributing to socio-economic multiplierseffects related to improved household self-reliance, efficiency of social capital and societalcohesiveness.

Moreover, in the view of the Committee,income support of this nature would assist thepoor to access Government services, therebyimproving the effectiveness of many servicedelivery programmes and social policies.

The Committee notes, however, that theconditions for an immediate implementation of


Consolidated Report

a Basic Income Grant do not exist. In particular,there is a need to first put in place appropriatecapacity and institutional arrangements to ensureeffective implementation. Therefore theCommittee recommends the gradualdevelopment of a comprehensive and integratedincome support that can underpin South Africa�scomprehensive social protection system.

5.6.3 A comprehensiveand integrated

medium- to long-termframework forincome support

Allowing for a diversity of policy preferences, theCommittee has identified three main optionsconcerning income poverty.

5.6.3.1 Option 1: Maintain the status quo

This option could be based on the view that

significant improvements, particularly incomepoverty interventions, cannot be accommodateddue to inflexible fiscal constraints. Alternatively,some may feel that new interventions are perhapsunnecessary because the suggested end-state willbe reached naturally over time withimprovements to the administration of currentprogrammes. There will probably beconsiderable support for this option, particularlyfrom those mainly concerned about possiblefinancial implications of any proposedinterventions.

The Committee, however, is of the view thatthis would be an ultimately short-sightedposition. The current programmes fail to satisfythe constitutional imperatives and thus make thestate vulnerable to Constitutional Courtchallenges, and are clearly inadequate. Further,the current system fails to address the socio-economic imperatives, and the social costs of theabsence of proper social protection will contribute

Figure 11Distribution of income with a Basic Income Grant.

Table 8 Social impact of the Basic Income Grant

Policy Poverty gap Additional peoplereduction freed from poverty

Current situation 23% �

Full take-up of existing grants 37% 0,8 million people

Basic Income Grant 74% 6,3 million people


Consolidated Report

towards a tendency of social instability. Thiswould force the state to increasingly criminaliseand judicialise essentially poverty-related socialconditions, thereby creating conditions that couldundermine the legitimacy of the democratic state.

In short, there is clearly a cost to not acting,which will sooner or later have to be borne bythe country. Intervention sooner would be bothsocially and economically prudent.

5.6.3.2 Option 2: Immediateimplementation of a comprehensivesocial protection system

The Committee has received many submissionsand representations, from a wide range of socialformations, which argued for immediateimplementation of measures to address thetremendous socio-economic challenges facingSouth Africa. It was argued that there areincreasingly large numbers of people indesperate need, and such people are not accessingany protective measures (such as emergencyshelter). In particular, such arguments tended tomotivate for a universal income grant that can,at the very least, alleviate persistent and pervasivelevels of destitution.

Many proponents of this view argued thatSouth Africa has considerable economicresources at its disposal, including low levels ofdebt and considerable space for increased taxation� and that such resources could be mobilised inthe short term.

The Committee, after due consideration ofthis view, and appreciation of the need foradditional social protection interventions, hasconcluded that the current institutionalarrangements do not easily allow for new andexpansive measures to be implementedimmediately. A comprehensive social protectionsystem requires the prior establishment of newinstitutional arrangements and measures. Sucharrangements, even in the most optimisticscenarios, would require several years to fullyplan, resource and implement.

Nonetheless, the Committee is of the viewthat, in the interim, there is a need to takeimmediate steps to expand certain poverty reliefand de facto schemes to assist those individualsand groups that can access such benefits.

5.6.3.3 Option 3: A phased approach towardsa comprehensive social protection system

This is the Committee�s preferred option. Thisoption seeks to progressively realise a

comprehensive social protection �end-state�. Atwo-phase approach is suggested, with the phaseddelivery balancing the need to allow time for thedevelopment of necessary institutionalarrangements and ensuring that the state is ableto meet important domestic and internationalcommitments related to poverty reduction. Indeveloping a long-term scenario which phasesin the progressive expansion of the socialassistance grant system the Committee has paidparticular attention to issues of fiscal feasibility.

5.6.3.4 Phase 1: Comprehensive socialprotection system prioritising the mostvulnerable

This phase would establish the preconditions fora comprehensive social protection system basedon the constitutional imperatives and apartheidbacklogs. It approaches the reform process in away that prioritises the most vulnerable, namelychildren up to the age of 18.

The key components of phase 1 are identifiedbelow.

5.6.3.5 Phase 1: Period 2002 to 2004

�Set minimum income level to eliminatedestitution

�Prioritise vulnerable and destitute groups(mainly children up to the age of 18)

�Put administrative systems in place

�Rationalise grants that are currently de factopoverty grants

�Ensure effective and efficient ways oftargeting the most vulnerable

�Simplify and eliminate the means testwhere they obstruct equity, administrativejustice and are costly to implement.

This phase uses the criteria of age as a parameterfor the income support grant. Since children areamong the most vulnerable and voiceless insociety, and children born into extreme povertyalmost never escape the poverty cycle theintroduction of an income support grant forchildren should be introduced.

The downside to this phase is that an exclusivechild-focused intervention does not addresshousehold poverty, and could create negativeunintended consequences. An exclusive focus onchild poverty, for example, would overlook thefact that the entire household will consumewhatever grant one gives for the child. Theaddressing of de facto household poverty in this


Consolidated Report

manner will exclude more than 1 million poorhouseholds without children, and may createperverse incentives regarding children. Theseproblems will be addressed in phase 2, however.

5.7.3.6 Phase 2: Comprehensive socialprotection for all

This phase completes the building of thecomprehensive social protection system initiatedin phase 1, by extending an income support grant,preferably called a �solidarity grant�, to all SouthAfricans.

Although more costly in the short to mediumterms than option 1, this alternative would havethe biggest impact on poverty and socialexclusion, thus providing various positive spin-offs for development and society in the mediumto long term. By the conclusion of this phase, thestate would have gone a long way in meeting itsWorld Social Development Summit (1995)commitments to eradicate absolute poverty by2015. The key components of phase 2 areidentified below.

5.6.3.7 Phase 2: Period 2005 to 2015

�Set a level to address absolute poverty

�Level of the grant determined in relation toa comprehensive package.

The two broad phases of option 3 are shown infigure 12.

Finally, the means-tested and universalmeasures, recommended in option 3, have beencosted (see chapter on Financial Framework forComprehensive Social security) and can be seento be financially feasible within the current fiscalframework.

5.7 ConclusionsIt is the view of the Committee that a morecomprehensive system of social assistance canplay an important role in achieving medium- tolong-term social and economic transformationin South Africa. First, it ensures theconstitutional and democratic imperatives aremet. Second, it indicates a commitment toaddressing the legacy of apartheid socio economicbacklogs. Third, it provides an opportunity tobalance social and economic policy goals byprioritising the most vulnerable in theimmediate term and ensuring in the long terman inclusive system of comprehensive socialprotection.

The Committee recommends that seriousconsideration needs to be given to fully utilisingfuture improvements in fiscal capacity to build

Figure 12Suggested phasing in of an income support grant system.


Consolidated Report

up this system of transfers.

The Committee also finds that although aBasic Income Grant is most able to eliminatedestitution and have a developmental impact onthe poorest, its implementation is constrained inthe short term due to fiscal and administrativeobstacles. It is, however, the view of theCommittee that fiscal and administrativecapacity exists for a phased and measuredintroduction of a comprehensive system ofincome support through social assistance, asdescribed in option 3. Within this context,households in clear distress should be givenpriority in the expanded framework.

An expansion of the social assistance grantsystem to adults living in destitution and povertycan have positive social and economicimplications, and is fiscally feasible if seen as along-term intervention.


Consolidated Report


Consolidated Report

Chapter 6Employment andUnemployment

6.1 IntroductionThe Committee has focused on unemploymentand the poverty and social consequences of low-wage, precarious employment, like much of thatfound in the informal sector.

Detailed analysis of employment andunemployment trends was provided earlier inthe Committee assessment of the socio-economic context. Key findings based on thatassessment are discussed below.

6.1.1 Key issuesThere is agreement that the number ofunemployed in South Africa has grownsubstantially over the past few years. In addition,many of their characteristics, and of thehouseholds in which they are to be found, can beascertained with some confidence:

�Formal sector employment levels are amatter of some disagreement. In recentyears, they may have risen (or fallen) alittle. The manner in which total formalemployment estimates are obtained (fromhousehold surveys) is problematic. Whatthe sector certainly has not done is to createjobs in the numbers so urgently requiredin South Africa.

�Informal employment appears to begrowing rapidly. This cannot, however, beconfirmed. Not only that, the (equallyshort) informal sector employment series isalso extremely volatile. It probably can besaid with some confidence that ifemployment in the sector is indeedgrowing, then most of the jobs created are

of the very low income (survivalist) variety.

�The official unemployment rate may havestabilised (at about 26 per cent), but thenagain, it may not � differences betweensuccessive estimates in the most recentperiod are not statistically significant, so itis not possible to say what has happened. Adiscontinuity in the official statistics alsogets in the way of our ability to interpretunemployment trends.

�Because economic activity rates appear tohave risen quite substantially between1999 and 2001, the numbers officiallyunemployed appear also to have risen(from about 3 million to 4 million).

�Unemployment rates according to theexpanded definition seem to have stabilisedat roughly their 1997 level (36-37 percent). It is not clear what impact, if any, the1999/2000 discontinuity has had on theseresults.

�The numbers of �expanded� unemployedappear to have increased substantially (bymore than 1 million since 1999, and bymore than 2 million since 1996). Onceagain, the explanation would seem to lie inrapidly rising economic activity rates. Whythe rates of economic activity are rising is amatter of conjecture at the moment.

�In 1999, among 2,6 million officiallyunemployed for whom usable data onduration of unemployment could beextracted, some 40 per cent of the 1,17million unemployed who had previouslybeen employed, had been out of a job for


Consolidated Report

more than a year. Among the 1,47 millionunemployed who had never previouslybeen employed, 45 per cent had beenunemployed for more than a year. There iswidespread agreement that the long-termunemployed constitute a particularlydifficult problem to resolve. The longerthey are out of work, the greater thedepreciation of human capital acquisitions.Among 4,58 million �expanded�unemployed Africans, only 1,24 millionhad previously been employed. Of the 3,34million who had never previously had ajob, considerably less than half (1,27million) were younger than 25 years. Thisis cause for concern.

�The circumstances of many of theunemployed are so poor as to make itextremely difficult to insert them gainfullyinto the labour market, under existingeconomic conditions. Certaincharacteristics, among them that of beingpoorly-educated, long-term unemployed,located in a non-urban area and so on, areamongst the structural impediments totheir obtaining jobs. Using thesecharacteristics, and mindful of the extremesensitivity of attempting to make such acalculation, the Committee has estimatedthat there may be as many as 1,5-2 millionpeople who fall into the category of�difficult-to-place�. Some commentatorsassert that these people are unemployable.While the Committee would notnecessarily want to go that far, the fact isthat these people are extremely difficult toreach with empowerment and skillsprogrammes. Their ability to earn even amodest living in the informal sector isprobably minimal. Clearly, their plight isclearly cause for grave concern. It is theview of the Committee that some form ofsocial assistance grant, concurrent withother socially integrative programmes,must be implemented for such persons.

�Despite dozens of Government povertyalleviation programmes at high cost, theavailable evidence suggests that poverty isgetting worse. The available evidence,though difficult to interpret, indicates thatthe number of �workerless� householdshad apparently risen dramatically between1995 and 1999. This, it would seem, is one

consequence of rising unemploymentlevels. In short, poverty alleviationprogrammes appear unable to keep pacewith unemployment-induced poverty.

6.1.2 Implicationsfor poverty

Poverty in South Africa is critically linked to thelabour market. Research carried out for theCommittee documents that in 1999 there were4,6 million South Africans in the pooresthouseholds (households where gross monthlyexpenditure was less than R400 per month). Afurther 5,7 million people lived in householdswhere expenditure was between R400-R800 permonth. Average monthly per capita consumptionexpenditure in the poorest households wastherefore, at best, in the region of about R100 permonth. In the next expenditure class (R400-R800) per capita, expenditure could not havebeen much more than R170 per month. Eventhe most conservative poverty datum line in thatperiod was set at between R300�R400 per monthper capita. All 10,3 million people discussedabove lived in households that contained noworkers, either formal or informal. Labourmarket failure was thus a key determinant of theirpoverty.

In 1999 there were roughly 2,6 millionunemployed in households in which there wasno worker present and in which monthly totalhousehold expenditure was less than R800 permonth. Of them, 1,4 million were women.Amongst them, 800 000 had given up searchingfor work, while a further 600 000 continued tosearch for jobs. Corresponding figures for the 1,2million men consisted of 560 000 discouragedunemployed � the remaining 590 000 sought work.Almost 96 per cent of these people belonged to theAfrican population group, i.e., extreme orchronic poverty has a disproportionate effecton them. All told, there were some 10,8 millionpeople (of whom 10,2 million were African)living in workerless households whereexpenditure was less than R800 per month.

The cost of unemployment goes further thanloss of income and even feelings of personalworth. South Africa�s failure to socialise manyyoung men, which is evident in the high rates ofcrime, alcohol abuse, violence against themselvesand women, and even in the spread of HIV/AIDS, creates a vicious cycle of familybreakdown.


Consolidated Report

6.2 Recommendations6.2.1 The appropriateform of social security

The current social safety net is inadequate to dealwith the immediate crisis of poverty andcontinuing alienation. The Committee thereforeagrees with Government that policies to addresspoverty and social exclusion should not belimited to the creation of a conventional socialsafety net. For, even after a net to stop people fromfalling into destitution has been constructed, thenegative effects of unemployment on socialcohesion will continue to be felt. Instead, thenotion of social protection has to be morecomprehensive and must provide an effectiveframework to design a system appropriate toSouth African needs.

Further the Committee proposes that theprinciple of social insurance, based onentitlements through contributory schemes,should where feasible, be extended to includeas many of the employed as possible. Thereare likely to be certain groups of workers whowill remain excluded from social insuranceschemes such as the UIF, because of theirlocation in the workforce. For these workers otherarrangements for providing social security areproposed.

Even if protection through social insurancecould be made available to all workers, it wouldnot solve the fundamental problems caused byunemployment. South Africa�s unemploymentproblem is more structural than cyclical. SinceUIF benefits are available for a maximum of sixmonths they cannot deal with unemploymentof this sort. The Committee�s research has shownthat because the majority of UIF claimants werepoorly paid when they were employed,replacement income levels are correspondinglylow. Benefit receipt durations are also very short(that is, benefit entitlements are quicklyexhausted). The degressive benefit schedule inthe new Act is an improvement, but it cannotsolve the problem of benefit exhaustion (asbenefits apply for a maximum of six months).

While unacceptably high levels of poverty andrelated problems require immediate Governmentintervention, the promotion of active labourmarket policies (and job creation policies, morebroadly conceived) is essential in South Africa.However, given the structural barriers,educational levels and other features of the global

economy, such policies on their own, are unlikelyto address the immediate social crisis. TheCommittee�s considered view, based on research,is that in the medium to long term, employmentgrowth and job creation will not be significantenough to ensure income security or earningsreplacement for low income workers in thesurvivalist sectors. In this context incomesupport through social assistance (or massivepublic works programmes) is necessary.

In short, social protection must comprise thetwo �direct� forms, as well as the set of �indirect�measures designed to facilitate favourable labourmarket transitions:

�Social insurance: This must be extendedwherever possible, with due considerationbeing given to the administrative feasibilityof providing such protection andrecognition of its limitations.

�Social grants: Coverage of these must beurgently widened to relieve the incomepoverty of the many who will not berescued by policies designed to stimulategainful labour market insertion.

�Indirect social protection: These approaches,through the facilitation of favourable labourmarket transitions, should be fostered by thedeployment of every policy instrument thatcan help to do so.

�Monitoring and evaluation: Each and everypolicy instrument to address the problemsof poverty and unemployment/employment must be monitored andevaluated.

Although close linkage of the direct(conventional social security measures) to theindirect (active labour market-type policies) isnot possible in the short to medium term inSouth African conditions, institutions to co-ordinate these policies so that their potentialrelationships are developed should beconstructed.

6.2.2 Socialinsurance: The UIF

Due to the highly unequal distribution ofearnings among those eligible to participate inSouth Africa�s UIF, and because of the highpropensity of poorly-paid workers to becomeunemployed, a contributor-funded scheme ofthis nature is vulnerable to fluctuations in thelevel of economic, activity, possibly to the point


Consolidated Report

where it is structurally non-viable. State supportmay be required to keep the scheme afloat duringcrises. It is also the case that business and labourhave on occasion agreed that increases in thecontribution rate (the pay-as-you-go rate[PAYG]) were desirable to secure the financialhealth of the scheme. The proposed increaseshave been blocked by Treasury. By comparisonwith similar schemes abroad, the PAYG rate inSouth Africa is low. The Committeerecommends that when agreement on the needfor an increase has been reached, that after dueinvestigation, such increase should bepermitted by National Treasury.

Maternity benefits proposed under the revisedlegislation are inadequate. It is proposed that theincome replacement rate should be raised on thedegressive scale. It is further suggested thatmothers become eligible for the full benefitpackage (17 weeks paid maternity leave) after 13weeks contributions. The possibility ofintroducing maternity-type benefits for those incasual, seasonal or insecure employment shouldbe investigated. It is proposed that an attempt toco-ordinate the many research initiatives onreproductive health and child health be made.

Domestic workers could be incorporated intothe UIF, on a similar basis to the IT-basedsolution proposed below. In return for bringingdomestic workers into the �banked� sector of theeconomy (the only way to create reliable contributorrecords) it is proposed that employers be granted amodest tax rebate.

Regarding the inclusion of Governmentemployees, the Committee is of the view thatGovernment workers should be allowed tochoose, through their representativeorganisations, whether or not to becomecontributors to the UIF.

The Committee has considered gaps in thenew UIF Act. A significant omission is theabsence of a clear principle for financing the UIF.

6.2.2.1 Financing principles for the UIF

The Committee has considered the complexitiesinvolved in the financing of the UIF and thatthe financing of the UIF has been a subject ofnegotiations in Nedlac and meetings between theDepartment of Labour and the National Treasury.The Committee�s notes current arrangements interms of which Government will effectively actas debt underwriter of the UIF. It is theCommittee�s view that although falling claim

levels and improved financial controls have easedthe financial crisis for the meanwhile, the UIFremains vulnerable. Until the special conditionsin which the UIF is required to operate areacknowledged, and appropriate Governmenttop up funds (in the form of a regularcontribution to the UIF) are provided, thisvulnerability will remain.

Noting that a start has been made innegotiations currently underway, theCommittee suggests that Government will haveto act as debt underwriter of the UIF, in thefinal instance, under conditions of (continuing)financial crisis. The Committee understandsthat Treasury support will be available for aperiod of some three years. During this periodthe UIF is required to take all steps necessaryto ensure that all the systems required to controla financial institution disbursing several billionare in place. At the end of the period theCommittee understands that the situation willbe re-evaluated. The Committee recommendsthat those concerned engage in a principleddiscussion over the nature of the desiredcontribution and benefit regime.

Unemployment �insurance� funds the worldover experience variations in their fortunes thatrelate in a complex way to movements in thebusiness cycle. In some funds, an attempt is madeto maintain a reserve from which to meetcontingencies. From time to time, even wealthyeconomies experience difficulties in keeping abalance between income and expenditure of thefunds. It is generally accepted that insuranceagainst unemployment is (actuarially) difficult,if not actually impossible. The Committee is ofthe view that insurance against massunemployment (on the scale experienced inSouth Africa) is out of the question.

The new Unemployment Insurance Actproposes the extension of coverage to thosecurrently excluded by virtue of the fact that theyearn more than the ceiling income. Claims of�high� income earners are likely to be so few asto make �value for money� for those newcontributors extremely low. Any tendency torespond to the discontent that this will evokeby an appeal to notions of social solidarity, shouldbe tempered by a realistic assessment of the extentto which such sentiments exist in South Africa.

The Committee has found that in South Africathe principles of social solidarity that underpin


Consolidated Report

or at least used to underpin systems in the UnitedKingdom (UK), the Scandinavian countries andGermany, are perhaps most remarkable by theirabsence. The proposed extension of coverage islikely to be seen by many of them for what it is �a tax, not an insurance premium. TheCommittee does not have a principled objectionto the levying of such a tax. From projectionsthat have been examined, it would appearhowever, that the additional funds so generatedwill still not be sufficient to keep the fund afloatin times of crisis.

6.2.3 Activelabour market and

job creation policiesIn the absence of the preconditions for a tightintegration of labour market and social securitypolicy there is a need, in the medium term, foran inter-departmental body to co-ordinate themany active labour market policies and job-creation initiatives. Since the linking of socialand economic policy goals are central to itswork, representation of the Social SectorCluster, and in particular the Department ofSocial Development on this body is essential.

The Committee supports initiatives presentlybeing developed by the Department of Labourto address youth unemployment through thecreation of several hundred thousand publicsector �learnership� opportunities, especially inthe provision of essential social services of apara-professional nature.

In the Committee, however, the argument wasmade that these are not enough. There is agrowing number of young and relatively healthyadults who cannot find jobs. It is far from clearthat the needs of these unemployed are met bycash grants or current active labour market/ jobcreation policies only. The unemployed needjobs that allow them to participate in society, tocontribute financially to their families and toincrease their self-worth.

The urgent social imperative is for more jobs,and the argument is clear: structuralunemployment requires changes to structures,and structures will not change unless they areexplicitly addressed at government level. Forexample, the Committee received a submissionthat more labour intensive methods could leadto over 300 000 additional jobs in theconstruction industry alone. This submission,

from experienced civil engineers, argued thatthese labour intensive methods would not leadto additional costs or compromise of quality. butdid require that government and industryundergo a change of mindset.

The change need apply not only to introducingmore labour intensive methods, but also toextending formal sector employment ingovernment, the private sector, and civil sector.

6.2.4 Publicwork programmes

The Committee endorses the proposal that asmany jobs as possible be created through suchprogrammes but cautions that public workprogrammes, by their very nature, do not offerlong term viable employment opportunities forthe unskilled structurally unemployed. Notingthe high praise that South Africa�s community-based public work programme has attracted, andthe success of certain special projects such as�Working for Water��, the Committee devotedconsiderable energy to the investigation of publicwork programmes.

They have the widely recognised advantageof being self-targeting � attracting only thoseso desperate for work (income) that any wageabove the opportunity cost of coming out towork would be accepted. It cannot beascertained, however, how many people fall intothis category.

Working out what wage would bring themout in the �appropriate� numbers (whatever oneunderstands that to mean), is also tricky. Almosttwo-thirds of domestic workers, and slightly lessthan half of informal sector workers earnedbetween R1-500 per month. What their meanearnings were is anybody�s guess. Suppose thatsome significant proportion of workers in thisincome class received only R200 per month(and there is abundant evidence that such lowwages are paid). A public work programmethat offered R300 per month, especially ifoffered for the foreseeable future should attractall of those being paid less than R300, give ortake whatever differences there were inopportunity costs. The public work programmewould therefore begin to address the povertyproblem as well. Given this complication (theintention of the programme is to address theproblem of unemployment), it really becomesimpossible to say how many jobs would have tobe provided.


Consolidated Report

It is probable that most of the 2,6 millionjobless people located in workerless householdsin 1999 where total household expenditure wasless than R800 per month, would accept the lowwages to be earned in public work programmes.It is not known how many such people there arein comparable households at present � since thenumbers of unemployed have risen, theirnumber could well have risen.

If R300 per month were offered to 2,6 millionpeople, the wage cost of the public workprogramme would be R9,4 billion per annum.Raising the wage to R400 per month would pushthis to R12,5 billion. Such programmes areexpensive to initiate and to run. Estimating themanagement and materials costs for a project asambitious as that being discussed (it is one thingto design a scheme employing a few thousand -it another matter altogether to do the same witha few million) takes the estimates even furtherinto the realm of speculation. If these costs werebetween 50 and 100 per cent of the wage cost, thetotal cost could lie between R14-25 billionannually.

Although large, it could be argued that this isnot an outrageous price to pay for a well-targetedscheme. To counter the charge that each �benefit�is delivered at very high cost (guessed at above aspossibly being between 50 and 100 per cent ofthe �benefit�) one could point out that societybenefits in two ways � useful work is done, andwelfare dependency is not created.

It should be noted that the other 4 millionunemployed are assumed by this approach to beable, one way or another, to take care ofthemselves. The 2,6 million catered for here willnot diminish in number until such time aseconomic growth can rescue them. In otherwords, the public work programme will have tobe sustained for many years. The sheermagnitude of such a task � finding projects totackle and the necessary managers to run suchprojects � is daunting. The question of theefficiency of such expenditure relative, say, to auniversal grant whose net cost would be roughlythe same, must be asked.

Certainly, public work programmes are worthimplementing. Although some of the projectsthat have been tackled have earned high praise,too little has been done to date. This, in itself,however, is probably an indication of howdifficult it is to organise such things. Anassessment of the potential of these programmes

must find that while they can do quite a bit torelieve poverty, they cannot be introduced on ascale large enough to do much more than dentthe surface of South Africa�s unemploymentproblem.

6.2.5 Informal social securityDespite the important role played by the formalsocial security system ( for example, social grants),it is clear that the formal social insurance systemexcludes and marginalises many of the poor whowork informally or who have to rely on informal(that is, family-/kinship-/community-based)social security mechanisms. The reasons for thisstate of affairs relate to:

�the formal employment bias andcategorical approach of the present socialsecurity system

�the urban bias of the present system

�the restricted family concept underlyingmuch of the formal system

�the limited concept of work which, as arule, does not encapsulate the productiveand reproductive work in which womenare involved, thereby marginalisingwomen in poor and traditionalcommunities.

It also has to be noted that the grant systemavailable under social assistance, despite theimportant role that it plays on an individual andhousehold level, is mainly restricted to old anddisabled persons and in respect of children underthe age of seven.

It is important not to impose a social securitysystem that will be detrimental to traditionalsupport mechanisms. Transformation of thepresent social security framework should,therefore, aim at supporting and strengtheningexisting informal social security with the view toenhancing solidarity.

In the first instance this requires consideringways to integrate currently excluded groups intoformal schemes. This includes embarking onpilot schemes aimed at supporting informalsocial security mechanisms; removingunnecessary legal restrictions in relation to accessto schemes; devising tailor-made schemes tocater for these excluded and marginalisedcategories and groups; introducingcomplusory membership of private or publicsocial schemes; and campaigns to promoteprivate insurance and savings.


Consolidated Report

Further, there is a need to consider broaderinterventions and programmes that bolster theoverall ability of communities and informalsystems to cope with and manage increased levelsof risk and hardship.

6.2.6 Institutionalarrangements

These recommendations must be seen within thebroader recommendations made later in thisreport on Institutional Arrangements. Anindependent Poverty, Social Exclusion and SocialProtection Studies Unit should form acomponent of the proposed Social ProtectionCommission (SPC) (See chapter 13). This unitwould take a lead in seeing that all usefulinformation (such as that generated by StatisticsSA) is extracted from the raw data and their policyimplications assessed for appropriate action.

Consideration should also be given to thecollection of data and information on the successor otherwise of all job creation policies ofGovernment and of skills training. Without thisinformation and policy tools to evaluate theimpact of Government�s job creation and humanresource development strategy, it will be extremelydifficult to determine efficiency and equity gainsand impediments to social security.

There are numerous non-profit organisations(NPOs) and non-governmental organisations(NGOs) engaged in the skills training and�empowerment� field. It is far from clear that thebenefits of their activities outweigh the costs. Therole and place of these institutions must beclarified.

6.2.7 Policy evaluationInstitutional capacity to undertake policyevaluation on an ongoing basis has to bedeveloped. In terms of available statistics, theOHS with appropriate modifications by StatisticsSouth Africa is essential. The LFS does notprovide all the data required to monitor, evaluateand make informed recommendations. It is notclear that the primary information required toevaluate policy can be obtained from the LFS.The five yearly population censuses are too farapart for evaluation purposes. These issuesshould be referred to the National StatisticsCouncil.


Consolidated Report


Consolidated Report

Chapter 7Protectingthe Children

7.1 IntroductionStudies have shown that a strong link exists betweenadverse circumstances experienced early in life andfuture success. The nature of this link is offundamental importance to overall social policy.Understanding and removing adversity for familieswith children must become the priority of anyGovernment. Some of the main issues highlightedby these studies indicate the following:

�There is a significant relationship betweenadverse social and economic conditions inchildhood and later success in life.Independent impacts are detectable fromfamily structure and income.

�A life of poverty is statistically associatedwith higher rates of activities detrimentalto individuals and society, such as crime,violence, underemployment,unemployment, and isolation from thelarger community.

�The chances of unemployment later in lifeare higher for children who experienceperiods of poverty than for those that do not.

�Increased incomes for single mothers withchildren through social transfers have asignificant impact on educationalperformance of children.

�The malnutrition and stunting of youngchildren is unacceptable in a relativelywealthy South African society.

As a result, the Committee is of the view thatstrategies to address child poverty must be part ofthe overall strategy to alleviate and reduce poverty.Therefore measures to address child poverty (andthe child support grant, in particular) are centrally

embedded in the Committee�s recommendationscovering �comprehensive and integrated medium-to long-term framework for income support�. Thischapter, then, deals with important gaps andadditional issues regarding child protection.

Finally, the Committee is aware that the SouthAfrican Law Commission is developing acomprehensive framework for children, and issupportive of that process. As a result, theproposals and recommendations of theCommittee within this chapter must beconsidered with that framework.

7.2 The Constitution7.2.1 Constitutional

obligationsSection 28(1) of the Constitution deals specificallywith fundamental rights of children. Unlike othersocio-economic fundamental rights that grant a�right of access� to the right, subject to progressiverealisation by the state within its availableresources, the rights pertaining to children(everyone under the age of 18 years) do not haveany such limitations. The special vulnerabilityof children and their protection is in this wayrecognised. Section 28 grants children inter aliathe right to the following core rights to:

�Family care or parental care, or toappropriate alternative care when removedfrom the family environments

�Basic nutrition, shelter, basic healthcareservices and social services

�Be protected from maltreatment, neglect,abuse or degradation

�Be protected from exploitative labour practices.


Consolidated Report

In all matters concerning children, the child�sbest interests are of paramount importance.(Section 18(2)) In terms of section 27 of theConstitution everyone has the right to haveaccess to social security, including, if they areunable to provide for themselves and theirdependants, appropriate social assistance.

7.2.2 Internationalinstruments

Various international law instruments pertainingto the welfare of children also bind South Africa.

7.2.2.1 African Charter on the Rights andWelfare of the Child

South Africa signed this Charter on 10 October1997 but is not yet under any obligations interms of the Charter as it has not yet come intooperation. State parties are required to takemeasures to eliminate harmful social andcultural practices. The responsibilities of thechild towards his/her family is also stressed.

State parties must see to the provision ofnecessary medical assistance and healthcare forall children, with particular emphasis on thedevelopment of primary healthcare.

7.2.2.2 United Nations Convention on theRights of the Child

South Africa signed and ratified this Conventionin 1995. A child, as defined in this Convention, isany human being under the age of 18, unless aparticular nation�s laws set an earlier age for theattaining of majority status. Article 6 of theConvention places state parties under an obligationto ensure the survival and development of the childto the maximum extent possible. This provisiongives rise to numerous derivative social securityrights, such as the right to healthcare necessary forsurvival and a standard of living that meets the needsfor food, clothing, shelter and education. The factthat this Convention has been ratified by SouthAfrica places South African courts and otheradjudicating bodies under an internationalobligation to comply with the duties placed onmember states. In terms of one of these duties thestate is expected to report regularly to the supervisingbody under the Convention on the Rights of theChild on the compliance with the duties bestowedon South Africa as a member state.

Amongst the various rights of children listedin the Convention, the most important for thepurposes of social protection are:

�Every child has the right to benefit fromsocial security, including social insurance,and the state should take the necessarymeasures to achieve the full realisation ofthis right in accordance with national law.Social security benefits should be granted,taking into account the resources and thecircumstances of the child and thoseresponsible for the maintenance of thechild (Art. 26).

�Linked to the above is the right of everychild to a standard of living adequate forthe child�s physical, mental, spiritual,moral and social development. The state�sduty is to (within the means available)assist the parents with this responsibility,by taking measures, which could includematerial assistance and supportprogrammes, particularly with regard tonutrition, clothing and housing (Art. 27).

�The state must undertake all appropriatemeasures for the implementation of therights contained in the Convention on theRights of the Child. As most of the rightslisted above are economic and social rights,the state is only required to undertake suchmeasures to the maximum extent ofavailable resources (Art. 4).

The UN Committee which monitorscompliance with this Convention has been fairlycritical about South Africa�s non-compliancewith many provisions of the Convention.Urgent attention needs to be paid to therecommendations of the UN Committee so asto ensure full compliance with the Convention.

7.3 Evaluationof current policiestowards children

Based on numerous submissions and documentsmade available to the Committee of Inquiry,substantial problems exist with the presentsystem of protection for children in South Africa.The inadequacy of the current policy frameworkappears a valid conclusion based on reviews ofthe socio-economic indicators above.

7.3.1 Social assistance7.3.1.1 Foster care grant

The foster grant is payable to a foster parent inrespect of a foster child who has been legally placed


Consolidated Report

in their custody in terms of the Child Care Act.

In order to qualify for a foster grant:

�The child must have been legally placed inthe care of the foster parent(s)

�The income of the foster child must notexceed twice the annual amount of thefoster child grant

�The applicant and foster child must beresident in South Africa at the time ofapplication.

Submissions to the Committee report thataccessing the grant is problematic for thefollowing reasons:

�To qualify for a foster grant, the child hasto be placed in the care of foster parentsthrough the children�s court. The courtprocess is lengthy and inappropriate formany families who are content with caringfor children who are not their own, butrequire some form of support.

�Problems exist with accessing foster grantsfor non-South African children because ofthe paperwork required.

Information was also provided to the Committeesuggesting that there are abuses of the system.Many parents who would otherwise be able tocare for their own children are abandoning theirchildren with relatives so that these relatives canaccess the foster grant.

Foster care is intended as a specialised servicefor children who have been removed from theirfamilies with the intent of returning them after aperiod of no more than two years. However, oneof the reasons why foster placements are renewedrepeatedly is the lack of financial support foradoption.

7.3.1.2 Care Dependency Grant (CDG)

A care-dependent child is a child between theages of one and 18 years who requires and receivespermanent home care due to his or her severemental or physical disability. The purpose behindthis grant is to enable parents or foster parents tocare for children with physical or mentaldisabilities in their homes.

The child must not be permanently cared forin a Government institution. The grants areawarded until the child is 18 years of age, or untilthe child is no longer cared for by his/her parents.

The following is a review of problems withthe current grant:

�The current purpose of the CDG to enablepermanent home care, only forpermanently disabled children, is limitingand inadequate, and open to differentinterpretations.

�There is a lack of clear definitions(disability, severe/moderate, and permanenthome care) in the current legislation. Thishas serious implications for inclusion/exclusion criteria and makes targetingextremely difficult.

�Currently the CDG benefits only severelydisabled children permanently at home, anddoes not cater for the many others with milderdisabilities or those in daycare facilities.

�While means-testing enables targeting of thepoorest quintiles, in practice it is rarely usedcorrectly, is administratively demanding andhas been reported as demeaning.

�The assessment test can be highlysubjective and open to the personalinterpretation of the medical officer.

The recommendations regarding the CDGare dealt with in the main disability section (seechapter 10).

7.3.1.3 Maintenance Act

The Maintenance Act No. 99 of 1998 makesprovision for the payment of maintenance by anyperson who has the legal obligation to support achild or children. There is also the obligation onthe part of employers to co-operate with themaintenance court orders in effecting deductionsthrough salaries.

The parental maintenance system in SouthAfrica appears to be in disarray. There is awidespread lack of responsibility shown by manyliable parents in terms of their obligations tosupport their dependants, especially wherechildren are brought up in single parenthouseholds. There is perception among somenon-custodial parents that custodial parents�abuse� the money they receive and spend it onthemselves rather than their children.

There is shortage of financial resourcesallocated to the maintenance system and lack ofpersonnel to deal with the vast number ofmaintenance cases and the lengthy delays. Dueto these restraints the Family Court PilotProjects, which were established by theDepartment of Justice in 1998, are not able todo much.


Consolidated Report

Other problems relate to the actual amount ofmaintenance to be paid by non-custodial parents.There is low and inadequate level of awards inparental maintenance. Great variations existbetween courts as to the monthly amount, whichis awarded.

There is also lack of proper tracing agents totrace liable persons and obtain accurateinformation about their income and means.Furthermore, the Maintenance Act does notaddress the problem of those who are self-employed or in informal sector, and whosefinancial state is often difficult to ascertain.

Despite being plagued by problems,however, the private maintenance system, , isan important system and as a result it shouldnot be discarded. There is a legal and moralduty on the part of both parents to provide fortheir child(ren). This duty operatesirrespective of whether the relationshipbetween the parents is still in existence or not.The state�s duty to provide enters the pictureonly when parents are unable to provide � notbecause they do not want to.

7.3.1.4 Adoption

Section 28(1)(b) of the Constitution guaranteesevery child the right to family or parental care orto appropriate alternative care when removedfrom the family environment. This is supportedby section 28(2), which states that the child�s bestinterests are of paramount importance in everymatter concerning children. Article 20 of theConvention on the Rights of the Child statesthat a child in need of care shall be entitled tospecial protection from the state. Such protectioncould include foster placement or adoption.Despite the recent adoption law reforms,introduced by the Child Care Amendment Act96 of 1996 and the Adoption Matters AmendmentAct 56 of 1998, certain key areas of adoption lawstill need reform.

Current South African adoption law iscontained in Chapter 4 of the Child Care Act, 74of 1983. The standard of adoption is still the bestinterest of the child. Some of the problem areasthat can be highlighted in this regard are (thissection is based on submissions made by theSouth African Law Commission �Project 110:Review of the Child Care Act � Adoption asSubstitute Family Care�):

�Section 10 of the Child Care Act does notkeep track of the child-rearing practices

especially in black communities whereinformal fostering is prevalent. Section 10prohibits anyone from receiving a childand caring for him/her for longer than 14days. This section does not have a penaltyfor the contravention but it remains a goodground for removal of a child.

�Transracial adoptions are allowed in termsof the current Child Care Act. The right toequality is subject to reasonable limitationsas can be justified in an open anddemocratic society. The best interests of thechild would be considered a reasonablelimit that might justify a race-sensitivecustody decision. In South Africa, becauseof the tradition and socio-economicrealities, relatively few black families adoptchildren. This results in �a-one-way-traffic-of-black-children-into-white-families�. An important consideration iswhether a child growing up in an adoptivefamily of persons who are different fromthem, in cultural background and physicalappearance, will still be able to develop apositive self-identity.

�Inter-country adoptions are still aproblematic issue in South Africa. SouthAfrica has not yet signed the HagueConvention on Inter-country Adoption.Law reform measures promoting inter-country adoption should recogniseadoption orders made in other countriesand encourage open adoption, making itpossible for a South African child adoptedin another country to maintain links withthe extended family members and with his/her cultural roots.

�Adoption is a private arrangement incustomary law in terms whereof twofamilies make an arrangement in order toensure an heir for a family head who hasno male progeny. Payment is sometimesmade to compensate the natural parentsfor rearing the child. This raises thequestion if customary adoption infringescommon law prohibition on traffickingin children and if it should take placeunder auspices of common law. This isarguable in light of the constitutionalrecognition of customary law as a systemof law and there is no reason whyrecognition should not be given to anadoption under customary law.


Consolidated Report

7.4 Policyrecommendations to

address problems, gapsand inconsistencies of

the current socialsecurity paradigm for

children7.4.1 Dealing with the

special needs of children7.4.1.1 Recommendations pertaining tofoster care grants

The foster care grants provide an importantsupport mechanism for children in compromisedfamily environments. However the followingissues need to be addressed:

�The process of foster care allocationshould be simplified. This will entail thatthe present children�s court procedure beamended in order to shorten the process.It must, however, be kept in mind thatthe procedure is extensive in order toallow for proper investigation by socialworkers to serve the best interests of thechild. Provision should also be made forfinancial support (interim foster carepayments while the court process ispending).

�Non-South African children should beallowed to access these grants. This meansthat the Department of Home Affairsshould ensure that all resident children arespeedily issued with appropriatedocuments.

�Provinces should have a uniform approachin granting this (and other) grant(s),especially as far as the eligibility ofchildren in daycare centres or LSENschools for the CDG are concerned.

�Programmes should be run and servicesbe made available at schools and hospitalsthat have frequent contact with candidateseligible for these grants.

7.4.1.2 Recommendations pertaining toorphans and child-headed households

There is no social assistance specifically intendedfor orphans. In the past, orphans that have notbenefited from social or private insurance payouts

have been cared for by other members of theirfamilies, foster parents or orphanages normally runby NGOs. The enormous increase in the numberof orphans as a result of adult AIDS deaths is goingto place overwhelming pressure on theseinstitutions.

Child-headed households are a rapidly emergingphenomenon in Africa (partly due to the HIV/AIDSpandemic). The extended family as social supportmechanism is eroded by factors, such as poverty,HIV/AIDS, urbanisation and over-stretchedresources. These eroded family structures areresulting in a shifted burden of care for childrenorphaned by AIDS. The burden falls on the elderlyor on other children, both who are ill equipped tocarry this responsibility, financially andemotionally.

If possible, child-headed households shouldbe incorporated into other households, or evenadult supervised institutions. The needs of thechildren, material and otherwise, should beparamount. Siblings should be allowed to livetogether. At the very least, the Department of SocialDevelopment must allocate responsibility for thehouseholds to some adult in the community.

The Department of Social Development must,as a matter of urgency, make provision to supportthe growing number of orphans, especially thoseleft in child-headed households. Short-termmeasures include the following:

�Extension of the CSG to all children 0-18years.

�Simplify access to this grant. In order toassist child-headed households, childrenshould be assisted by community-basedorganisations (CBOs) and NGOs in orderto allow for adult supervision inapplication and spending of the grant.

�Develop skills and vocational training forthese children or participation indevelopment projects, e.g. gardeningproject, small business management etc.

Long-term measures would include:

�Projects aimed at prevention andintegration of these children into society

�Encourage of home- or community-basedcare

�Simplify foster and adoption process,where a mechanism is developed for�informal carers� of children to accessfoster child grants


Consolidated Report

�Involve CBOs and NGOs in theidentification, assessment and care ofvulnerable children.

Set up of childcare centres to increase thesechildren�s learning opportunities andpsychological skills.

7.4.1.3 Recommendations pertaining tochildren living with HIV/AIDS

One of the tragic effects of HIV/AIDS is the vastnumber of children infected and affected by thedisease

These children currently fall outside of thesocial safety net. They would seldom qualify fora CDG (unless in cases of terminal stages of thedisease where they might require permanenthome care), and if they are older than 7 yearsthey would no longer be covered under the ChildSupport Grant. As HIV/AIDS orphans are oftencared for by family on an informal basis, they donot apply for foster care and do not make use ofthe foster child grants.

Submissions to the Committee suggested thatthe situation be addressed in the followingmanner:24

�Extension of the CSG to all children 0-18years

�Allow for free health services for allchildren

�Projects aimed at AIDS awareness andprevention should continue and beexpanded

� Simplify foster process, where a mechanismis developed for �informal carers� of childrento access foster child grants.

7.4.1.4 Recommendations pertaining to theMaintenance Act

Parents who shy away from their duty tomaintain their children, as required by law, placea heavy burden on the state�s social services.When parents take responsibility and providemaintenance for their children, the burden onthe state is eased. It therefore follows that toensure that state resources are directed to thosechildren who are vulnerable, not becausesomebody is not taking responsibility, but whoare in need (for example, because their fatherpassed away), the private maintenance system isvery crucial.

The following suggestions by Law, Race andGender Research Unit (UCT) to the Committee

of Inquiry are supported by the Committee:

�Increased personnel are indeed required todeal with the vast number of maintenancecases and lengthy delays. It is a known factthat many private maintenance claimantsor would be claimants have lost faith in thesystem. In addition many elect to stop tryingto claim maintenance due to lack of taxi faresto frequent the magistrate�s courts.

�Specialised tracers be employed to traceliable persons. The suggestion of settingspecialised tracers on liable persons hasalready been raised by the LundCommittee. Using tracers to track downthose liable parents who are avoidingresponsibility is to be welcomed.

�The maintenance division of the familycourt could be required to take direct actionon arrears without waiting for charges to befiled by the complainant and make itmandatory, unless good reason is given, forarrears to be recovered with interest.

�Summons and subpoenas used to orderrespondents to come to court should bestandarised. This is more than necessary inthe light of the level of illiteracy in SouthAfrica.

�A campaign modelled along theMasakhane Campaign lines to instil aculture of responsibility towards theirchildren in the general population,proposed by the Lund Committee, shouldbe supported.

�There is a need for clear policy guidelinesfor those involved in the handling ofmaintenance claims. One issue, which canbenefit from the proposed policyguidelines, is the question of secondfamilies and multiple parenting. The issueof second families has raised a variety ofquestions of both legal and moral nature.Should, for example, courts take the non-custodial parent�s new obligations to anewly established family intoconsideration when dealing with amaintenance claim?

7.4.1.5 Recommendations pertaining toadoption

A means test is applied in terms of s17 of theChild Care Act, in terms whereof adoption willbe allowed if the adoptive parents possess adequate


Consolidated Report

means to maintain and educate the child. Thismeans test bears potential prejudice towardspoorer applicants for adoption. It should be notedthat this means test serves a different purpose thanthe normal means test. A means test is normallyinstituted to protect the interests of the state. Inthe case of adoptions, the means test is applied inorder to serve the best interests of the child. Alsoof relevance here is the fact that foster parentswho are financially needy can receive a statefoster care grant, but adoptive parents who havethe same financial difficulties are not eligible forany grants. This may prevent foster parents fromadopting a child in their care and this impactsnegatively on permanency planning for the child.

In order to overcome these problems, the LawCommission proposed that the possibility of anadoption subsidy should be made available. Thepurpose thereof is to enable poor people to adoptchildren. The relevance of such a subsidybecomes less apparent in the case of a universalchild grant, (i.e. until the age of 18 years).

Subsidised adoption may have certainnegative implications:

�Parents might tend to adopt children forthe wrong reasons

�Adoptive parents would then be treateddifferently from biological parents

�One would have to bear in mind thatdisallowing poor parents to adopt childrenis in order to serve the best interests of thechild.

7.4.1.6. Measures to enhance the CSG

There was argument in the Committee that theeffectiveness of the CSG is undermined by a lackof nutrition information and inappropriatenutrition education. The Committee thereforerecommends that the CSG be supplemented byan appropriate nutrition and child care supportprogramme. In this regard, caregivers of HIVpositive children will need particular help.

7.5 ConclusionIt is the constitutional and internationalobligation of the state to provide social securityto children. Through providing social assistance,and with the concurrent development of servicesand development programmes, the state can andmust attempt to improve the standard of livingof children.

This chapter, however, does not purport toprovide a comprehensive package of socialsecurity for children. It is envisaged that the SALaw Commission in its proposedComprehensive Child Care Statute will makeprovision for that.


Consolidated Report


Consolidated Report

Chapter 8Health

8.1 IntroductionTransformation of the health system in SouthAfrica has been and remains an urgent priorityfor the democratic Government since 1994.South Africa has introduced significant policyshifts and institutional changes to deal with theunderlying problems of an inequitable race basedsystem. Through its primary healthcareapproach, the Government prioritised the needsof women and children by extending freehealthcare for children under the age of six andpregnant women. The free healthcareprogramme was thereafter extended to all SouthAfricans using public primary healthcarefacilities. At hospital level, payment for servicesis means tested, and indigent citizens are entitledto receive free services.

Health interventions are critical indetermining how Governments address issuesof capability poverty. In the design of socialsecurity reform, health indicators are used toidentify the extent to which deprivation andexclusion from essential health services affectthe life chances of people. This section of thereport addresses the key problem areas in thecurrent policy context, and sets out a long-termstrategy to address the underlying challengeswith particular emphasis on those aspects relatedto social protection.

Healthcare provision constitutes an essentialcomponent of a minimum package of goods andservices for the development and advancementof people. Moreover, given the history of unequalallocation of resources, levels of poverty andunemployment, a central policy objective is toachieve equity in and access to the allocation of

state resources to address health needs. Therelationship between public and private healthprovision and the roles and responsibilities thatare located in these environments are examinedby the Committee to ensure the sustainable,equitable use of resources in the interests of all.

An overview of developments in South Africa�shealth system indicates that the reform directionand approach developed and proposed in the1995 National Health Insurance (NHI) Paperremains a valid point of departure for ongoingreform. This requires that South Africa moveultimately toward a NHI system over time thatintegrates the public sector and private medicalschemes within the context of a universalcontributory system.

8.2 Findings8.2.1 Problems identified

with the existingstrategic framework

The existing structure of the health system hascertain endemic perverse cycles that need to bereversed through interventions at an institutionallevel. The central contributors to this negativecycle are identifiable in four areas:

�Cover: The public sector is faced with anincreasing population, both low-incomeand indigent, while the private sectorpopulation is not increasing. The publicsector also has to provide cover for sickerand less healthy groups traditionallycovered by the private sector. This lattershift is induced through risk-selection


Consolidated Report

within and uncontrolled cost increases.

�Burden of disease: The public sector isfacing a worsening burden of disease as aresult of HIV/AIDS as well as increasinglevels of diseases of poverty. The privatesector is attempting to shift HIV/AIDSpatients and chronic patients onto thestate system, as part of the risk selectionprocess.

�Finance: Despite an increasing populationand disease burden, the public sector healthsystem faces a constant or declining realbudget allocation. The private sector, bycontrast, increases its expenditure atroughly double the annual inflation rateon a per capita basis. As costs increase inthe private sector, so does the effective taxsubsidy.

�Providers: In the face of an increasedpopulation to cover, an increased diseaseburden, and a declining budget, the publicsector is losing clinical personnel to theprivate sector. As such, the private sectoreffectively drains resources from the stateto provide cover to a relatively healthy andyounger population. The private sectoreffectively receives a tax subsidy ofapproximately R7,8 billion to reinforce thistrend.

Taking account of the above, Government needsto adopt a strategic approach to reforming thehealth system that engages fully with both thepublic and private sectors. The objective wouldbe to achieve jointly what each cannot realisealone.

8.2.2 Role and scope ofGovernment involvement

The ultimate responsibility for the overallperformance of a country�s health system lieswith Government, which in turn should involveall sectors of society. Government has theresponsibility for establishing the best and mostequitable health system possible with availableresources. The oversight and effective regulationof the private sector has to form part of the overallGovernment response and must be high on thepolicy agenda.

Central objectives

�Increased risk pooling: Risk pooling needsto be encouraged through the use of acombination of instruments. These would

include the tax system, the creation of riskequalisation mechanisms within bothpublic and private sectors, Governmentmandates, and the reinforcement ofcommunity rating.

�Benefits: Government policy needs toprovide a framework that results in coverfor a minimum level of essential benefitsirrespective of whether it is provided inthe public or the private sectors.

�Efficiency: Given the existence of perverseincentives in unregulated markets forhealthcare, any regulation must paycareful attention to the incentivesgenerated. The use of mixed systems forcovering and providing healthcarecombined with the correct elements ofchoice is the best approach to balancinghealthcare objectives with the need foroperational efficiency.

Role of the public sector

The public sector system must remain thebackbone of the overall health system andshould be protected from chronic under-funding.

Role of the private sector

The private sector can provide an effectiveenvironment for achieving increased levels offunding over and above tax-based allocations.However, as the private market for healthcaresuffers from chronic market imperfections,public sector involvement is required to ensurethat funding levels are socially optimal and notmerely what the market will bear.

8.3 Recommendations8.3.1 Reform strategy

The Committee recommends that South Africamove toward a NHI system based on multiplefunds and a public sector contributory environmentas defined in the 1995 NHI Committee Paper.Initially the environment would remaindifferentiated between a private contributoryenvironment and a general tax fundedpublic sector environment. Over time thisstrict differentiation should diminish with abroader contributory environment emerging,replacing general taxes as a revenue source.The ultimate elimination of general taxes as akey revenue source is unlikely for a fairly longtime (figure 13).


Consolidated Report

The reform process has to take into accountthe need to develop a phased approach wherebykey enabling measures are implemented and thebase established for the longer-term reforms.

Four phases are envisaged defining importantlinked reform measures. The phases guide theevolution of health system toward theachievement of a universal contributory system(figure 13).

8.3.1.1 Phase 1: Development of theenabling environment

The current health system is incompatible withthe introduction of, or integration with,contributory environments. The overall systemof cross-subsidies is fragmented and notstructured in accordance with strategic policygoals. Furthermore, the strict partitioningbetween the public and private sector spheresis resulting in substantial and unsustainableprivate sector cost escalations. This occursbecause private medical schemes are technicallybarred from officially contracting for and usingpublic sector services.

The priorities within phase 1 therefore focuson an enabling environment for moresubstantive future policy reforms. Central to

this process is a focused improvement of publichealth facilities and their management.

(a) Preparation of the public hospitalsystem:

i. Decentralise public hospitalmanagement

ii. Implement a coherent uniformpolicy with respect to enhancedamenities

iii. Investigate financial injectionoptions to enhance public sectoramenities

iv. Establish a process to develop andimplement minimum servicerequirements for the public system

v. Revise the human resourceenvironment as it relates to healthpersonnel to improve managementand incentives to perform.

(b) Consolidation of medical schemereforms to remove any residual risk-selection and to increase coverage:

i. Expand prescribed minimumbenefits to include chronic

Figure 13Reform strategy and approximate timeline.


Consolidated Report

conditions and other essentialservices

ii. Phase out benefit options or,alternatively limit the degree towhich they can be differentiated

iii. Phase out medical savings accounts

iv. Refine the late-joiner penalties

v. Significantly improve the regulatoryframework dealing withintermediaries.

(c) Development of an effective policyprocess on defining and implementingbasic essential services: Ultimately boththe public and private sectors will needto ensure coverage for an equivalentminimum core set of services. Withinmedical schemes these would beregulated as prescribed minimum benefits.Within the public sector a similar processwould occur and be framed as minimumnorms and standards.

(d) Development of an integrated subsidysystem:

i. A process needs to focus on rectifyingstructural deficiencies within andbetween the existing risk-poolingmechanisms. These include:

1. Inequity in the allocation of publichealth services

2. The tax subsidy to medicalschemes

3. Risk-equalisation between medicalschemes

4. Unfair penalties applied within themedical schemes environment.

ii. The public sector budget systemneeds to be revised to ensure that theregional allocation of health servicesis equitable. Furthermore, thesubsidy provided to the private sectormust at no time exceed that providedto people covered through the publicsector.

iii. The tax subsidy currently runscounter to the achievement of healthpolicy objectives and must bereformed. It is recommended that itbe converted into an explicit income-and risk-adjusted subsidy. Thissubsidy could ultimately be funded

from an earmarked tax, althoughinitially it should be funded fromgeneral tax revenue.

iv. It is essential that a system of risk-equalisation between medicalschemes be introduced. This fundwould also serve the function ofallocating any appropriatelystructured risk-adjusted subsidy tomedical schemes provided byGovernment.

(e) Measures to contain private sector costincreases need to be more explicitlytargeted by Government policy. Theseshould include the use of:

i. Direct controls on the supply ofservices

ii. Various market-related measures

iii. Improved regulation of competition.

8.3.1.2 Phase 2: Implement preparatoryreforms

These need to focus on the creation of regulatedrisk pools, and changes to the regulation andsubsidisation of the medical schemesenvironment. The objective is to improve thequality and cost-effectiveness of cover withinthe voluntary contributory environment(medical schemes).

The phase 2 reforms serve to enhance thevoluntary contributory environment in order tofacilitate the establishment of a mandatoryenvironment emphasised in phases 3 and 4. Thegreater the degree of cover, and the acceptabilityof the contributory environment, the less thedisruption involved in establishing any futuremandatory environment.

Key reform elements would include the finalimplementation of:

�The risk-equalisation fund (begun inphase 1)

�The risk-adjusted subsidy to medicalschemes (begun in phase 1)

�The state-sponsored medical scheme

�A mandatory environment for civilservants.

8.3.1.3 Phase 3: Implementation of theinitial mandates

Once the preparatory reforms of phase 2 aresubstantially in place, the groundwork would


Consolidated Report

have been established for the implementation ofthe first statutory mandates. Given the incomedistribution in South Africa, the mandatesshould begin with higher income groups. Wherelower income groups are concerned, this phaseshould focus on further active encouragementand development of the voluntary contributoryenvironment.

Phase 2 would have seen the initiation of astate-sponsored medical scheme. Phase 3 shouldfocus on the development of a contributoryscheme for non-medical scheme members in additionto the state-sponsored medical scheme. This willhelp to establish the institutions in Governmentthat would ultimately manage a public sectorcontributory scheme within a NHI framework.Thus two contributory mechanisms will exist:the first based on medical schemes (includingthe state-sponsored medical scheme); and thesecond a dedicated Public Sector ContributoryFund (PSCF). The non-contributory portion ofthe health system would continue to be fundedfrom general taxes.

8.3.1.4 Phase 4: Implementation of an NHI

The last phase envisages the implementation ofa universal contributory system that would, to asubstantial degree, replace general tax funding

as a source of revenue (figure 14). General tax asa supplementary source of revenue maynevertheless prove desirable.

The final phase essentially envisages theestablishment of a contributory environment forall groups and individuals assessed to be in aposition to contribute toward the health system.These contributions would not replace medicalscheme contributions, but rather fund thesubsidy provided to medical schemes. In otherwords, medical scheme contributions would beregarded as a top-up contribution to the subsidy.

All contributions and general tax allocationswould be made directly to a Central Equity Fund(CEF) which would in turn allocate them to thepublic sector and medical schemes based on arisk-adjusted equity formula.

A PSCF would become the national fundingauthority for the public health system. Thiswould either operate as a dedicated unit withinthe national Department of Health, or exist as aseparate parastatal reporting to the Minister ofHealth. Phases 1 through 3 would have seen thecentralisation of the health budget, and theestablishment of capacity to fund provinces viasubstantial improvements in the capacity tomanage and apply the conditional grant system.

Figure 14Institutional framework for a universal contributory system.


Consolidated Report

The end phase of these enhancements wouldsee the creation of the PSCF that would takeresponsibility for, and manage, the allocation offunds from general tax revenues andcontributions allocated through the CEF.

All residents of South Africa should becomeentitled to a subsidy equivalent to the risk-adjusted per capita average of all contributionsand revenue received into the CEF. This subsidysystem should evolve from the reforms in phases1 through 3.

8.4 Strategicfinancial framework

Unlike retirement provision and other forms ofinsurance and social assistance, the health systemcomprises both a financial framework as well asa provider system. The proposed strategicframework involves the development of threerisk-pooling systems.

The first is the universal per capita subsidy,funded from general taxes and enhanced througha redirection of the existing employer tax-subsidy.This system begins as entirely non-contributory

Figure 15Strategic financial framework for the South African health system.

(funded from general taxes � phases 1-3) andconverts to a contributory fund in phase 4.Contributors toward the universal per capitasubsidy can choose to utilise this subsidy throughthe PSCF and obtain an enhanced public sectoramenity, or to subsidise their contributions to amedical scheme. This system becomes the basisfor entrenching income cross-subsidies withinboth the non-contributory and contributoryfinancial systems.

The second major system is the medicalschemes environment. This remains voluntaryfor high income groups for phases 1 and 2, afterwhich it becomes mandatory. The third majorrisk pooling system involves the establishmentof a state-sponsored medical scheme targeted atlow-income groups, the informal sector, andmiddle-income groups who wish to obtain morecost-effective cover (figure 15).

8.5 CoverageCoverage changes over the four general phaseswith the gradual expansion of the contributorysystem (table 9). The public sector basic amenityis the non-contributory environment offered free


Consolidated Report

Table 9Summary of coverage by broad income category

Phase 1 Phase 2 Phase 3 Phase 4Poor o Public sector: o Public sector: o Public sector: o Public sector:

basic amenity basic amenity basic amenity basic amenity(free) (free) (free) (free)

Low-income o Public sector: o Public sector: o Public sector: o Public sector:basic amenity basic amenity basic amenity basic amenity(user fee) (user fee) (free) (free)

o Medical Scheme o Public sector o Public sector(voluntary) contributory contributory

fund fund via NHI(voluntary) contribution

o Medical (mandatory)Scheme o Medical(voluntary) Scheme

(voluntary)

Middle- o Public sector: o Public sector: o Medical o NHIincome basic amenity basic amenity Scheme contribution

(user fee) (user fee) (mandatory) (mandatory)o Medical o Medical o Medical

Scheme Scheme Scheme(voluntary) (voluntary) (mandatory)

High- o Public sector: o Public sector: o Medical o NHIincome basic amenity basic amenity Scheme contribution

(user fee) (user fee) (mandatory) (mandatory)o Medical o Medical o Medical

Scheme Scheme Scheme(voluntary) (voluntary) (mandatory)

to all below a certain income level. Higherincome groups move from a voluntarycontributory environment into mandatoryoptions for both medical scheme membershipand a final NHI contribution.

By phase 3 the user fee system for public hospitalsis eliminated and replaced by a combination ofmandatory medical scheme membership and avoluntary contributory system for an enhanceddifferential amenity. Middle- and upper-incomegroups will be compelled to join a medical schemeduring this phase. Public sector schemes will beable to contract for the differential (enhanced)amenity. Phase 4 creates a mandatory contributoryenvironment that includes low-income groups.From that stage on, low-income contributors willaccess enhanced amenity services.

8.6 Concluding remarksThe various phases outlined in this frameworkreflect the need for careful planning andprioritisation of interventions. The reform

process is complex and multi-dimensional.Significant technical work and consultation willbe required in virtually every phase and step ofthe process. This complexity should berecognised as inherent to health systems� reformand a degree of openness and flexibilitypermitted to fully develop the reforms forimplementation.

It is recommended that the Department ofHealth engage in a consultation process to fullyrefine and develop this framework.


Consolidated Report


Consolidated Report

Chapter 9Retirementand Insurance

9.1 IntroductionThe main social policy objective relating to theprovision of retirement and insurance benefits isto ensure that people have adequate provisionfor their old age and other risks and contingenciesthat may befall them during their financial lifecycle. The institutional policy objective is anintegrated benefit system that ensures that peopleare able to survive hardship and risk irrespectiveof their circumstances or life chances.

South Africa�s private pension and insurancesectors are estimated to be the largest in the worldrelative to gross national product (GNP). Majorachievements have been noted:

�The State Old Age Pension (SOAP) hasreached racial parity.

�The Government Employee Pension Fund(GEPF) is fully funded and managed inaccordance with sound accounting andactuarial practice.

�Trustee legislation has been passed toensure democratic and effectivemanagement.

Given the size of the sector, it is not surprisingthat there is significant diversity within it. Thediversity in itself is to be welcomed, but questionsarise as to the effectiveness of legislation (bothsupervisory and tax related), the interfacebetween private and public institutions andbetween the benefits they pay.

9.1.1 CoveragePension funds account for R600 billion ofinstitutional investor assets, being the majorprovider of capital for the equity listed on the

Johannesburg Stock Exchange. The FinancialServices Board (FSB) 1998 annual report showsthat formal group retirement funds for which theywere able to provide statistics, have almost 11million members (but 1,7 million are retired andthere are a number of duplications).

Pension fund contributions are recorded asamounting to R54,3 billion a year - 14 per centof total personal remuneration in South Africa.Another R8 billion in contributions to retirementannuity funds, administered by insurers can beadded. A large proportion of the R27 billion ofregular premium life assurance is written asendowment policies that are also intended tomature at retirement.

The total accounts for almost 20 per cent ofpersonal remuneration. This is considerablyhigher than necessary if no withdrawals were evermade. The figures do, however, tend tocorroborate the Smith Committee estimates thatsome 80 per cent of formally employed workersare covered by retirement funds.

9.1.2 Proposedstrategic framework

Reforming retirement provision within SouthAfrica requires consideration of a holisticapproach that integrates social assistance, socialinsurance options and private cover (figure 16).

9.2 Findings andrecommendations

The Committee is of the view that the currentdistinction, in the Income Tax Act, betweenpension, provident, and retirement annuity


Consolidated Report

funds should be eliminated. All retirement fundsshould have the characteristics set out below.

9.2.1 Compulsionand preservation

9.2.1.1 Minimum mandatory contributions

In line with the discussion in section 3.6, theCommittee recommends that all peopleemployed in the formal sector (including allcasual and part-time workers) be required tocontribute a prescribed minimum percentage oftheir income for retirement saving. (In line withlife cycle needs, this could be redirected to repaya housing bond if they are under the age of 40.) Anexemption could also be given where the monthlycontributions are so low that administrative costsexceed say 25 per cent of the contributions.

The Committee believes that the SouthAfrican Revenue Service (SARS) should be ableto identify non-contributors, and recommendsthat an appropriate enforcement mechanism beinstituted to ensure that all employers contributeto an approved retirement fund. Provision shouldbe made to take action against employers whofail to comply.

9.2.1.2 Compulsory preservation andunemployment

Compulsory contributions will not have the

desired effect if members continue to withdrawtheir benefits when changing jobs. There ishowever, a need to make the funds available ifthe member becomes unemployed.

The Committee recommends that retirementfunds� rules be amended so that benefits may betransferred on a member�s withdrawal to their newemployer�s fund, or a retirement fund of themember�s choice, if the member remains inemployment in the formal sector. If no longeremployed in the formal sector monthly paymentsa monthly amount, no greater than 60 per cent ofincome before becoming unemployed, should bepayable � once the member is no longer entitled tofurther UIF benefits.

9.2.1.3 Annuitisation and inflation

The Committee recommends that all approvedretirement funds be required, in their rules, toprovide pensions that are intended to compensatefor future inflation, and able to adapt to futurechanges in the rate. The rules must provide a specificmechanism for deciding on pension increases:either a formula or the use of a disinterestedarbitrator. Up to one third of the value of the pensioncould be commuted for a lump sum.

9.2.1.4 Compulsory survivors� and disabilitycover

The Committee recommends that survivor�s and

Figure 16Proposed strategic framework for retirement provision in South Africa.


Consolidated Report

disability cover be mandated as one of the benefitsfrom a retirement fund. The level could be of theorder of pensions of 20% of income for orphans,and 50% of income in the event of disability.

If such compulsory benefits are introduced, astrong argument could be made for theabolishment of the Road Accident Fund and thevarious forms of compulsory coverage foremployment accidents � as they would beredundant. The Committee was divided overwhether such a recommendation could be made,with some members being convinced a publicsystem should remain to provide benefits overand above the minimums suggested.

9.2.2 Administration9.2.2.1 Boards of management

Most boards are now equally divided betweenemployer and employee appointed trustees,whose primary fiduciary duties are to members.Not many trustees are financial or legal experts,and they are largely reliant on expert help for theadministration of their funds and the investmentof assets. This has led to proposals that fundsappoint professional trustees to ensure theirboards have adequate expertise. This has apparentadvantages, but clearly weakens the position ofthe lay trustees in board meetings. If trustees wantindependent advice, they would be advised tocontract others to provide it, rather than surrendersome of their powers.

The Committee recommends that theprinciple of lay trustees, with democraticaccountability to the members, be encouraged.All funds, including those umbrella funds andretirement annuity funds currently exempted,should be required to have elected trustees.Payment to trustees for services rendered shouldnot be permitted.

9.2.2.2 Administration costs

Smaller retirement funds are considerably moreexpensive to administer, and less likely to be ableto afford expert advice.

The Committee recommends that responsibilityfor rationalising the over 15 000 retirement schemesbe left in the hands of trustees. It is furtherrecommended that provincial Governments,together with unions and local businessorganisations and retirement fund administrators,investigate the possibilities of establishing regionalfunds, which will allow for both democratic electionof trustees and greater efficiency.

9.2.2.3 Allocating benefits to dependants

The process of allocating benefits to dependantson death and divorce is problematic from thepoint of view of efficiency and equity.

The Committee recommends that the matterbe given a high priority by the Financial ServicesBoard (FSB). The rights of divorced spousesrequire particular attention.

9.2.2.4 Fiduciary duties

The fiduciary duties of trustees and directors ofinsurance corporations require ongoingmonitoring. Conflicts of interest arise frequentlyin the choice of service providers, of investmentsand in the allocation of benefits. The Committeerecommends that the FSB be required to includein its annual report the steps it has taken to ensurehigher standards are met.

9.2.3 Retirementfund investment

The Committee recommends that Governmentexplore wage-linked bonds as a measure that ispossibly more appropriate to the needs ofretirement funds, and to its own risk managementstrategies.

The Committee recommends that retirementfunds initially be required to invest a proportionof their assets within a defined socially desirableinvestment universe. The initial yield on theseinvestments should be in line with the return onmarketable investments.

The Committee recommends that no furtherincreases in international investments bepermitted until a performance index divided intolocal and foreign exposure, for both foreign andlocally listed equities, is created, and trustees areable to gauge the exposure of the funds to foreigncurrency risks.

Trustees� lack of expertise makes it difficult forthem to judge the appropriateness of investmentperformance benchmarks. The Committeetherefore recommends that the FSB developappropriate investment benchmarks for use byretirement funds, and investment managers berequired to measure their performance againstthese benchmarks.

The Committee recommends that trustees, ortheir delegated agents, be required to apply theirminds to the major issues of corporategovernance such as full accountability toshareholders, the appropriate structure and


Consolidated Report

functioning of the board, and the control ofmanagement remuneration, and to use the voteattached to their shares.

The Committee also recommends that all com-panies be required to amend their articles of asso-ciation in order to achieve representation of signifi-cant minority shareholders (including minoritiesindirectly held through pyramids and cross hold-ings) on their boards of directors. As both Govern-ment and union controlled funds are significantinvestors, they will be entitled to representation ona number of boards. This will therefore boost therepresentivity of boards of directors.

9.2.4 Taxation ofretirement funds

The Committee is of the view that the currenttax structure should be retained. It can howeverbe significantly simplified.

9.2.4.1 Deductibility

The Committee recommends that totaldeductible contributions should be limited to 20per cent of the employee�s taxable income. Theallowance for arrear contributions should bescrapped as representing an unnecessarycomplication.

9.2.4.2 Tax on investment income

The Committee recommends that the taxationof retirement funds be brought within the ambitof the Income Tax Act and that income could betaxed in the same manner as insurance funds.Generous deductions for administrationexpenses could reduce criticism that such a taxfalls heavily on the poor, without openingopportunities for much tax planning.

The current basis of taxation allows for the taxpaid to exceed real earnings. This can be the casewith tax on interest-bearing instruments and withCGT. This significantly undermines the securityof all saving, and views were expressed within

the Committee that the basis should be amendedto prevent this occurring. The opinion wasexpressed that government was obtaininginappropriate tax advice from low inflationOECD countries.

Some members of the Committee felt thatabolishing STC and CGT on realised shareprofits, and increasing the corporate rate tocompensate, could significantly simplify thetaxation of investment income and increase theliquidity of the market. No agreement wasreached on this matter.

9.2.4.3 Tax on benefits

The current tax position encourages lump sumwithdrawals, lump sum death benefits and thecommutation of pensions on retirement. It isinconsistent to prevent people from withdrawingmost of their savings from retirement funds andthen give them tax incentives to withdraw themaximum possible.

Consideration should be given to accordinglump sums with no special status and taxingthem as income in the year in which they arereceived. This approach is simple and wouldmean that people would pay more tax when theychose lump sums over pensions, and is preferredby the Committee.

9.2.4.4 Tax on life assurance

The taxation of life assurance benefits also leadsto a preference for lump sum rather than annuitybenefits.

The Committee recommends that benefitspaid out of the taxed life assurance fund beexplicitly freed from taxation as annuities in thehands of beneficiaries.

9.2.5 Defined benefit vsdefined contribution

The Committee did not wish to take a positionon this issue. It was however concerned that

Table 10Summary statistics on the retirement industry

Funds Members Contributions Assets(000�s) (R�billion)

Self-administered 2 682 2 963 18 987 278

Under-written by Insurers 13 127 6 416 20 339 132

Government 7 1,438 14 888 194

Industrial 8 76 100 1

Total 15 824 10 893 54 314 605


Consolidated Report

greater attention be paid to finding appropriateinvestments for retirement funds. Theinvestment recommendations set out on page 96derive from this concern. It is also concerned thatmembers are fully informed of the present valueof their benefits.

It therefore recommends that all funds berequired to inform their members annually ofthe present value of contributions made byemployers on their individual behalf, and of theincrease (or reduction) in value of theirretirement benefits. It also recommends that anemployer guarantee of the benefits payable fromthe fund or of its investment returns be the subjectof a separate contract between the fund and theemployer.

The Committee�s brief also requires it toconsider question of impact on government asemployer. It is recommended that, if the GEPFwere converted into a DC fund, the assets shouldprimarily be inflation and wage linked bonds. Itwould be inappropriate for government to havetoo large an influence on private sectorbusinesses should the GEPF become a majorshareholder in local companies.

9.2.6 Consumerprotection issues

Powerful vested interests control the insuranceand related industries. There is limitedregulatory capacity to monitor compliance withtrustee laws across 15 000 funds. Consumerprotection is sometimes inadequate, and there islittle real competition. The following suggestionsare made.

9.2.6.1 Excessive sales and lapses, but fewcomplaints

The Committee has not been able to obtaindefinitive statistics, but it appears that about halfthe working population appear to have littlepersonal insurance, while the other half appearover-insured. Thirty per cent of life policies resultin a lapse, causing loss to the policyholder. TheCommittee�s research suggests that mostpolicyholders do not appreciate the losses theysuffer, and believes that the problem lies with thecurrent commission system.

The Committee recommends thatcommission could continue to be payable toagents of an insurer, who make it clear that theyare selling policies, not giving advice. Suchcontracts should either offer a defined set of

insurance benefits, or be investment productswith restricted charges - as with stakeholderpensions in the UK.

9.2.6.2 Competition

The Committee was concerned at the apparentlylow level of competition in the financial servicesmarkets - with four or five groups dominatingthe insurance and banking industries. Evidencefrom the UK was that even their much largerand open markets were not regarded assufficiently competitive. Responsibility forregulating these industries is divided between theFSB, the South African Reserve Bank (SARB)and the Competition Commission. It seems thatthere might be inadequate coordination andinsufficient monitoring of possible abuses ofmarket power.

The Committee recommends that theregulating bodies be required to perform a regularreview of competition within these markets andinvestigate ways of increasing competition.Suggestions were made that life insurancecompanies be compelled to make comparableexpense and performance statistics available. Thepublication of investment performance statisticsalso needs monitoring. In particular, all costsshould be fully disclosed. Other possibilities arethat more encouragement should be given tosmaller mutual societies to develop into realcompetitors in the market.

9.2.6.3 Low-cost national savings scheme

Representations to the Committee have includedsuggestions for a low-cost national savingsscheme that can cater specifically for the growingnumbers of workers excluded from formal,regular employment retirement funds. Such ascheme, preferably administered through a non-profit public entity, would cater specifically forworkers with unstable incomes and irregularcontribution patterns. It is not clear, however, thatsuch a scheme would be able to function at alower cost level than existing savingsarrangements offered by banks and insurers. Itmight well however compete with existingschemes and so promote much neededcompetition in the industry.

The Committee therefore recommends thatsuch a national scheme be investigated. Thisinvestigation should be carried out jointly or inconsultation with the state development financeinstitutions, the FSB, and SARB, DevelopmentBank of South Africa, Department of Labour and


Consolidated Report

other relevant stakeholders.

9.2.6.4 Guarantee fund

Many countries offer some guarantee to smallerinvestors in publicly supervised institutions suchas banks, insurers and retirement funds. TheCommittee recommends that the SARB and theFSB investigate such a scheme.

9.2.6.5 Improved adjudication

Retirement fund members can currentlycomplain to the Life Assurance Ombudsman, thePensions Fund Adjudicator, the Labour Court,the FSB or the High Court � or to a range ofbodies representing professional advisors. ThePensions Fund Adjudicator believes that SouthAfrica requires a single, specialised, well funded,properly staffed Pension Complaints Tribunalwith exclusive jurisdiction over all pension funddisputes arising from whatever quarter and inrelation to all pension fund matters.

The Committee recommends this suggestionbe investigated in line with the Committee�srecommendations in 13.3.6.

9.2.6.6 Unfair contracts

Policyholders often fall prey to small print in theircontracts, and are often deprived of the benefitsor protection that they anticipated. This problemis particularly serious given the relativelyconstrained levels of education and awarenessamong consumers on such matters. TheCommittee believes that the introduction of anUnfair Clauses Act, similar to that introduced inthe UK, would assist in protecting the interestsof consumers. It is recommended that FSBexplore the introduction of such legislation.

9.2.7 Missing beneficiariesIt appears that a large proportion of retirementand life assurance benefits are not claimed. TheCommittee recommends that the FSBcoordinate a national initiative to find the missingbeneficiaries. This should apply to all deferredpensioners, policyholders no longer payingpremiums, and shareholders not receivingdividends. The campaign should includeextensive advertising in all media, and a collationof the records of the different retirement fundsand insurers. Where there is no address, and norecord of the beneficiary in the records of theDepartments of Home Affairs, SocialDevelopment or SARS, then the benefits couldbe legally forfeited to the state. Provision can bemade for funds and insurers to have recourse in

the event of a beneficiary subsequently claimingtheir benefits.

9.2.8 State Old Age PensionThe number of people qualifying for a pensionis expected to increase by some 50% in the next15 years and will increase the annual cost toalmost R30 billion in 2002 terms in the currentpolicy environment. This is because the meanstest is not effective in capturing lump sumproceeds from retirement funds.

The Committee�s recommends recoupingcosts via the income tax system, rather than usingmeans tests. If the recommendations above areaccepted, then taxation revenue will rise bothfrom the increased payment of annuities and thegreater taxation of lump sums. Payments fromretirement funds are easily monitored for taxpurposes.

If the net costs are still too high, considerationwill have to be given to increasing the retirementage in line with the longer life expectancy ofpensioners. For constitutional reasons, this mayhave to involve increasing the retirement age ofwomen to equal that of men.


Consolidated Report


Consolidated Report

Chapter 10Disability

10.1 IntroductionThe 1997 White Paper on an Integrated NationalDisability Strategy (INDS) noted that there is alack of reliable information on disability in SouthAfrica. However, although available data cannotprovide a perfect portrait of disability in SouthAfrica, it can provide a rough sense of the scopeof disability25. Internationally, it has beensuggested that as much as 10 per cent of theworld�s population live with disability. Estimatessuggest that moderately to extensively disabledpeople constitute around 5 per cent of developingcountry populations.

A special disability survey of South Africaconducted in 1998 by the Department of Healthand the Community Agency for Social Enquiry(Case) estimates that approximately 5,9 per centof South Africa�s population is comprised ofdisabled people. This compares with an estimateof 5,2 per cent from the 1995 OctoberHousehold Survey (OHS), 12,8 per cent of theNational Health & Population Development:South African national survey of 1996 and 6,6per cent from the 1996 census, and 3,7 per centfrom the 1999 OHS. The variance of thesemeasurements illustrates the difficulty of making areliable estimate of disability; since the underlyingpopulation characteristics are unlikely to havechanged dramatically since 1995, differences inprevalence estimates mostly reflect differences inthe survey instruments and techniques.

The broad scope and the many socio-economic repercussions of disability make it,by any measure, an important policy issue forSouth Africa. The impact of disability extendswell beyond the disabled themselves. Disability

touches the lives of friends, family, and fellowcommunity members. Indeed, while individualsbear the brunt of illness and impairment,disability also creates hardships for those whocare for and depend on disabled family members.At least 16 per cent of South Africans arethemselves disabled or live in a household witha person who is disabled, according to the 1999OHS � and this is possibly a conservativeestimate.26 A first step in formulating a nationalpolicy framework to address disability is toacknowledge the scope of the problem.

10.2 The demographicsof disability

Disability tends to follow certain patterns withrespect to age, income, and employment. Olderpeople are more likely to have a disability. This isdue in part to increased probability of serioushealth problems and in part to accumulated riskof suffering a disabling accident over the courseof their lifetimes. While the elderly account forthe largest share of the disabled population,children account for a disproportionately smallportion of national disabilities. Disability amongchildren is, however, a major concern. Theimpact of disability on the lifetime outcomes ofchildren can be extraordinarily high. Childrenwith disabilities tend to have lower schoolattendance rates, less education, and ultimately,additional barriers to independent living andengagement with society. According to the 1999OHS, nearly 30 per cent of school-age childrenwith disabilities were not attending school or notattending full time, compared with 10 per centof children without disabilities.


Consolidated Report

Poor individuals make up a disproportionatelylarge share of the disabled population. Disabilitytends to be more common among poor peoplefor two reasons:

�First, poverty increases vulnerability todisability, chiefly through poor nutrition,difficulty accessing adequate basichealthcare, lack of knowledge aboutprevention, and the greater concentrationof poor workers in dangerous jobs.

�Second, disability increases vulnerability topoverty: lower education, discriminationin the labour market (both active andinstitutional), special disability-relatedcosts, and in some cases the need for otherhousehold members to spend time andresources supporting disabled familymembers increases the likelihood thatdisabled people will remain � or become �poor.

The 1999 OHS suggests that while less than2 per cent of individuals living in households withmonthly incomes above R10 000 are categorised asdisabled, the disability rate was more than twice ashigh for individuals living in households withmonthly incomes below R1 200.

Men are slightly more likely to live withdisability after accounting for age effects, probablydue to the dangerous jobs many men hold.Although men may be marginally more likelyto experience disability, the qualitativeexperiences of women with disabilities may bemuch more difficult. The 1997 White Paper onINDS acknowledges that disabled women mustendure the oppression of a patriarchal society, whichcan be magnified for those women who cannotperform even the traditional roles of motherhoodand homemaking. Women who bear disabledchildren may even in some cases be subject to thescorn and rejection of the community.

10.2.1 The imperative to actNot only is the impact of disability widespreadbut it is likely to increase for the foreseeablefuture. This trend has been noted at theinternational level, where violence and ageingpopulations are driving forces. In sub-SaharanAfrica the situation is considerably worse, duein large part to the projected impact of HIV/AIDS on the health and well-being of theregion. Therefore, not only must we recognisethe scope of disability as it affects society, but sotoo must we recognise the urgent need to put in

place measures to address the growing threat ofdisabling disease.

These imperatives to action are not new.Indeed, South Africa has for some timerecognised the need to provide special protectionto people living with disabilities. The foundationfor this protection is established in section 9 ofthe Constitution, which prohibitsdiscrimination on the basis of disability. Thesenotions have found affirmation in the decisionsof the Constitutional Court and in a series ofinternational agreements to which South Africais committed27.

The 1997 White Paper on an INDS provides thebasis for national disability policy:

�Disability affects a significant share of thecountry and places a significant drain onthe human resources of people withimpairments as well as their families.

�The incidence of disability in South Africais set to rise, in line with internationaltrends.

�South Africa has a clear legal commitmentto address disability. This commitment isenshrined in the Constitution, affirmed bythe Constitutional Court and reaffirmedby international agreements to whichSouth Africa is a signatory.

10.3 Findings10.3.1 Conceptualisation

and definition ofdisability and implications

for national policyThe present disability definitions that underpincurrent social security provisioning are forvarious reasons problematic. They are (a) basedentirely on the medical model; (b) areconstructed in such a way as to undermine thepolicy objective of maximising full participationin the world of work by creating a disincentive towork; and (c) do not take into account social andlabour market barriers, as well as broader socialand environmental factors which inhibit labourmarket participation.

10.3.2 Consequencesof disability

Disability bears certain economic consequencesand is closely related to poverty because it makes


Consolidated Report

households more vulnerable to poverty andpoverty-creating shocks and disabled peoplethemselves face additional barriers to education,employment, and access to basic services (liketransport) that conspire to keep them poor. Theopposite is also true, namely that povertyincreases the risk of disability.

It also has social consequences in theawareness and attitudes of the public. Sometimesovertly, sometimes unintentionally, societydiscriminates against people with disabilities.Legal measures, and even fundamentalconstitutional rights, cannot alone redress theprejudices of the public.

10.3.3 Progress and gaps inthe current framework

Progress in the present system is mixed.According to the INDS the present socialsecurity legislative framework, itsadministration and allocation systems tend tobe discriminatory, punitive, insensitive to thespecific needs of people with disabilities,uncoordinated, inadequate and riddled with highlevels of fraud. In defence of the state it must besaid that the State has, however, managed to createand maintain a system of reasonable coveragefor a relatively large group of adults withdisabilities. The provisioning for children withdisabilities is far less successful.

Of most immediate concern to Governmentis the need to improve access to official support,basic services, and also effective means ofadjudication when disabled people are deprivedof these rights. Among the most consistentcomplaints in the social security system are thoseconcerned with the poor administration of publicsupport for disability.

10.4 Integratingdisability measures into

social security10.4.1 Fundamental

principles and legislativeframework

It must be recognised that people with disabilitiesare not a homogenous group, but have a widerange of needs and circumstances that contributeto their well-being and opportunities in life. Evenpersons with similar disability types have

completely different social, financial and physicalenvironments that directly impinge on theircapacity to function at their maximum potential.This must be recognised when designing asensitive and holistic social security system thatattempts to meet the needs of this group.

10.4.1.1 Currently the different pieces oflegislation regarding the various schemes of socialsecurity are fragmented, sometimes contradictory,and make for gaps in provisioning. Hence theattempt to arrive at a comprehensive system mightnecessitate one overriding piece of legislation, suchas a Social Security Act, which would incorporatethe concept of social security for disabled people, itsaims and objectives, as well as highlighting thepurposes and eligibility criteria of each scheme,including the social assistance programmes.

10.4.1.2 There may still be need for separatelegislation and regulations to guide each of theseschemes, but these must be consistent with thefundamental principles embodied in the Act.There needs to be some �linking� andcohesiveness between the different socialinsurances and the legislation. For example, atpresent the Employment Equity Act calls for theemployment of persons with disabilities, whilereintegration into the labour market is not a goalof COIDA. This mix of policy and legislationshould ensure that the guidelines set out in thevarious policies are achieved and enforcedthrough the necessary legislation and regulations.

10.4.1.3 Particular attention should be paidto clear definitions in the Act(s) that can beoperationalised in the regulations, withaccompanying guidelines for theirimplementation.

10.4.1.4 With regard to legislation forchildren, both the Child Care Act and the SocialAssistance Act should provide for social securityfor children with disabilities. A combinedapproach, with the Child Care Act determiningthe rights and the package of benefits, while theSocial Assistance Act incorporates the finerpractical details and regulations, might be usefuland ensure a comprehensive approach, but maybe cumbersome for implementation.

10.4.2 Concept of socialsecurity for disabled people10.4.2.1 Social security systems should be

seen not merely as safety nets and povertyalleviating measures, but also as measures to


Consolidated Report

promote self-sufficiency and independence.

10.4.2.2 Social security should protectsocietal members from and compensate for, thefinancial consequences of a number of socialcontingencies or risks, including thosepreventative and rehabilitative measures. Itshould ultimately aim at societal solidarity, andat the full development, equality and participationof persons with disabilities (UN Committee onEconomic and Social Rights - GeneralComment No. 3. Para 11)

10.4.3 Definition of disability10.4.3.1 In light of the differing definitions

and measurements of disability used in thevarious pieces of legislation regarding socialsecurity, it is suggested that a broad concept ofdisability be used. This could be adapted for morespecific definitions in each scheme, dependentupon the purpose and coverage of each.Obviously the definitions must be�operationalised� in the assessment tools, whichmust accurately translate the concepts within thepurpose into simple and measurable criteria.

10.4.3.2 It must also be stressed that thesystem should not define beneficiaries accordingto the disability, but should rather determineprovisioning in response to need.

10.4.4 Purposesand eligibility criteria

Within the broad concept of social securitymentioned above, there could be specificpurposes of each of the social security measures.For example, the purpose of COIDA could be toprovide for the compensation of an injuredperson or his/her dependants for work-relatedincidents resulting in injury, death, or anoccupation disease, in order to enable their fullrehabilitation, retraining and re-integration intothe labour market.

10.4.4.1 Persons with disabilities, physical,sensory, mental and intellectual, who cannotprovide for their basic needs, should be eligiblefor the disability grant. In addition, it is suggestedthat persons with chronic illnesses, includingHIV/AIDS, should also qualify for the grant.

10.4.4.2 Eligibility should not be based onthe person�s �incapacity� to work, as often theirlack of work is due to the poor economic climateand prejudice in the work place, as opposed to theirphysical or mental inability to perform the job.

10.4.4.3 Eligibility should be determined bya Needs-based Assessment. This should replacethe current means testing.

10.4.5 Assessmentprocedures

10.4.5.1 The assessment procedure shouldencompass a �needs-assessment� whichconsiders not only the type and severity ofdisability or illness, but other social, economic,physical and environmental factors. Persons withthe same disability can have very different needs,depending on all these factors and on the supportstructures and resources available to them.

10.4.5.2 The assessment should also focuson the applicant�s capabilities, rather than onlyon the degree of disability, as well as theirpotential for re-training and re-employment.Relevant training and reintegration measuresshould also form part of the package of socialsecurity.

10.4.5.3 The International Classification ofImpairment, Disability and Handicaps (ICIDH-2), which is currently being developed by theWorld Health Organisation, may be useful inindicating the main categories and indices formeasurement.

10.4.5.4 The assessment form must includeall the disability categories, i.e. physical, mental,sensory and intellectual (currently it onlyincludes physical and mental).

10.4.5.5 An appeal mechanism (such as aReview Tribunal) is necessary for those rejectedapplications, and must consist of relevantintersectoral representatives.

10.4.6 Targeting10.4.6.1 The issue of means testing versus

universal provisioning is complex and representsthe contradiction, or struggle, between thefundamental rights to social security and theavailable resources.

10.4.6.2 Obviously resources are notinfinite and personal or company income taxsystems are exhaustible as sources of financingfor social security systems. Thus efficientallocation of resources to suit the presentingneeds of the population is required. In the face oflimited resources, some form of targetingmeasure, to identify the most in need, is essential.However, this must be viewed within the rights-based framework as stipulated by the South


Consolidated Report

African Constitution and the variousinternational instruments, which stress the basicrights of persons with disabilities to social securityand social assistance, with progressive realisationand within the constraints of available resources.

10.4.6.3 It is suggested that a thoroughsystem of �needs-assessment� as described above,would include analysis of a person�s financialsituation and their need. Some threshold level ofincome, in relation to need, would have to bedetermined. It is suggested that the DisabilitySector and economists undertake this.

10.4.6.4 It is important that the tools oftargeting be sensitive and accurate indetermining �need� versus purely a medicaldiagnosis emphasising categories of disability.

10.4.6.5 This assessment should beundertaken at regular intervals, so as to re-assessthe level of need and to adjust the benefitsaccordingly. Sudden termination of grants isstrongly discouraged. There should be adequatewarning of the gradual �phasing� out of payments.

10.4.7 Benefits10.4.7.1 A system making use of a needs-

based assessment as described above, would thenprovide a sliding scale of benefits, to suit the rangeof presenting needs within available resources.This should incorporate cash transfers and otherindirect forms of social security.

10.4.7.2 The provision of cash transfers is anessential means to alleviate poverty, to smooth theincome cycle, to meet those special needs due tothe disability, and to overcome barriers that manypersons with disabilities face in maximising theirdevelopment and potential. For these reasons thedisability grant and the CDG must be maintainedand kept at their current level, if not increased.

10.4.7.3 The Committee recommendspersons with disabilities and in poverty receive basicincome as a first step in the package of benefits.Thereafter, consideration should be given to theirspecial needs and provision be made in the form of�topping up� in relation to cash benefits, in-kindbenefits and other essential services.

10.4.7.4 It is recommended that the Grant-in-Aid be re-examined and its usefulness andrelevance determined. In addition, a brief analysisof the Department of Social Development�sprovisioning of �personal assistants� should beundertaken. Some scope of choice in personnelby the beneficiary would be advised.

10.4.7.5 Changes in thinking over recentyears have led to a lower emphasis on institutionalcare for those with disabilities. It is widely recognisedthat effectice de-institutionalisation requiresadequate support at a community level. In the caseof residential care, a major barrier to discharge isthat many patients have no effective family linksand, if discharged, would be without shelter. Thisprocess must therefore be properly resourced.

A submission to the Committee from theDepartment of Health expressed concern forthose with mental disabilities particularly. In theirinstance, co-ordination is essential between theDepartment of Health, Labour (which runsprotective and sheltered workshops) and Prisons(as many as 30% of prisoners may suffer frommental illnesses).

10.5 Short-termmeasures

�Suggested immediate amendments to theSocial Assistance Act and regulations andto the Department of SocialDevelopment�s administrative structures.

�Remove the clause of �permanent homecare� for eligibility of the CDG.

�Extend the CDG to children withmoderate disabilities and those in specialschools or day centres.

�Revamp the current medical andassessment forms. Include sensory andintellectual disabilities. The disabilitysector could assist with this process.

�Utilise a multi-disciplinary panel forassessments.

�A disability representative should bepresent on all the boards examiningclaims for insurances.

�Develop clear eligibility criteria andguidelines for assessors.

�Remove the criteria of spouse�s income inthe means test. Only the income of theperson with the disability should bemeasured, not the �household� income.

�Provide free health services to personswith disabilities.

�Establish a review process for cases atregular intervals.

�Establish an appeal mechanism.


Consolidated Report

�Increase the back-pay to 6 months.

�Speed up the time of processing claims forgrants and insurances.

�Educate the public on the social securityavailable to them.

10.6 RecommendationsIn general, policy should stay focused onimproving the preconditions for equalparticipation, even as more general efforts atbolstering the capacity for governance andadministration in the system of social protectionmove forward. The emphasis should be onpromoting independent living, not institutionalcare. Specific and achievable policyrecommendations are needed to promote actionand not just further deliberation.

�Retain existing social assistance disabilitybenefits until such time as income supportmeasures are universally implemented,with the real value of the current disabilitygrant being taken into account as well asother government measures.

�In-kind benefits should be de-linked fromcash benefits. If people with disabilitiestake jobs and become disqualified fromreceiving public income support, theyshould not be required to give up access tonon-cash support. De-linking cash andnon-cash benefits may help to reduce thestrength of the welfare trap around themeans test.

�The present disability definitions are forvarious reasons unacceptable, as they areconstructed in such a way (a) as to serve asa disincentive to work; and (b) that theyoveremphasise capacity for labour marketparticipation, and do not take into accountsocial and labour market barriers, as well asbroader social and environmental factors. Itis, therefore, proposed as a short-termmeasure that the said definitions beamended so as to reflect an interactiveapproach, which takes into account bothmedical condition and social andenvironmental factors.

�It is further proposed that the definitions bewidened in order to include four maincategories of disability, namely physical,mental, sensory and intellectual disability.

�It is also suggested that the ICIDH-2

approach to the definition of disabilitycould be helpful, as it stresses three mainelements, namely: (i) anatomicalmalfunctioning; (ii) the impairment ofnormal human functions as a result of thecondition; and (iii) the question whetherthe condition hamper or impair the abilityto socially integrate.

�An amended definition should serve twomain purposes, namely, firstly, to identifywhether the person concerned is indeedcovered according to the revamped(wider) definition and, secondly, toidentify the appropriate range and levelof transfers (cash or in-kind benefits,goods and/or services) required toaddress the needs of the particular personaccording to the particular kind ofdisability suffered by the individual.

�Provision should be made for thepurchasing of essential assistive devices.Procedures for acquiring simple assistivedevices should be simple and costsshould be subsidised.

�Simplify administrative procedures.Although administration needs greaterresources and capacity building, it shouldalso be an aim of government to reducebureaucratic complexities that furthercomplicate applications for social assistanceand impede access to entitlements.

�Establish a series of quantitativeindicators and benchmarks inconsultation with civil society. Thesebenchmarks can provide measurabletargets for policy as well as standards bywhich civil society can monitor theprogress of reforms.

�It is recommended that the regulatoryenvironment and enforcement of theSocial Assistance Act and its concomitantregulations be adapted in order to complywith the tenets of administrative justiceas required by the Constitution, thestatutory law, and well-founded commonlaw principles. It is further recommendedthat officials responsible for dealing withmatters pertaining to grants, for example,when approving, turning down orreviewing applications, be made aware ofthe obligations on the state and the rightsof applicants in this regard.


Consolidated Report

�The discriminatory elements in theprovisioning of grants and insurancecoverage should be removed. In particularis it recommended that the citizenshiprestriction for purposes of entitlement tosocial assistance grants be removed. All thatshould be required is lawful residence inSouth Africa.

In order to avoid further unnecessaryfragmentation, it is recommended thatadministrative and institutional arrangementsconcerning people with disabilities be includedin a new national framework developed for socialsecurity delivery.


Consolidated Report


Consolidated Report

Chapter 11Road AccidentFund (RAF)

11.1 BackgroundThe RAF�s purpose is to compensate victims forloss or damage wrongfully caused by driving ofmotor vehicles. In turn, the driver of the motorvehicle is indemnified against liabilityincurred for loss or damage wrongfully caused.The RAF has unlimited liability and thereforeall damages proven must be paid by the Fund.However, claims of certain categories ofpassengers are limited against their own driver(to R25 000). The principles regardingapportionment of fault do apply. The RAF istherefore a public compensation/insurancesystem based on fault.

The RAF is financed through a levy on fuelsold. This type of financing is uniqueinternationally. The current system ofcompensation can be described as social benefitswith elements of insurance.

The Committee has considered the followingdifficulties relating to the RAF:

�Some attorneys are said to unnecessarilydelay claims, running up high costs andgrossly overstating claims. Similarly,medical specialists are often said toprepare medico-legal reports exclusivelyfor claimants. Some assessors are said toassist in lodging fraudulent claims and arerendering inflated and false accounts.

�The legislation is found to be verycomplex, with the result that the wholesystem has become extremely legalisticand virtually incomprehensible to theaverage member of public.

�The many confusing and antiquatedprovisions result in real or perceivedunfairness, encouraging expensivelitigation.

�The delictual basis of a claim requiresextensive investigations and these are alsocostly.

�The claims procedure is said to becumbersome, time consuming and veryexpensive to administer. A claimant hasthree years in which to institute a claimagainst the RAF. In the case of a hit-and-run claim, the period is two years. TheRAF then has to conduct an investigationand more often than not it is difficult toobtain documents and evidence.

�The settlement of claims is held to be themost painstaking procedure of it all.

�The high accident rate seems to be one ofthe major causes of the RAF�s desperatefinancial position.

�One of the main reasons for the depletionof available funds is the payment of hugemedical expenses. This benefits inparticular those who make use of theprivate hospital system.

11.2 ConclusionsThe Committee is not in a position to make finalrecommendations concerning future directionsfor the RAF. After consultations with theSatchwell Commission, and a review of some ofthe processes underway, the Committee haschosen to make certain observations that canserve as inputs into more substantive processes.


Consolidated Report

In some countries there is only one accidentcompensation system, covering bothemployment-related and road accidents, andoften also other accidents outside theemployment and road spheres. Where separateschemes exist for occupational injuries anddiseases and road injuries, certain problems canarise:

�The possibility of double compensation

�Victims shopping around for the bestpossible compensation

�Different definitions of quantum ofdamage creates uncertainty

�Double administration costs.

However, in South Africa the long history of twoseparate schemes probably prevents thelikelihood at this stage of one integrated schemefor all instances of compensation as a result ofinjury.

It is, however, clear that some ambiguitiescould be eliminated by, for example, adoptingone model of assessing damages, eliminatingfault as requirement for liability in bothschemes, and creating an integrated computerdata basis as to eliminate double claims. Assuggested in Chapter 9, a case can be made forthe introduction of an integrated system thatoffers life, disability and health insurance coverfor all accidents and diseases.


Consolidated Report


Consolidated Report

Chapter 12Coverage AgainstEmployment Injuriesand Diseases

12.1 BackgroundCompensation for occupational injuries anddiseases is covered by the Compensation forOccupational Injuries and Diseases Act(COIDA). COIDA provides a system of no-fault compensation for employees who areinjured in accidents that arise out of and in thecourse of their employment or who contractoccupational diseases. Fault continues to play arole, however, since an employee is entitled toadditional compensation if he/she can establishthat negligence of the employer (or certaincategories of managers and fellow employees)caused the injury or disease.

The main problems that currently exist as far ashealth and safety measures are concerned are:

�Large numbers of persons are excludedfrom the operation of COIDA, mainlydomestic workers, and those involved innon-standard forms of work - such as theinformally employed, the self-employed,and so-called dependant contractors.

�Labour market (re)integration is not apriority, as little general provision exists inthis regard. Similarly, prevention in thesecases does not seem to receive anyparticular attention from policy-makers.

�Finally, a lack of linkage with other socialinsurance and social assistance schemesleads to duplication of payments (double-dipping), thereby seriously eroding thefinancial soundness of the respectivepublic insurance funds and the sourcefrom which social grants are paid. It does,of course, also serve as a disincentive toaccess or return to the labour market.

12.2 Fragmentedstatutory framework

There are several significant pieces of legislationin South Africa that provide for preventive safetymeasures. Apart from COIDA, these are theOccupational Health and Safety Act (OHSA),the Mines Health and Safety Act, and theOccupational Diseases in Mines and Works Act(ODMWA).

In addition, while COIDA requires employersto contribute to a centralised state fund (contracompulsory insurance policies with privateinsurers), there are two important exceptions. TheRand Mutual Assurance Company Limited thatoperates in the mining industry and the FederatedEmployer�s Mutual Association, that operates inthe building industry, are allowed to perform thesame functions as COIDA.

There are major differences between COIDAand ODMWA as far as benefit structure andentitlements and other matters are concerned.Health and safety standards in mines are alsodealt with differently from the same standardsin other workplaces. The Mine Health andSafety (MHSA) Act, which is enforced by theDepartment of Minerals and Energy, requiresof the owner of every worked mine to ensure,as far as reasonably practicable, that the mine isdesigned, constructed, equipped and operatedin such a way that employees can perform theirwork without endangering the health and safetyof employees or of any other person. TheOccupational Health and Safety Act spells outthe duties of employers and employeesrespectively in other workplaces and makesprovision for a number of offences if the Act is


Consolidated Report

contravened. Major differences exist as far asthese two laws are concerned.

There is a clear need for some degree ofalignment of the different laws, and theirintegration within the broader occupationalhealth and safety, and social securityframework.

12.3 Employeeprotection and interests

It is recommended that appropriate departmentalmeasures be adopted to ensure that COIDA andthe officials employed there sufficientlyappreciate the fact that employer and employeeinterests are to be respected alike. Furthermore,the protection of employee interests requiresthat significant steps be taken to enforcecompliance by employers of their statutoryduties, and that claims processing bestreamlined in order to deal speedily andefficiently with claims by employees and theirdependants.

12.4 Internationalstandards

A set of general principles can be deduced fromthe ILO Conventions passed on this subject:

�Financing of employment injury benefitsmust be by employers

�Periodic payments should be madeavailable rather than lump-sum benefits

�Coverage: The appropriate scheme�sscope must extend to at least half of thenational workforce or 20 per cent ofresidents

�Minimum compensation levels must beprovided for

�Migrant workers must receive equaltreatment.

South Africa has not ratified Convention 121of 1964 (the Employment Injury BenefitsConvention) yet, but is in a position to do so. It isrecommended that steps be taken to effect theratification of this Convention, and that theratification of the other Conventions in this fieldbe seriously considered. By doing so, SouthAfrica will be seen to adhere to basicinternational standards informing policy-making in an area fraught with problems.

12.5 PreventionIt is the view of the Committee that, unlikeoverwhelming precedent in this regard, nocomprehensive strategy has yet been developedto incorporate prevention as part of the overallsystem of employment injury and diseaseprotection. The recommendation made by theReport of the Committee of Inquiry into aNational Health and Safety Council, namelythat prevention policy must be developed as partof a national strategy, is supported. Allcompensation agencies, including the mutualassociations, should participate in developingthis policy.

12.6 ReintegrationCOIDA is not strong on reintegrationmeasures. In contrast with the positionelsewhere, there is no provision in COIDA,which specifically attempts to enforcereintegration measures � such as compulsoryrehabilitation or vocational trainingprogrammes.

It is, therefore, especially in the area ofreintegration measures that the system isextremely deficient. One would have to suggestthat policy-makers should, as a matter of priority,consider the introduction of measures whichwould give effect to the principle of labour marketintegration. Rehabilitation, vocational trainingand, where appropriate, linking entitlement tobenefits payment to participation in suchprogrammes, should serve as minimalmechanisms to attain this goal.

12.7 BenefitsThe recommendation made by the Report ofthe Committee of Inquiry into a NationalHealth and Safety Council is endorsed. Inparticular there is an urgent need for a thoroughinvestigation of benefits provided by thecompensation system. This includes concernsrelated to the type of benefit, the basis for awardingcompensation, and access to benefits.

12.8 ExclusionThe exclusion of domestic workers, the self-employed, and dependent contractors may befound to constitute a violation of section 9(1)of the Constitution (equal protection andbenefits of the law) and section 9(3) (indirect


Consolidated Report

discrimination against, for example black women(as domestic workers) as a particularly vulnerablegroup). An employee would also have a claim interms of the Employment Equity Act of 1998 ifthe employer does not provide all employeeswith adequate protection in the event ofemployment injuries or diseases, especiallybearing the definition of �employee� in the EEAin mind. Its definition of �employment practicesand policies� in terms of which no unfairdiscrimination may take place and in terms ofwhich barriers must be removed, includesliterally all phases, stages and elements of theemployment relationship.

As far as benefits to dependants are concerned,preference is given to a civil law wife at the expenseof an indigenous law wife, a wife according tocustom, and a cohabitant (if the spouse wasmarried to more than one). This isconstitutionally challengeable � also in view ofthe fact that no distinction is made betweenchildren born out of these relationships.

It is suggested that the possibility of voluntaryregistration in terms of COIDA should beconsidered, if compulsory coverage is not foundto be feasible.

12.9 Commutinginjuries

It is recommended that urgent attention be paidto enlarging the scope of accidents coveredunder COIDA so as to include commutinginjuries on a wider basis than presently foreseen.

12.10 Definingaccidents and diseases

Employees need to be made aware of their rightsunder COIDA, in particular as far as claimingin respect of occupational diseases is concerned.The Compensation Fund should be activelyinvolved in broad public awareness campaigns.

12.11 AdministrationThe critical gaps and concerns mainly relate to:

�Responsibility for compensation beingdivided between different bodies withdifferent administrative criteria forassessing claims and making awards,resulting in an inequitable system.

�The administrative backlogs ofcompensation systems in resolvingcompensation claims submitted by and onbehalf of workers have resulted ininefficient compensation service providedby the state, which is prejudicial toworkers affected by an occupationalinjury or disease.

It is suggested that these problems should beaddressed within the framework of developinga comprehensive national occupational healthand safety policy.

In the interim a more efficient administrationof the current compensation system needs tobe established, while indicators for theassessment of progress in this regard have to bedetermined.

12.12 Civil suitsagainst employers

In Jooste v Score Supermarket Trading (Pty) Ltd theConstitutional Court found that section 35 ofCOIDA does not violate the right to equalprotection and benefit of the law in section 9 ofthe Constitution. The question whether or notan employee ought to have retained thecommon law right to claim damages, either overand above or as an alternative to the advantagesconferred by the Act, represents a highlydebatable, controversial and complex matter ofpolicy, according to the Court. The Court statedthat such a contention represents an invitationto the court to make a policy choice under theguise of rationality review, an invitation that thecourt firmly declined.

Another alternative would be to allow tort-based civil claims to be brought in respect ofthe damages not covered in terms of the presentcompensation systems. The currentdispensation operating on a no fault basis wouldthen provide limited benefits, as is the casepresently. In fact, workmen�s compensationschemes by their nature do not provide fullcoverage. In addition thereto, and only if faultcan be established, employees and theirdependants are allowed to sue employersdirectly (also for general damages, such as forpain and suffering). This combination ofworkmen�s compensation and employer (tort)liability is most common in Europe. Of course,if the employer is sued directly, any amount paidout by the compensation system should be


Consolidated Report

deducted from the damages award, as theprinciple should remain that the employee/dependant should not receive more than his/her actual damages.

12.13 Financial issuesAn actuarial assessment of the entirecompensation system should be undertaken toassess the feasibility of increasing (and/orindividualising) employer premiums,abolishing the rebate system in order to improvebenefits awarded to employees and fully coverall costs related to administration of COIDA.

The staff component of COIDA should bestrengthened in order to be able to target non-complying firms sufficiently.

12.14 ConclusionThe Committee recommends that the issuesdiscussed above be dealt with within theframework of a comprehensive nationaloccupation health and safety policy. As withthe earlier discussion of the Road AccidentFund, this should include consideration of thecosts and benefits of amalgamating thedifferent systems in a single comprehensivesystem that offers life, disability and healthinsurance cover to all employees for allaccidents and diseases.


Consolidated Report


Consolidated Report

Chapter 13InstitutionalFramework

13.1 IntroductionThe term �institution� refers to more than anorganisation. Institutions also refer to regularisedpatterns of behaviour or systems of legitimateenforceable rules embedded in social relations.The development of both institutions andorganisational structures is primarily determinedby the existing social, political and economiccontext.

Major shifts become possible, as in the case ofSouth Africa, with changes from undemocraticto democratic forms of Government and throughthe introduction of a Constitution. Such shiftsmay preface the development of new policy,organisations and legislation that previouslycould not evolve due to imbalances in politicaland economic power. Nevertheless, the degreeto which institutions and organisations change,and the manner in which they change, dependson the influence different stakeholders exerciseon policy-making.

Examples occur when programmes ostensiblyfor the poor are starved of adequate funds or theirbudget are consumed by administrators. Evenapparently well-intentioned efforts to support thepoor can be so clouded by ignorance of the realposition of the poor that they create greaterobstacles:

�Means tested grants create poverty trapsthat penalise formal job efforts

�It is possible that support for informalsector and marginally viable productionmean that effort is directed into risky andlow income jobs that never lift people outof poverty

There is no simple solution. The Committeerecommends that a number of mechanisms beutilised:

�The views of vulnerable and marginalgroups should be sought wheneverprogrammes are being evaluated. They canalso be encouraged to participate indecision making, but it must be recognisedthat their representatives will be open toco-option as a result of their vulnerability.

�Formal performance evaluation (fromwithin departments and by externalauditors) should be made publiclyavailable.

�Periodic reviews should be held that seek toincorporate the widest possible range oflocal and international views.

Regarding the substance of institutionaltransformation, there are three importantconsiderations. First, due to the many domesticstructural and capacity imbalances, one cannotreasonably expect markets to lead thistransformation process. Indeed, with povertybeing as persistent and pervasive as it is, buildingthe capacity of the state and its public sectorinstitutions is an essential starting point forinstitutional transformation. Second, the desiredinstitutional structure must follow the chosenstrategic functions and priorities set by the SouthAfrican developmental state. Third, socially andeconomically sustainable social securityarrangements need to seek to increase theintegrity of households and communities to copewith socio-economic challenges, reducedependency on grants, and effectively rationaliseexisting grants.


Consolidated Report

13.2 Challenges13.2.1 Overview

Social security systems typically embody a rangeof organisational and institutional arrangements,each with different operational and regulatorychallenges. The three key arrangements are publicsector provision, social insurance organisations,and regulated private markets (table 11).

13.2.2 Public sectorA central challenge affecting service deliverywithin the public sector is operational inflexibility.Its impact can be felt on everything from humanresource management to procurement. Poorlystructured hierarchies, over-centralised decisionmaking � particularly with respect to basicoperational matters � and the lack of appropriateperformance evaluation and remuneration leadsto poor morale and ultimately poor service delivery.

In addition, organisational capacity needs tobe addressedby in the im plementation of effectiveand disciplinedmanagement process.

13.2.2.1 Budget allocation mechanism

Large complex functions within Governmentrequire careful attention to policy determinationand financing. The processes required toformulate policy may not do so in a manner thatlinks effectively to the budget determination andallocation process. There is therefore a need tobetter link strategic prioritisation, planning andbudgeting processes.

13.2.2.2 Decentralisation of operationaldecision-making

The public sector operates with a large array ofinstitutions directly responsible for service

delivery. Many suffer from chronic under-performance, amongst other things, due to thelack of a clear model for decentralising themanagement of these facilities or institutions.One solution may lie in moving away from thestandard hierarchical civil service model ofreporting and management toward moredecentralised approaches. This involvesintroducing revised governance models,emphasising the role of Boards withcomprehensive responsibilities and chiefexecutive officers with appropriate accountabilityrequirements. Decentralised responsibility forprocurement, human resource management, andfinancial management is important. Suchapproaches do not involve the decentralisationof policy, merely the carrying out of policy.

There is also a need to create optimumopportunities and processes for interaction with,and feedback from, beneficiaries and potentialbeneficiaries. This will promote betteraccountability and hence greater operationalefficiency.

Where these powers have been delegated toindependent Boards, great care must be taken thatthey are independent and representative, and thatmembers have the necessary expertise and timeto fulfil their responsibilities. They mustthemselves be subject to stringent auditingstandards particularly, and be monitored by thecentral department.

13.2.2.3 Targeted services

Services that are targeted at selected vulnerablegroups suffer from the standard problem ofhaving to find and make allocations to thoseeligible for the support. If managed badly, manyeligible people could be excluded from access to

Table 11Summary of key issues affecting the operational and policy effectiveness of

different organisational options within the social security systemPublic sector Social insurance Regulated private marketso Fragmentation in policy o Statutory authority o Strong and independent

making o Independent board with regulatory authoritieso Linkages between policy fiduciary responsibility o Efficient enforcement

making and resource o Chief executive officer mechanismsallocation (accounting officer) o Appropriate corporate

o Efficiency of the budget o Independence from civil governanceallocation mechanism service HR and o Consumer protection

o Decentralisation of procurement o Unfair discriminationoperational decision-making requirements

o Targeting of servicesand grants


Consolidated Report

goods, services, or grants to which they areentitled. Attention therefore has to be paid tostreamlining targeting criteria or the provisionof certain goods and services on a universal basisif the inefficiency of a targeting system is so greatas to prove dysfunctional.

13.2.3 Social insuranceSocial insurance is typically offered via astatutory institution lying outside of theconventional civil service structure. In manycountries this results in a significantimprovement in operational efficiency soessential for direct service delivery to the public.In South Africa certain of the social insuranceinstitutions (RAF, COIDA, UIF) are operatedas if they were part of the civil service structure,despite the fact that these institutions are engagedin specialised functions with a need for a highdegree of operational flexibility.

13.2.4 Regulatedprivate markets

Social security provision in South Africa has, inmany instances, placed significant responsibilityfor financing and delivery onto a regulatedprivate market. This has created a number ofchallenges. These include the requirement forstrong regulatory authorities, streamlinedenforcement mechanisms, regulated governancerequirements, regulation protecting consumersfrom abuse, protections against unfairdiscrimination and requirements for protectingsocial solidarity principles when these areimportant, e.g. healthcare and retirement.Without ensuring that these issues are adequatelyaddressed it is likely that policy objectives will beundermined.

13.2.4.1 Regulatory authorities

Regulatory authorities need to be fullyindependent statutory authorities. It is importantto protect such authorities both frominappropriate Government interference andregulatory capture by private stakeholders. Suchauthorities need good capacity and sufficientscope to offer attractive career opportunities tohigh quality staff.

13.2.4.2 Enforcement

Regulated markets require efficient methods ofenforcement and arbitration by public regulatoryauthorities. Without this the process of ensuringcompliance with legislation or contracts (in the

case of functions outsourced by the public sector),or protecting people from abuse, may become avery slow and arduous process, undermining theeffectiveness of whatever legislation has been putin place. Having to revert to the High Court forlitigation may substantially diminish the speedof enforcement. In this regard, more efficientoptions need to be considered, particularly theestablishment of dedicated social securityadjudication structures and possibly courts.

13.2.4.3 Corporate governancePoor corporate governance will result in fraudand poor organisational decision making. Eachpolicy area that relies on the private sectorrequires a coherent approach to corporategovernance. In the case of pension funds, forexample, elected boards of trustee and theelimination of conflicts of interest are essentialto protect against fraud and corruption. This goestogether with a strong regulatory authority andstreamlined enforcement.

13.2.4.4 Consumer protection

Consumers need a fair degree of protection inareas ranging from defective products to corruptagents and intermediaries. Relying on consumerwatchdogs or following up complaints will beinsufficient where consumers are placed at astructural disadvantage relative to private agents.The whole area of consumer protection needs tohave a clear policy focus where any aspect of thesocial security system relies on the private marketfor delivery.

13.2.4.5 Unfair discrimination

Where an aspect of the social security systemrelies on the private market for delivery,legislation typically has to be introducedenforcing minimum solidarity and cross-subsidisation requirements. Without theseprotections, reliance on the private market willbe undermined through unilateral decisionsmade by the market to exclude certain groupsfrom cover.

13.2.5 Policy co-ordinationSince 1994 the government has made significantstrides in policy co-ordination throughinstitutions such as the cluster of cabinetministers. There are, however, still a number ofpolicy areas in which policy co-ordination islacking. For instance, with regard to retirementand old age, the Department of SocialDevelopment develops policy for old age grants,


Consolidated Report

whereas the National Treasury develops thepolicy for private old age provision. These twoenvironments are not viewed holistically.

The specific areas of social security identified bythe Committee as having no clear over-ridingpolicy responsibility or lead Ministry are:

�Old age and retirement

�Disability

�Maternity benefits and support

13.3 Recommendations13.3.1 Integratedinstitutional and

organisational frameworkThe Committee recommends that considerationbe given to a significant revision of theorganisational framework and institutionsgoverning social security (figure 17). Thisrevision should focus on a clarification of rolesand responsibilities within the following areas:

�Policy determination

�Organisational framework for social

security and protection

�Governance structures for social insurancefunds

�Private sector regulation

�Enforcement.

13.3.2 Socialsecurity board and agency

A revised operational framework is required toensure effective service delivery in both socialassistance and social insurance. The keyrecommendation here is for the implementationof a social security board (reporting to theMinister of Social Development) and agency(reporting to the social security board) tooperationalise various social security functionsoutside of the civil service. Policy developmentand determination will remain with theDepartment of Social Development.

The Committee proposes that considerationbe given to the following functions of the socialsecurity agency (figure 18):

�Social assistance: The agency will have thefunction of managing the non-contributory social assistance fund,

Figure 17Recommended integrated organisational framework.


Consolidated Report

including budget determination and grantadministration.

�Social insurance: The agency willbecome the oversight authority for allsocial insurance funds operating inSouth Africa. This will not extend topolicy control, as this function will restwith the lead ministers responsible forparticular policy areas.

�Intermediary services: The importantinterface between the general public andall areas of the social security system,whether contributory or non-contributory, would become theresponsibility of the social securityagency. The agency may be developedeventually into an intermediary betweenthe general public and relevantGovernment departments (e.g. HomeAffairs) or social assistance and socialinsurance institutions (e.g. UIF,COIDA).

13.3.3 Governance structuresfor social insurance andrelated organisations

The existing social insurance funds require areconsideration of their governance structure toensure their operational efficiency. It isrecommended that new decentralisedgovernance structures be introduced for existingand future social insurance structures. Theyshould ultimately report to the social securityboard (although policy responsibility forparticular funds will remain with designatedministers, e.g. the policy environment for the UIFwill be determined by the Minister of Labour).

13.3.4 SocialProtection Commission

It is recommended that a standing SocialProtection Commission (SPC), representing keystakeholders, be established. The SPC should relateto NEDLAC. The SPC would have a mandate toreview all issues relevant to social protection.

Figure 18Proposed structure of a social security board and agency.


Consolidated Report

This reporting structure is required to preserveits independence in any research, monitoringand investigation it carried out.

In addition to its broad mandate, it isrecommended that the SPC be given specificterms of reference to monitor and review publicand private sector social security institutions andregulatory structures. The specific requirementwould be to monitor the extent to which socialsecurity objectives are either achieved orundermined. Another key area is to develop andmonitor employment policy.

13.3.5 Privatesector regulation

Apart from the regulation of medical schemes,there is very little recognition withinGovernment at present of the need to achievekey social objective through the regulation ofprivate markets as exist in other nations. Thereis also the constant risk of regulatory capture,which requires that Government introducemeasures to guarantee the ongoingindependence of regulatory authorities. It istherefore recommended that the SPC developspecialised capability and have funding tomonitor and evaluate the performance ofregulatory authorities. To preserve theindependence of the regulatory authorities, itis important that this monitoring and evaluationprocess focus specifically on the achievement ofsocial protection policy and consumer protection.

13.3.6 Social securityadjudication mechanism

The present system providing for complaints andappeals against decisions taken by social securityproviders has many shortcomings:

�There is little consistency as differentbodies or officials are called upon to hearcomplaints and appeals in respect ofdifferent parts of the social security system

�Undue delays are the order of the day

�The powers of the courts to deal with thesematters are unsatisfactory

�The normal courts of the country areapparently not specialised enough to dealeffectively with social security matters

�Access to the courts is limited, in particularas far as the indigent are concerned

�Cases are often dealt with on a purely

technical and legalistic basis, with littleregard to broader fairness considerations

�Court proceedings tend to be prohibitivelyexpensive.

One of the guiding principles in devising anappropriate social security adjudication systemis the need to ensure that an institutionalseparation exists between administrativeaccountability, review and revision, and a whollyindependent, substantive system of adjudication.

The Committee recommends that a uniformadjudication system be established to dealconclusively with all social security claims. Itshould, in the first instance, involve anindependent internal review or appeal institution.

It should, in the second place, involve a court(which could be a specialised court) which hasthe power to finally adjudicate all social securitymatters, and that this court has the power todetermine cases on the basis of law and fairness.

The jurisdiction of this court should cover allsocial security claims, whether under the newUIA, the RAFA, the COIDA and all the otherbenefits (including the Social Assistance Act)emanating from the social securitysystem(including claims falling under thejurisdiction of the Pension Funds Adjudicator).


Consolidated Report


Consolidated Report

Chapter 14Financial Frameworkfor ComprehensiveSocial Protection

14.1 IntroductionThere is no single approach to financing ordelivering social security. A range of approachesincluding user charges, earmarked taxation ofone form or another, Government regulationor general tax financing exists. In someinstances, non-contributory social assistance iscombined with contributory social insurance toachieve greater institutional integration.

Mixed financing options for public and semi-public goods and services are not the exclusivedomain of social security. Establishing theseprinciples generically for Government is,however, essential quite aside from itsapplicability to social security. In this chapter ofthe Report consideration is given to:

�The Constitution, particularly where itrefers to aspects of Government finance

�An overview of South Africa�s socialsecurity system and the key issues insocial security financing

�The principles underlying the use orselection of particular forms of taxation,levy or user fee for the social securitysystem as a whole, or individualprogrammes

�Accounting for alternative revenuesources in the national accounts, and theirrelationship to the National RevenueFund

�Financial management issues wheremixed financing options are considered.

14.2 Constitutionalprovisions affecting

social security financingAn important issue for the Committee toconsider was whether any �prohibition� to theuse of earmarked taxes or contributoryfinancing existed in terms of the Constitution.These forms of funding social security are wellestablished internationally and will need to beintroduced more extensively in South Africaover the medium-to long-term.

Section 185 of the Constitution prescribesthat a National Revenue Fund be established,�into which shall be paid all revenues, as maybe defined by an Act of parliament, raised orreceived by the national Government, and fromwhich appropriations shall be made by an Actof parliament, raised or received by the nationalGovernment, and from which appropriationsshall be made by parliament ��

Section 186 stipulates that an annual budgetreflecting the estimates of revenue andexpenditure shall be laid before the NationalAssembly for each financial year. Provision issimilarly made in section 159 as amended, forprovincial revenue funds in each province:

� into which shall be paid all revenueaccruing to the provincial government, and allfinancial allocations � made by the nationalgovernment to such a provincial governmentand to local governments within the provinceof such a provincial government.


Consolidated Report

The Constitution stipulates, in effect, that thenational Government and provincialGovernments should receive revenue into, andmake appropriations from, single general purposefunds. The Constitution lays the basis forconsolidated national and provincial revenue andexpenditure accounts, thereby contributing to thetransparency and the effective accountability tothe legislative authorities of the public finances.

According to the Katz Commission,although the notion of �revenues� in section185 is not defined and the sections of theConstitution dealing with financial mattersare open to various possible interpretations,these provisions appear to bar the extra-budgetary assignment of national orprovincial Government revenues to specialpurpose funds.

The Katz Commission also notes that althoughthe constitutional provisions:

� do not altogether preclude the earmarkingof nationally collected taxes for the(conditional) financing of specific schedule 6services within the budgetary process, it is clearthat such arrangements could only beconsidered as an integral part of the broaderinter-governmental financial framework.(Katz Commission, par.3.5.4).

The Constitution also provides for theimposition by provincial legislatures of usercharges, taxes, levies and duties other thanincome tax, value added tax or other sales taxes,and surcharges on taxes. These may notdiscriminate against citizens who are notresidents of the provinces concerned.

The Committee finds that the Constitutionin no way prohibits the creation of dedicatedtaxes, earmarked funding or for tax and non-tax forms of revenue for public or quasi-publicinstitutions. Choices about revenue or spendingmechanisms fall into the realm of policydetermination and their merits have to beevaluated on a case-by-case basis.

14.3 An overviewof South African social

security financingThis section provides a review of specific issuesand concerns brought to the attention of theCommittee. These came by way of written andoral submissions, and reports made available.

14.3.1 Determination of theallocation to the social

sector functionsThe determination of the social sector allocationis said to be based largely on available funds ratherthan an objective relationship between prioritiesand available resources. The issue in question isnot whether or not financial constraints actuallyexists, but rather whether the constraint is relatedto objective criteria that takes into account socialimpact and relative priorities.

The constitutional obligations imply that aproper budgetary assessment must be carriedout which takes into consideration all theavailable resources of the state, not simply thosein the relevant department�s budget. If resourceconstraints prevent the state from dischargingall of its obligations then it must give priorityto the most vulnerable sections of thecommunity. Further, it must demonstrate thatevery effort has been made to use all of theresources that are its disposition.

14.3.2 Budget prioritisationThe process of prioritising the allocation ofthe Government�s budget is a complexpolitical and institutional exercise. From 1994a substantial change in the process and thepolicy direction of Government occurred.These changes included a number of attemptsto improve budget prioritisation andplanning. Substantial improvements haveoccurred in areas such as transparency andexplicit links between intentions and fiscalresources.

However, residual problems appear toremain. A key concern relates to the fact thatthree social policy areas of major nationalsignificance, social security, health andeducation, are budgeted for at a provincial level.The link between national policy determinationand provincial decision-making is consequentlyweak. This affects the degree to which provincesadhere to national policy, and the extent towhich financial resources are allocatedaccording to national priorities.

The process by which the largest and mostimportant social allocations are determined isboth indirect and fragmented. As aconsequence, changes in global and inter-provincial allocations are not explicitlydetermined.


Consolidated Report

The loss of explicit budgeting control overnational priorities appears to result in particularproblems in the following areas:

�Poverty alleviation

�Social transfers

�Inter-provincial co-ordination andplanning of healthcare services

�The achievement of equity in the physicalallocation of health resources.

14.3.3 Mixedfinancing principles

No clear framework, consistent with generallyaccepted public finance principles, exists inrelation to non-general tax revenue sources offinance. Alternative financing will always existfor goods and services that are quasi public orquasi private in nature. Sources will derive fromlevies, user charges and earmarked taxes of oneform or another. The approach used to priceand operate public entities that are fully orpartially funded on such a basis needs to beconsistent and uniformly applied.

14.3.4 Allocationof the horizontal division

The formula used to divide up the allocationbetween provinces was cited as a problem.Mismatches can occur between the funds madeavailable through the formula and actual needsresulting from mandates established at thenational level. A key example is the socialassistance system, whereby national legislationestablishes the entitlement but the provincialallocation system does not guarantee that fundswill be available.

14.3.5 Medium-termExpenditure Framework

The Medium-term Expenditure Framework(MTEF) is intended to improve budgetplanning. The MTEF process is clearly in itsinfancy in terms of achieving more advancedand complex budget planning and prioritisationobjectives. To date it has assisted in collatingbudgets within fiscal envelopes. These werelargely devoid of relative prioritisation.

14.3.6 Conditionalgrants and special projects

Conditional grants are allocations voted fortransfer to a lower level of Government subject

to certain conditions being met. South Africahas used this fiscal mechanism since theintroduction of the fiscal federal system in 1997for a number of special programmes andspillover problems. The largest system ofconditional grants occurs within the healthsector, with around R6 billion voted to thenational Department of Health for allocationto provinces with supra-regional services andteaching and research activities.

A number of the smaller grants linked topoverty alleviation, HIV/AIDS, hospitalrehabilitation, primary school nutrition, all faceddifficulties at various stages in fully utilising theavailable funds. Problems appear to have beenrelated to the following:

�Onerous application requirements areplaced on accessing the funds.

�Treasury, in advance of adequate planningin departments, often determines theavailability of funds. This results insignificant delays in complying withplanning and tendering requirements andhas an impact on the utilisation anddisbursement of the allocations.

14.3.7 Allocationof the vertical division

According to the 2001 Budget Review thenational provincial and local spheres ofGovernment provide different services, so thatthe allocation of resources between the spheresalso reflect the prioritisation of differentservices.

The Budget Review explains that theConstitution establishes three distinct spheresof Government. It identifies the responsibilitiesof each sphere and requires an inter-Governmental fiscal system that meets theserequirements. According to Treasury, local andprovincial Governments are responsible fordelivering social and municipal services as wellas a range of services that contribute to theeconomic and social well-being of SouthAfricans:

A system of concurrent or jointresponsibilities applies between national andprovincial governments for functions likeschool education, health, welfare, housing,agriculture and urban and rural development.This in practice means that national governmentdetermines policy and regulates compliance,


Consolidated Report

while provincial governments are responsiblefor implementation. Exclusive functions forprovinces include provincial roads and traffic,ambulance services, planning responsibilities,abattoirs, liquor licences etc.

This leaves national government largelyresponsible for policy and regulatoryfunctions over school education, health,welfare, housing and agriculture, resulting insmall budgets for these departments. Onlyeducation has a large budget, but this is fortransfers to institutions of higher education.

Treasury currently sees a division ofresponsibility between national and provincialGovernment whereby national focuses onpolicy while provinces deal with delivery. Thisstrict division of responsibility is a policyconstruct and does not derive from theConstitution. As such the rationality of therelationship between the determination ofnational policy and its ultimate achievement inprovincial allocations needs to be assessed.

14.4 Key issuesin social security

financingIn the design of a social security system fourkey issues around financing need to beconfronted.

Firstly, there is the issue of how much societywants to or should spend on social security. Answersin this regard depend on the values of the societyand the nature of its economic system. This alsorelates to issues of affordability andsustainability. Clearly the question of how muchshould be spent on a social security systemcannot be approached only from a narrowfinancial or fiscal perspective. Such a perspectivedoes, however, enable us to highlight certainkey questions. These relate to the implicationsof reprioritising in favour of social securityspending within the current fiscal framework,the macro-economic implications of a lessconstrained fiscal stance and the micro-economic implications of larger fiscal transfersbased on larger tax revenues.

A second set of issues has to do with choosingbetween the variety of different mechanisms forfinancing social security. A first choice is betweenpublic and private financing, but each routeholds a number of alternatives. Public financing,

for example, can be from general revenuesor through earmarked taxes. A range ofoptions are possible and indeed in operationinternationally. The issue is clearly related toquestions about the appropriate balancebetween public and private provision but alsorelates to the particular role of the state inensuring availability of social security. Taxexpenditures (including tax deductibility ofcontributions) provide a mixture of publicand private financing with Governmentproviding �subsidies� in the event of privateprocurement of social security cover (such asmedical and retirement insurance). Indeedthe whole tax framework with regard to socialsecurity benefits becomes an important issue.

The possible use of earmarked taxes alsoraises the third issue of how social securitybenefits and its financing should fit into thebudgeting process and the inter-governmental systemin the context of fiscal federalism. At what levelshould the function lie and how is budgetingdone given the centralisation of the majorrevenue sources at the national level.

A prominent issue surrounding statefinancing of social security benefits, andretirement provision in particular, is theadvisability or not of pre-funding benefits in contrastto a pay-as-you-go system . While ageingpopulations and projected actuarial deficits instate retirement schemes have led, in manycountries, to arguments for pre-funding as acorrective pre-measure, there are questionsabout whether pre-funding necessarily providesthe appropriate response in South Africa.

14.5 The existingframework for

financing social securityin South Africa

An important consideration for mandating thisinvestigation into the South African socialsecurity system is the fragmented nature of thecurrent system. This fragmentation is mirroredon the financing side.

A recent ILO estimate puts the total socialsecurity expenditure in South Africa in the1998/99 financial year at R147,8 billion. Thiscomprised 22,6 per cent of Gross DomesticProduct (GDP). Social security expenditureincluded in this �social budget� comprises:


Consolidated Report

�Retirement and disability benefits � a totalof R72,7 billion, or just less than 50 percent of social security expenditure. Nearly69 per cent of these benefits were fromprivate, mostly occupational benefitsfinanced from employer and employeecontributions, either voluntary or interms of conditions of employment orunion agreements. About 8 per cent relateto occupational pensions for civil servantsand 17 per cent to means tested old ageand disability grants funded from generalrevenues.

�Unemployment benefits, employmentinjury benefits and road accident benefits� a total of R5,9 billion or 3 per cent ofsocial security. Work injury is financedthrough contributions levied onemployers (industry-specific and risk-rated) and unemployment insurancethrough employee and employercontributions � these are essentiallyearmarked payroll taxes. The RoadAccident Fund (RAF) is financed througha dedicated proportion of the fuel levy.

�Healthcare spending � R51,2 billion or 35per cent of social security expenditure.Forty-five per cent of this expenditure isin the public health sector, financed fromgeneral Government revenues. �Private�expenditure is financed through private oremployer based medical schemes andsome out-of-pocket expenditure.

�Family benefits � R2,3 billion (1,6 percent of social security expenditure), allmeans tested and financed from generalGovernment revenue.

�Other benefits, primarily social assistanceand housing benefits (R12,7 billion), alsofunded from general Governmentrevenue.

Of the above social security expenditure R58,4billion (about 9 per cent of GDP) can be seenas publicly funded, the bulk from generalrevenues (R52,5 billion) in contrast to dedicatedlevies or charges. Private and occupationalfunding then comprises nearly R90 billion.

To the above estimates must be added taxexpenditure related to favourable treatment ofcontributions to retirement annuity funds(estimated at R11 billion per year by the KatzCommission). Employer contributions are

made to medical schemes on behalf ofemployees (estimated to be approximately R7,8billion). These tax expenditures imply largefiscal subsidies to individuals and are regressivein the sense that higher income earners benefitfrom them to a larger extent.

A key issue in the South African debate isthe relatively large role, particularly inretirement provision, of private funding and theabsence of a comprehensive, compulsory firsttier retirement insurance mechanism.

The current fragmentation of funding sourcesalso raises the possibility of generating greaterefficiency through establishing a more integratedsystem financed to a greater extent throughtaxation, either general taxes or earmarked taxes.

14.6 Overall allocationsto social security

For a number of reasons it is difficult to judgethe appropriateness of current levels of socialsecurity expenditure. Not only are internationalcomparisons fraught with difficulties becauseof institutional differences and gaps in the databut it is also difficult to decide which appropriatecountries to use for comparisons.

With regard to social service expenditure as awhole (education, health and welfare) it hasoften been argued that relative to GDP, SouthAfrica compares relatively well, particularlyamong developing countries. Social outcomesare, however, not in line with such high levelsof expenditure because of spending oninappropriate services, insufficient targetingtowards the poor and inefficiency in servicedelivery. Other factors that must also be taken intoaccount are the apartheid backlogs, the lack ofwaged work and overall levels of income poverty.

A similar conclusion may be justified withregard to social security. While 23 per cent ofGDP towards social security expenditure istowards the lower end among OECD countriesit is fairly respectable among developingcountries. Because of the dominance of privateprovision and substantial tax expenditure inSouth Africa, however, benefits are substantiallyskewed towards the wealthy and less subject tocost containment. This calls for careful scrutinyof the current system and the identification ofoptions that will be more redistributive, cost-effective and pro-poor.


Consolidated Report

Given this diagnosis three avenues forreforming social security can be identified:

1. Restructure within the current socialsecurity spending envelope;

2. Reprioritise towards social securitywithin the current spending framework;and

3. Allocating more resources towards socialsecurity and fund it through increasing taxto GDP ratios or relaxing deficit targets.

In many quarters the distribution of publicexpenditure is not seen as the key problem withregard to social security reform. It is oftenargued that self-imposed fiscal constraints(particularly tax to GDP ratios and deficittargets) translate into sub-optimal levels ofpublic spending. Not only, it is argued, willsubstantially increased spending on socialsecurity directly alleviate poverty and improveincome distribution (and so satisfyconstitutional mandates), it will also havedynamic benefits leading to higher growth andincreased socio-economic stability. Wideninginequality and poverty is seen as fuelling socio-economic instability and development prospectsand must therefore be turned around urgently.

On the other hand, proponents of fiscalconstraint maintain that conservative fiscaltargets are indicated by:

�International norms in a globalisingeconomy, in order to ensure positiveexpectations, business confidence andinternational competitiveness.

�The need to reduce indebtedness in orderto avoid a debt trap, to release resourcesfor increased social and economicinvestment and contribute to decliningreal interest rates.

�The fact that given public sector capacity,increased levels of expenditure will notnecessarily translate into improved levelsof service delivery.

�The possibility that increased spendingwill not enhance growth but more likelytranslate into macro-economic imbalanceand inflationary pressures.

Indeed, fiscal restraint is argued not to be anend in itself but motivated by the desire toenhance economic growth and make moreresources available for human and socialdevelopment. While recent South African

growth experience has been disappointing,current 3-year projections indicate animprovement in the fiscal situation with debtreduction and stronger revenue projectionscreating the fiscal space for possible significantreal growth in service expenditures.

Progress in reconciling these differentperspectives will depend to some extent onmore detailed analysis of the macro- and micro-economic implications of alternative fiscalstances and different tax regimes.

14.6.1 The budgetframework and

inter-governmental systemThe category �welfare services� is a concurrentfunction of national and provincialGovernment. Currently national Governmentis mainly responsible for setting overall policyand monitoring implementation whileprovincial Governments deliver the actualservices (primarily grant payments). Provincesalso budget for grant expenditure from theirequitable share of national revenue. Thisequitable share is based on a formula, usingweighted demographic structure, reflectingprimarily the relative demand for social servicesbetween the provinces. The currentarrangement has been criticised on a numberof points:

�Separation between responsibility for policyand for budgeting leads to inappropriateincentives. On the one hand, policy couldbe made without sufficient consideration ofcost implications while, on the otherhand, budget gaming may ensue becausewelfare expenditure is seen as a nationalmandate.

�The current formula is seen as notredistributive enough, particularly in thelight of substantial backlogs in someprovinces.

�The financing mechanism (based on therelative demand for social services and nota set of costed norms) results in unfundedmandates on provincial Governments.

�Given uncertainty about likely take-up ofgrants and the absence of substantialprovincial own revenues, individualprovinces cannot absorb the riskassociated with different possiblescenarios.


Consolidated Report

In addition to the above, provincial budgetprocesses has been criticised in the past fornot establishing sufficient certainty aboutallocations and for not accepting thelegislative force of entitlements to socialgrants. These institutional and process issuesmust clearly be addressed in a reformed socialsecurity system.

14.6.2 Sourcesof financing

As indicated above, current social securitybenefits in South Africa are financed from arange of different sources. Key characteristicsare the large proportion coming from privatefinance and, of public funding, the dominanceof funding from general revenues.

The current balance between public andprivate funding is chiefly due to the large extentof private/occupational provision for retirementand medical cover subsidised via the tax system.This balance would be altered through theintroduction of mandatory first-tier retirementinsurance and/or the introduction of socialhealth insurance in one or other format.

The bulk of state-provided income supportis currently funded from general revenues.General tax-financed benefits (social grants)dwarf the insurance type benefits(unemployment and work accident insuranceand RAF) which are funded from earmarked(payroll) taxes and dedicated taxes.

This situation, coupled to the extensive useof earmarked social security taxes in manycountries, raises the issue about the appropriatebalance between earmarked and generalrevenue funding in South Africa.

A large number of earmarked taxes, levies anduser charges are currently in place in SouthAfrica. Most of these were instituted prior to1994 and serve a number of functions. Someare used to fund regulatory bodies (e.g. theNational Electricity Regulator, IndependentCommunications Authority of South Africa,etc) and industry ombudspersons. Others fundGovernment service delivery andprogrammes. Examples of the latter are theunemployment insurance fund (through anearmarked payroll tax), provision forworkplace accidents and the RAF (throughan earmarked portion of the fuel levy).

The period post-1994 saw the introduction

of some new earmarked taxes as well as levies.The most important example is the skillsdevelopment strategy which is funded througha levy grant system on all private sectoremployers, introduced in 1999. There has alsobeen a consistent trend towards setting uppublic entities at arms length fromGovernment, mainly to regulate a variety ofindustries. In many cases these regulatorybodies are funded through levies on theindustry/consumers. Examples are theCouncil for Medical Schemes and regulatorsin the aviation industry. There has also beenan increase in the use of user charges,particularly through more extensive use of tollroads and creating an enabling environmentfor the charging of fees by schools.

In recent years there has been increasingefforts from Government departments andagencies to secure dedicated sources of fundingoutside the normal Government budgetaryallocations from general revenues.

The merits and demerits of allowingdedicated or earmarked funding sources areagain being raised in the context of theCommittee. Specifically, the funding ofhealthcare through a social health insurance taxand funding a Basic Income Grant to addresspoverty through some dedicated tax have beenraised. The solvency problems of the UIF alsoraise the issue of how and when earmarked taxesshould be adjusted. An additional aspect ofsocial security reform that has been consideredis the gains to be had from consolidation of thecurrent fragmented �social security taxes� andthe extension of this financing mechanism to abroader set of social security benefits.

Arguments against such taxes and levies arethat they fragment and complicate the tax systemand that they allow departments and agenciesto escape the discipline of the budget process.In addition it could be argued that dedicatedfunding sources allow agencies to avoidprioritisation through the budget and politicalprocess. This aspect could be addressed withappropriate changes to the institutionalframework and an integrated policy frameworkwith clear political and financial oversight.Proponents of earmarked taxes point to greaterfunding certainty resulting from dedicatedfunding and a more direct relationship betweenpayments and benefits that may enhance bothequity and efficiency.


Consolidated Report

14.6.3. Net burden offinancing an income grant

In evaluating the costs of a Basic Income Grantfinanced by increasing income taxes or increasesin the value added tax, it was argued in thepresentations to the Committee that one shouldtake note of the difference between �grossburden� and the �net burden� of the incomegrant. For example, although the additional costsof a R100 monthly Basic Income Grant to all(given existing grant obligations) could be R46billion, evidence was given that R22 billion ofthis could be clawed back through the tax system,this means that the net additional funds neededwould amount to a much reduced R24 billion.

Similarly, the Committee was presented withthe argument that, if universal income grantsshould be funded out of increases in the VATrate, the net additional burden of a R100 grantwould be about R13,5 billion.

The Committee took note of these arguments,but since it is not within its brief to considerpotential tax increases, the calculations in the restof the chapter are based on gross additional costs.

14.7 The tax system andsocial security benefits

Reference has previously been made to the taxtreatment of contributions to retirement funding.The tax structure related to retirement provisioncan be characterised as �exempt-tax-tax�, withcontributions being tax exempt and investmentincomes and benefits being taxed. This system ofcontribution deductibility provides a positiveincentive to make private provision for retirement.Other issues related to taxation of retirement thatneed to be addressed are the taxation of investmentand trading income of retirement funds and thecurrent treatment of lump-sum benefits.

Current deductibility of a maximum of two-thirds of medical aid contributions in the hands ofthe employer provides an incentive for privateprovision as well as favouring those with higherincomes disproportionately. The appropriateness ofthis arrangement need to be reviewed along withissues related to the treatment of savings accounts.

14.8 Review of thecurrent tax dispensation

User charges by Government and Government

agencies currently amount to 2 per cent of GDP.Some fees contribute to general Governmentrevenue, some are retained in trading accountsor extra-budgetary entities, and some fees arenot reflected in public sector accounts at all.They include departmental sales, registrationand inspection fees, hospital fees and charges,motor licence fees, payments for research andother sales of research councils, road tolls,university and technikon tuition and residencefees, public school fees, state water schemetariffs and sales of various other Governmententerprises.

Dedicated levies not reflected in national andprovincial budget appropriations currentlyamount to about 1,5 per cent of GDP or about 6per cent of total tax revenue. Internationalaccounting standards count statutory levies asGovernment revenue and their spending aspublic expenditure. The largest of these aresocial security taxes, i.e. unemploymentinsurance contributions and workmen�scompensation levies (R4 billion), fuel leviesfor the RAF (R2 billion) and for thesubsidisation of Sasol and Mossgas (R0,8billion), and regional services councils levieson turnover and payroll for financing localinfrastructure (R3 billion).

Levies also finance various industrial, researchand regulatory bodies, such as the South AfricanTourism Board, Financial Services Board andSouth African Bureau of Standards. Televisionlicence fees are assigned to the South AfricanBroadcasting Corporation. Local property taxesand surpluses on municipal water and electricitytrading accounts are earmarked for municipalservices.

14.9 Review ofprinciples underlying theclassification and use ofmixed financing options

This section reviews the current evaluation andprinciples underlying user-fees, earmarkedtaxes and levies and proposes a way ofproviding guidance on their use within publicpolicy and the overall system of social security.This is done where appropriate with referenceto views expressed by the Katz Commissionand the Treasury Department. Extensive usehas been made of the views of Herber (1975),


Consolidated Report

Fisher (1996) and Gildenhuys (1993) as partof the review.

14.9.1 General taxationBroad-based taxes such as the personal incometaxes such as personal income tax, value-addedtax, and corporate taxes are most appropriate forfinancing those public goods for whichconsiderable difficulty arises in applying theexclusion principle.

Earmarked taxes, user fees, and administrativerevenues are all capable, at least to some extent ofutilising or approximating the exclusionprinciple when used as financing techniques.

According to Herber it should be emphasisedthat the �presence� or �absence� of strong traitsof �publicness,� whether deriving from someexternality or from some other source, doesprovide a logical �tie-in� with the institutionalsector, public or private, which is likely to be themore efficient in influencing the allocation of aparticular economic good. None-the-less, a�case-by-case� approach is still; required.

If �Government� production and/ordistribution of a quasi-public good is considereddesirable, a variety of financing techniques areavailable to the unit of Government providingthe good. Pure public goods, which are notsubject to the exclusion principle, cannot beallocated by the commercial principle.

Quasi-public goods can be allocated accordingto at least a degree by utilising commercialprinciples as their benefits are often partiallysubject to the exclusion principle. Quasi-publicgoods can also be allocated and financed throughgeneral taxation. The choice between generaltaxation and the application of some commercialprinciple as financial allocative techniques isrelevant for quasi-public goods but irrelevant forpure public goods.

The case for general taxation as a means offinancing quasi-public goods rests upon severalrelated points.

�General tax financing is preferred in thoseinstances where the short-run marginalcost of an additional unit of output is verylow or zero and the price elasticity ofdemand of the good is highly elastic. Thelow or zero marginal cost means thatadditional units of the good do notwithdraw resources in any way fromalternative uses. A very inelastic good

would imply little purpose in using price asa rationing technique.

�General tax financing would also bepreferred in the case of quasi-publicgoods where important jointconsumption characteristics exist that maycause a serious supply shortage of thesegoods (e.g. tuition costs).

�General tax financing of a quasi-publicgood would also be preferred in instanceswhere the collection costs of user fees aresubstantial (administration costs). Severeinconvenience to users from the collectionsystem would also be important(consumption disutility).

�General tax finance may also be preferredwhere certain distributional objectives arebeing pursued. For instance, medicalservices or school lunches may not beavailable in adequate quantities to certainlow-income people if they are availableonly on a direct pricing basis.

Under certain conditions, mixed financing usingboth user prices and general tax revenues wouldconstitute the most rational alternative forfinancing quasi-public goods. Under othercircumstances, it may provide negativeallocational and distributional non-neutralities.The use of mixed financing may be rational ifthe good possesses both substantial jointconsumption effects, which benefit the societyas a whole and which remain outside theexclusion principle, and also private benefits. Taxfunds would finance the community or socialbenefits while user prices would finance theindividual or private benefits.

The use of general fund financing to coverthe losses associated with Government pricing,e.g. the case of optimum social output underconditions of decreasing production costs, isrational if it is collectively determined by thecommunity that fiscal means of this sort shouldbe used to redistribute real income byincreasing the allocation of the quasi-publicgood in question. The combined use of generalfund financing and user pricing to financepublic university education meets thisrationale since the benefits of education areboth social and private in nature and, inaddition, improving the education of the pooris an effective means of improving their long-term real income position.


Consolidated Report

The absence of sufficient joint consumptionbenefits to justify tax (or debt) subsidisation ofany loss, or the absence of a sufficientcommunity-approved redistribution objective,would make the mixed financing techniqueirrational. This could be true because privateusers would derive most or all of the benefits fromthe consumption of the good and few, if any, socialbenefits would result. Yet, general tax fundscollected from society as a whole would subsidisepart of the cost of the private consumption. Theresults would be both a redistribution of incomein favour of private consumers of the quasi publicgood as well as an allocational distortion.

Herber also raises the issue about what heterms the size of the �relevant interacting group.�The term �relevant� here suggests theconnotation of �inter-dependent� consumption.Essentially, it asks the question: is the group smallenough to reach a market-type agreement withoutencountering a serious �free-rider� problem?

14.9.2 User chargesUser charges should operate as benefit taxeswith an individual�s charge depending both onbenefit (use) and cost of provision. Theprincipal rule for economic efficiency requiresthat marginal benefit equal marginal cost. Forservices that primarily benefit the directconsumer then, the price charged should equalmarginal cost.

The arguments in favour of user charges arelargely the converse of those used to motivategeneral taxes in the case of quasi-public goods.The argument for use of the commercialprinciple is found in those quasi-public goodswhose economic effects are mostly subject to theexclusion principle. The following issues areimportant considerations:

�The absence of significant positiveexternalities may suggest that user pricingmay be preferable. Significant negativeexternalities may, however, require highuser prices to discourage consumption � tothe extent that the exclusion principle canbe applied to the good.

�Where the cost of collecting user fees islower than general tax administration forthe same revenue yield, user fees arepreferred. (This was the case with primarycare clinics in South Africa until 1996.)

�Distributional goals may be better met

through user fees than tax financing. Thisis particularly important for goods such aselectricity or water where anapproximation of the benefit principle isimportant.

The pricing of user-charges face a number ofalternatives depending upon the nature of thegood and the goals to be pursued. These willinclude:

�Profit maximising pricing

�Average cost pricing

�Marginal cost pricing.

Negative and distributional non-neutralities tendto be reduced as output is expanded towardmarginal cost equals average revenue equality(the social welfare optimal allocation point).However, �isolated� examples of marginal costpricing in a society where imperfect marketsprevail do not necessarily constitute an optimalallocation solution, though in many cases theywould constitute an improvement in allocation.

The strongest case for marginal cost pricingwould likely centre around an importanteconomic good possessing significant jointconsumption characteristics over a large group and/or the presence of decreasing production costs atthe relevant output scales. An administrationproblem of effectively applying the exclusionprinciple may nevertheless still occur. Theimplicit danger remains that pursuit of themarginal cost pricing rule for public-type goods,at a time when it is not being followed generallywithin the economy as a whole, will irrationallyexpand the supply of public versus private goods.Nevertheless, the use of the marginal cost pricingtechnique in the allocation of quasi-public goodsmay be considered, at times, as an acceptable�second-best� solution in a world inextricablyassociated with imperfect market structures, jointconsumption, and externalities.

Use by Government of the profit maximisingprice, as determined by the intersection ofmarginal cost and marginal revenue, would bestserve only the revenue goal unless society needsto reduce consumption of an undesirable good.

Average cost pricing is the preferred alternativeto profit maximising pricing for governmentalpricing of quasi-public goods as it helps to reducenegative allocational and distributional distortions.

Marginal cost pricing appears to be the mostdesirable apart from the problems already mentioned.


Consolidated Report

Gildenhuys draws attention to a commonclassification error in which user charges, leviesand consumer tariffs are confused. In most of theliterature a common term is used to describe allthree, namely user charges. There are in factfundamental differences.

User charges and consumer tariffs, however,have certain common characteristics, namely:

�The absence of compulsion � theirpayment is voluntary because the user orthe consumer has a choice of buying theservices.

�They are both based on the benefit-received principle, which means that theuser charge or consumer tariff which has tobe paid is based on the direct benefit of theservice to the user or consumer; in otherwords there is a direct quid pro quo whichmeans that the user or consumer receivesactual value in services to the amount paid.

�The user charge or consumer tariff isestablished according to the costs fordelivering the service. Consumer tariffs,therefore, are comparable to prices of goodstraded in the private sector, while usercharges are levied to recover additionaloperational (direct) costs incurred onbehalf of a specific user of a service.

Gildenhuys proposes that consumer tariffsshould be used in the case of quasi-collectiveservices to pay for the extra operational costsincurred of quasi collective services. In this waythere can be an efficient and effective allocationof the costs of public services. This also relievesthe pressure on tax revenue because if consumertariffs and user charges are not imposed, the costof all particular and quasi-collective services hasto be met from taxation.

User charges do not cover the full costs ofquasi-collective services. If users were expectedto pay the full cost of collective services theywould in all probability be beyond the means ofthe larger part of the population.

User charges have the advantage of raisingadditional revenue. However, they also serve theimportant function of limiting the misuse ofquasi-collective services and improved equityresulting from direct pricing (i.e. non-users donot subsidise users, e.g. toll roads, foreign users).Furthermore, user charges can register andrecord public demand for services. User chargescan also serve to correct price signals in the

market. This is particularly important inconserving a resource which otherwise wouldappear free.

Fisher points out several general principles ofefficient user charges:

�User charge financing becomes moreattractive as the share of marginal benefitsthat accrues to direct users increases.

�User-charge financing requires that directusers can be easily identified and excluded(at reasonable cost) from consuming theservice unless the charge is paid, assumingthat most of the benefits of a service orfacility go to direct users.

�The efficiency case for user-chargefinancing is stronger when demand ismore price elastic. In the special case of aperfectly inelastic (vertical) demand, pricedoes not matter. No inefficiency wouldresult if consumers underestimate cost.Obviously, the more price elastic demandis, the greater the potential for inefficiencyif consumers do not face true costs.

�Marginal benefits, not total benefits, matterfor the determination of user charges.

�The costs for construction of a publicfacility should be paid by those groups insociety who will benefit directly from theexistence of the facility, which may bedifferent from those who benefit fromusing the facility directly. Two reasons aregiven for this:

�The existence of a facility providesindividuals the option of use in the future,should their demands change.

�Individuals who are not direct users alsomight benefit if the facility generatesspillovers in the form of additionaleconomic activity.

If all residents as well as users should pay all orpart of the long-run production costs of publicfacilities, these charges should be independentof the amount of actual use of the facility. Somecharges could be applied to everyone to coverthat part of the capital cost that benefits all, anddifferent charges could be applied to everyone tocover their share of the capital costs.

In practice, however, standard recommenda-tions on the pricing policy guidelines prove quitedifficult to implement. For instance, in theUnited Kingdom (UK) the following guide-


Consolidated Report

lines were provided in various White Papers theinstructions to public corporations were:

�To balance their accounts, taking one yearwith another over a period of five years,after providing for interest and depreciationat historic cost (1961 White Paper).

�To adopt pricing policies so that �revenuesshould normally cover their accountingcosts in full� and take into account bothshort-run and long-run marginal costs, i.e.costs of producing an additional unit ofoutput (1967 White Paper).

�To take the opportunity cost of capital intoaccount in pricing their output (1978White Paper).

The pricing policies of nationalised industrieshad been based on the following generalprinciples:

�Consumers should pay the true cost of theprovision of goods where it can beidentified.

�Cross-subsidisation between profitable andunprofitable services within the industrymight be permitted in some circumstances.For example, British Rail could use asurplus earned by commuter services tosubsidise loss-making rural services.

�Below cost prices could be charged whenthere was a surplus capacity and areduction in price would stimulatedemand.

�Differential prices might be charged whendemand fluctuated and was heavier atcertain times, so that prices were higher atpeak times and lower at the off-peak time.

�Multi-tariffs could be introduced, listingcharges that were related to the volume ofgoods transported or purchased andcustomers could be offered reducedcharges.

British corporations had found implementationof marginal cost pricing policies difficult. BritishRail could not price some of its services on thebasis of marginal costs. The Post Office estimatedthat marginal cost pricing would have added toits costs as it would have required theestablishment and operation of computer models,and results would not have been substantiallydifferent from those of a pricing policy based onaverage costs.

14.9.3 Consumer tariffsThe primary and only objective of consumertariffs is that they should yield enough incometo pay the full costs of supplying such services toindividual consumers. For this reason it isnecessary to keep full operational accounts foreach particular service. The purpose is for eachconsumer to pay the full cost of each unit of aparticular service consumed. The full costincludes the fixed production cost as well as thevariable operational cost for supplying eachservice unit.

As is the case with the prices of privateconsumer goods, consumer tariffs, therefore, fulfilthe same function as prices of private consumergoods by allocating consumer spending todifferent factors of production. Consumer tariffshave no consumer regulation function, nor aredistribution of wealth function.

Governments are supposed to deliverparticular public services at cost with nodeliberate profit element built into the consumertariff structure � the tariff is supposed to coveronly the per unit of the services supplied. Casualsurpluses or deficits on the operational accounts ofparticular services should be carried forward to thenext financial year so that the surplus or deficit ofthe previous year can be taken into account whenestablishing the new tariff for the ensuing year.

Surpluses and deficits are inevitable as it isimpossible to make absolutely correct budgetforecasts. The proper way of dealing with casualsurpluses and deficits is to establish a tariffstabilisation fund for each particular service andto credit or debit any casual surpluses or deficitsagainst such a fund.

Where local Governments are allowed tocompile their tariffs in such a way as todeliberately induce a surplus, this becomes anindirect tax on particular services with aredistribution of wealth effect which isunacceptable as the benefit-received principle isviolated. There is no moral justification forconsumers of particular services to be taxed inorder to subsidise the users of collective services.It is unfair because not all users are tax payers.These arguments apply in all instances, also atthe central government level where consumertariffs for particular services are pooled with allother revenue in the State Revenue Fund.

To comply with the objective of a consumertariff, an operational account should be


Consolidated Report

established for each particular service, while thestructure of the consumer tariff should be of sucha nature that it covers the expenditure of suchservice without surpluses or deficits.

If certain consumers need to be subsidised forsome acceptable reason, they should be openlyand directly subsidised from the tax revenue.Governments should not hide their socialpolicies behind indirect and invisible subsidiesfrom consumer tariffs.

14.9.4 Nominallevies and sundry revenue

There are some sources of revenue, such asnominal levies and sundry charges, that aresimilar to user charges and sometimes based onthe benefit-received principle. Nominal leviessometimes do not recover the full cost of theservice rendered but, just as in the case of usercharges, imposing nominal levies is an effort torelieve the burden on the ordinary taxpayer bytrying to recover part of the cost from thebeneficiaries of a special service.

Nominal levies partially compensateGovernments for the costs of special servicesrendered on request to identifiable individuals,or for special paperwork, such as extracts fromofficial records and the issue of various certificatesand documents. The services related to nominallevies are not continuous services offered for saleon a regular basis but are delivered sporadicallyat the request of an individual or businessenterprise.

Nominal levies are seldom charged for publicgoods or services in the normal sense of theirmeaning and cannot therefore be regardedconsumer tariffs or user charges.

Sometimes a nominal levy is paid for aprivilege or right granted by Government to anindividual or business enterprise. The enjoymentof such a right may cause expenses for theGovernment and the purpose of a nominal levyis then to recover all or some of these expenses.Various form of licence fee fall into thiscategory.

The fees for trading licences with a regulationfunction should be no more than a nominalamount to cover only the application andregistration costs, because the regulation ofbusinesses is a collective service for the protectionof the public, which should be financed fromtaxation.

Fines, forfeitures and traffic fines are casualsources of revenue which should not bedeliberately budgeted for in order to balance theoperational account of some service.

Administrative revenues are collected by a unitof Government from individuals as part of theperformance of general governmental functions.These governmental functions are primarilyregulatory in nature. It must also provide a certainbasic framework within which private economicactivity will take place. In the performance ofthese and other general functions, Governmentfrequently charges a fee, levies a fine, escheat, orotherwise collects revenue from individuals. Thecorrelation between the payment ofadministrative revenue by the individual isusually broad and imprecise. Only in a generalsense, therefore, may it be said that a quid proquo relationship exists in the case ofadministrative revenues.

14.9.5 Earmarked taxesThere is some disagreement concerning theeconomic efficiency results of earmarked taxes.Some suggest that earmarking tends to reducethe willingness of taxpayers to approveexpenditures on specific public services. Othersargue that earmarking is important as a deviceto generate taxpayer support for the expansionof certain governmental services.

Earmarking may increase allocationalefficiency by insuring more rational individualchoice since, with earmarking, the individual canappraise more closely the relevant costs andbenefits of a particular project. The individual isthus able to adjust the amount consumed of eachpublic good in order to attain his or her mostpreferred consumption position. This is not truein general fund financing which is similar to a�joint-product sale� in the sense that to get onecommodity the consumer must also purchaseanother. The individual consumer of quasi-public goods, in the latter case, is subject to anallocational distortion since his or herindependence of choice is reduced.

General fund financing will tend to attract agreater supply of publicly-supplied economicgoods with elastic demands than will earmarkedfinancing. Thus, when �general fund financing�is used, society receives a greater proportion ofthose economic goods with highly elasticdemands since they are tied in with theacquisition of other goods, and it receives a


Consolidated Report

smaller proportion of those goods which possessless elastic demands.

It is sometimes argued that the social securityprogrammes should be financed from �general�instead of �earmarked� taxes since negativeallocational non-neutrality can result from anexcise tax on the wage alone, but not on capital. Itis claimed, for example, that payroll taxes, byincreasing the cost of hiring new employees,encourage the substitution of capital for laboursince capital is not subject to the payroll tax.Therefore, it is argued that labour will be forcedinto �non-covered� employment while capitalis drawn to employment �covered� under socialsecurity. It is thus concluded that important non-neutrality takes place.

Consideration can be given to the general taxfinancing of social security benefits from theprogressive personal income tax. This wouldmake the financing of social security moreequitable in ability-to-pay terms. This would bea move away from any initial �private insurance�intent for the social security programme andwould recognise, instead, that its benefitsprimarily represent a �redistributional�programme toward lower-income groups ratherthan a precise quid pro quo exchange of costsand benefits. (The schedule of benefits may bedesigned to favour workers whose lifetimeearnings are below average. However, part of thisredistribution could also reflect a transfer fromthe �youthful poor� to the �elderly poor�.) Suchconsiderations may be quite pertinent to state runhealth and unemployment insuranceprogrammes.

The overall focus of the Treasury Departmenton user charges, levies and earmarked taxes islimited to offering definitions and generalpositions on desirability.

The review of the public finance literaturelooks at the issue from a different direction. Itlooks primarily at the relative �publicness� or�privateness� of a good as a basis for assessingoptions with respect to finance and provision.The particular character of the good or servicewill also determine what financing technique isfeasible or available to policy-makers. A user feeor earmarked tax option may be desirable, butexcluded as a policy option due to technicalreasons (collection costs, inability to applyexclusion principle, etc).

The differentiation between user charges andlevies is unclear in the Treasury documentation.

For instance it recommends that �a policypreference [should favour] user charges abovelevies on economic efficiency and equitygrounds� (Department of Finance, 1998). Aslevies are used in very different circumstances touser charges a general preference is inappropriateas the options are not substitutes. A similarinconsistency arises with comments such as:�Levies are a last resort and should only beconsidered where costs are borne by the targetedgroup benefiting from the service to be financed.�

The idea of a �quasi-user charge�, to be usedwhere a �user charge� proves unattainable, isdescribed by the Treasury as �essentially a levy�.Television licence fees are used as an example.This is confusing and probably should beconsidered as a levy without any reference to auser charge. Levies are determined by statute andbenefits are not directly linked to use. The use ofa term such as �quasi user charge� in this contextdoes not serve to improve clarity.

The Treasury does not make the distinctiondrawn by Gildenhuys between �user charges�and �consumer tariffs�. Consumer tariffs wouldbe used in the case of �quasi-collective-services�to pay for the extra operational costs incurred ofsupplying services to individuals. User charges,by contrast, do not cover the full costs of suchservices. This would occur in services where full-cost collective services would exceed theaffordability of the larger part of the population.An example would be primary healthcareservices.

�Consumer tariffs� would therefore seek torecover the full cost of each unit of a particularservice consumed. An example in South Africawould be the billing of private medical schemesby public hospitals. The full cost would includefull production and variable costs. A separateoperational account is appropriate in suchcircumstances.

In the conclusions offered by the TreasuryDepartment the proposal is made that inconsidering levy proposals, �the agency sofunded is accountable to the constituency whobears the charge�. This proposal is too broad andconsequently difficult to apply.

�No distinction is made between a usercharge, a consumer tariff, or anadministrative levy of one form or another.Clearly the circumstances in each of theseinstances would vary so substantially that theterm �accountability� is confusing.


Consolidated Report

�If it is assumed that this is a reference to anadministrative levy and not a user charge,then it is difficult to understand what ismeant by constituency and how this mightpractically be applied to a specific instance.Particularly if the levy has a regulatoryfunction. The levy may be imposed in thebroader interests of the country and beopposed by a particular constituency.Distinguishing between broader andnarrower interests lies in the domain ofGovernment policy.

�Thirdly, accountability to narrowerinterests, if it assumed they can bedisentangled in some democratic manner,could have perverse consequences, e.g. inthe case of regulatory authorities or bodieswhere regulatory capture is facilitated.

There are therefore numerous instances where alevy must be applied and administered in amanner that is accountable to Government in atransparent manner, but for which clearseparation from the paying public is required. Acase-by-case assessment is required. For instance,a distinction should be made between a licencefee paid to a commercial television station andlevies used to finance a regulatory authority.

The Katz Commission suggestion that thedistinction between user charges and generaltaxes on the basis of the benefit received principlecan be blurred as �many taxes can be based onbenefits received� is not correct. The distinctionis typically made between benefits that arereceived in exchange for payment, e.g. anelectricity and water charge, and those that arereceived collectively, e.g. through a defence force.The distinction is not based on the existence of abenefit, but rather on how directly the benefit islinked to the person paying.

The Katz Commission refers to instanceswhere an �impost was merely used as a revenuegeneration device� and as such the �charge ismore akin to a tax�. It should be pointed out thatcrude examples of this, such as �traffic tickets,�will have to be distinguished from a policy whichhas as its objective to make a particular class ofperson pay directly for services which theypreviously received free. This would haveredistributional consequences but would notnecessarily be regarded as a tax. In all instancesGovernment authorities should clearly andtransparently implement policy, making clearwhat its distributional and administrative

objectives are. Poor policy making should be seenas a separate issue from the a priori appropriatenessof a particular form of provision or finance.

Consideration needs to be given to theprinciples required to determine the initialgrounds for the application or non-applicationof a particular form of finance or provision.Where grounds exist, certain pre-requisitesshould then be satisfied before implementation.

Social security systems typically make use ofvariety of funding sources and provider systems.These range from general tax-funding toregulated private markets, with many options in-between. South Africa has a fairly under-developed social security framework, and makesuse of either direct state provision or partiallyregulated private markets. There is a need thereforeto broaden the range of social security instruments,and to modernise existing approaches.

14.10 Genericfinancial frameworkfor social security

Social security systems usually have thefollowing features as reflected in their financialframework (see figure 19):

�Non-contributory: This reflects benefitsprovided directly by the state, either as socialtransfers or benefits in kind. These benefitscan be either means-tested or universal.

�Contributory: These benefits are fundedthrough some form of contribution. Thesecan be voluntary, as in the case of medicalschemes and private retirement, ormandatory where social insurance schemes(UIF, COIDA) are involved. Mandatorycontributions can either provide universalcover (National Health Insurance) or belimited to contributors only (Social HealthInsurance). Voluntary contributions caneither occur in relatively unregulatedenvironment (private short-terminsurance) or in regulated environments(medical schemes in South Africa).

14.10.1 Principlesunderpinning social security

financial managementA clear need has emerged to specify the underlyingpublic finance principles that underpin SouthAfrica�s social security system, both from a technical


Consolidated Report

and a social point of view. The former refers totechnically efficient approaches to managing fundsappropriated and spent differently from theconventional Government budget. The latter refersto redistributional and social solidarity goals. Theoverall objective is to harmonise technicalefficiency with social solidarity requirements, sothat the needs of the one do not undermine the other.

The principles underlying mixed financingoptions should arise logically from the characterof the good or service and the key related policyobjectives. Where variations from the generalbudget approach occur, operational requirementsshould be fairly easy to determine.

14.10.2 RecommendationsMixed financing options, which include usercharges, consumer tariffs, levies, and earmarkedtaxes, are of relevance only in cases of quasi-publicor quasi-private goods and services. Once a goodclearly fits into this range, efficiency gains are verylikely through a move toward a mixed system.

As the grounds for mixed financing optionsrelate primarily to the specific circumstancesand nature of the good or service in question,no predisposition for or against such optionscan exist. For this reason, options should beassessed on a case-by-case basis. The level offunding always remains a policy decision basedon Government priorities.

Once the nature of the good or serviceestablishes that significant efficiency gains areachievable through mixed financing approaches,then a consistent operational environmentshould be established noting the following:

14.10.2.1 Use of alternative revenue sources

In all instances where user charges, consumertariffs, or levies are charged, the relevantinstitution or authority should maintain separateoperational accounts.

14.10.2.2 Financial accountability

Financial accountability should be delegated tothe lowest appropriate level where separateoperational budgets exist.

14.10.2.3 Earmarked taxes

Earmarked taxes should not be considered as analternative to the general budget but rather beused only in specific instances where the quasi-public nature of the good or service requires adirect relationship to be established between thecontributor and the good or service to be provided.Insurance of one form or another and retirementcontributions, where compelled by the state,would fall into this category.

Where earmarked taxes are considered,separate operational budgets are recommendedto ensure consistency between the funds raisedand the entitlements to be funded.

Figure 19Generic financial framework of a social security system.


Consolidated Report

14.10.2.4 Redistributional goalsAs far as possible, specific redistributional goalsshould be achieved through general tax andbudget allocations. It would not be inconsistent,however, for certain redistributional goals to beachieved amongst those paying a dedicated tax.

14.10.2.5 Trading accountsWhere trading accounts are in place, appropriatemechanisms must be established to managesurpluses or deficits through either:

�A stabilisation account used to equalisebalances on a multi-year basis

�An approach for topping up and repaying(i.e. lending) money from the generalbudget, where a surplus or deficit occurs

�The establishment of consistent criteria asto how redistributional elements interactwith the general budget.

14.10.2.6 Governance structuresAn appropriate governance structure must be putin place if funds are managed outside of the usualpublic sector framework.

14.11 Social budgetImportant to the revision of the social securitysystem, is the development of a broaderunderstanding of the inter-relatedness of allareas of social security, whether public, socialinsurance or private. The financial system isessentially a reflection of the institutionalframework of social security. For this reason

social budgets, which measure all of socialsecurity expenditure, and not merely the on-budget items, have become important measuresto evaluate the performance of such spendingwithin a nation.

An evaluation of the South African system ofsocial security shows (estimates for 2001) thatoverall 30,1 per cent of GDP is spent on bothcontributory and non-contributory benefits(including education) in both the public andprivate sectors. Around 12,6 per cent (R112,8billion) occurs within the non-contributorypublic sector environment, while 17,6 per cent(157,8 billion) occurs in the contributoryenvironment (figure 20 and table 12). Most socialsecurity expenditure or contributions occurwithin the largely unregulated privatecontributory environment (by this is meant thatsocial solidarity principles are not protected).

14.11.1 Fiscal capacityand the prioritisation of

social security expenditureAccording to the 2001 Budget Review the totalbudget for 2001/02 is R258,3 billion up fromR255 billion in 2000/01. Overall this allocationimplies a reduction in Government expenditurefrom 28,8 per cent of GDP in 2000/01 to 27,4 percent of GDP by 2003/04. Despite the reducedpercentage of GDP, these allocations involveslight real increases over the period of the MTEF.

The 2001/02 financial year provides for theintroduction of a contingency reserve of R2

Figure 20Social security expenditure in South Africa (estimates for 2001).


Consolidated Report

billion in 2001/02 but increases substantially toR8 billion by 2003/04. This contingency reserveis an unallocated amount.

Although interest payments appear to berising, in real terms they are decreasingsignificantly over the period of the MTEF. Intheory this should improve the capacity forimproved Government services.

Social services receive a declining shareof overall GDP over the MTEF period, from13,3 per cent of GDP in 2000/01 to 12,7 percent in 2003/04. Of all the most significantsectors , this decl ine i s the largestreprioritisation (table 13).

Over the period interest payments declinesignificantly as a percentage of GDP, from 5,3per cent of GDP in 2000/01 to 4,5 per cent in2003/04 (figure 21). If Government expenditurewere to remain a constant proportion of GDP,this in itself would have provided an additional0,9 per cent of GDP for Government services.

Overall consolidated expenditure has alsodeclined as a proportion of GDP, reflecting theposition that increased economic growth shouldnot be translated into additional public goods andservices.

When the national allocation is added to thecontingency reserve, and expressed as a percentageof the overall consolidated expenditure, the nationalallocation rises from 31,7 per cent in 2000/01 to34,8 per cent in 2003/04. This substantially outstripsthe changes in the provincial allocation whichreduces from 46,3 per cent of the total consolidatedexpenditure in 2000/01 to 45,4 per cent in 2003/04.

Of interest is the decision to leave a substantialportion of the gains from the reduced interestpayments unallocated. According to themedium-term budget, all reductions in debtservice costs are effectively added to theContingency Reserve. (The space calculatedhere is an estimate, and could in fact be larger orsmaller depending on what in fact transpires.)

Table 12Financial estimate of the total expenditure

within the South African social security systemTotal Contributory Non-contributory

R� billion 270,6 157,8 112,8% of GDP 30,1% 17,6% 12,6%

Contingency Voluntary Mandatory Means tested Universal

R� billion 270,6 152,4 5,5 54,8 58,0% of GDP 30,1% 17,0% 0,6% 6,1% 6,5%

Not Regulated Social National Social In-kind Social In-kindregulated Insurance Insurance transfers benefits transfers benefits

Total R� billion 270,6 117,4 35,0 3,3 2,1 22,8 32,0 0,0 58,0 % of GDP 30,1% 13,1% 3,9% 0,4% 0,2% 2,5% 3,6% 0,0% 6,5%

Education R� billion 52,8 52,8% of GDP 5,9% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 5,9%

Health R� billion 68,5 0,7 35,0 0,3 0,5 32,0% of GDP 7,6% 0,1% 3,9% 0,0% 0,1% 0,0% 3,6% 0,0% 0,0%

Housing R� billion 5,2 5,2% of GDP 0,6% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,6%

Retirement R� billion 62,5 50,2 12,3% of GDP 7,0% 5,6% 0,0% 0,0% 0,0% 1,4% 0,0% 0,0% 0,0%

Disability R� billion 14,1 8,7 0,3 0,8 4,2% of GDP 1,6% 1,0% 0,0% 0,0% 0,1% 0,5% 0,0% 0,0% 0,0%

Children R� billion 6,4 6,4% of GDP 0,7% 0,0% 0,0% 0,0% 0,0% 0,7% 0,0% 0,0% 0,0%

Adult R� billion 0,0poverty % of GDP 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0%

Unem R� billion 32,4 29,8 2,6-ployed, % of GDP 3,6% 3,3% 0,0% 0,3% 0,0% 0,0% 0,0% 0,0% 0,0%

Survivors R� billion 28,8 27,9 0,1 0,8 % of GDP 3,2% 3,1% 0,0% 0,0% 0,1% 0,0% 0,0% 0,0% 0,0%


Consolidated Report

Table 13Consolidated provincial and national expenditure from

2000/01 to 2003/04 (nominal prices) R� million1997/98 2000/01 2001/02 2002/03 2003/04

R� million Estimated Revised Medium-term estimatesoutcome estimate

Protection services 31 214 40 975 45 778 49 221 52 031

Social services 99 230 116 577 126 242 135 444 144 156

Economic services 18 123 19 589 22 645 24 409 25,574

General Governmentservices and unallocatedexpenditure 17 020 22 276 25 046 26 795 28 760

Interest 38 820 46 186 48 138 49 651 51 022

Subtotal: Votes andstatutory amounts 204 407 245 603 267 849 285 520 301 543

Contingency reserves 0 0 2 523 4 378 8 766

Consolidated expenditure 204 407 245 603 270 372 289 898 310 309

Figure 21Consolidated provincial and national expenditure from 2001/02 to 2004/05.

Overall there is evidence of sufficient fiscal capacityfor improved social sector spending without adversemacro-economic impacts.

The following potential sources of improvedfiscal capacity (expressed in nominal terms) arenoted by the Committee:

�Economic growth: If consolidatedexpenditure was to remain at 28 per cent ofGDP for the MTEF, an additional R3,7billion would be available in the 2002/03financial year, and R6,1 billion in the 2003/04 financial year.

�Reduced debt servicing: Effectively this isthe allocation provided within theContingency Fund. In 2002/03 thisamounts to R4,4 billion, while in 2003/04it rises to R8,8 billion.

�Expenditure increased as a percentage ofGDP: If Government were to increasetargeted expenditure to 29 per cent of GDP afurther R9 billion would be available in2002/03 and R9,7 billion in 2003/04.

Thus, in 2002/03 the combined additional fundswould amount to R8,1 billion and in 2003/04


Consolidated Report

R14,9 billion without any increase in expenditureas a percentage of GDP. This increases to R17,1billion in 2002/03 and R24,5 billion in 2003/04if expenditure is allowed to increase to 29 percent of GDP (figure 22).

14.12 Reform proposalsand their financial

implicationsThis section provides an estimate of the financialimplications of major policy recommendationsof the Committee. The evaluation is provided forthe period 2001-2015. The assessment is providedboth in 2001 prices and expressed as a percentage ofGDP. A summary of the policy areas evaluated

14.12.1 AssumptionsThe following assumptions underpin thefinancial evaluation presented below:

�The policy evaluation occurs over a period2001 to 2015

�GDP growth is assumed to be 2 per centper annum from 2001 to 2005 andthereafter until 2015 to be 3 per cent

�GDP in 2001 is taken as R897,9 billion

�Government expenditure in 2001 is takenas R245,6 billion

�Costs are assumed to be in constant 2001prices.

Due to the complexity and inter-relatedness ofthe policies under review a central scenario hasbeen chosen for illustrative purposes, based bothon recommendations of the Committee as wellas scenarios assessed.

�Education: Here it is assumed thatGovernment expenditure on educationwill remain a constant percentage of GDPconsistent with the current ratio.

�Health: The policy framework provided forin the section above on Generic FinancialFramework for Social Security is modelledhere, reflecting the introduction of low-cost medical schemes, the conversion ofthe current tax rebate to an explicit subsidyfor medical scheme members, the gradualmandating of cover, and the ultimate shiftaway from general tax funding to auniversal contributory system.

�Housing: Here it is assumed thatGovernment expenditure will remain aconstant percentage of GDP consistentwith the current ratio.

�Retirement: The policy framework outlinedin the section above on Reform Proposalsand their Financial Implications is modelledhere, reflecting the shift of funding fromunregulated to regulated and mandatorycover. The removal by 2005 of the means teston the state old age pension is also assumed(funded from the removal of the inequitabletax rebate for people over 65).

Figure 22Fiscal capacity over and above the existing MTEF with

no negative macro-economic implications.


Consolidated Report

�Disability: Here it is assumed that allcurrent areas covering disability willremain a constant proportion of GDP.

�Children: The extension of the CSG inaccordance with the medium- to long-termincome support scenarios examined earlier isassumed here. It is assumed that a gradualextension of the CSG will occur includingolder age categories. Coverage for childrenup to 18 years is assumed to be implementedby 2005 (with a delayed take-up).

�Adult poverty: Scenarios consistent with themedium- to long-term income supportscenarios are provided with a �solidaritygrant� introduced in 2006 (with a delayedtake-up).

�Unemployment: Here it is assumed thatGovernment expenditure onunemployment will remain a constantpercentage of GDP consistent with thecurrent ratio.

�Survivors: Here it is assumed thatGovernment expenditure will remain aconstant percentage of GDP consistentwith the current ratio.

14.12.2 ResultsThe reforms show that overall expenditure onsocial security (both public and private) increasesin real terms from R270,6 billion (30.1 percentof GDP) in 2001 to R427,8 billion (32,4 per centof GDP) in 2015, largely in accordance withgrowth in the economy (figures 23 and 24, andtable 13). Overall there is an additional 2,3 percent of GDP spent on social security, primarilyas a result of expected increases in the cost of privatevoluntary cover for social security benefits.

The most significant change over the periodinvolves the increase in mandatory forms ofcontributory social security cover and universalbenefits offered directly by the government. Theformer primarily involve healthcare and

Table 14Social security expenditure in 2015 if all the maximum scenarios

evaluated by the Committee are implementedTotal Contributory Non-contributory

R� billion 427,8 266,7 161,0

% of GDP 32,4% 20,2% 12,2%

Contingency Voluntary Mandatory Means tested Universal

R� billion 427,8 175,0 91,8 6,2 154,8% of GDP 32,4% 13,3% 7,0% 0,5% 11,7%

Not Regulated Social National Social In-kind Social In-kindregulated Insurance Insurance transfers benefits transfers benefits

Total R� billion 427,8 145,0 30,0 31,1 60,6 6,2 0,0 69,7 85,1 % of GDP 32,4% 11,0% 2,3% 2,4% 4,6% 0,5% 0,0% 5,3% 6,5%

Education R� billion 77,5 77,5% of GDP 5,9% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 5,9%

Health R� billion 88,6 0,0 30,0 0,3 58,2% of GDP 6,7% 0,0% 2,3% 0,0% 4,4% 0,0% 0,0% 0,0% 0,0%

Housing R� billion 7,6 7,6% of GDP 0,6% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,6%

Retirement R� billion 96,1 47,4 26,4 22,3% of GDP 7,3% 3,6% 0,0% 2,0% 0,0% 1,4% 0,0% 1,7% 0,0%

Disability R� billion 20,7 12,8 0,5 1,2 6,2% of GDP 1,6% 1,0% 0,0% 0,0% 0,1% 0,5% 0,0% 0,0% 0,0%

Children R� billion 26,2 26,2% of GDP 2,0% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 2,0% 0,0%

Adult R� billion 21,2 21,2poverty % of GDP 1,6% 0,0% 0,0% 0,0% 0,0% 0,0% 0,0% 1,6% 0,0%

Unemployed R� billion 47,6 43,8 3,8% of GDP 3,6% 3,3% 0,0% 0,3% 0,0% 0,0% 0,0% 0,0% 0,0%

Survivors R� billion 42,3 41,0 0,1 1,2 % of GDP 3,2% 3,1% 0,0% 0,0% 0,1% 0,0% 0,0% 0,0% 0,0%


Consolidated Report

Figure 24Social security policy evaluation for the period 2001 to 2015:

expenditure expressed as a percentage of GDP.

Figure 23Social security policy evaluation for the period 2001 to 2015

expenditure in constant 2001 prices.


Consolidated Report

retirement, while the latter involve the removalof the means test on the state old age pension,and other poverty measures becoming moreuniversal. Overall, when phased in over a longperiod, the strategic reform of the system ofsocial security involves a significant shift towardsocial security funding allocated to funds thatincorporate social solidarity principles with littlenet affect on the fiscal framework.

The reform path assessed is affordable whenseen from a long-term perspective, as allimprovements in the social security systemoccur broadly within current macroeconomicconstraints. No significant changes in theproportion of GDP allocated to social securityare required if these scenarios are implemented.In particular, the implementation of a universalsystem of social assistance grants in key areasbecomes both feasible and affordable.Importantly, the restructured social securitysystem removes gaps in coverage that exist in thepresent, without changing the overall proportionof GDP used.


Consolidated Report


Consolidated Report

15.1 BackgroundAddressing social needs and problems by way ofsocial protection measures within the region isof paramount importance to achieve the goalsand aspirations of economic and socialintegration and co-operation. As starkdifferences in the quality of life are the drivingforce behind human migration, a regionalapproach on social collectivity and equality istherefore in South Africa�s interest. Effectivemeasures should be put in place to ensurecompliance with treaties and the protocols.

A distinct social security paradigm is alsonecessary in order to regulate at a regional levelsocial insurance measures for particularcategories (e.g. the employed) and, with thispurpose in mind, to develop measures of co-ordination in the region. Finally, it may benecessary to develop baseline standards thatapply across the board in the region, but areimplemented with reference to the particularsocio-economic status of each of the membercountries. These baseline standards may rely oneither internationally accepted norms and/orgenerally applicable human rights norms.

Analysis of regional aspects highlights, amongstothers, the extent of the inadequacy social securityarrangements in the region (table 15). Studies inthe region further indicate the apparent failure ofdomestic social security measures to addresspoverty alleviation meaningfully. Social inclusionand participation is not an option for largenumbers and significant categories of people. Thisflows from the fact that most country specific socialsecurity systems cater exclusively for the wholeor part of the formally employed, therebymarginalising the non-employed workforce, the

self-employed, the informally employed, and theunemployed.

In addition, the growing interdependence inthe region, and the more extensive migrationof the region�s workers and residents, suggeststhe need for a common response.

15.2 Regionalimplications

for South AfricaProvisions in South African social assistance andin some social insurance laws distinguishbetween nationals and non-nationals.

It is, therefore, necessary, firstly, to considerthese distinctions between South Africancitizens and citizens of other Southern AfricanDevelopment Community member states.Secondly, to develop a common framework andcharter on social protection and to ensure aconsistent approach is implemented. Thirdly,it will be necessary for South Africa as a SADCmember state to engage actively in promotingthe social protection dimension of SADCintegration and interdependence. Fourthly,active involvement in developing acceptablebaseline standards in the area of social protectionfor the region is required. These standardsshould be implemented with reference to theparticular socio-economic status of each of themember countries, as suggested above.

Finally, it will be necessary for South Africato adopt measures aimed at co-ordinating itssocial security system with those of the otherSADC member states. This can be done eitherbilaterally and/or (preferably) multilaterally.

Chapter 15Social SecurityWithin aRegional Context


Consolidated Report

15.3 FindingsIt is clear from the analyses above that most ofthe social security schemes across SouthernAfrica mainly focus on protecting people whoare employed in the formal sector. Coverage oftargeted populations tends to be narrow,excluding the most vulnerable across the region,in particular those in rural areas, without anyform of social protection. The benefits paid bymany schemes are inadequate to meet basicneeds. In the case of non-contributory schemes,a heavy reliance on general tax revenues strainsGovernment financing, keeping benefits at lowlevels in most countries. Moreover, the socialwelfare schemes (or non-contributory schemes)are still in an embryonic stage, and the numberof beneficiaries has been low as well as thebenefits.

It is also clear that the systems, as is the casewith the underlying socio-economic,

administrative and political profiles of thecountries, are hugely diverse in nature. This, ofcourse, makes it difficult to develop baselinestandards for the region and to adopt measures toco-ordinate the various country social securitysystems. This is exacerbated by the fact that manyof the country systems have not been wellresearched.

Co-ordination of social security is presentlyalmost totally absent in the region. The fewexamples that do exist do not function satisfactorily,while attempts to enter into more comprehensivearrangements still have to bear fruit.

Administrative inertia and inefficiency in thearea of social security delivery are, with somenotable exceptions, apparently major obstacles.And yet it would appear that tailor-madesolutions, for example, relying on NGOs andtraditional authorities, to assist in this regard, havebeen relatively successful.

Bot

Les

Mal

Mau

Moz

Nam

RSA

Swa

Tan

Zam

Zim

Note: X denotes the existence of a scheme.

Table 15Social security systems in Southern Africa, a comparative assessment

Type ofScheme ContingenciesMandatory Old age X X XSavingsschemes(mainly by way Disability X X Xof NationalProvident Death X X XFunds) (beingreplaced)Non- Old age X X X XContributory Disability X X XSchemes Widowhood X

Orphanhood/Children X XWar veterans X X X

Social Insurance Old age X X X X XDisability X X X X XSurvivorship/ X X X X X XDeath MSD UIFSickness X X X

MSD UIFUnemploy- Xment UIFMaternity X X

MSD UIFAdoption X

UIF


Consolidated Report

Most of the countries in the region haveembarked on restructuring processes. Some ofthese have already yielded interesting results,such as a clear transition from national providentfund to pension fund systems and theintroduction of short-term benefits as a first steptowards developing the social security systemholistically. More research and policy analysisis required to understand the role andfunction of informal social securitymechanisms and how they link to the formalsystem in the region as a whole.

The aim of economic integration as is evidentin most of the protocols and the resultantmovement of labour from one SADC countryto another, requires that provisions be adoptedto co-ordinate current social security systems inthe SADC countries. Although the Charter ofFundamental Social Rights contains more directprovisions in this regard, specific standards stillneed to be developed to assist the member statesto fulfil their duties as expected at a regional level.


Consolidated Report

Chapter 16ConclusionSouth Africa faces a continuing challenge ofalleviating poverty and meeting basic needs.The Government has committed itself toreducing inequalities through providing basicservices to poor households. Progress has beenmade in terms of delivering some of theservices. Severe challenges remain, however.There is no income support programme forchildren between 7-18 years, adults between 18-59 years and no general assistance forhouseholds where no one is employed. Over13 million people live below the poverty lineand have no access to social security. As such,South Africa�s social security system is neithercomprehensive nor adequate.

The Committee has found that much of whatwe refer to as �social security� derives from theEuropean concept, which took as its basicassumption that social security would developaround formal sector employment. The reality isthat in the developing world formal sectoremployment may never become the norm that itis in Europe. As a result, the Committee has foundthat the development paths of African economies,and third world countries in general, require afresh look at social protection systems moreappropriate to their environments and needs.

In this regard, the Committee has found thata lack of policy to address income poverty hasbeen a constraining feature of South Africa�ssocio-economic programmes. The Committeetherefore recommends that an appropriate socialsecurity concept for South Africa must prioritisethe needs of people without any incomes, withinsufficient incomes or who are engaged ininformal activities.

In addressing this gap, the Committee hasformulated a comprehensive social protectionsystem that enables the attainment of �positive-sum� policy interventions, rather than narrowpolicy trade-offs between social and economicpolicy objectives. The Committee is of the viewthat by creating greater income security thepoor, who are currently trapped in survivalistand low-income informal work, becomeempowered to risk pursuing higher-returnactivities that can break their cycle of poverty.

In addressing the full spectrum of socialprotection policy, the Committee paidconsiderable attention to matters concerninghealth, retirement, unemployment, children,disability, constitutional and legal requirements,and institutional and financial feasibility.Detailed research analysis has informed theCommittee�s recommendations in these areas.

To address income poverty, and underpin theComprehensive Social Protection framework asa whole, the Committee has recommended acomprehensive and integrated medium- tolong-term framework for income support. TheCommittee is of the view that such aComprehensive Social Protection frameworkwould simultaneously address the constitutionaland socio-economic imperatives throughemphasising two aspects:

�Ensuring that all citizens have a minimumacceptable standard of living through asocial protection package, enabling themto participate and advance in social andeconomic life, thus reducing socio-economic constraints to sustainablegrowth in South Africa.

�Providing people with their socio-economic rights, thus enabling them toenjoy their democratic rights. As suchcomprehensive social protection, byproviding universal coverage, thusembeds an important form of socialcitizenship � and could be seen to form acentral component of the democraticState�s �contract with the people�.

Finally, the Committee recommends that animplementation phase for the ComprehensiveSocial Protection framework be designed withofficials from the various Social ClusterDepartments, a core group of Committeemembers and two international experts.


Consolidated Report

ReferencesAFReC, Public Hearings on the MTEF Figures

for 2001/2002, Submission to the SelectCommittee on Finance, 23 August 2001.

Adams, J. 2000. Quarterly Sectoral Report:Welfare. Budget Brief no. 47. Institute forDemocracy in South Africa. Cape Town.

African National Congress. 1994. TheReconstruction and DevelopmentProgramme: A Policy Framework.Johannesburg.

Ardington, E and Lund F. 1995. Pensions andDevelopment: Social Security asComplementary to Programmes ofReconstruction and Development. In:Development Southern Africa, vol. 12, no.4.

Baber, R. 1998. The Structure of Livelihoodsin South Africa�s Bantustans: Evidencefrom Two Settlements in NorthernProvince. D.Phil Thesis in Economics.University of Oxford.

Bhorat, H, and Hodge, J. 1999. DecomposingShifts in Labour Demand in South Africa.In: South African Journal of Economics,vol. 67, no. 3: 348-380.

Budlender, D. 2000. Human Development. In:May, J. (ed.). Poverty and Inequality inSouth Africa: Meeting the Challenge.David Philip.Cape Town: 97-140.

Butcher, K and Rouse, C. 2000. A Study of WageImpacts of Unions and Industrial Councilsin South Africa. Working Paper no. 442,Industrial Relations Section. PrincetonUniversity.

Carter, M and May, J. 1999a. Poverty, Livelihoodand Class in Rural South Africa. In: WorldDevelopment, vol. 27, no. 1: 1-20.

Case, A and Deaton, A. 1998. Large CashTransfers to the Elderly in South Africa In:Economic Journal, vol. 108, no. 450: 1330-1361.

Cichello, P, Fields, G and Leibbrandt, M. 2000.Which African Workers Have GottenAhead and by How Much? The Story ofKwa-Zulu Natal, South Africa: 1993-98.Unpublished Report, School ofEconomics. University of Cape Town.

Constitutional Court of South Africa(Judgement). 4 October 2000. TheGovernment of the Republic of South Africa etal Vs Irene Grootboom et al.

Department of Finance 1998 PrincipledApproach to User Charges and Levies,Unpublished.

Department of Finance. 2000. Budget Review2000. Department of Finance, SouthAfrica.

Duflo, E. March 2000. Grandmothers andGranddaughters: The Effects of Old AgePension on Child Health in South Africa.Mimeo, Massachusetts Institute ofTechnology.

Financial and Fiscal Commission,Recommendations 2001-2004 MTEFCycle, May 2000.

Fischer, Ronald C. 1996. State and Local PublicFinance, Second Edition, Irwin Chicago.

Gildenhuys, JSH. 1993. Public FinancialManagement, J.L. van Schaik, First Edition.

Herber, Bernard P 1975. Modern PublicFinance, the Study of Public SectorEconomics, Third Edition, Irwin, Illinois.

Katz, M. 1995. Third Interim Report of theCommission of Inquiry into certain aspectsof the Tax Structure of South Africa.Pretoria. Government Printers.

Kingdon, G and Knight, J. April 1999b. LinksBetween Unemployment and Poverty inSouth Africa. Presented at the Conferenceof Poverty in Africa, Centre for the Studyof African Economies. University ofOxford.

Klasen, S and Woolard, I. 1999. Levels, Trendsand Consistency of Employment andUnemployment Figures in South Africa.Development Southern African, vol. 16,no. 1.

Leibbrandt, M, Bhorat, H and Woolard, I. May1999. Understanding ContemporaryHousehold Inequality in South Africa.Development Policy Research Unit,University of Cape Town, DPRU WorkingPapers No.99/25.


Consolidated Report

Le Roux. P. June 2001. The Basic Income Grant.Unpublished paper prepared for theCommittee of Inquiry into aComprehensive Social Security System.Pretoria.

May, J et al. 1998. Poverty and Inequality inSouth Africa. Report prepared for theOffice of the Executive Deputy Presidentand the Inter-Ministerial Committee forPoverty and Inequality. Pretoria.

Meth, C. June 2001. Social Exclusion and SocialProtection. Unpublished paper preparedfor the Committee of Inquiry into aComprehensive Social Security System.Pretoria.

McGrath, M. 1983. The Distribution ofPersonal Income in South Africa inSelected Years Over the Period from 1945to 1980. Unpublished Ph.D. dissertation,University of Natal. Durban.

McGrath, M, Janisch, J and Horner, C. 1997.Redistribution Through the Fiscal Systemin the South African Economy. Paperpresented to the Economics Society ofSouth Africa Conference. Potchefstroom.

Moll, P. 2000. Discrimination is Declining inSouth Africa But Inequality is Not. In:Studies in Economics and Econometrics,vol. 24, no. 3.

Naidoo, R. November 2000. The Concept ofComprehensive Social Security.Unpublished paper prepared for theCommittee of Inquiry into aComprehensive Social Security System.Pretoria.

National Treasury, Budget Review 2001, ISBN:0-621-30598 7, RP: 27/2001.

Nattrass, N. 2000. The Debate aboutUnemployment in the 1990s. In: Studiesin Economics and Econometrics,vol. 24, no. 3.

Nattrass, N and Seekings, J. 1997. Citizenshipand Welfare in South Africa:Deracialisation and Inequality in a Labour-Surplus Economy. Canadian Journal ofAfrican Studies, vo1. 31, no.3: 452-81.

OECD. June 1998. Recent Labour MarketDevelopments and Prospects: SpecialFocus on the Patterns of Employment andJoblessness from a Household Perspective.In: Employment Outlook: 1-30.

Olivier, M et al. 1999. Social Insurance andSocial Assistance: Towards a CoherentApproach. Report for the Department ofWelfare.

Olivier, MP, van Niekerk, R, Kuppan, GY andMpedi, G. June 2001. �The LegalFramework of Social security in SouthAfrica� Unpublished paper prepared for theCommittee of Inquiry into aComprehensive Social Security System.Pretoria.

Olivier, MP. June 2001. ConstitutionalFramework of Social Security in SouthAfrica. Unpublished paper prepared for theCommittee of Inquiry into aComprehensive Social Security System.Pretoria.

Posel, D. 2000. Altruism, Kin Selection andIntra-Family Transfers: Evidence fromRemittance Behaviour in South Africa.Unpublished paper, Princeton University,Princeton. New Jersey.

Standing, G. 1999. Global Labour Flexibility.Seeking Distributive Justice MacMillanPress Ltd. London.

RSA. 1994. South African Statistics: 1994.Government Printer. Pretoria.

Samson, M, et al. June 2001. The Social,Economic and Fiscal Impact of a BasicIncome Grant for South Africa.Unpublished paper prepared for theCommittee of Inquiry into aComprehensive Social Security System.Pretoria.

Schultz, TP and Mwabu, G. February 1998.Wage Premia for Education and Location,by Gender and Race in South Africa.Economic Growth Centre, Yale University,Center Discussion Paper no. 785.

Schultz, TP and Mwabu, G. 1998b. LaborUnions and the Distribution of Wages andEmployment in South Africa. In: Industrialand Labour Relations Review, vol. 51, no.4, pp. 680-703.

Seekings, J. 2000. Visions of Society: PeasantsWorkers and the Unemployed in aChanging South Africa. In: Studies inEconomics and Econometrics, vol. 24, no.3: 53-72.

Simbi, T and Aliber, M. September 2000. TheAgricultural Employment Crisis in South


Consolidated Report

Africa. Paper presented at the TIPS PolicyForum. Muldersdrift.

South Africa: Transformation for HumanDevelopment-2000. South Africa HumanDevelopment Report, United NationsDevelopment Programme. Pretoria.

Standing, G. June 2000. Unemployment andIncome Security. International LabourOrganisation. Geneva.

Statistics South Africa. June 2001. �Survey ofTotal Employment and Earnings�:Statistical Release P071. Pretoria.

Taylor, V. November 2000. Poverty in SouthAfrica. Unpublished paper prepared for theCommittee of Inquiry into aComprehensive Social Security System.Pretoria.

The Constitution of the Republic of SouthAfrica, Act 108 of 1996.

Trotman-Dickenson, DI 1996 Economics of thePublic Sector, First Edition, Macmillan,London.

United Nations. 13-23 February 2000.Enhancing Social Protection and ReducingVulnerability in a Globalising World:Report of the Secretary-General. 39th

session of the Commission for SocialDevelopment. New York.

Van den Heever, A. June 2001. South Africanbudgetary process. Unpublished paperprepared for the Committee of Inquiry intoa Comprehensive Social Security System.Pretoria.

Van der Berg, S. 2000. An Analysis of FiscalIncidence of Social Spending in SouthAfrica, 1993-97. Report of the Departmentof Finance.

Whiteford, A, Posel, D and Kelatwang, T. 1995.A Profile of Poverty, Inequality and HumanDevelopment. Human Sciences ResearchCouncil. Pretoria.

Whiteford, A and Van Seventer, D. 2000.Understanding ContemporaryHousehold Inequality in South Africa.In: Studies in Economics andEconometrics, vol. 24, no. 3.


Consolidated Report

1 Survey of Total Employment and Earnings,Statistics South Africa.

2 This trend was evident and discussed at ameeting of technical experts representingapproximately 15 countries contributing toa Forum on Progressive Governance inStockholm in June2001.

3 Dorrington et al. 2001. The impact of HIV/AIDS on adult mortality in South Africa.Medical Research Council: Tygerberg.September 2001.

4 The Stats SA publication Unemploymentand Employment in South Africa states that32 per cent of African householdscontained no employed people. The totalnumber of African households(5 950 992) comes from Leibbrandt,Woolard and Bhorat (2000, p.49). Theseauthors used the 1995 OHS The totalnumber of households in the Leibbrandtet al piece (8 801 993) accords well withthe 1996 population census number of9 060 000 (Report No. 03-01-12 [1996],p.86. Note that this excludes institutions andhostels. The number of workerless Africanhouseholds in 1999 comes from a reportusing the 1999 OHS. by Michael Samson ofthe Economic Policy Research Institute.There were 3 069 897 such households,2 859 167 of them containing workingadults, and 210 730 containing eitherpensioners only or pensioners and children(skip-generation households).

5 This information is drawn from the SouthAfrica Human Development Report 2000,South Africa Transformation for HumanDevelopment 2000, United NationsDevelopment Programme, Pretoria.

6 USAID, 2001. �Payment for municipalservices� Pretoria.

7 Constitutional Court of South Africa(Judgement). 4 October 2000.

8 Standing 2000, p.9.

9 Giddens 1998, 2000. Giddens (1998,p.100) says that: �Social democrats have toshift the relationship between risk andsecurity in the welfare state, to develop asociety of �responsible risk takers� in the

spheres of Government, businessenterprise and labour markets. People needprotection when things go wrong, but alsothe material and moral capabilities to movethrough major periods of transition in theirlives.�

10 These pitfalls were identified in a WorldBank Study on Best Practices in SocialProtection draft document, April 2001.

11 UNSD, 2000.

12 Based on May et al, 2000.

13 Section 8(1).

14 In terms of section 8(2) a provision in theBill of Rights binds a natural or a juristicperson if, and to the extent that, it isapplicable, taking into account the natureof the right and the nature of any dutyimposed by the rights.

15 Given the distinct constitutional dutiesplaced on the state and organs of state tofulfil, promote, protect and respect socialsecurity rights, and the constitutional rulesof interpretation of these rights, all existinglegislation, as well as common law andcustomary law must be scrutinised andbrought in accordance with the right toaccess to social security, as well as othersocial security-related rights.

16 World Bank, Poverty Reduction Sourcebook.

17 The SALDRU survey takes its name fromthe University of Cape Town�s SouthAfrican Labour and Development ResearchUnit, which undertook a nationalhousehold survey in 1993 in cooperationwith the World Bank.

18 Payment Extraction Report for Pay PeriodApril 2001, SOCPEN system �Department of Social Development, 5April 2001. The figure counts beneficiariesfor the Child Support Grant as the actualnumber of grant recipients, not the numberof children. In March 2001, there were 842892 beneficiaries, receiving grants for 1 084659 children.

19 Finance Minister Trevor Manualacknowledged the State Old Age Pensionsystem as one of Government�s mostimportant poverty alleviation programmes

End notes


Consolidated Report

(Budget Speech 1997/98), a fact which issimilarly recognised in the White Paper(1997): �The number of elderly SouthAfrican beneficiaries has stabilised, withfairly good coverage (80%), but there arestill particular pockets where many eligiblepeople do not get a grant. The impact of agrant income on household income forpeople in poverty is dramatic. The majorityof people in poverty who are not white livein three-generation households, and thegrant is typically turned over for generalfamily use. In 1993, there were 7,7 millionpeople in households that received a stategrant. For black South Africans, eachpensioner�s income helped five otherpeople in the household.� See alsoCOSATU (1996), Ardington & Lund(1995), and Haarmann (2000).

20 Haarmann (2000) summarises the findingsof the task team�s report (Schneider &Marshall, 1998): �The task teamrecommends changing the test by movingfrom assessment of functional capacity onlyto evaluation of a range of needs andeconomic factors and hence developing a�profile of needs� of the applicant. Thisprofile should, besides the medical andfinancial indicators, also include indicatorslike the costs related to the specificdisability, the support mechanisms, and asocio-economic profile of the area andpossible vulnerability to discrimination.The rationale for this recommendation isthe appreciation that each disability createsa range of needs. This is especially the casein the South African situation where othersocial security measures like accessiblehealthcare, re-training, vocationalrehabilitation and transport are largelyabsent. The task team inter aliarecommends the employment of�evaluators� in each district for evaluatingthe needs of people with disabilities, animprovement in the administration andinformation system of the grant and astronger inter-sectoral collaboration of thedifferent departments. Strategies for peoplewith disabilities that were already set outin the White Paper ranged fromimprovement of accessibility to the welfaresystem, to training opportunities, transportand the labour market.

21 Finance Minister Trevor Manuel, BudgetSpeech 2001.

22 See for instance significant researchanalyses on this undertaken by Pieter leRoux (2001).

23 COSATU (1998); Hazelhurst (2000);Samson, Babson, Mac Quene, van Niekerk(2000).

24 These submissions include (and are basedon) information supplied by the ChildHealth Policy Institute, UCT � SpecialFocus on Social Security for Childreninfected and affected by HIV/AIDS�, 2001.

25 Gathering accurate data on disability isnotoriously difficult. This is for manyreasons, but is in large part due to the manycompeting definitions of disability itselfand to practical difficulties with identifyingpeople with disabilities. Definitions ofdisability differ from survey to survey;people answering householdquestionnaires respond to questionsdifferently depending on their personalnotions of disability; the stigma associatedwith disability may cause some people toconceal their status; since the disabilitytends to more prevalent in poorerhouseholds and communities, preciselythose communities where it is mostdifficult to gather data, it can be moredifficult to capture people with disabilitiesin standard household surveys. Suchproblems are common to many efforts atobtaining a quantitative profile of peoplewith disabilities in developing countries

26 This figure is derived from the 1999 OHS,which provides the lowest estimate ofdisability prevalence among availablesurveys. For example, Elwan [1999] p.8reports that evidence from the 1990 UnitedStates census suggested that nearly 29.2 percent per cent of the population had a familymember who was disabled in some respect.

27 These include, among others, the UniversalDeclaration of Human Rights (1948), theAfrican Charter on Human and People�sRights (1981), the United Nations StandardRules on the Equalisation of Opportunitiesfor Persons with Disabilities (1993), andThe Convention on the Rights of the Child.

committee for the comprehensive reform of social … report.pdf · iv transforming the present Œ...

Documents