Social Pensions in Southern Africa in the Twentieth Century

Literature › Works

Stephen Devereux (2007). Social Pensions in Southern Africa in the Twentieth Century. Journal of Southern African Studies, Vol. 33, No. 3. Taylor & Francis.

Type	journal article
Scope	Wider apartheid literature
Method	archival narrative
Period studied	1928–2000
Regions	South Africa, Namibia (South West Africa), Botswana
Themes	origins segregation, bantustans removals influx, inequality poverty unemployment, macro fiscal finance debt, politics ideology elections, transition post apartheid, demography urbanisation
Citation key	devereux2007

Argument

Devereux reviews the political and administrative history of the social (old-age) pension in South Africa and Namibia, with a short account of Botswana. The pension began as a welfare measure for poor whites (1928). Extended to Africans in 1944 at lower rates, it was then used under apartheid as a tool of influx control, homelands policy and, in Namibia, 'winning hearts and minds'. Racially differentiated payments persisted until the 1990s, when equity and antipoverty goals took over. The article closes by arguing that fiscal, administrative and political sustainability depends mainly on political commitment. The case rests on parliamentary debates, statutes, commissions and secondary literature.

Evidence and method

Archival work in the National Library of South Africa (Cape Town) and the Library of Parliament (Windhoek): parliamentary debates, statutes and ordinances, commission reports, government policy papers, newspaper reports, and interviews with Namibians (e.g. a former Owambo Administration paymaster). Also secondary literature (Iliffe, Sagner, Posel, Mamdani, Case and Deaton) and World Bank, ILO and SAIRR data and fact papers.

Place in the debates

• Why the social pension was extended to Africans and why payment rates were racially differentiated (see Living standards, health and social policy). Rejects a purely welfarist reading. Argues the pension served instrumental apartheid aims (de-urbanisation, influx control, homelands resettlement, co-opting elites), though it also reflected Hofmeyr's liberal equity concerns and Smuts' 'enlightened self-interest'.
• Why the pension was extended to black Namibians in 1973 (see Living standards, health and social policy). Weighs two explanations: bolstering separate development/influx control and 'winning hearts and minds'. The regional coverage data cast doubt on the second (northern, SWAPO-leaning regions had the lowest coverage), and the motive is undocumented.
• Theories of welfare-state emergence applied to South Africa (see Living standards, health and social policy). Favours diffusion theory over the convergence hypothesis, with elements of welfare industrialisation and interest-group theory, to explain the 1920s pension for poor whites.
• Affordability and sustainability of social pensions (see Living standards, health and social policy). Argues sustainability is as much political as fiscal. He cites Beattie and McGillivray that there is 'no economic law' preventing greater old-age spending, and warns that means testing risks eroding middle-class political support.

Findings

• The 1928 Act gave pensions only to whites and coloureds; Africans and Indians, and all residents of South West Africa, were excluded, partly on the view that 'native custom' provided for dependants. (p. 4)

  Significant omissions from the Pensions Act were 'natives', 'Indians' and all residents of Namibia

• The 1944 extension to Africans was tiered by city, town and rural residence, and designed to discourage migration to cities for pensions. (p. 6)

  'These tests were designed to keep people from coming to urban areas to get a pension'.

• Pension policy was explicitly used as a lever of influx control and homelands policy; in 1965 the three African rates were merged into one to remove an incentive to urbanise. (p. 7)

  'the object of introducing a uniform rate was to avoid providing an incentive to move to cities'.

• The nominal value of the pension grew far more slowly for Africans than for other groups between 1944 and 1963. (p. 8)

• Racial discrimination in payment levels narrowed from the 1970s; by 1992 the gap was slight, and the pension was equalised after 1994 using means testing. (p. 9)

  By 1992, the monthly payment to whites (R345) was just 1.1 times the payment to coloureds and Indians (R318) and 1.2 times the payment to Africans (R293).

• By the late 1990s pension receipt was very widespread among Africans, making it a major antipoverty instrument. (p. 10)

  By 1999, the total number of social pensioners had reached 1.8 million: 16 per cent of age-qualified whites and 90 per cent of age-qualified Africans.

• Namibian coverage by region suggests the colonial administration favoured politically supportive areas rather than winning over the Owambo north. (p. 12)

  This heavily skewed pattern of coverage casts doubt on the argument that the social pension was a sop aimed at 'winning the hearts and minds' of the Owambo people.

• Namibia's 1990 pension was steeply racially graded, with the lowest black rate a small fraction of the white rate. (p. 13)

  At independence in 1990 some black Namibians received a pension of just R55 per month, while whites were paid R382,66 a ratio of almost 7:1

• South African pension costs rose sharply after equalisation, but the real value then declined under conservative macroeconomic policy. (p. 19)

  Between 1991 and 1995, the social security budget increased by 22.6 per cent per annum, mostly due to the equalisation of pension payments across race groups.

• Sustainability is framed as political commitment rather than an objective fiscal limit. (p. 21)

  There is no economic law that prevents societies from deciding to allocate more resources to old-age security and less to some other expenditure.

Evidence ledger

Sources used

Source	Kind	Period	How used
Old Age Pensions Act No. 22 of 1928 (Statutes of the Union of South Africa: 1928)	legislation gazette	1928	Establishes the original pension eligibility (whites and coloureds aged 65, means-tested) and the exclusion of natives and South West Africa.
Amending Acts to the Old Age Pensions Act: No. 34 of 1931, No. 34 of 1937, No. 48 of 1944, No. 35 of 1949, Act No. 38 of 1962	legislation gazette	1931-1962	Traces the repeal of the recovery-from-children provision, the lower female age, the 1944 extension to Indians and natives with tiered rates, and the 1949 extension to white South West Africans.
Commission on Old-Age Pensions and National Insurance (1926) and Interdepartmental Committee on Poor Relief and Charitable Institutions (1937), cited via Sagner and Human Awareness Programme	commission or official report	1926-1937	Quoted to show the official rationale for excluding natives (native custom) and the later arguments that kraal support was eroding.
Social Security Commission 1943 (Hofmeyr) and the Transvaal Indigency Commission 1908 (as described)	commission or official report	1908; 1943	Used for the origins of the poor white problem and the 1943 recommendation to extend pensions to Indians and natives.
House of Assembly Debates (South Africa) 1965a, 1965b, 1973a, 1973b, 1973c	parliamentary debates	1965-1973	Supplies ministerial and MP statements on racial differentials, de-urbanisation, the homelands policy, work disincentives, and the extension of pensions to Namibia.
Debates of the National Assembly (Namibia) 1992a, 1992b, 1999	parliamentary debates	1992-1999	Documents the post-independence equalisation, the means-testing and 'basic state grant' debate, and the elite attitudes towards pensioners.
Social Pensions Ordinance No. 2 of 1965 and Social Pensions Act No. 37 of 1973 (Laws of South West Africa)	legislation gazette	1965; 1973	Establishes the stages of extension of social pensions to coloured Namibians (1965) and to all Namibian residents (1973).
National Pensions Act 1992 (Namibia), Government Gazette 395; Basic State Grant Amendment Act (draft) 1998	legislation gazette	1992-1998	Shows the equalisation of age and payments, the means-testing provision, and the proposed renaming to 'Basic State Grants Act'.
Report on an Investigation into the Possible Establishment of a Contributory Pension Scheme for South Africa (1976)	commission or official report	1976	Cited as a source for Table 1 payment rates and for the Commissioner of Pensions' 1957 concerns about rising pensioner numbers.
SAIRR Fact Papers: Draper 1964 'Social Pensions in South Africa'; Suttner 1966 'Revised Social Pensions in South Africa'	secondary literature compilation	1964-1966	Supplies calculations of the relative growth in pension values by race from 1944 to 1963 and the Finance Minister's 1965 uniform-rate statement.
Table 3 source: UNICEF-Namibia, A Situation Analysis of Children and Women in Namibia (1991)	secondary literature compilation	1990	Provides the 1990 ethnically differentiated monthly pension schedule (R382 whites to R55 Owambo).
Subbarao 1998 (World Bank Policy Research Working Paper 1996) regional pension coverage data, 1993	secondary literature compilation	1993; 1998	Table 2 coverage by region is used to test political explanations for uneven pension delivery.
Author's interviews with Namibians, including a former paymaster in the Owambo Administration Directorate of Welfare	interviews oral history	c. 2000s (unspecified)	Provides competing explanations for the 1973 extension of pensions to black Namibians (compensation versus winning hearts and minds).
Archival research in the Library of Parliament, Windhoek and the National Library of South Africa, Cape Town	state archives	1928-2000	The stated basis for the political and administrative history, mainly Hansard and legislation.
Newspapers: The Namibian (13 March 1998; 28 Feb 2001), New Era (28 July 1994), SAPA (27 Feb 2000)	newspapers periodicals	1994-2001	Used for ministerial statements on pension freezes and means testing, the white pensioners' reaction, the tender scandal, and Manuel's comment on affordability.
Secondary statistical compilations on beneficiaries and spending (Van der Berg 1998; Fultz & Pieris 1999; Naidoo 1996; Legido-Quigley 2003; Barrientos & Lloyd-Sherlock 2002; Willmore 2001)	secondary literature compilation	1947-2000	Provide pensioner numbers, racial payment ratios, coverage rates, and the cross-country Table 4 comparison.
Namibian government policy documents: Integrated Poverty Reduction Strategy 1997, First National Development Plan 1995, Food and Nutrition Policy 1995, NEPRU 1996, World Bank Public Expenditure Review 1995	commission or official report	1995-1997	Source for cost estimates, GDP share, and official sustainability concerns about the pension.
Botswana government and consultant documents: National Policy for Rural Development 1973, Vision 2016, Old Age Pension Unit 1999, Ernst & Young 1998, BIDPA 2002	commission or official report	1973-2002	Used to describe the introduction, coverage, level and fiscal concerns of Botswana's Old Age Pension.
South African policy documents: White Paper for Social Welfare 1997; Taylor Committee Report 2002	commission or official report	1997-2002	Used for post-1994 coverage and cost escalation and the Basic Income Grant proposal.
World Bank, World Development Report 2000/2001	national accounts macro	1999	Provides 1999 GNP per capita figures to argue that wealthier countries sustain social pensions.

Inference

Quantification: Descriptive tables.

Historical-political narrative based on legislation and parliamentary debates, supplemented by a few interviews and secondary statistics. Tables compare racial payment ratios over time, Namibian regional coverage and regional programmes, and motives are inferred by reading ministerial statements and comparing coverage patterns across regions.

Key claims

Claim	Basis	Strength
South Africa's pension was used as a tool of influx control and de-urbanisation, evidenced by differentiated native rates (city/town/rural) and their abolition in 1965 to support the homelands policy. (p. 543)	1944 and 1962 Acts, Hansard 1965, Human Awareness Programme and SAIRR papers, Table 1.	Partial data
Racial discrimination in pension payments was persistently wider than the stated 4:2:1 principle, though it narrowed from the 1970s. (p. 545)	Table 1 compiled from Acts and the 1976 report; SAIRR calculations; Fultz and Pieris for 1992 rates.	Systematic data
Pension take-up among Africans grew from about 54 per cent of the eligible in 1947 to 81 per cent of all pensioners by 1993, despite National Party attempts to restrict numbers. (p. 544)	Native Affairs Department estimate via Sagner, Van der Berg, and the Commissioner's 1957 remarks.	Partial data
The extension of pensions to black Namibians in 1973 had unclear motives, plausibly influx control or 'winning hearts and minds', supported by interviews and Hansard. (p. 547)	Author's interviews, a 1973 Hansard quotation, and the Morgan and FAO literature.	Illustrative or anecdotal
The 1993 regional coverage pattern in Namibia suggests the colonial administration withheld resources from politically oppositional northern regions rather than favouring Owambo. (p. 549)	Table 2 from Subbarao (World Bank), read against regional political geography.	Partial data
At independence Namibia moved to a uniform pension, set at a low level (N$120 in 1994) because of fiscal constraints, and the real value was subsequently eroded, as the 1990 differentials ran up to 7:1. (p. 550)	UNICEF 1990 rates (Table 3), the 1992 Act, ministerial statements, Figure 1 and Schleberger's calculation.	Partial data
The social pension is under fiscal, administrative and political pressure in all three countries, with costs of about 1.4-2 per cent of GDP, real value declining, and weakening elite support. (p. 557)	Secondary compilations (Ardington and Lund, Barrientos et al.), Namibian and Botswana government documents, and Hansard.	Partial data

The author's caveats

Notes that the motivation for the 1973 extension to Namibia is undocumented, that Botswana's reasons for introducing the pension were never explained, and that the Table 2 coverage evidence is only circumstantial. It also notes Draper's and Table 1's slightly different percentages.

Gaps in the evidence

The motive claims rest on selective Hansard quotations and a few interviews, with no ministry or administration files (e.g. Department of Bantu Administration, South West Africa Administration) to show actual policy deliberation. The Namibian coverage inference uses 1993 post-independence data to explain 1970s-80s politics and has no pre-1990 regional administrative data or controls (e.g. population, remoteness) that archives of the second-tier authorities might supply. Pensioner and expenditure series by race and year, which the Department of Social Welfare and Pensions published, would test the growth and fiscal claims better than secondary compilations.

Replicability: Data described but not public.