The origins of the 'welfare state' in pre-apartheid South Africa¶
David Duncan (1993). The origins of the 'welfare state' in pre-apartheid South Africa.
| Type | working paper |
| Scope | Wider apartheid literature |
| Method | archival narrative |
| Period studied | 1919–1948 |
| Regions | South Africa, Johannesburg, Cape Town, Natal |
| Themes | origins segregation, business and state, mining migrant labour, bantustans removals influx, inequality poverty unemployment, politics ideology elections, demography urbanisation |
| Citation key | duncan1993 |
Argument¶
The paper argues that in the 1930s and 1940s the South African state, under external and domestic pressure and following British and American models, began to lay the foundations of a welfare state for African workers. Its commitment was never more than lukewarm. The Smuts wartime government used commissions and planning councils to appear active and to contain unrest, then diluted or shelved the proposals once the war ended. The reasons were cost, white voters' fear of higher taxes, mining-industry opposition to measures that would disturb low-wage migrant labour, and electoral competition from the National Party. The argument is built from case studies of housing, pensions, unemployment insurance and health services, with attention to conflicts between government departments. Civil servants are shown as sometimes progressive but constrained by short-term political interests.
Evidence and method¶
State Archives (Pretoria) files of the Public Health, Native Affairs and Labour departments. Commission and committee reports (Smit, Thornton, Gluckman, Social Security Committee, SEPC, Native Economic Commission). White Papers, Hansard debates, Union Year Books, Statutes, Race Relations journal articles and Shula Marks's essay. It also draws on the author's PhD thesis (Queen's University, 1990).
Place in the debates¶
- Origins and character of South African state social policy (welfare as legitimation and control vs genuine reform) (see Living standards, health and social policy). Welfare measures were largely window-dressing and legitimation, aimed at controlling unrest and protecting whites from disease. Some reforming civil servants did push for more.
- Cheap labour thesis / mining capital's influence on state policy (see Living standards, health and social policy). Mining capital, backed by the Mines and Native Affairs departments, secured exclusion of migrant mineworkers from unemployment insurance. This protected low-wage migrancy and the influx-control system.
- Whether apartheid marked a break in social policy (see Living standards, health and social policy). The Smuts government had already abandoned its plans for improving African workers' conditions before 1948. The National Party inherited a worsening problem rather than reversing a reform trajectory.
Findings¶
-
The central thesis: wartime welfare planning was largely appearance, used to contain unrest. (p. 1)
It cynically manipulated the more progressive elements within the State to give the appearance of action and thus contain unrest during World War II.
-
Early housing loans produced very little housing for Africans and coloureds, and much of it was barrack rooms. (p. 3)
By the end of 1930, only 7,609 dwellings had been built for Africans and coloureds with Central Housing Board loans.
-
The urban housing shortage was large and growing, and had worsened by the 1936-42 period. (p. 4)
Whereas the 1920 Housing Committee estimated a shortfall of 10,000 houses, the 1936 Public Health Department Report put the figure at 16,000 in the eight largest cities alone.
-
The 1944 pension law first covered Africans but at rates far below the minimum the SEPC envisaged, and far below white rates (£60, £54, £48 in the notes). (p. 5)
established three levels of £21, £9 and £6 per annum, depending on whether the recipient lived in a city, town or rural area.
-
The 1946 Unemployment Insurance Act excluded most categories of African workers, notably mineworkers housed and fed by employers. (p. 6)
The 1946 act specifically excluded African gold and coal workers who were provided with accommodation and food, along with farm labourers, domestic servants, Africans employed in rural areas (but not in factories), and casual labourers.
-
The Native Affairs compromise exempted the mines from contributions, protecting both mine profits and the migrant labour system. (p. 6)
that the mines should be spared from paying contributions for migrants whose return fares were provided by the employer.
-
By 1940, health provision for Africans was very thin relative to population. (p. 8)
There were 2,000 doctors in the country in 1940 for a total population of over 10,000,000.
-
The Smuts government rejected the Gluckman national health scheme on cost grounds, leaving only modest health centres. (p. 9)
But the Smuts government rejected its report's proposals for a national health scheme as too expensive.
-
Cost and white voters' attitudes were the main reason the state avoided responsibility for African welfare. (p. 9)
White voters were afraid of disease among Africans but would not have welcomed substantial tax increases for black social services.
-
Planners' proposals were pared back heavily once cost was considered. (p. 10)
recommended increasing social security and health spending by over 200 per cent (from E17.5m, to £52m).
-
The Smuts government was a hard-pressed coalition rather than a reformer, choosing containment over a costly welfare state. (p. 7)
The Smuts regime in its last years was not a reforming government, set on implementing a new deal for all its citizens.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| State Archives (Pretoria), Public Health Department files (GES series), e.g. GES 593/13, GES 82/33, GES 1/62, GES 7/26A, GES 4/5A | state archives | 1921-1952 | Internal memoranda and correspondence show Public Health officials' motives, reluctance to press municipalities, and handling of housing, clinics and medical examinations. |
| State Archives (Pretoria), Native Affairs Department files (NTS 24/349A pt 2; NTS 133/62) | state archives | 1946-1947 | Documents the interdepartmental conflict over the Unemployment Insurance Bill and the NAD's pension estimates. |
| Wits Archives, AD843 (Rheinallt Jones Papers) | state archives | 1938 | Shows liberal views of Public Health Secretary E H Cluver as an improvement on predecessors. |
| Union Year Book (Vols 13, 15, 18, 20, 23) | official statistical series | 1930-1946 | Supplies housing loan totals, hospital bed counts for non-Europeans, uses of the Slums Act, and Chamber of Mines health aid. |
| Report of the Inter-Departmental Committee on the Social, Health and Economic Conditions of Urban Natives (Smit Report), 1942 | commission or official report | 1942 | Evidence on shacks, overcrowding, housing shortage and recommendations for housing finance and health visitors. |
| UG 8-40, Thornton Committee report on peri-urban areas, 1938-9 | commission or official report | 1938-1939 | Shows officials' dissatisfaction with local administration of urbanising areas. |
| Report of the Commission of Enquiry on the City of Johannesburg Public Health and Native Affairs Departments, 1935 (GES 82/33) | commission or official report | 1935 | Finding that Johannesburg cleared slums without building enough location housing. |
| UG 9-1943 SEPC Report No.1 and UG 14-1944 Social Security Committee Report / SEPC Report No.2 | commission or official report | 1943-1944 | Source for planners' social security proposals and cost figures against which government legislation is judged. |
| SC 10-1944 Report of the Select Committee on Social Security | parliamentary debates | 1944 | Used for the 1944 House resolution and for cuts to the SEPC spending plan (£32.5m). |
| Government White Paper 1945 (Memoranda on Poor Relief and on Government's Proposals regarding Social Security) | commission or official report | 1945 | Shows the government's proposals to restrict spending on pensions and unemployment insurance and its coverage estimates. |
| Report of the National Health Services Commission (Gluckman Report), UG 30-1944 | commission or official report | 1944 | Evidence on inadequacy of health services, divided jurisdiction and costs of a national scheme (£12m). |
| Report of the Native Economic Commission, UG 22-32 | commission or official report | 1932 | Describes rural environmental degradation and disease conditions for Africans. |
| Department of Public Health annual reports (UG 58-1948; 1936 report) and National Housing and Planning Commission Report (UG 45-1949) | official statistical series | 1936-1948 | Housing shortfall estimates and 1947 assisted-dwelling figures; evidence of officials passing responsibility to municipalities. |
| Statutes of the Union (Acts 49/1944, 45/1945, 48/1944, 53/1946, 15/1911) | legislation gazette | 1911-1946 | Provisions on pension rates, unemployment insurance exclusions, housing powers and labour medical regulations. |
| House of Assembly Debates (Vol 34 1939; Vol 51 1945; Vol 61 1947) | parliamentary debates | 1939-1947 | Nationalist and ministerial statements on cost of African welfare and health. |
| Rand Daily Mail (1935) article on native townships | newspapers periodicals | 1935 | Illustrates white public concern about disease from African locations. |
| Race Relations (SAIRR journal) and Hellmann (ed), Race Relations Handbook (1949) | newspapers periodicals | 1937-1949 | Evidence of liberal pressure and background data on urban areas and social welfare. |
| Shula Marks, 'Industrialization, Rural Health and the 1944 National Health Services Commission' (1992), drawing on Public Health Department files | secondary literature compilation | 1920s-1944 | Section iii relies heavily on Marks's evidence about Public Health Department behaviour and the Gluckman Commission. |
| Duncan PhD thesis, 'The Mills of God' (Queen's University, 1990) | secondary literature compilation | 1918-1948 | The paper summarises a thesis chapter; the thesis rests on archival research. |
Inference¶
Quantification: Illustrative figures.
Archival, documentary case study of state bureaucracy: reads departmental files, commission reports and parliamentary debates to trace policy shifts over time and interdepartmental conflict, and sets this against international welfare-state models. Figures are used illustratively.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| The Smuts government used commissions and councils to appear committed to welfare while holding back on cost, and abandoned the plans after the war. (p. 1) | Commission and White Paper proposals set against the legislation actually passed and the Select Committee cuts. | Partial data |
| State housing provision for Africans was minimal before 1939 because it depended on discretionary municipal uptake of loans. (p. 3) | Housing Act and loan totals from Union Year Books and Public Health files. | Partial data |
| Pension and unemployment legislation gave Africans far less than planners proposed, with low rates and mine and rural worker exclusions. (p. 5) | Statutes, White Paper, NAD memoranda and Social Security Committee reports. | Partial data |
| The Chamber of Mines, backed by Mines and Native Affairs officials, secured exclusion of mine workers from unemployment insurance. (p. 6) | NAD file NTS 133/62 on the 1946 Bill, including Chamber, Labour and NAD memoranda. | Partial data |
| The Smuts government rejected the Gluckman national health scheme as too expensive, and health provision rose only marginally. (p. 9) | Gluckman Report cost estimates, Marks's account, Union Year Book bed counts. | Partial data |
| Cost and white political fears were the chief reasons the state avoided responsibility for African welfare. (p. 9) | Hansard speeches by Nationalists, cost comparisons of SEPC, Select Committee and Gluckman proposals. | Illustrative or anecdotal |
Gaps in the evidence¶
The argument about government motives (cynical manipulation, appearance of commitment) rests on selective reading of departmental memoranda and Hansard, with little systematic expenditure data. Racially disaggregated budget and expenditure series on housing, pensions and health, and Cabinet or Treasury records, would test the claims better, and these exist in the archives. Data on the actual coverage and take-up of pensions and unemployment insurance would also show the real incidence of benefits.
Replicability: Data described but not public.