Migrant Labour and South African Economic Development¶
Jill Nattrass (1976). Migrant Labour and South African Economic Development. South African Journal of Economics, vol. 44, no. 1. Economic Society of South Africa.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | descriptive quantitative |
| Period studied | 1936–1970 |
| Regions | South Africa, Homelands (KwaZulu, Transkei, Ciskei, Lebowa, Bophuthatswana), White areas, Foreign labour source areas |
| Themes | mining migrant labour, bantustans removals influx, labour markets wages, demography urbanisation, education human capital, methods data sources, inequality poverty unemployment |
| Citation key | nattrass1976b |
Argument¶
Nattrass argues that South Africa's development has entrenched a massive structural imbalance between where jobs are and where Africans live, sustained by a system of temporary oscillating migration. She estimates the size and characteristics of the migrant stream from 1970 census data, using a 'standard population' method, and shows that migration grew steadily from 1936 to 1970. The migrants are young and better educated, which drains the Homelands. The modern sector benefited from avoided housing and social overhead costs and from lower wages, while Homeland households became heavily dependent on remittances. She concludes that markets alone will not correct the imbalance and recommends government intervention on both the White-sector and Homeland sides.
Evidence and method¶
1970 Population Census (Report 02-02-02), 1960, 1946 and 1936 census reports, Department of Bantu Education and Bantu Administration annual reports, Transkei labour statistics, Hansard figures on foreign Africans, and the Lombard and van der Merwe and Wilson estimates for comparison. Chi-square tests are used. Remittance estimates draw on manufacturing wage rates and the Marais Commission. The conference discussion with the author is also included.
Place in the debates¶
- Cheap labour thesis: did migrancy depress African wages and benefit employers? (see Cheap labour thesis and the function of the reserves). Presents the case that temporary migration held Black wages lower and weakened organisation as an argument ('it can also be argued'), and lists housing cost savings to the modern sector. She is cautious and does not test the claim.
- Market forces vs state/institutions in creating spatial imbalance (liberal vs revisionist framing) (see Homelands, separate development, decentralisation and removals). Holds that the imbalance 'has largely resulted from the operation of market forces but has been strengthened by certain institutional elements', so government intervention is needed rather than markets alone.
- Homeland development and decentralization policy (see Homelands, separate development, decentralisation and removals). Supports continued decentralization, Homeland urban housing and cautious Homeland industrialization, and phasing out migrancy over the long run. This is reformist and accepts the separate-development framework.
- Employer preference for migrant vs stabilised labour (see Cheap labour thesis and the function of the reserves). Finds no marked employer preference for or prejudice against migrants outside mining, since migrants from most groups obtain high-wage jobs at similar rates.
Findings¶
-
Migrants were a large share of the modern-sector labour force in 1970. (p. 2)
In 1970 migrant workers in South Africa formed 28 per cent of the total male work force and 59 per cent of the African male work force enumerated in White areas.
-
The estimated migrant stream in 1970 was about 1.49 million men and 260,000 women (net out-migration, so an understatement). (p. 1)
the total number of migrants of 1 490 000 men and 260 000 women in Table 1, understates the size of the migration stream
-
Migration grew steadily over 1936-1970 rather than being a temporary adjustment. (p. 3)
the number of migrants from rural areas grew at an average compound rate of 3,1 per cent per annum. In addition, the male migration rate rose by 36 per cent
-
Migration selects the better educated, which lowers the social return to Homeland education. (p. 3)
although sixty per cent of all African children in school are in schools in the Homelands seventy-five per cent of the African men with primary school education and two further years
-
Migrants are younger and have shorter modern-sector working lives than stabilised workers, raising turnover and training costs. (p. 4)
The average length of the working life in the modern sector of a migrant labourer is 19 years, whereas that of his stabilized colleague is 23 years
-
The migrant system saved the modern sector housing and social overhead investment, and may have held wages down. (p. 5)
the system of temporary migration has held Black wages in South Africa at a lower level than would otherwise have been the case.
-
Outside mining, migrants are distributed across industries much like settled workers; Transkeian and foreign migrants are concentrated in low-wage sectors, especially mining. (p. 7)
Oscillating migrants from the Transkei and the areas outside the Republic make a substantial contribution to the labour supply of
-
Migrant labour was a large share of all African male workers and drained the Homelands of young, educated men. (p. 7)
43 per cent of all the economically active African men in South Africa are migrant workers.
-
Homeland households depended heavily on remittances (KwaZulu, 1970). (p. 7)
these remittances in 1970 were the equivalent of forty-one per cent of KwaZulu's Gross Domestic Product, eighty-eight per cent of the area's subsistance sector income
-
The author warns that the system could become unmanageable without large-scale policy action. (p. 8)
the social, political and economic problems created by the continued and increasing reliance by the South African economy on the migratory labour system could well become insurmountable.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Population Census 1970, Report 02-02-02, Bantu Age, Occupation, Industry, School Standard, Birthplace | population census | 1970 | Basis for the standard-population estimate of migrants (Table 1), age-specific migration rates, education profiles and age distributions of workers. |
| Population Census 1960 Vol. 7 No. 1, Characteristics of the Population in Each Magisterial District and Economic Regions, Age, Marital Status and Home Language | population census | 1960 | Used to estimate the number and rate of African temporary migrants from rural areas in 1960 (Table 2). |
| Seventh Census of the Population of the Union of South Africa, Vol. II, U.G.60/1950 (1936 and 1946 censuses) | population census | 1936 and 1946 | Age data on the African population used to estimate migrant numbers for 1936 and 1946 in Table 2. |
| Department of Bantu Education Annual Report 1971, RP31/1973 | commission or official report | 1971 | School enrolment data used to compare educational profiles of populations with school-going populations (Table 3) and age distributions. |
| Annual Report of the Department of Bantu Administration and Development 1971, RP 41/1973 | commission or official report | 1971 | Source for industrial distribution of Tswana, Pedi and Ciskei migrants placed through labour bureaux (Table 6). |
| Minister of Bantu Affairs reply in House of Assembly, 5 March 1971, Hansard cols 5384-386, quoted in SAIRR Survey of Race Relations 1971 | parliamentary debates | 1971 | Provides the industrial distribution of foreign African migrants in Table 6. |
| Transkei Government Department of Interior, Statistical Labour Report 1971 | official statistical series | 1971 | Source for the industrial distribution of Transkeian migrants in Table 6. |
| Lombard and Van der Merwe, Central Problems of the Economic Development of the Bantu Homelands, Finance and Trade Review, June 1972 | secondary literature compilation | 1967 | Ratio of commuter to migrant income in 1967 used to estimate the number of homeland commuters and the share of modern-sector workers living outside the sector. |
| Wilson (1972) Migrant Labour in South Africa and Department of Statistics estimate (via South Africa 1974) | secondary literature compilation | 1970 | Alternative migrant estimates used to check the reasonableness of the author's 1970 estimate. |
| South Africa 1974 / South Africa 1974, Department of Information (GDP, subsistence and agricultural output) | national accounts macro | 1970 | Supplies GDP, subsistence and agricultural output figures against which estimated KwaZulu remittances are compared. |
| Marais Commission, R.P./84 1970 (wage rates) | commission or official report | 1970 | Wage-rate data used for estimating remittances by sector (low-wage rate taken as 43% of high-wage). |
New data¶
The author built estimates of the number and age-specific rates of African temporary migrants from rural areas at 1936, 1946, 1960 and 1970 census dates, by standard-population method, with ethnic-group, educational and industrial breakdowns for 1970. Details are not published, only available from the author.
Inference¶
Quantification: Statistical analysis.
Indirect demographic estimation: a standard population from the global population of each ethnic group is imposed on rural areas and the actual adult population subtracted to get migrants. Descriptive tables over census years and groups, chi-square tests of group differences, and an illustrative cost calculation for housing savings. Policy conclusions are normative.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| About 1.49 million male and 260,000 female oscillating migrants were in the modern sector in 1970, making 43% of economically active African men migrants. (p. 2) | Standard-population estimate from the 1970 census (Report 02-02-02), cross-checked against Wilson and Department of Statistics figures. | Partial data |
| The migrant stream grew steadily from 1936 to 1970, with male migration rate rising from 25% to 34% and 3.1% annual compound growth. (p. 3) | Table 2 estimates from the 1936, 1946, 1960 and 1970 censuses using the same method. | Partial data |
| Migration is selective of the better educated; out-migration from Homelands reaches about 90% at primary plus 4-5 years of schooling. (p. 3) | Comparison of census educational profiles with school enrolments in Bantu Education Annual Report 1971; no direct data on migrants' education. | Partial data |
| Migrants are younger and have shorter working lives than stabilized workers (19 vs 23 years). (p. 4) | Table 4 age distributions estimated from the 1970 census and the Bantu Education report; chi-square test of difference. | Partial data |
| Migration patterns are highly consistent across Xhosa, Zulu, Pedi and Tswana, but industrial distribution differs: Transkeians and foreign migrants concentrate in mining and the low-wage sector. (p. 6) | Table 5 from census; Table 6 from labour bureau, Transkei, Hansard and census surrogates; chi-square tests (Zulu assumed to follow Pedi). | Partial data |
| The modern sector gained from avoiding housing costs for migrants' families: R393m (4% of GDP) in 1970, rising to 8% if compounded. (p. 11) | Illustrative calculation using assumed marginal housing costs, assumed return differential and GDP at 1963 prices. | Illustrative or anecdotal |
| Remittances to KwaZulu in 1970 were equivalent to 41% of KwaZulu GDP and 142% of agricultural output, so Homelands depend on the migrant system. (p. 7) | Preliminary remittance estimates from migrant numbers, manufacturing wage rates and Marais Commission data; compared to Department of Information output figures. | Partial data |
| Migration has held Black wages down and undermined organisation, benefiting profits. (p. 5) | Argued from theory and unspecified earlier arguments; no data presented. | Asserted |
The author's caveats¶
The author notes the estimates are unreliable owing to lack of adequate statistical data and that Table 4 is not wholly consistent. Net out-migration understates the stream. There is no direct evidence on migrants' education, the Zulu industrial distribution is a surrogate, the data are highly aggregated and remittance estimates are preliminary.
Gaps in the evidence¶
The estimates are indirect and rest on the assumption that ethnic-group populations are representative of rural populations. Direct migrant surveys, labour-bureau and Chamber of Mines recruitment records, and pass-law or influx-control registers would test the estimates better; some existed then, but the author did not use them. Claims about wage suppression and benefits to capital lack wage series or firm data, and the method details are unpublished.
Replicability: Data described but not public.