Shifting Understandings of Skills in South Africa: Overcoming the Historical Imprint of a Low Skills Regime¶
Simon McGrath; Azeem Badroodien; Andre Kraak; Lorna Unwin (2004). Shifting Understandings of Skills in South Africa: Overcoming the Historical Imprint of a Low Skills Regime. HSRC Press.
| Type | book |
| Scope | Wider apartheid literature |
| Method | mixed |
| Period studied | 1890–2003 |
| Regions | South Africa, Cape Colony, homelands |
| Themes | education human capital, labour markets wages, origins segregation, manufacturing industrial policy, business and state, transition post apartheid, inequality poverty unemployment, comparative theory |
| Citation key | mcgrath2004 |
Argument¶
This edited HSRC volume traces how 'skill' has been understood and institutionalised in South Africa since industrialisation. It argues that race, social control and the migrant-labour system produced a polarised, racialised 'low skills regime' in which black skill formation was shaped by welfare and discipline rather than economic development. Late-apartheid training reforms (Wiehahn/Riekert, NTB/HSRC) were market-led, voluntarist and constrained by continued racial schooling, so they entrenched a low-skill equilibrium. The post-1994 chapters assess the National Skills Development Strategy, FET colleges and SMME policy against the international 'high skills' literature, and propose a hybrid high/intermediate/low skills model. The apartheid-era chapters are mostly historical and policy analysis; the later ones are policy description and survey synthesis.
Evidence and method¶
Government commission reports (De Villiers, Wiehahn, Riekert, Kleu, NTB/HSRC investigations 1984/1989/1991), Cape and Union education department documents, SANAC 1903-5 evidence, archival items, education enrolment statistics, enterprise training surveys (HSRC 2000, World Bank 2000, BMR 2002), Department of Labour reports, and secondary literature.
Place in the debates¶
- Low skills equilibrium vs high skills route (social institutionalist skills literature) (see Education, Bantu Education and skills). Accepts the institutionalist framework but argues the high skills thesis exaggerates the spread of high skill production and neglects intermediate skills and middle society. Proposes a hybrid, differentiated model and sees a positive low-skills job strategy (Ashton) as legitimate.
- Role of market vs state in training (voluntarism vs coordination) (see Education, Bantu Education and skills). Kraak argues market-led training under late apartheid reflected the triumph of ideology over evidence, and that market solutions alone are insufficient; the post-1994 levy-grant and SETA system is a state-led corrective.
- Liberal vs revisionist views of apartheid and the colour bar (see The job colour bar, white labour and its erosion). Cites Terreblanche & Nattrass and Beinart, who hold that the Second World War made the colour bar economically redundant. It stresses skill as bound up with racial ideology, white labour protection and social control, with employers floating the colour bar upwards from the 1960s.
Findings¶
-
Early industrialisation relied on immigrant white craft skills, which allowed the state and employers to neglect indigenous skills formation. (p. 22)
Crucially, it allowed the state and employers to pay less attention to the role of indigenous skills formation than would have been necessary otherwise.
-
Skills for white South Africans were tied to social policy and social control from the 1890s. (p. 22)
Skill thus became infused with notions of social control and of the value of industriousness over notions of skills being about economic development
-
The migrant labour system fostered the perception that African labour was homogeneous and interchangeable, so low skill and short-termism became central to thinking about labour. (p. 22)
This system led to the perception amongst employers and government officials that African labour was homogenous and interchangeable.
-
The Second World War brought rapid black entry into manufacturing and strained racial definitions of skilled work. (p. 46)
of the 125 000 people absorbed into manufacturing and construction employment during the Second World War, only 15 per cent were white.
-
Apartheid-era import substitution, the minerals-energy complex and the small white market produced bifurcated labour demand and neglect of craft skills. (p. 25)
The result was an industrial strategy that produced an unusually bifurcated demand for labour between a high skill segment and a far larger low skill segment
-
The 1977 Wiehahn/Riekert/Kleu reforms marked a shift towards market-led policy, which shaped training reform. (p. 70)
All three reports emphasised that the time had come for the state to play a less important role in shaping the economy
-
The late-apartheid reform proposals ended up reinforcing a low skills equilibrium. (p. 80)
the reform proposals of the late apartheid period both perpetuated and reinforced a low skills equilibrium predicated on market regulation
-
Apartheid-era employer surveys showed little structured training or career pathing for black workers. (p. 63)
66 per cent admitted that they had no recognisable career paths for their workers, and 32 per cent acknowledged training by the 'sit-by-Nellie' type
-
Bantu Education expanded enrolments but cut per-pupil spending and confined most pupils to primary level. (p. 100)
contributing to a decrease in expenditure per pupil from R17.08 in 1953 to R12.70 in 1965
-
Post-1994 enterprise training remains racially and gender skewed. (p. 161)
In 2000 about 71 per cent of employees trained in the managerial/professional category were white, and only 16 per cent were African
-
Jobless growth persisted after 1994 despite macroeconomic stabilisation. (p. 236)
employment has declined, with 1.3 million jobs lost in the formal sector during this period
-
Employer participation in the levy-grant system is weak. (p. 153)
only 65.5 per cent of firms actually pay the levy as of September 2002
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| De Villiers Commission (Report of the Commission on Technical and Vocational Education of 1948) | commission or official report | 1948 (published 1949) | Establishes official thinking that black technical training should focus on teaching discipline and 'how to work' and that apprenticeship rules excluded black youth. |
| Annual Reports of the Union Education Department (1925, 1928, 1938, 1941–45, 1946–7, 1952) | official statistical series | 1912–1952 | Supplies institutional transfers (19 state and 23 state-aided industrial institutions in 1925), COTT training figures and policy statements on poor-white and black training. |
| Malherbe (1977) Education in South Africa Vol. 2, and Malherbe (1932) Carnegie Commission volume on Education and the Poor White | secondary literature compilation | 1910–1975 | Relied on for enrolment figures, education spending, white urbanisation and rural white schooling statistics. |
| Chisholm (1989 PhD thesis; 1992 Technical Colleges policy options) | secondary literature compilation | 1882–1992 | Source of counts of black technical institutions in 1955 and historical account of reformatories/industrial schools. |
| Botha Commission (Report of the Coloured Education Commission 1953–6) | commission or official report | 1953–1957 | Shows the commission's recommendation of manual and reformatory-style vocational training for coloured and Indian learners. |
| Callinicos, Terreblanche & Nattrass, Beinart, Duncan (secondary accounts of WWII manufacturing and labour) | secondary literature compilation | 1939–1948 | Supplies wartime manufacturing growth and the 125 000 manufacturing/construction jobs of which 15 per cent were white. |
| Riekert (RSA 1979), Wiehahn (1982) and Kleu (RSA 1983) Commission reports | commission or official report | 1977–1983 | Evidence of the late-apartheid shift to market-led, employer-led training and the Manpower Training Act 1981. |
| NTB/HSRC Investigations (1984 Training of Artisans; 1989 Skills Training in the RSA; 1991 National Training Strategy) | commission or official report | 1984–1991 | Principal evidence for the critique of apprenticeship, fragmentation of certification, employer-led training and reform proposals; includes survey findings on career paths and sit-by-Nellie training. |
| De Lange Commission report (HSRC 1981) and DNE (1984) report on a national certifying council | commission or official report | 1981–1991 | Used to document certification fragmentation (nine examining bodies administering up to 90 exams per year) and non-articulation of academic and vocational sectors. |
| COSATU position papers (1991) and NTB guidelines (1986, 1990, 1995) | union or movement records | 1986–1995 | Shows union criticism of the NTS and nominal union representation on Industry Training Boards. |
| South African Native Affairs Commission 1903–1905 evidence (5 vols) and Cape Archives/Transvaal Archives files | commission or official report | 1903–1905; 1940–41 | Witness testimony and archival correspondence (Moila Reserve file NTS 8537) to show colonial and employer attitudes to African industrial and agricultural education. |
| Newspapers and periodicals (Cape Times, Cape Argus, Christian Express, Cape Mercury, The Journal) | newspapers periodicals | 1887–1911 | Quoted to illustrate contemporary debates on white poor and African education. |
| Department of Public Education (1929) The Native Primary School; Bot (1936) Empire Exhibition; Huss (1928); Rogers (1949); Loram (1917) tables | commission or official report | 1917–1936 | Curriculum specifications and school/learner counts (Tables 1 and 2) underpin claims about the agricultural/manual character of African schooling. |
| Keiskammahoek Rural Survey 1952 | author survey or fieldwork | 1952 | Evidence on lack of water, fencing and equipment in rural schools blocking agricultural teaching. |
| Hurwitz (1964) occupational distribution of African males 1936 and 1951 (Census-derived) | secondary literature compilation | 1936–1951 | Table 3 shows peasant farmers falling from 34 to 16.2 per cent of African male employment while general labourers rose. |
| Unterhalter (1991) enrolment and expenditure data on Bantu Education | secondary literature compilation | 1953–1987 | Provides enrolment growth, per-pupil expenditure decline and homeland primary share data. |
| South African Senior Certificate Examination 2001 data on agricultural science (Paterson & Arends 2002) | official statistical series | 2001 | Table 4 shows 82 per cent of schools offering agricultural science to Grade 12 are in former homelands. |
| DoL policy documents: Green Paper 1997, NSDS 2001, Skills Development Act 1998, Levies Act 1999, Synthesis/Implementation Reports 2002, Preliminary Annual Report 2001/2002 | commission or official report | 1996–2003 | Basis for descriptions of SETAs, learnerships, levy-grant system and for measuring progress against NSDS success indicators. |
| Ntsika (1999) The State of Small Business in South Africa 1998 | official statistical series | 1997–1999 | Table 6 distribution of enterprises by size and employment, showing 72 per cent micro and 43 per cent of employment in large firms. |
| HSRC Baseline Survey of Industrial Training in South Africa (Kraak, Paterson, Visser & Tustin 2000) | household or labour survey microdata | 2000 | Provides training rates by sector, race, gender and occupation (Tables 8 and 10) and short course duration. |
| World Bank/Greater Johannesburg surveys (Chandra et al. 2000a, 2000b), BMR (Martins & Van Wyk 2002), P-E Corporate Services (2001) | author survey or fieldwork | 1999–2002 | Compared in Tables 7 and 9 to estimate aggregate training rates (20-30 per cent) and mode of training. |
| Statistics South Africa Labour Force Survey and HRD Review 2003 compilations (McCord, McCord & Bhorat, Devey et al.) | official statistical series | 1994–2002 | Unemployment rates and jobless growth statistics used to frame the limits of the NSDS. |
| South African FET and HRD policy documents (DoE 1997, 1998, 1999, 2001, 2002; White Papers 1995, 1998; DoE & DoL 2001 HRDS; RSA 1995b Small Business White Paper) | commission or official report | 1995–2002 | Documentary policy analysis of skills for enterprise development and FET coherence. |
| Brown, Green & Lauder (2001) High Skills comparative data (export shares Table 11) | secondary literature compilation | 1994 | Table 11 export shares by skill band underpin the claim that intermediate skill sectors matter. |
| Press and DoL commentary on SETAs (Mail & Guardian, Sunday Times, Business Day, Pretoria News 2002) and DoL SETA evaluations 2001 | newspapers periodicals | 2001–2002 | Evidence of employer resistance to levy-grant and SETA capacity weaknesses. |
Inference¶
Quantification: Descriptive tables.
Edited volume mixing documentary policy analysis, historical narrative drawn largely from secondary literature and official commission reports, meta-analysis of existing enterprise training surveys and DoL monitoring reports (Chapter 6), and theoretical argument using the high-skills literature; no original datasets or regressions.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Technical and vocational provision for blacks before 1970 was oriented to discipline, social control and welfare rather than skilled trade training, while white provision became certified and differentiated. (p. 41) | De Villiers Commission 1948, Botha Commission 1956, Union Education Department reports and Chisholm/Malherbe secondary data, read documentarily. | Partial data |
| Late-apartheid training reforms (NTB/HSRC reports, Manpower Training Act amendments, ITBs) were voluntarist, employer-led and fragmented, entrenching a low skills equilibrium. (p. 70) | NTB/HSRC 1984, 1989, 1991 reports, Riekert/Wiehahn/Kleu commissions, COSATU critiques; reading of policy texts. | Partial data |
| Industrial education for Africans in the Cape was manual and agricultural in character, with only under 1 per cent of African pupils in mission industrial institutions and funding for them falling from £6 387 in 1890 to £1 338 in 1910. (p. 77) | Loram 1917 and Rogers 1949 tables, grants-in-aid data, SANAC evidence, 1929 curriculum specifications. | Partial data |
| Agricultural curricula persist: 82 per cent of schools offering agricultural science in 2001 were in former homelands, and 91 per cent were formerly African schools. (p. 92) | SASCE 2001 candidate and school data compiled by Paterson & Arends (Table 4). | Systematic data |
| Enterprise training under the post-1994 NSDS remains low, concentrated in large firms and racially and gender segmented, with a mean training rate of roughly 20–30 per cent. (p. 148) | Five enterprise surveys (HSRC 2000, World Bank 2000, BMR 2002, P-E 2001), DoL levy data, Ntsika firm-size data. | Partial data |
| The NSDS faces limits from jobless growth, capacity constraints and employer treatment of the skills levy as a tax; only about 10 per cent of levy-paying firms claim grants. (p. 237) | DoL quarterly reports, McCord/Altman & Mayer macro analysis, press reports, DoL SETA evaluations. | Partial data |
| Apartheid industrial strategy (import substitution, minerals-energy complex) produced polarised skills with a neglected intermediate segment, so a hybrid high/intermediate/low skills model fits South Africa. (p. 219) | Secondary literature (Fine & Rustomjee, Altman & Meyer, Brown et al.), theoretical typology and Brown et al. export shares. | Asserted |
| FET college policy suffers from incoherence across DoE, DoL and DTI, and colleges achieve only modest labour market outcomes (34 per cent of 1999 graduates employed after two years). (p. 164) | NBI/DoE quantitative overview, Cosser tracer survey, policy document reading, SESD fieldwork interviews. | Partial data |
The author's caveats¶
Badroodien notes South Africa lacks a coherent national data collection point on training, SETA-level detail is unavailable, the studies each have unique limitations, DoL figures aggregate cumulatively across quarters and percentages lack underlying counts, and the HSRC survey covers only formal-sector firms. McGrath notes NBI data lacked cost per learner and employment rates and that colleges disputed data quality. Kraak notes that informal workers' qualifications are not captured.
Gaps in the evidence¶
The historical chapters rest mostly on a few commission reports and secondary compilations (Malherbe, Chisholm, Rickett) rather than systematic census, apprenticeship-registration or enrolment series by race, so claims about the reach and labour-market effects of training are hard to test. Firm-level panel data or SETA administrative records linking training, race, wages and later employment would test the NSDS and low-skills-equilibrium arguments; some of this exists in later Department of Labour and Statistics South Africa data but was not available to these authors. Archival evidence from employers and the Department of Manpower on actual apprenticeship intakes by race in 1970–94 would strengthen the late-apartheid claims.
Replicability: Data poorly documented.