South Africa's Segmented Labour Markets: Skill Formation and Occupational Mobility Under Apartheid, 1979-1993¶

Literature › Works

Andre Kraak (1995). South Africa's Segmented Labour Markets: Skill Formation and Occupational Mobility Under Apartheid, 1979-1993. Work, Employment & Society, Vol. 9, No. 4. SAGE / British Sociological Association.


Type	journal article
Scope	Economics and economic history of apartheid
Method	synthesis review
Period studied	1970–1993
Regions	South Africa, Bantustans/homelands, Pretoria-Witwatersrand-Vaal (PWV), urban metropolitan areas
Themes	labour markets wages, education human capital, business and state, unions strikes, inequality poverty unemployment, bantustans removals influx, transition post apartheid, comparative theory, manufacturing industrial policy
Citation key	`kraak1995`

Argument¶

Kraak argues that South African labour markets cannot be explained by neoclassical theory. He proposes a 'macro-institutional' model that combines American segmented-labour-market theory, the Cambridge internal/external labour market school and post-Fordist insights. The late-apartheid reforms of 1979-93 (Wiehahn, Riekert, De Kock, De Lange) diluted statutory racial barriers and turned a mass of unskilled Africans into semi-skilled operatives in a unionised 'core'. This resegmented the labour market rather than dissolving it. At the micro level, racial exclusion persisted in new forms: word-of-mouth recruitment through White-dominated 'extended internal labour markets', White worker and middle-management resistance, 'qualification inflation', and ideological 'work ethic' training. He warns that post-apartheid Affirmative Action and export-oriented post-Fordist growth strategies risk deepening segmentation between a skilled core and a peripheral majority.

Evidence and method¶

Secondary literature (Wolpe, Legassick, Hindson, Webster, Sitas, Human and Hofmeyr, Franks, Lundall and Kimmie), official statistics (South African Statistics 1980, 1982 and 1993; NATED education statistics), the NTB/HSRC 1985 apprenticeship recruitment survey, state commission and President's Council reports, COSATU and NUM documents, and the author's own earlier studies of metal-industry skills and training courses.

Place in the debates¶

Neoclassical/human-capital vs structural accounts of labour market discrimination (see Racial wage gaps, wage convergence and wage determination). Rejects the neoclassical view (e.g. Human and Greenacre) that discrimination is a market distortion caused by racist policy; sees segmentation as a structural feature of South African capitalism shaped by state, capital and labour strategies.
Cheap labour thesis (Wolpe/Legassick) and its critics (see Cheap labour thesis and the function of the reserves). Accepts its insight on cheap African labour and the reserves but criticises it as treating capitalist transition monolithically. Follows Hindson, Morris, Kaplan and O'Meara in stressing differentiated labour power, with an urban stabilised proletariat alongside migrants.
Did the 1979-80 reforms end racial discrimination in the labour market? (see Racial wage gaps, wage convergence and wage determination). Reform diluted racial occupational barriers and created a unionised African semi-skilled core, but discrimination was reproduced through informal and micro-level mechanisms and has had 'little effect' on the racial occupational structure at the top.
Post-apartheid policy: Affirmative Action and export-led post-Fordist growth (see Post-apartheid economic policy, governance and corruption). Sceptical of voluntarist Affirmative Action and of export-led growth. Favours COSATU-style 'active' labour market policies and democratisation to avoid a new core/periphery divide.

Findings¶

Wolpe and Legassick's cheap labour thesis is summarised as the Marxist structuralist baseline for labour market analysis. (p. 7)

Their basic thesis was that capitalist expansion in its initial phase was totally dependent on the creation and maintenance of a supply of cheap African labour power
The Wiehahn Commission (1979) recommended dismantling job reservation, indenturing African apprentices and recognising black unions. (p. 9)

recommended the abolition of almost all forms of job reservation, the legalisation of the indenturing of African apprentices (something previously prohibited), and the recognition of black trade unions
The state implemented the four reform commissions' market-oriented recommendations, relinquishing some economic control. (p. 10)

The Apartheid state accepted most of the recommendations of these four Commissions of Inquiry.
Africans moved into semi-skilled operative jobs in manufacturing from the late 1960s and 1970s, reversing earlier racial shares in the metal industry. (p. 10)

by 1977, Africans constituted 67 per cent of all operative labour in the metal industry while Whites occupied 16.4 per cent of semi-skilled jobs.
By the late 1980s African semi-skilled operatives had become the largest stratum of the African working class (citing Hindson). (p. 12)

by the late 1980s the number of African operative workers exceeded two million
African educational attainment rose markedly in 1980-91, though large inequalities remained. (p. 12)

The percentage of African people with no formal education decreased from 41.1 per cent in 1980 to 25.1 per cent in 1991.
Homeland poverty and spatial inequality remain extreme: huge per capita GDP gaps between homelands and other areas (Abedian and Standish). (p. 13)

14 million people lived in the homelands in 1985, but they generated an annual per capita GDP of only R370, compared to R6 662 and R3 927
Formal abolition of job reservation had little effect on the racial occupational structure, as new forms of exclusion emerged. (p. 15)

The abolition of the statutory mechanisms of job reservation and racial exclusion in the labour market in the early 1980s has had little effect on the racially-determined occupational structure.
African artisans increased but Whites kept the high-skill trades and most apprenticeships in the 1980s. (p. 18)

African artisans increased from 5.8 per cent of the total number of artisans in 1981 to 13.8 per cent in 1989
Whites retained most apprenticeship places in 1981-89 (citing Lundall and Kimmie). (p. 18)

Whites continued to constitute 68.6 per cent of total apprentices recruited in the period 1981-1989
African managers remained a small share until the late 1980s, when they reached 9.6 per cent. (p. 18)

It was only in the late 1980s and early 1990s that a significant increase in the employment of African managers occurred, growing to 9.6 per cent of the managerial job market.
'Qualification inflation' replaced job reservation on the mines after the 1988 repeal. (p. 20)

In reality, black miners are now required to possess standard eight certificates before entering these occupations.
Employer training aimed to reshape black workers' values towards free-market norms, which Kraak reads as an ideological reconstruction of skill. (p. 21)

I concluded that much of this training was aimed at re-orientating the values of black workers towards the free-market system.
Export-led post-Fordist growth risks a new divide between primary-market workers and the secondary sector, informal sector and unemployed. (p. 27)

It is therefore very likely that the new growth path will see the emergence of a new industrial divide

Evidence ledger¶

Sources used¶

Source	Kind	Period	How used
Department of Statistics, South African Statistics 1980; Central Statistics Services, South African Statistics 1982 and 1993 (occupational structure by race, 1970, 1980, 1991)	official statistical series	1970-1991	Table 1 shows the racial recomposition of occupational categories, with African and White shares of employment, to document resegmentation.
Central Statistics Service, South African Statistics (reports for 1982, 1990 and 1993) - managerial employment by race	official statistical series	1970-1991	Table 5 traces the number and share of African and White managers to show slow black advancement into managerial jobs until the late 1980s.
Department of National Education (1993), Education Realities in South Africa 1993, NATED Report 02-300	commission or official report	1980-1991	Table 2 gives the educational attainment distribution of Whites and Africans, showing African gains and persisting inequality.
NTB/HSRC (1985) survey of employer apprenticeship recruitment, Investigation into the Training of Artisans in South Africa	author survey or fieldwork	1985	Table 4 reports the share of employers using each recruitment method, taken as evidence of reliance on local/employee-network recruitment that favours Whites.
Lundall and Kimmie (1992), apprenticeship and artisan employment data, South African Labour Bulletin	secondary literature compilation	1981-1989	Supplies figures on African share of qualified artisans (5.8% to 13.8%) and White share of apprentices recruited (68.6%) to show limited black advancement.
Human and Greenacre (1987), survey results on income by race, gender, education and age in manufacturing	secondary literature compilation	c.1980s	Used as example of neoclassical evidence showing three income segments by race and gender.
Abedian and Standish (1992) regional and homeland income/GDP statistics (including Simkins estimates)	secondary literature compilation	1980-1987	Used to show regional and urban-rural disparities in per capita income and homeland poverty.
Hindson (1991) estimates of operative workers and urban-integrated population	secondary literature compilation	1970s-1980s	Supports claims that African operatives exceeded two million by the late 1980s and 56% of the population is integrated into the urban economy.
Reports of the Wiehahn, Riekert, De Kock and De Lange Commissions (1977-1980)	commission or official report	1977-1980	Described as the reform impetus for abolishing job reservation, recognising black unions, and expanding urban rights and black education; known via secondary sources.
President's Council (1989) Report of the Committee for Economic Affairs on a Strategy and Action Plan to Improve Productivity	commission or official report	1989	Quoted to show state emphasis on promoting free-market values and work ethic among black workers.
Walters Commission (DEC 1990), The Evaluation and Promotion of Career Education; and Education Renewal Strategy (DNE 1991a,b)	commission or official report	1990-1991	Used to show state emphasis on 'work socialisation' attributes and recurriculation along free-market lines.
National Training Board Annual Report 1990 and NTB/HSRC investigations (1986, 1989a, 1989b, 1991)	commission or official report	1986-1991	Cited for state concern with trainees' values and attitudes, and with white resistance and black marginalisation.
National Manpower Commission (1987) High Level and Middle Level Manpower in South Africa	commission or official report	1987	Cited for the claim that high-level manpower jobs were opened to Whites with low qualifications.
Newspapers: Sowetan (30 January 1990); Financial Mail (13 February 1987); Argus (11 August 1987)	newspapers periodicals	1987-1990	Evidence on racially biased apprenticeship advertising, old-boy networks, and new entry requirements for formerly reserved mining jobs.
COSATU discussion papers (1991a, 1991b) and NUM document (Hermanus 1989)	union or movement records	1989-1991	Union statements used as evidence of persisting discrimination in recruitment, training and qualification inflation on the mines.
Kraak (1991a) study of enterprise training courses (e.g. the '6-M' course)	author survey or fieldwork	c.1991	Author's own study used to claim employers seek ideological and social skills over technical competence and train workers toward free-market values.
Kraak (1987, 1988, 1989) and Sitas (1983) studies of the metal industry and skill requirements	secondary literature compilation	1960s-1980s	Supports the figures on African share of operative labour in the metal industry in 1977 and the apprentice entry requirement changes.

Inference¶

Quantification: Descriptive tables.

Largely a theoretical and synthetic argument combining segmentation and Cambridge labour market theory, illustrated by descriptive tables built from published official statistics, a few secondary studies and state/union documents; no formal statistical testing, with the author calling the micro section 'impressionistic'.

Key claims¶

Claim	Basis	Strength
Between 1970 and 1991 Africans moved into occupations previously dominated by Whites (professional, managerial, clerical and sales), diluting formal racial occupational segmentation. (p. 665)	Table 1 from South African Statistics 1980, 1982, 1993 occupational counts by race.	Systematic data
The African unskilled proletariat was transformed into semi-skilled operatives, a 'core' subordinate primary segment, mainly from the late 1960s and 1970s. (p. 665)	Metal industry figures from Kraak 1987/Sitas 1983 and Hindson 1991; state statistics too aggregated.	Partial data
African educational attainment rose sharply in the 1980s but large racial inequalities in education remained. (p. 667)	Table 2 from DNE (1993) Education Realities; Std 8 comparison 90% vs 20.5%.	Systematic data
Black advancement in artisan and managerial jobs was minimal in the 1980s because of White worker and management resistance, and African managers only grew markedly after 1985. (p. 672)	Lundall and Kimmie (1992) data, Table 5 from CSS statistics, and attitude studies by Human and Hofmeyr and Franks.	Partial data
Recruitment through employee and local networks creates a White-dominated 'extended internal labour market' that perpetuates discrimination after the end of job colour bars. (p. 672)	1985 NTB/HSRC employer survey (Table 4), plus Webster's newspaper comments and Moore's theory.	Partial data
After formal job reservation ended, employers and unions raised entry qualifications (qualification inflation) for black workers in metal and mining occupations. (p. 675)	Kraak 1988/1989, press reports (Financial Mail, Argus), NUM document (Hermanus 1989), COSATU papers.	Illustrative or anecdotal
The state and capital sought to ideologically reconstruct black workers' 'skill' toward free-market work ethics in the late apartheid era. (p. 676)	Kraak's 1991a study of training courses, NTB 1990, President's Council 1989 report, Walters Commission 1990 and ERS documents.	Partial data
The urban/rural and regional divide means resegmentation benefited only a minority 'core', with massive impoverishment in the homelands. (p. 668)	Abedian and Standish (1992) statistics, Simkins estimate and Hindson 1991.	Partial data

The author's caveats¶

The author notes that figures for semi-skilled production workers are not sufficiently disaggregated in state statistics, that tables cover only Whites and Africans, and that the micro-level account of internal/external labour markets is 'merely impressionistic'.

Gaps in the evidence¶

The argument about micro-level exclusion (extended internal labour markets, qualification inflation, ideological skill) rests on one 1985 employer survey, secondary studies, press items and union statements, with no firm-level or establishment data on recruitment, promotion or wages by race. Disaggregated census and labour force data on semi-skilled operatives, wages and mobility, and case studies from company personnel records, would test the segmentation claims more directly; some Manpower Survey and census microdata existed then. Longitudinal data on individual occupational mobility are needed to show actual movement between segments, which Table 1 stock figures cannot.

Replicability: Data described but not public.