Labor Markets during Apartheid in South Africa¶
Martine Mariotti (2009). Labor Markets during Apartheid in South Africa. Working Papers in Economics & Econometrics, No. 503, School of Economics, College of Business and Economics, Australian National University. Australian National University.
| Type | working paper |
| Scope | Economics and economic history of apartheid |
| Method | mixed |
| Period studied | 1948–1985 |
| Regions | South Africa |
| Themes | origins segregation, labour markets wages, unions strikes, education human capital, manufacturing industrial policy, business and state, politics ideology elections, methods data sources |
| Citation key | mariotti2009 |
Argument¶
The paper argues that the desegregation of semi-skilled jobs in the late 1960s and early 1970s was driven by white workers' own economic interests, not by political pressure, capital or state intervention. Apartheid education raised white attainment, especially among Afrikaans speakers, who moved into skilled jobs. Skilled whites then used the flexibility of the Industrial Conciliation Act (1956) to open semi-skilled jobs to Africans so as to keep their skill premium. A two-sector CES model, census and manpower-survey data and manufacturing-census regressions are used to show that desegregation did not track output growth, wages, capital intensity or African education. The author suggests that this white-led change may unintentionally have helped set apartheid's unravelling in motion, for example the 1973 labour unrest.
Evidence and method¶
Individual-level 1970 and 1980 population censuses; Manpower Surveys (1969-85); Manufacturing Censuses (1950-85, 23 industries); the Ganzeboom, De Graaf and Treiman socio-economic occupation (SEO) index; labour agreements as reported by Horrell; Rand Daily Mail 1973; SAIRR surveys.
Place in the debates¶
- What caused the relaxation of job reservation / desegregation of the workplace (political pressure, capital, state, or white workers) (see The job colour bar, white labour and its erosion). Argues that white workers' economic incentives were the primary cause. It rejects explanations based on political pressure, capitalist pressure or state intervention, and rejects changes in African characteristics.
- Becker vs Heckman on whether discrimination is competed away or sustained by collective action (see The job colour bar, white labour and its erosion). Sides partly with Heckman: white workers did act collectively, but their preference for higher incomes overrode their preference for segregation, even with explicit state support for segregation.
- What ended apartheid (international and domestic political opposition vs earlier economic forces) (see Why apartheid ended: reform, transition and elite bargaining). Challenges the conventional view that mid-1970s and 1980s pressure explains the unravelling. It argues that labour-market desegregation began earlier, at the height of apartheid, for economic reasons.
- Capital and apartheid (liberal view that apartheid was uneconomic vs political power of capital) (see Liberal vs revisionist: was apartheid functional or dysfunctional for capitalist growth?). Argues that political power did not rest with capitalists. Capital's returns were lower under segregation, yet transformation came late.
Findings¶
-
Rising white education reduced resistance to opening semi-skilled jobs to Africans; the change reflected economic preferences, not softer attitudes. (p. 2)
Increased educational attainment among whites reduced resistance to opening semi-skilled jobs to Africans.
-
The Industrial Conciliation Act's segregated unions gave white unions the power to negotiate job reservation. (p. 6)
The consequence of this trade union segregation was that white unions were able to negotiate the exclusion of Africans from occupations that attracted large numbers of whites, a process known as job reservation.
-
Job reservation was renegotiated in industry agreements as white supply declined, e.g. in the 1967 metal and engineering agreement. (p. 7)
In 1967 the trade unions agreed to classify some of the high demand occupations as lower-skilled so that they could be opened up to non-whites (Horrell 1969).
-
Africans and whites were still far from proportional representation in occupations, but African access grew. (p. 15)
Column 1 of table 5 shows that in 1969 only 12.5 % of the occupations listed in the industry-wide Manpower survey employed Africans and whites in a proportion equivalent to that in the population.
-
African access to higher-skilled occupations rose between 1969 and 1985. (p. 16)
access to higher skilled occupations as demonstrated by the increase in the weighted average African SEO skill rank from 27.22 in 1969 to 34.07 in 1985.
-
The white educational advantage over Africans was extreme and persisted, protecting skilled whites. (p. 14)
whereas 85% of Africans had never been enrolled in high school (see Table 1).
-
The occupational rank of white males rose, consistent with upward movement of whites into skilled work. (p. 15)
I find that the weighted average rank for white males was 46 in 1970 and 49 in 1980
-
The white skill premium is judged to have been maintained over 1970-80, which fits the claim that skilled whites enabled desegregation to protect it. (p. 17)
therefore remained constant in the ten year period, consistent with the claim that skilled whites enabled desegregation in order to maintain the skill premium.
-
Manufacturing regression: a post-1970 shift in the growth of the African share of production workers, not explained by output, capital-labour ratio or wages. (p. 18)
In contrast, the coefficient on the dummy variable for 1970 suggests an increase in the growth rate of 6 % from 1970.
-
African education explains little of the rise in African occupational rank. (p. 19)
I find that education is responsible for only 22% of the change in African occupation rank using the 1970 coefficients.
-
Oaxaca-Blinder decomposition: most of the African-white occupational gap is explained, which points to pre-labour-market discrimination in education. (p. 19)
discrimination accounted for only 15.8 % of the gap in 1980, down from 26.5 % in 1970.
-
The change in white behaviour was economic rather than ideological. (p. 20)
Changes in white characteristics do not imply a change in racial ideology; rather, white workers began to be affected by the economic constraints imposed by apartheid.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| South African Population Census 1970 (individual-level micro data; separate census for whites/Coloureds/Indians and a shorter one for Africans) | population census | 1970 | Documents white and African educational attainment by birth cohort, occupational rank and income, and supports regressions of white-collar premium and African occupation rank. |
| South African Population Census 1980 (individual-level micro data) | population census | 1980 | Provides the 1980 comparison point for occupational rank, incomes and education, used in the stacked 1970/1980 wage regressions and the African occupation-rank regressions. |
| South African Manpower Surveys (biennial 1969-1985 except 1979), 271 occupations by race and gender | official statistical series | 1969-1985 | Shows the growing share of occupations with proportional African employment and the rising weighted African skill rank, evidencing desegregation of semi-skilled jobs. |
| South African Manufacturing Census 1950-1961, 1963, 1965, 1967, 1970, 1972, 1976, 1979, 1982, 1985 | industrial mining census | 1950-1985 | Industry-level panel of employment by race, wage bills, capital and output used in the growth regression (Table 7) testing whether desegregation was associated with economic conditions. |
| Ganzeboom, De Graaf and Treiman (1992) Standard International Socio-Economic Index of Occupational Status (SEO index) | secondary literature compilation | 1992 (index based on data from 16 countries) | Supplies occupation skill ranks assigned to census and manpower-survey occupations to measure skill level and occupational advancement. |
| Industrial Conciliation Act of 1956 (incl. Section 77 job reservation) | legislation gazette | 1956; Section 77 implemented 1959 | Establishes the institutional mechanism by which white unions and employers negotiated job reservation, and anchors the time dummies in the regression. |
| Industry labor agreements, e.g. 1967 metal and engineering industry agreement (as reported by Horrell 1969) | employer association records | 1967 and late 1960s | Illustrates the renegotiation of job reservation and reclassification of occupations as lower-skilled so they could be opened to non-whites. |
| Rand Daily Mail (January-May 1973) | newspapers periodicals | January-May 1973 | Listed in references, apparently for the January 1973 labor unrest mentioned in the conclusion; use is not explained in the text. |
| South African Institute of Race Relations Annual Surveys 1967-1973 | secondary literature compilation | 1967-1973 | Listed in references as background on race policy and labor developments; no specific use in the text. |
| Central Statistical Service, South African Statistics (biennial 1970-1990); SA Reserve Bank Quarterly Bulletin; Mid-year Population Figures 1980 | official statistical series | 1970-1990 | Listed as references for aggregate and population statistics (e.g. racial population shares); specific use is not stated. |
| Ballinger (1969) population shares; Abasebenzi 1973-1975 journal | secondary literature compilation | 1969; 1973-1975 | Ballinger is used for the African population share (80.7% non-white); Abasebenzi is listed but its use is not described. |
Inference¶
Quantification: Statistical analysis.
A simple theoretical CES model of white skill-premium maintenance is combined with descriptive census tables and OLS regressions (white-collar wage premium 1970/1980, industry panel growth regression with time dummies, African occupation rank on education plus an Oaxaca-Blinder decomposition) to rule out alternative explanations; no causal identification, so inference rests on timing, comparison and elimination of alternatives.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| White educational attainment, especially among Afrikaans speakers, rose sharply across cohorts and converged toward English speakers by 1970, while African education improved only modestly. (p. 13) | 1970 population census education by birth cohort (Tables 1-3). | Systematic data |
| Whites moved into higher-ranked occupations between 1970 and 1980, with Afrikaans speakers converging toward English speakers. (p. 14) | 1970 and 1980 census occupations mapped to the SEO index and weighted by employment (Table 4). | Systematic data |
| African access to occupations and to higher-skilled jobs rose from 1969 to 1985, indicating desegregation of semi-skilled occupations. (p. 15) | Manpower Surveys 1969-1985, share of occupations with African employment proportional to population and weighted SEO rank (Table 5). | Systematic data |
| Skilled whites kept their skill premium after desegregation, consistent with the argument that whites opened semi-skilled jobs to maintain it. (p. 15) | Stacked 1970/1980 census OLS of log white male income on white-collar status, education, year and interactions (Table 6); the interaction is significant only in the manufacturing specification without education. | Partial data |
| Desegregation was not driven by industry output, capital-labor ratios, wages or employment growth; only a 1970 break in the growth of the African production-worker share is notable. (p. 17) | Manufacturing census industry panel OLS with time dummies and industry fixed effects (Table 7, 327 observations). | Partial data |
| Changes in African education explain only about a quarter of the rise in African occupational rank between 1970 and 1980, so African characteristics were not the cause of desegregation. (p. 18) | 1970 and 1980 census regressions of African SEO rank on education, with the education change plugged into the coefficients (Table 8). | Partial data |
| White workers, not capital, the state or political pressure, initiated the opening of semi-skilled jobs, as shown through the IC Act's job reservation mechanism and the government's continued commitment to segregation. (p. 3) | Institutional account of the IC Act and industry agreements (Horrell), plus an argument from the government's stated commitment and capital's lower return under segregation; no direct data on negotiations. | Illustrative or anecdotal |
The author's caveats¶
Notes that the 1970 African census had far fewer questions and no income data, that occupation, industry and education codings had to be harmonized and corrected between 1970 and 1980, that incomes are categorical (midpoints used), that the 1958 manufacturing observation is excluded for a data discontinuity, and that comparing African and white skill-rank changes is difficult. Acknowledges that census data cannot fully characterize white progression and that individual labor contracts would give more detail.
Gaps in the evidence¶
The central claim that white workers drove job reservation changes is not tested with direct evidence, such as the individual industrial council agreements, union minutes or negotiation records, which the author defers to future work and which exist in archives. There is no data on wages by detailed occupation across the 1960s, since the censuses start only in 1970, so the pre-1970 timing of educational shifts and the skill premium is inferred; the regressions are time-trend and dummy comparisons that cannot separate white supply shifts from employer or state pressure. Establishment-level or firm records on employer behavior, and fuller manpower survey microdata, would give a better test.
Replicability: Data described but not public.