Race, class and inequality in the South African City¶
Jeremy Seekings (2010). Race, class and inequality in the South African City. CSSR Working Paper No. 283 (to appear in Bridge and Watson (eds), The Blackwell Companion to the City). Centre for Social Science Research, University of Cape Town.
| Type | working paper |
| Scope | Wider apartheid literature |
| Method | synthesis review |
| Period studied | c.1950–2008 |
| Regions | Cape Town, Johannesburg, South African metropolitan areas, bantustans |
| Themes | origins segregation, bantustans removals influx, demography urbanisation, inequality poverty unemployment, transition post apartheid, politics ideology elections |
| Citation key | seekings2010 |
Argument¶
The apartheid city was an extreme case of urban social engineering. Statutory racial classification, Group Areas removals and influx control produced a caste-like hierarchy within which class positions were formed. Segregation also served the interests of mining houses, commercial farmers and white workers. After 1994 the class structure changed little, and inequality persisted. Racial deracialisation was confined to the upper and middle classes, while most of the city stayed segregated, partly for economic reasons and partly because of housing allocation and social preference. Seekings rejects the critical view that neo-liberalism has reproduced apartheid-style inequality in the city. He argues that redistributive financing and social grants have meant rising decommodification for poor urban residents.
Evidence and method¶
Synthesis of secondary literature (Christopher, Crankshaw, Besteman, McDonald, Lemanski and others). Class-structure figures come from Seekings and Nattrass (2005) for 1993 and from the first wave of the National Income Dynamics Study for 2008. Cape Town census and survey figures on housing and refuse removal, schematic maps of Cape Town, and a novel (Matlwa's Coconut) used as illustration.
Place in the debates¶
- Is the post-apartheid city a 'neo-apartheid' city, with neo-liberalism reproducing apartheid-era inequality? (see Urban segregation, Group Areas and the apartheid city). Partly rejects it. Segregation and class inequality persist, but the neo-liberal critique ignores redistributive financing, improved services and grants, and underestimates desegregation in richer areas.
- Economic versus racial/social explanations of persistent residential segregation (see Urban segregation, Group Areas and the apartheid city). Economic barriers and white racism matter, but they are insufficient. Segregation also persists among equally poor African and coloured groups, through public housing allocation to informal-settlement 'communities' and through social preference.
- Does caste or class structure South African urban inequality? (see Urban segregation, Group Areas and the apartheid city). Under apartheid caste shaped the range of class positions and class shaped the caste system. After 1994 class is increasingly decoupled from race at the top, while the lower classes remain entirely African.
Findings¶
-
Group Areas removals displaced hundreds of thousands, especially in the 1960s. (p. 3)
Hundreds of thousands of people were forcibly removed, especially in the 1960s, into racially-designated
-
Apartheid urban inequality was caste-like, with race limiting class position. (p. 5)
inequality in South African cities took on a caste-like form, with largely impermeable boundaries between castes.
-
Segregation served the interests of particular capital fractions and white workers, so class shaped the caste system. (p. 5)
some aspects of racial discrimination and segregation served the interests of some sections of the bourgeoisie (notably the gold-mining houses and many commercial farmers)
-
Urbanisation was displaced into bantustan slums, which grew explosively. (p. 6)
whose population grew from 24,000 in 1970 to (officially) 342,000 in 1991
-
Extreme inequality meant the poverty gap was small relative to national income. (p. 7)
A perfectly targeted transfer of only 5 percent of national income from rich to poor would have sufficed to eliminate poverty, even using the higher South African poverty line set above US$1 per person day.
-
On the eve of the 1994 election a small upper class held a highly disproportionate share of income. (p. 9)
In 1993, on the eve of the first democratic elections, the upper classes accounted for 12 percent of the total population of households but 45 percent of the national income.
-
Inter-class income gaps narrowed somewhat by 2008 but remained large. (p. 9)
households in the upper category have, on average, incomes three and a half times larger than households in the middle category, and approximately eight times larger than households in the lower category.
-
Mass unemployment, especially of an underclass, persists at a very high level. (p. 10)
with the unemployment rate standing at about 35 percent, down from a peak of 40 percent in the early 2000s
-
Deracialisation is confined to the upper classes; the bottom remains entirely African. (p. 10)
The lower cluster of classes remains entirely African, but the higher cluster of classes has continued to diversify in racial terms.
-
Early studies underestimated desegregation in richer Johannesburg neighbourhoods. (p. 13)
somewhere between one-quarter and one-third of the residents (excluding domestic workers)
-
Service delivery to poor African households in Cape Town expanded sharply after 1994. (p. 16)
Over this period the actual number of African households with weekly refuse removal doubled, even though the proportion only rose by about 20 percentage points.
-
Social assistance is the main channel of post-1994 decommodification and redistribution. (p. 16)
After 1994 the ANC government expanded this system, such that it entailed the redistribution of approximately 3.5 percent of national income by the late 2000s.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| 1993 household income data as analysed in Seekings and Nattrass (2005), Class, Race and Inequality in South Africa (class structure of 1993; poverty and inequality measures) | secondary literature compilation | 1993-1994 | Supplies the 1993 class structure shares of households and income and the poverty and poverty-gap figures at the transition. |
| National Income Dynamics Study (NIDS), wave 1, 2008 | household or labour survey microdata | 2008 | Used to calculate the 2008 class structure (Figure 2), comparing household and income shares of upper, middle and lower classes with 1993. |
| Population census occupational data, 1980-2001, as analysed by Borel-Saladin and Crankshaw (2009) for Cape Town | secondary literature compilation | 1980-2001 | Shows deindustrialisation and a shift from middle-income artisan and operator jobs to clerical, sales and service jobs, which mutes polarisation. |
| Crankshaw (2008) analysis of Johannesburg residential and class composition, 2001 | secondary literature compilation | 2001 | Evidence that desegregation in richer Johannesburg neighbourhoods was larger than early studies suggested, and that the middle class was evenly divided by race. |
| Christopher (2001, 2005) studies of post-apartheid urban segregation (1996-2001 census-based) | secondary literature compilation | 1991-2001 | Establishes that post-apartheid desegregation was very slow and the city remains mono-racial in most areas. |
| Phuthaditjaba population figures 1970 and 1991 (via Nattrass, 2000) | secondary literature compilation | 1970-1991 | Illustrates displaced urbanisation into bantustan slums. |
| Government social assistance (grant) coverage and expenditure figures, late 2000s | official statistical series | late 2000s | Supports the claim that social grants redistribute about 3.5 percent of national income and reach about half of metropolitan households. |
| Post-apartheid poverty and inequality trend studies (Van der Berg et al. 2008; Leibbrandt et al. 2009; Seekings 2009) | secondary literature compilation | 1994-2007 | Supports the claim that poverty and inequality rose after 1994 and that poverty declined in the early 2000s because of welfare expansion. |
| ANC (1994) Reconstruction and Development Programme; 1996 Constitution; Government reviews (South Africa 2003, 2007) | commission or official report | 1994-2007 | Documents the stated post-1994 policy commitments to poverty reduction and the government's own assessment of progress. |
Inference¶
Quantification: Descriptive tables.
Synthetic review essay. It draws on the author's earlier household class analysis (1993) and on 2008 NIDS data, compares class and income shares across 1993 and 2008, and cites secondary urban studies, maps of Cape Town and a few service-delivery statistics to argue a contested interpretation against the neo-liberal critique.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| The apartheid city was a caste-like system of racial segregation, with forced removals and influx control, that also served the interests of mining houses, commercial farmers and white workers. (p. 5) | Secondary literature (Western, Davies, Lemon) and the schematic maps of Cape Town; no primary data presented. | Asserted |
| At the transition, income was extremely unequal: the top decile earned about half of national income and a 5 percent transfer could have eliminated poverty. (p. 5) | 1993 household income data as analysed in Seekings and Nattrass (2005). | Partial data |
| The class structure changed only modestly between 1993 and 2008: the upper classes held their position, the middle shrank, the lower grew, and inter-class income gaps narrowed but remain large. (p. 9) | Comparison of 1993 data with NIDS wave 1 (2008), which the author acknowledges are not precisely comparable. | Partial data |
| Racial segregation persists in most of the post-apartheid city and is reproduced in newly built neighbourhoods, with exceptions such as inner Johannesburg and Delft. (p. 9) | Christopher's studies, Cape Town case studies and the schematic Figure 3; no systematic segregation indices. | Illustrative or anecdotal |
| Economic factors do not fully explain segregation; public housing allocation by informal settlement 'community' and social preference also reproduce racial separation. (p. 12) | Case studies (Besteman, Lemanski, Crankshaw) and the author's reasoning on housing allocation; no data. | Asserted |
| Access to services by the poor improved after 1994 and rests on redistributive financing, so the neo-liberal critique overlooks substantial decommodification. (p. 14) | Cape Town 1996-2003 household service percentages and census household counts, plus assertions about municipal financing. | Partial data |
| Social assistance is the most important element of decommodification, redistributing about 3.5 percent of national income, and reaches about half of metropolitan households. (p. 14) | Unreferenced government grant data and survey data; no source cited at this point. | Asserted |
The author's caveats¶
The author notes that delineating class structure is not straightforward given the limits of the available data and theory. He also states that the 1993 and 2008 data are not precisely comparable.
Gaps in the evidence¶
The argument leans on author-authored earlier work and secondary case studies rather than systematic segregation measures. Census-based dissimilarity indices at suburb or sub-place level, drawn from the 1996, 2001 and 2011 censuses, would test the persistence claims and the claims about desegregation in richer areas. Municipal finance and tariff data are needed to test the redistributive-financing and service-delivery claims. Comparable 1993 PSLSD and 2008 NIDS class coding would sharpen the change-over-time claims, and both datasets exist.
Replicability: Data described but not public.