The emergent middle class in contemporary South Africa: Examining and comparing rival approaches

Figures & data

Table 1: Characteristics of classes when categorised based on occupational skill levels, 2008

Class	Share of total Population (%)	Share of black population (%)	Share of white population (%)	Black share of each class (%)	Mean Income (Rands)	Mean Age (Years)	Mean educational attainment (Years)
HH not economically active	14.1	15	14.4	84.4	1060	43.5	7.2
Unemployed	16.1	18.5	3.3	91.2	551	33.4	8.9
Working & occ info missing	25.1	26.1	23.8	80.2	1803	39.3	8.7
Low skilled	10.1	10.9	1.8	86.3	902	38.4	7.1
Medium skilled	21.7	20.8	20.2	76.1	2001	36.0	10.0
Highly skilled	13.0	8.8	36.6	54.0	4919	39.0	12.8

Source: NIDS 2008. Household classifications are based on the highest skill level within the household. The mean age and education attainment relates to the household member with the highest occupational skill level.

Table 2: Characteristics of classes when categorised based on vulnerability approach, 2008

Class	Share of total population (%)	Share of black population (%)	Share of white population (%)	Black share of each class (%)	Mean income (Rands)	Mean age	Mean educational attainment (years)
Lower class	22.6	28.3	–	98.1	323	38.1	6.4
Vulnerable class	37.5	45.5	0.2	95.2	639	38.2	7.5
Middle class	26.8	23.8	22.6	69.7	1 925	36.5	9.9
Upper class	13.1	2.4	77.3	14.6	7 308	43.0	12.7

Source: NIDS 2008 and 2010. Based on the predicted probability of a household to be non-poor in 2010, households are classified as lower class (probability <10%); vulnerable (10 to 50%); middle class (50 to 90%) or upper class (>90%). The mean age and education attainment relate to the household member with the highest occupational skill level.

Table 3: Characteristics of classes when categorised using the income polarisation approach, 2008

Class	Share of total population (%)	Share of black population (%)	Share of white population (%)	Black share of each class (%)	Mean income (Rands)	Mean age	Mean educational attainment (years)
Lower class	60.6	70.7	4.7	92.2	376	37.7	7.6
Middle class	28.5	25.2	31.7	69.8	1 763	38.4	9.6
Upper class	10.9	4.1	63.7	29.4	9 574	43.1	12.5

Source: NIDS 2008. The classification of classes is based on the polarisation approach as described in Esteban et al. (1999). The mean age and education attainment relates to the household member with the highest occupational skill level.

Table 4: Characteristics of classes when categorised using self-reported assessment, 2008

Class	Share of total population (%)	Share of black population (%)	Share of white population (%)	Black share of each class (%)	Mean income (Rands)	Mean age	Mean educational attainment (years)
Lower class	33.9	37.5	13.3	87.7	771	38.4	7.6
Middle class	58.6	56.3	74.2	75.9	1 922	38.7	10.0
Upper class	7.5	6.2	12.5	65.2	4 434	38.6	11.3

Source: NIDS 2008. Households identifying themselves as standing on ladder rungs one and two were classified as lower class, those on ladder rungs three and four were classified as middle class and those on the top two ladder rungs were classified as upper class. The highest identified ladder rung was used to classify the household. The mean age and education attainment relates to the household member with the highest occupational skill level.

Table 5: Self-reported social class in South Africa, 1995, 2001, 2006 and 2013

Social class	1995 (%)	2001 (%)	2006 (%)	2013 (%)
Lower class	42.5	26.4	43.9	44.9
Working class	24.1	30.8	18.9	24.9
Lower middle class	16.6	21.0	19.5	17.3
Upper middle class	15.4	18.5	15.2	11.5
Upper class	1.4	3.2	2.4	1.4

Source: World values survey 1995, 2001, 2006 and 2013.

Figure 1: Comparing trends in rival approaches for 1993, 2008, 2010 and 2012 (or closest years)

Notes: NIDS 2008, 2010 and 2012; and SA World Values Survey for 1995, 2006 and 2013. We report interpolated WVS values for comparison with the NIDS 2010 estimates.

Esteban, J, Gradín, C & Ray, D, 1999. Extensions of the measure of Polarization, with an application to the income distribution of five OECD countries. Working Paper No. 218, Maxwell School of Citizenship and Public Affairs, Syracuse University, New York.

 Google Scholar