Talking in class: Subjective indentification in South Africa

SOUTH AFRICAN SOCIAL ATTITUDES SURVEY

  
Ben Roberts examines the extent to which South Africans identify with different classes. He also investigates the relationship between subjective class, objective indicators of socio-economic status and attitudes to inequality.

This article focuses on the subjective component of measuring ‘class', meaning defining class in relation to how South Africans view themselves rather than in terms of more objective criteria such as occupational status, educational attainment or income.

In South Africa, where exceptionally high inequality in living standards persists, a sizable share of the adult population places itself in the lower class.

The data used for this study come from the 2008 round of the South African Social Attitudes Survey (SASAS). The survey consists of a nationally representative probability sample of 3 305 South African adults aged 16 years and over living in private households.

Respondents were asked the following common social class identification item: ‘People sometimes describe themselves as belonging to the working class, the middle class, or the upper or lower class. Would you describe yourself as belonging to the...?'. Respondents were offered five ordered response codes: ‘lower class', ‘working class', ‘middle class', ‘upper middle class', and ‘upper class'.

National-level class placement

Virtually all respondents were able to identify with one of the five classes, with the exception of 3% that indicated that they did not know what social class they belonged to. Using the self-rated class scale, 37% identified with the lower class, 22% with the working class, 29% with the middle class, 7% with the upper middle class, and 2% with the upper class. From a comparative perspective, this is an interesting finding. In practice, especially in developed nations, there is a tendency for the public to identify with either the middle class or working class, with marginal shares placing themselves in top or bottom class positions. In South Africa, where exceptionally high inequality in living standards persists, a sizable share of the adult population places itself in the lower class.

Although South Africans are clearly able to identify themselves as members of specific social classes, this does not appear to be a major source of their identity or group belonging.

The survey also contained an item probing for the three main sources of group-based identity people use to describe themselves. The results reveal that most South Africans deem their family or marital status as the most important form of self-identification (43%), followed by race or ethnicity (14%), current or previous occupation (12%) and gender (8%). Only 2% of South Africans identified themselves primarily in terms of their social class, and it also hardly features as a secondary or tertiary form of self-identity.

Therefore, although South Africans are clearly able to identify themselves as members of specific social classes, this does not appear to be a major source of their identity or group belonging. However, this does not mean class is inconsequential, as class position may assume significance in terms of the bearing it has on other attitudes and behaviour. This theme will be returned to later.

Attributes of subjective class

How well does subjective social class relate to some of the other objective indicators typically associated with welfare and class in South Africa, such as race, education, geographic location and income?

With regard to racial groups, less than 10% of African respondents placed themselves in the top two classes (upper middle and upper classes), compared to a third (33%) of white respondents (Figure 1). Conversely, a large share of African respondents considered themselves as being in the lower class (45%) compared to a mere 3% of white respondents.

At a time when education increasingly matters for getting ahead in life, achievement at school remains intricately connected to one's self-described class. Of those with no schooling, 78% describe themselves as lower class with barely a tenth placing themselves in the top three classes. Among those with a primary level or some secondary education, relatively large shares feel they are in the lower class (59% and 44% respectively). For those with matric, a third represent themselves as middle class, with approximately a quarter placing themselves in each of the working and lower classes. For the privileged minority with a tertiary education, 26% describe their position in the hierarchy as being upper middle or upper class, with a further 40% in the middle class.

How well does subjective social class relate to some of the other objective indicators typically associated with welfare and class in South Africa, such as race, education, geographic location and income?

Income is also tightly associated with subjective class. Those with a monthly household income of less than R1 000 are more than 50 times as likely to be in the lower class than someone with a monthly household income exceeding R10 000. At the other end of the spectrum, 29% of those with a monthly household income greater than R10 000 place themselves in the upper middle or upper classes, compared to 4% of those with less than R1 000. A similar gradient was found with regard to the Living Standards Measure, which is a wealth index based on household assets. People with low living standards were found to be much more likely to identify with lower classes than those with medium and high living standards.

Residential location also informs the social classes people belong to. In both formal metropolitan areas and other urban localities, a smaller than average proportion of residents classify themselves as lower class, with larger than average shares placing themselves in the middle and working classes. More than half of residents in informal settlements and rural areas self-identify with the lower class. In the case of residents of informal settlements in metropolitan areas, a staggering 62% declared membership of the lower class.

Subjective class and attitudes

Does class position matter in terms of the shaping effect it has on attitudes and behaviour? A full examination is beyond the scope of this article. Instead, select variables were chosen, touching on personal well-being and attitudes towards inequality.

Subjective class exhibits a positive association with subjective well-being, as measured using a conventional 5-point satisfaction with life-as-a-whole scale (Table 1). Therefore, the higher the self-described class, the greater is the likelihood of reporting higher life satisfaction. In 2008, three-quarters (77%) of South Africans that identified themselves as upper middle or upper class expressed satisfaction with life. By comparison, life satisfaction among the self-identified lower class was on average 50 percentage points lower (26%). The pattern is much weaker in relation to retrospective and future evaluations of life improvements.

Similar class-based differences are observed in relation to specific forms of redistribution, with only 45% of those in the upper middle and upper classes believing that there should be preferential hiring and promotion of black South Africans, in contrast with 77% of the lower class.

Several attitudinal items are included in the survey on inequality and redistribution. The statement ‘income differences in South Africa are too large' produced responses demonstrating a high level of dissatisfaction with the perceived level of income inequality in the country (Table 2). An estimated 84% of respondents ‘strongly agreed' or ‘agreed' that incomes are too unequal. South Africans therefore appear to be generally intolerant of the level of inequality in the country, an attitude that seems consistent across subjective classes.

These low levels of tolerance of inequality appear to have produced a relatively strong demand for governmental redistribution, with two-thirds (65%) of respondents believing that government should be responsible for reducing income inequality. Support for this sentiment is more strongly felt by the lower class (76%) with those in the upper middle and upper classes expressing a supportive but less favourable attitude (50%).

Similar class-based differences are observed in relation to specific forms of redistribution, with only 45% of those in the upper middle and upper classes believing that there should be preferential hiring and promotion of black South Africans, in contrast with 77% of the lower class.

Concluding reflections

This exploratory mapping of the subjective class structure in the country has illustrated that South Africans are able and willing to describe themselves in terms of class membership, even if most do not see this as a fundamental part of their self-identity. That a significant proportion of South Africans locate themselves in the lowest class is unsurprising given the burden of old and new forms of inequality and deprivation that are faced by a sizable contingent in many facets of their lives.

The results from the initial analysis presented here tend to reaffirm the view that class does matter. Subjective class is distinctly associated with household income, living standards, and subjective well-being. While class based-differences seem to diminish in relation to intolerance for income inequality in our society, they reappear with regard to levels of support for redistributive measures such as affirmative action.

Future analysis will need to focus on a more encompassing set of socio-economic and political attitudes and behaviours. Doing so will allow for a nuanced account of commonalities and differences between self-described classes, while discerning levels of support for specific social policies, such as whether a class-based redress agenda focusing on poor South Africans would be better favoured than a race-based affirmative action agenda. Considerations such as these are important as the 2009 round of SASAS goes to field, including as it does a specialised, in-depth international module on attitudes to social inequality.

Benjamin Roberts is a research specialist in the programme on Child, Youth, Family and Social Development (CYFSD).