Southern African Regional Poverty Network (SARPN) SARPN thematic photo
Country analysis > South Africa Last update: 2008-12-17  

Statistics South Africa

Income and expenditure of households 2005/2006: Analysis of results

Statistics South Africa


SARPN acknowledges the Statistics South Africa website as the source of this report - the website should be consulted for a fuller set of data.
Copies may be ordered from:
[Download complete version - 353Kb ~ 2 min (45 pages)]     [ Share with a friend  ]

Main findings

  1. Highlights

    • Expenditure on housing, transport and food continues to dominate household consumption in the South African economy (close to 60% of the total).

    • Transport (20% of total) grew fastest as a result of accelerated purchases of motor vehicles in 2004– 2006.

    • There was a substantial reduction in expenditure on food in relation to total consumption from 2000 to 2005/2006.

    • Real income per capita increased in all income deciles from 2000 to 2005/2006.

    • Ten percent of the total population continues to receive over half of household income (based on income from work and social grants).

    • Social grants play an increasingly important role in reducing inequality as they are a major source of income for the poor.

  2. Consumption expenditure patterns

    With two notable exceptions, expenditure shares in total consumption in South Africa are broadly similar for the three survey periods – 1995, 2000 and 2005/2006. The two significant exceptions are transport (substantially higher share in 2005/2006) and food and non-alcoholic beverages (substantially lower share in 2005/2006). Both of these should be seen in the context of the economic growth and rise in incomes that South Africa experienced between 2000 and 2005/2006.

    While the rand values of household expenditure on both food and transport tend to increase as income increases, they typically move in opposite directions when measured as a proportion of total consumption expenditure. When incomes increase food expenditure tends to fall as proportion of the total and transport expenditure tends to rise as a proportion of the total.

    In the case of food expenditure as a proportion of total consumption, the decline from 2000 to 2005/2006 was evident across all income groups, across all population groups, and across all provinces. This report discusses the changes in measurement methodology which contributed to the magnitude of the decline. However, within the food category itself the percentage contributions of the main groupings (e.g. meat, vegetables) were relatively stable over the three surveys.

    In the case of transport expenditure the main contributor to its much higher share in 2005/2006 was the purchase of motor vehicles. These purchases increased rapidly in the period 2004–2006. As the economy grew in real per capita terms and employment increased, many more households acquired their own vehicles and found the means to finance them. The increase in transport’s proportion was evident across all income groups, across all population groups, and across all provinces.

    Apart from transport there were also sizeable increases in expenditure on communication (e.g. telephone equipment and services, postal services), recreation and culture, insurance, and financial services (e.g. banking costs) as a proportion of total consumption.

  3. Consumption expenditure by income group, population group and province

    As expected the poorest 20% of households allocated a considerably higher proportion of their expenditure to food and non-alcoholic beverages and clothing and footwear than the richest 20% of households. The magnitude of the spread, however, is a matter of social concern. The former allocated approximately 37% of their consumption expenditure to food and non-alcoholic beverages while the latter allocated approximately 10%.1

    The richest 20% of households allocated at least twice as much of their relative expenditure to recreation and culture than the poorest 20%.

    The different spending patterns of the different population groups are largely explained by the persisting income differences among them. Black African households allocated a considerably higher proportion of their expenditure to food and non-alcoholic beverages and clothing and footwear than white households, and a considerably lower proportion of their expenditure to transport, recreation and culture and miscellaneous goods and services than white households (insurance constituted over half of the miscellaneous category in 2005/2006). In these respects, the results of the three surveys are remarkably similar.

    Expenditure by province reveals substantially different spending patterns among provinces. As expected inter-provincial differences in income appear to play an important explanatory role. Expenditure on food and non-alcoholic beverages (as a proportion of consumption expenditure) was highest in the low-income provinces of Limpopo and Eastern Cape, and lowest in the high-income provinces of Gauteng and Western Cape. This pattern is consistent with the analysis of expenditure by income group, and a similar pattern is evident in the case of clothing and footwear.

  4. Share of consumption expenditure by population group

    From 2000 to 2005/2006, black African households’ share of consumption expenditure rose from 42,9% to 44,3% (and their share of the population rose from 78,3% to 79,4%). White households’ share of consumption expenditure fell from 44,1% in 2000 to 42,9% in 2005/2006 (and their share of the population fell from 10,1% to 9,2%). Black African households experienced increases in their share of expenditure in each expenditure category except for miscellaneous goods and services, with the largest percentage point increases occurring in:
    • food and non-alcoholic beverages (+8 percentage points);
    • furnishings, household equipment and maintenance (+10,7 percentage points); and
    • communication (+9,7 percentage points).

    There was little change in the shares of the coloured and Indian/Asian population groups (share of expenditure as well as share of population).

  5. Income and inequality

    Mean real per capita income increased in all income deciles between 2000 and 2005/2006 (based on income excluding imputed rent). But the increases were uneven, with above-average increases occurring in deciles 1, 2, 3 and 10, and below-average increases occurring in deciles 4 to 9 (where decile 1 refers to the 10% of the population with the lowest income and decile 10 refers to the 10% of the population with the highest income).

    Of the various sources from which households derive their income, the largest by far continues to be income from work, including employment, self-employment and business income. The importance of grants as a source of income among lower-income households is increasingly significant.

    While 10% of the population continues to earn more that 50% of household income in the country, the poorest 40% of the population accounts for less than 7% of household income, with the poorest 20% accounting for less than 1,5% of income (based on income from work and social grants).

    Inequality continues to remain high between population groups and within individual population groups. The Gini coefficient based on disposable income (from work and social grants) for the whole country was 0,72. Within individual groups the Gini coefficient was highest at 0,63 among black African households, with the other population groups ranging between 0,56 and 0,59. If social grants and taxes are excluded, the Gini coefficient for the whole country would be 0,80 rather than 0,72, i.e. the reduction of inequality through redistributive policies reduces the Gini coefficient by 8 percentage points.

  1. As explained in the report, these percentages exclude imputed rent.

Octoplus Information Solutions Top of page | Home | Contact SARPN | Feedback | Disclaimer Get Adobe Reader