How to square the inequality circle
South Africa has evidently made limited progress in reducing income inequality since the end of apartheid. Since 1994 the Gini coefficient increased somewhat to 0.69 in 2011 and 61.3% of aggregate consumption expenditure comes from the richest 20% of citizens.
It is crucial to determine the progressiveness of the major fiscal policy instruments to establish whether government spending and taxation, separately and together, are ameliorating or worsening the degree of inequality between individuals.
To make this determination one must ask two questions. First, who bears more of the burden of taxation and who benefits relatively more from the various forms of social spending – the poor, those in the middle or the rich? Second, what is the combined effect of taxes and spending patterns on inequality?
Method of analysis
A study by an international group of researchers assesses how taxes and social spending programmes redistribute income to groups at different income levels.
This involves comparing the income of individuals in terms of:
- Market income: income before paying taxes and receiving transfers; and
- Final income: income after taxes, transfers and social spending – in other words, after redistributive fiscal efforts.
Market income is defined as the income a person receives from wages, salaries, capital, private transfers and contributory pensions before paying taxes – and not counting benefits from government spending. Final income is derived in steps (see Figure 4).
The study interprets transfers in their broadest sense: in addition to cash transfers such as social grants, it includes an estimated monetary value of the receipt of basic services such as water and electricity, as well as the benefits of education and health spending. These services can be regarded as in-kind transfers.
The tax system is mildly progressive
With a progressive tax individuals with higher taxable incomes pay progressively higher proportions of their income in tax. For the population as a whole, a tax is progressive if the combined share of the tax paid by, for example, the poorest 10% of the population is lower than their share of the aggregate income while the tax share of the richer income groups is higher than their share of the income.
The results, reflected in Table 1, indicate, first, that direct taxes (especially personal income tax) are progressive, because richer income groups pay a proportionally higher share of total direct tax collections than their share of market income. The wealthiest 10% of individuals earn about 63.7% of total market income, but they pay 86.9% of total personal income tax. Fifty percent of the population – the poorest deciles 1 to 5 – do not earn enough to pay income tax, and only a small proportion of individuals in deciles 6 to 8 pay income tax and a very small share of total income tax.
Second, indirect taxes – VAT, excises on alcohol and tobacco and the fuel levy – are slightly regressive, notably in the bottom half of the income distribution (see Table 2). In 2010, the poorest 40% of people contributed 5% of total indirect tax collections, compared with their share of 4.8% in total disposable income.
This regressiveness of indirect taxes at the lower end of the income distribution is largely a result of the effect of excise taxes on alcohol and tobacco. On the other hand, fuel taxes and VAT are slightly progressive. The fact that VAT is not regressive can be explained by the zero rating of basic food items, as the poor tend to spend a greater percentage of their income on VAT-free items, and similarly for the VAT exemption of public transport.
Taken together, the mix of progressive direct taxes and slightly regressive indirect taxes generates a mildly progressive tax system. Corporate taxes are not included in the analysis.
Spending programmes are strongly progressive
Government spending is progressive when the share of the monetary value of government goods and services going to the poor exceeds their share of aggregate market income and the richer groups receive transfers of a value less than their share of market income.
The results show that government spending – social spending in particular – through direct and in-kind transfers are strongly progressive.
Cash transfers, especially the child support grant, are the most progressive of all spending programmes. The poorest 40% of individuals receive 69% of all cash transfers. These transfers have significant value: for the poorest 10% of South Africans, they are worth 10 times more than their market incomes.
In-kind transfers in the form of free basic services – water, electricity, sanitation and refuse removal – are strongly progressive when targeted at the poor, and they are slightly progressive when provided as an equal subsidy to everyone. The reality probably is in between because some municipalities target services to the poor but most say they lack the capacity to do so.
Education and health spending are progressive overall. Figure 1 illustrates how the poorest households comparatively receive much more benefit from spending on education than their share of market income. It was not possible to analyse the quality of services.
Most types of education are strongly progressive. The exception is tertiary education, which is only slightly progressive, as is the case in many countries.
Health spending is strongly progressive. It is well targeted at the poorest, given that low income groups are more likely to attend public health facilities than private ones. Figure 2 shows how the poorest households comparatively receive more benefit from health spending than their share of market income.
Jointly, taxes and spending are very progressive
Not only are South Africa’s main fiscal instruments progressive overall, the degree and structure of progressiveness is such that these instruments achieve significant reductions in income inequality.
The combined effect of taxation and government spending on income inequality can be estimated in terms of the reduction in the Gini coefficient as a result of taxation and social spending. South Africa’s Gini coefficient using market income is estimated to be about 0.77.
However, for final income – when direct and indirect taxes (excluding corporate taxes) as well as key categories of government spending are taken into account – the Gini coefficient is estimated to be about 0.59, a reduction of 0.18 Gini points. This is a big change.
Income inequality moves from a situation where the combined market income of the richest decile is more than 1 000 times that of the poorest to one where the final income of the richest decile is 66 times higher than that of the poorest.
Moreover, this is the largest reduction in inequality of 12 middle-income countries analysed using the same method. Figure 3 illustrates that South Africa’s Gini reduction of almost 0.18 points compares favourably with Brazil’s at 0.14 or Mexico’s at 0.08. But is the reduction in inequality enough?
The evidence from 2010 shows the fiscal system in South Africa is progressive: the tax system is mildly progressive and government spending is highly progressive. In other words, the rich bear relatively more of the tax burden than the poor and the government redirects these resources more towards the poorest.
This significantly raises their final incomes. Without such a system, income inequality would be much higher than its current level.
Despite this progress, the level of final income inequality remains unacceptably high. Our level of income inequality after redistributive efforts is still higher than income inequality in all the other countries in the study before they apply redistributive fiscal policies.
Figure 4 shows how the combined effect of taxation and government spending reduces South Africa’s Gini coefficient from 0.77 (for market income) to 0.59 (for final income). But it also shows that our Gini coefficient of 0.59 after fiscal redistribution compares poorly with the second most unequal country before fiscal redistribution – Brazil’s Gini of 0.57 for market income.
Policy options and limitations
Inequality in South Africa remains stubbornly high, even though progressive taxation and progressive government spending reduce inequality significantly. The analysis may be interpreted to suggest that even greater fiscal redistribution is required.
Yet the country’s fiscal deficit and debt indicators signal that there is limited fiscal scope to spend more to achieve even greater redistribution.
As a result of the problematic global economic climate since 2008, the implementation of countercyclical policy has meant South Africa has increased its net debt burden of 22.9% of gross domestic product in 2008-2009 to 39.7% in 2013-2014.
Some would argue that South Africa should tax the rich even more, or target government spending on the poor even better. In this way, the fiscal system would become more progressive. Further research is needed to determine the consequences of such policies on growth, unemployment, poverty and economic efficiency.
Addressing income inequality in a way that is consistent with fiscal sustainability will need higher-quality and more efficient public services.
It will also require interventions to achieve more inclusive economic growth to address the need for employment and higher incomes at the lower end of the income distribution. Fiscal redistribution alone is unlikely to achieve the desired reductions in inequality.
Ingrid Woolard, Rebecca Metz and Mashekwa Maboshe are at the University of Cape Town, Gabriela Inchauste and Catriona Purfield are at the World Bank and Nora Lustig is at Tulane University. This is an edited version of an article that first appeared on Econ3x3.org.