Op-Ed: We can’t tax ourselves out of inequality
The raw facts for those really concerned with social justice and wrenching levels of inequality is that the tax system we have cannot do much more than it already is.
The budget speech has come and gone. It was a mixed bag: Reductions in funding available for investment, the provinces and local government. At the same time, some sort of funding for tertiary education for students from poorer backgrounds who defied the odds that condemn most poor people to a shoddy education, and made it to tertiary level. There were increased taxes, a higher marginal rate for the well-off, and bracket creep for the rest of the taxpayers. The big change was the increase in VAT to 15%. This is expected to raise an additional R20 billion per annum, about the same amount needed to fund free tertiary education.
The increase in VAT was controversial. Unlike other taxes, it is an indirect consumption tax that everyone pays as one consumes food, services and other VAT-able items. The change, it is charged, hits the poor more than the rich. Arguably, this is correct but there are some charges then that the budget, as a whole, is anti-poor. Gilad Isaacs, an academic at the WITS Corporate Strategy and Industrial Development School of Economic & Business Sciences, had this to say: "Despite wealth inequality in SA being extreme – the top 10% of South Africans hold at least 90-95% of its wealth, while the top 1% holds 50% or more of its wealth – taxation on wealth, or income from wealth held, is low". Could this be true?
As a country at this level of development, the amount of tax raised is relatively high. Some of our neighbouring countries have higher tax rates, but that is more a result of transfers from South Africa through our customs union with them. This time next year, we’ll see whether the additional taxes result in increased revenues. Perhaps South Africa, at its stage of development, has hit the peak amount of tax that can be extracted.
However, we are a very unequal society. A 2016 World Bank study
entitled The Distributional Impact of Taxes and Transfers: Evidence from Eight Low- and Middle-Income Countries, shows that in South Africa, between 2006 and 2011, the proportion ofthe population living below the national poverty line fell from 57.2% to 45.5%. Still, the top 20% of South Africans in 2011 accounted for 61.3% of national consumption, while the bottom 20% accounted for just 4.3%. Using the international poverty line of US$2.50 per person per day, South Africa’s poverty head count ratio was 34% in 2011, whereas it was 11.7% in Brazil and 5% in Costa Rica the same year.
However, using Stats SA figures, the authors also show that since the end of apartheid, the government spends far more that other middle-income countries on social assistance programmes, health and education. By 2014, total government spending stood at 32% of GDP, and more than half of this was devoted to social spending. While income inequality has increased in the post-Apartheid years, social spending policies have undone about 40% of the increase in income inequality between 1993 and 2008.
From a tax point of view, the paper shows that the South African tax system generates considerable resources by middle-income country standards. In 2010/2011, these amounted to 27% of GDP. About half of the tax collected came from direct taxes, personal income tax (PIT), corporate income tax, and payroll taxes in the form of unemployment insurance and the skills development levy. An updated summary from National Treasury in respect of projected tax revenues into the 2018/2019 suggests the share of different taxes to the total tax take has remained about the same.
The World Bank study also showed that South Africa’s social spending is among the highest in the study. The data shows that over a half of South Africa’s total general government expenditure in 2010/11 was devoted to social spending and as much as 3.3% of GDP was dedicated to social grants in 2010/11. In addition, another 0.5% of GDP covered “free basic services” such as power, sanitation, water supply, and refuse removal, which are provided for free to low-income households. South Africa’s total social expenditures are more than twice the median level among low- and middle-income countries (World Bank 2009). They point out that the number of beneficiaries receiving the various kinds of social grants doubled from 8-million in 2003/04 to 15.8-million in 2013/14, and have since increased to around 17-million.
Moreover, in 2011, around 12.6% of GDP was spent on in-kind transfers through health (at 4.1% of GDP) and education (at 7% of GDP). A further 1.5% of GDP was devoted to housing and other in-kind transfers.
The study also showed that the public health care system in South Africa serves mostly the poor, with primary health care available free of charge and hospital services provided at relatively low cost, with a sliding tariff scale calculated according to income level. Social grant beneficiaries were automatically exempt from paying for any public health services. Education spend amounted to 7% of GDP in 2010/11. While the government provides all public schools with a grant to finance their operational costs and teacher salaries, schools in poorer neighbourhoods are designated “no fee” schools, which receive a slightly higher state subsidy to compensate for the absence of school fees.
In 2011, 78% of learners attended these no-fee schools. Furthermore, municipalities are required to provide relief for the poor from charges for municipal services (water, electricity, and sanitation and refuse removal). The national government funds about half of total municipal spending through an “equitable share formula”. This transfer amounted to about 1% of GDP in 2010/11 and about 75% of the equitable share transfer reflects that part of the formula used to cover the operating costs of providing basic services to poor households in each municipality, using an estimated number of poor households in each area. In addition, many of the larger municipalities use surpluses generated through the sale of electricity to cross-subsidise free basic services.
While the authors of the study describe the difficulty in monetising the effect of spending on health and education because of a highly variable quality of the services provided. The outcome is described in the graph below: