There really is no arguing that there are huge wealth disparities in SA, so it seems not only morally righteous but also logical and politically pleasing to soak the rich.
But, it’s a terrible idea, not necessarily in general but as it specifically relates to South Africa. There are countries around the world where the very wealthy get away with outrageous inequity. And there are countries in the world where the rich are not paying their fair share. The problem is that SA is in neither of those categories.
Because SA’s tax system is steeply progressive already, as it should be, with the top 10% of earners paying basically all of SA’s personal tax, adding to this level of progressivity garners increasingly small quantities of tax.
The most egregious thing about this fantastically unoriginal idea is that it is somehow presented as though SA’s heavily socialist government has never thought of it before. Yet, SA’s government has been imposing steadily higher taxes for decades, and doing so in the context of declining economic growth. In 2000, SA’s tax rate in relation to its GDP was 22.4%. That has risen steadily over the past two decades, and it reached 29.1% in 2018. This is more or less the same as the current OECD average. It was then that Finance Minister Pravin Gordhan imposed, you guessed it, a wealth tax.
Compare this to the rest of the continent and the numbers are really eye-popping. In 30 African countries with more or less reliable statistics, the average tax-to-GDP ratio is 16.5%. There are only two African countries with higher tax-to-GDP ratio’s on the continent, Tunisia and the Seychelles. What is more, the increase in the ratio has been happening faster in SA than on average on the continent.
The momentum to impose a wealth tax has gained some academic credentials with the publication this week, not accidentally, of the World Inequality Lab report. The Lab is a Paris-based research centre associated with anti-inequality campaigner Thomas Piketty.
The lab proposes either a once off or a recurring wealth tax aimed at the top 1% set at about 3% to 7% of GDP of which it estimates would garner somewhere between R70-billion and R160-billion. It estimates a 30% evasion rate, suggesting the tax would raise 2.8% of GDP or about R134-billion.
You really can’t fault the moral integrity of the report’s authors, but these numbers are just consummate fantasy. But what really irks me about the report is that it does not even address SA’s previous attempt at a wealth tax, or record what happened.
So, back to Pravin Gordhan’s 2017 wealth tax. It simple enough; it constituted an increase of the maximum marginal rate – the rate of tax that applies to the top income bracket – from 41% to 45%. He also increased the withholding tax on dividends from 15% to 20%, and limited the adjustments for bracket creep (when salaries rise with inflation but the tax brackets do not).
The expectation at the time was that this would affect only about 103000 people earning over R1.5-million a year. The estimated income from those three measures was supposed to be around R23-billion. It was a full-on assault against the rich, because, as we all know, the rich are all evil cheats who gained their wealth not through skill, or hard work or innovation, but through skiving the system somehow.
So what happened? We had to wait a year to see, but the truth came out in the 2018 budget. Not only did SA not get R23-billion, SARS scooped R20.4-billion less than expected in personal tax. Instead of the R483-billion from personal income tax anticipated, it collected only R462-billion. Increasing the tax rate garnered exactly nothing.
Now, there could be all kinds of reasons why that happened. For a start, these were the years when state capture was at its height – and that is not an irrelevant context by any means. Anecdotally, I suspect, the willingness to pay tax is highly correlated with the sense that taxes are well spent. I also suspect it was at this point that rich South African’s began taking active measures to get their investments out of SA, partly from the somewhat paranoid fear that “Venezuela is coming”.
Whatever the case, the Davis Tax Commission was tasked with looking at the issue again, and decided that the ceiling on wealth taxes had probably been reached.
“The possibility of tax reform now is limited. As the Davis tax committee reported, while a wealth tax would add to the legitimacy of the tax system in a country with such vast inequality, it would require significant institutional capacity that can’t just be switched on like a light,” the head of the commission Denis Davis wrote later.
The reason why wealth taxes don’t work is not only the issue of institutional capacity but also the problem of how incentive systems operate. To state the absolutely obvious, but at a 15% personal tax rate, the level of tax avoidance if the tax is increased by, say, 5% is likely to be very small. But if you are paying 40% of your income in tax, and tax is increased by 5%, the incentive to take action is higher, etc.
This is the logic behind the much-derided Laffer curve, the notion that people will adjust their behaviour in the face of the incentives created by income tax rates. At a certain point, tax increases result, in certain circumstances, to lower overall tax.
Lots of Republican-controlled states in the US have tried the opposite proposition: that lowering tax rates will lead to a higher overall income for the state – often with disastrous results. The Laffer curve idea is, in fact, just too simplistic. There are lots of things that affect tax income besides the tax rate. Yet, it’s worth noting that in the SA case, the theory did actually held true.
In which case, would there not be an argument to decrease taxes in SA in order to gain more tax income, and pay for the vaccines in that way?
Just asking. DM168
- This total personal tax collected in 2018 has been corrected
Tim Cohen is the editor of Business Maverick.
This column first appeared in our weekly Daily Maverick 168 newspaper which is available for free to Pick n Pay Smart Shoppers at these Pick n Pay stores.