Here is something that one hears quite often (the below comes from this article):

After two decades of democracy, economic inequality in South Africa remains very high. Survey comparability issues make analysing trends tricky, but it is clear that overall income inequality has not fallen; the country has long been one of the highest ranked by Gini coefficient, a widely used measure of income inequality. Its peak of about 0.72 in the late 2000s is one of the highest ever recorded.

As an aside, I’ve explained the GINI coefficient in this post: That Gini Might Give You A Headache. But, really, the important thing to know is that the Gini is a number between 0 and 1, where “0” is perfect equality (where everyone earns the exact same basic income), and “1” is perfect inequality (where only one person earns all the income). So the closer the Gini is to 1, the more unequal things are.

And here is a map, showing how badly South Africa was doing on the income inequality front in 2014:

So, you know, not good.

Except that this map only shows pre-tax income inequality – it does not take into account the impact of progressive taxes, wealth transfers, and all the benefits that are supplied through government efforts to address income inequality.

Here are some graphs and tables showing how that impacts (and I’m going back to the paper that the Davis Commission keeps citing*: The Distributional Impact of Fiscal Policy in South Africa):
*I keep referencing the Davis Commission because that process is rather firmly operating in favour of a more leftist approach to tax collection – which at best makes them neutral about the impact of wealth transfers; and at worst, incentivises them to understate their impact. So I’d rather use their numbers than those from any studies generated by the anti-tax groups on the Right.

So putting the impact of free basic services aside, the after-tax income of the poorest decile is increased by around 10,000%, while the after-tax income of the richest decile is reduced by almost a third. And that’s after all taxes, not just personal income tax.

The paper also makes some comparisons to similar middle-income countries, in which South Africa does a lot of outstripping on the direct cash transfers front:

Then there are the free basic services:

And health-care spending:

Here’s a summary of how these impact the Gini Coefficient:

Interestingly, it seems that government subsidies of tertiary education increase inequality rather than decrease it (!). So I’m not sure whether to admit that I was in favour of #FeesMustFall or not…

But that aside, here is how South Africa’s efforts to reduce inequality compare to similar countries:

And South Africa’s efforts to reduce poverty:

And finally, the study makes this comment:

The thing is: I’m not trying to say that we’ve fixed inequality, or that we’re not doing enough, or that we’re hitting the Goldilocks “just right” temperature of redistribution policies. I think that is a different conversation.

But what I am trying to say is that the talk around South Africa’s income inequality seems bizarre because it outrightly ignores the impact of the policies that are currently in play to reduce it.

And surely that has to count for something?

PS: I do realise that I’m talking about income inequality rather than wealth inequality. I will write about wealth inequality again – but basically, I suspect that wealth inequality is a far trickier beast, mainly because it seems far more likely to reverse itself over time – which I wrote about in this post: When Land Reform Works, and When It Reverses. 

PPS: also, apologies for the confusion with yesterday’s post. I published the original post, then the WordPress app on my phone unpublished it, and then when I re-published it, it reverted to an earlier incomplete draft. I’ve finally corrected it now (I hope!) – but very frustrating. 

Rolling Alpha posts about finance, economics, and sometimes stuff that is only quite loosely related. Follow me on Twitter @RollingAlpha, or like my page on Facebook at www.facebook.com/rollingalpha. Or both.