There is an old joke that rich people are, in fact, very giving. Very giving. What they love to give is advice. Particularly if it's unsolicited. And from the perspective of the poor, there is this: money, they say, can't buy love. Poor people would probably agree. But I think they might like to do their own research.
And so on. Poor and rich; rich and poor. It's one of the most profound and complicated cleavages in human society, for as long as we have craved faster chariots and large golden adornments. It's endless, multifaceted. Obvs.
As complex as the relationship is between poverty and riches, it's weirdly tricky to understand the way the different classes in our society perceive wealth. Rich and poor are fabulously prone to stereotyping, especially of each other. And then there is the middle, and their views of both. It's a morass.
So I was intrigued recently by a survey published jointly by a group called Patriotic Millionaires and the NGO Oxfam South Africa, which purports to show that nearly 64% of South Africa’s dollar millionaires support a 2% wealth tax.
Well, you know, strike me down with a feather, but for all kinds of reasons, that just doesn’t seem likely to me. I was particularly sceptical just looking at the name of the publisher, “Patriotic Millionaires”. Doesn’t the outcome chime with the name of the sponsor? Start there. But the story on the topic in Daily Maverick was written by Ferial Haffajee, a veteran journalist, who is renowned as scrupulous, smart and not prone to manipulation.
Ferial Haffajee
Anyway, I reached out to Oxfam SA, upfront about my scepticism. I asked a couple of questions and then for the raw data from the survey. Much to my surprise, Oxfam came back to me after a couple of days, answering my questions and providing the results of the survey. The way the information was presented was not perfectly rigorous, nor even accurate, to my taste, but it was certainly much better than my own prejudices would have suggested.
The first point to make is that Oxfam itself did not conduct the survey, but by a reputable survey organisation, Survation, which is a member of both the British Polling Council (BPC) and the Market Research Society (MRS), meaning it adheres to their professional codes of conduct and transparency rules. It has a decent track record, particularly for major UK political events.
The survey was compiled using online interviews of individuals living in South Africa with investable household assets worth more than $1-million; the sample size was 543, which compares to the 40,000 or so dollar millionaires in SA. This was not, as Isobel Frye, Senior Policy Advisor of Oxfam South Africa, acknowledges, a “representative” survey.
The participants were randomly selected, but it's small compared to the total sample. And I presume there is also an element of self-selection by the people prepared to answer this kind of survey. And by the way, how would you answer if someone asked you publicly if you support education for the poor (one of the prompts for the idea of the tax)? So technically it's an indicative survey, not a formal study, as Oxfam acknowledges.
So far so good.
But it's at this point the priors start creeping in. The question on the wealth tax runs like this: “A 2% wealth tax on the very richest individuals in South Africa could generate billions of rand annually, enough to fund social protection, education, or the energy transition. To what extent would you support or oppose a 2% wealth tax if it would pay for these services?” 70% said they would support, either “strongly” or “somewhat”, that suggestion.
The problem with this question is not what is asked but what is not explained. The question doesn’t specify if this wealth tax would go directly to solving the problems of social protection, education or the energy transition, or would aim to do that via the fiscus or a special-purpose philanthropic trust. In the article, Frye said she was personally against something she described as “ring-fencing”, saying all public services and the funding should be subject to parliamentary scrutiny and oversight. In other words, she interpreted the answer as rich people agreeing to a 2% tax increase or 2% wealth tax.
I’m willing to bet a large amount that this is not what South African millionaires were thinking of when they answered the question affirmatively. And actually, this is absolutely the nub of the issue, because in my experience, rich people (South African and everywhere else) are not as miserly and detached as left-wing politicians often suggest. But they are very conscious of waste, transparency and effectiveness, because apart from anything else, that is how they got to be rich.
Notice too, the question references a “wealth tax”, so we are not sure whether that is a tax on income or wealth. The difference is huge: theoretically, a realistic behaviour-adjusted wealth tax would raise around R80-billion as a once off. A 2% increase in income tax on the rich would garner maybe R8-billion a year.
I’m not granting rich South Africans an absolutely clean bill of health here. There are statistical studies out there, which suggest the rich are actually less generous than the poor in terms of the proportion of income donated. A large chunk of the very poor in the US, for example, donate in double-digit percentages of their income.
Cord Blomquist
On the other hand, the poor are the only quintile where fewer than half make donations, meaning that those who do donate do so very generously. Obviously, in absolute terms, the quantity of donations made by the very rich dwarfs all other quintiles combined. So what does this tell you? Actually, not much. But it is interesting that the total amount donated by rich and poor, in the US at least, has quintupled in real terms over the last 75 years.
The same indecisiveness is true of the supposition on the part of the righteous left that the rich are less “pro-social” than the poor. There is a very famous study by Paul Piff et al. from 2012 called the “Rich Drivers Study” which found that drivers in snazzier cars appeared more likely to “cut off” other drivers or pedestrians more often at busy four-way intersections - three times more!
The study echos that famous song by Johannes Kerkorrel en die Gereformeerde Blues Band called BMW, which goes, “Ons ry a BMW … Ons gaan elke jaar oorsee … Ons sal jou vokol gee, … Polina, gaan maak vir die miesies tee,” and so on. Very funny. (We drive a BMW; every year we go overseas; we will give you fuck all; Polina, go and make the madam tea”). Kinda funny.
But this has been one of many “studies” in the psychological and cognitive sciences that subsequent efforts have found hard to replicate. A larger study by Jung, Smeets, Stoop & Vosgerau in 2023 used larger sample sizes (≈ 2.5 times the original), applied inflation-adjusted vehicle-value scales, and tried to mimic the original methodology while updating for contemporary conditions.
Guess what: they found no significant relationship between inferred car value and the likelihood of cutting off other vehicles or pedestrians. The finding that higher social class predicts increased unethical behaviour rests on the notion that social class and motor car value are identical. And we all know that is not strictly true.
As so often is the case, the original finding probably reflects more the prejudices of the researchers than the views of the subjects.
There is one other thing. The “Patriotic Millionaires / Oxfam poll also questioned wealthy individuals across G-20 countries, announcing in January last year that 58% supported a 2% wealth tax for people with more than $10-million.
Although the 2% figure is the same, we are talking about a higher level of wealth in the G-20. But it's interesting that the proportion of the rich who notionally support this idea is much higher in SA than it is internationally (70% as opposed to 58%). That makes sense to me: the need is greater here.
But the big problem is this: it's not what the Oxfam study shows that is important, it's what it does not show. I would like to see the empirical study by Oxfam or anyone else that collects data before and after a 2% wealth tax has been implemented globally (or in a substantial jurisdiction) and then measures the effects on inequality, poverty reduction, revenue mobilisation, or social outcomes. In other words, there is no real-world “trial” of a 2% wealth tax in Oxfam’s published record.
Oddly, we do have a kind of a record in SA that answers this question, and that was in the 2016/17 and 2017/18 Budgets. National Treasury introduced a series of above-inflation tax hikes targeted at higher-income groups, increasing higher marginal income-tax brackets for top earners, not fully adjusting for bracket creep, and increasing dividends-withholding tax (from 15% to 20%). This wasn’t a “wealth tax” as such because it focused on income not wealth, but still, the idea was to soak the rich. The result was disastrous: instead of increasing tax, the increase reduced the tax base, and undermined tax collection. Instead of raising R4.4-billion a year, the total increase was more like R400-million - a decline in real terms. It should have been obvious this was going to happen: 1% of South Africans pay 60% of all personal tax.
The issue is not only the blistering 45% top rate of tax in South Africa, but also the credibility of the expenditure system. And that’s the real problem with surveys like this: people want to do good. And that applies to the poor and the rich.
Poverty fosters empathy but limits ability; wealth expands ability but I would guess increases sensitivity to methodology and outcomes. Corruption is not constituted only by what it stolen, but by what is lost to the system as a whole, and indicative surveys by left-leaning ideologues seldom address that issue sufficiently IMHO. 💥
From the department of the bear necessities of life ...
Justin McCurry
From the department of the not necessarily bear necessities ...
Mark Walker
From the department of possibly losing your bearings ...
Nick Lichtenberg
Thanks for reading - please do share if you have a friend (or enemy!) you think would value this blog and ask them to add their email in the block below - it's free for the time being. If the sign-up link doesn't appear, you'll find it on the site.
Till next time. 💥
Tim Cohen
Join the conversation