In any debate, Africa Check verifies statements of fact, not opinions. Whether there has been “spectacular” racial transformation in South Africa – as Leon Louw from the Free Market Foundation has argued – falls outside Africa Check’s remit. Our goal is to keep debate honest by reviewing the evidence an argument is built on.
Africa Check also aims to fact-check both sides of a dispute. For example, former Economic Freedom Fighters member of parliament, Andile Mngxitama, stated earlier this year: “Whites are only 10% of the economically active population but occupy more than 60% of the top management positions.” Louw wrote: “Blacks in top management… doubled.”
Who’s right? Both – by presenting only a specific slice of the data.
We accept that by analysing the evidence we won’t convince everyone. Mngxitama and Louw were both harsh in their reaction to our fact-checks. In a thinly veiled reference to Africa Check, Louw suggested in a column that we were racist for finding fault with some of the facts he cited, while Mngxitama claimed we have a “racist ideology”.
It seems Mngxitama and Louw share common ground after all.
The source Louw provided was a study carried out by the University of Cape Town’s Unilever institute of strategic marketing called “Four Million and Rising”. It stated that South Africa’s black middle class grew from 1.7 million in 2004 to 4.2 million in 2012 – a 147% increase over that period.
According to the institute’s management consultant, Paul Egan, someone qualified as middle class if they were older than 18 and lived in a household with an income of between R16,000 and R50,000 per month at the time of the study. Two of the following criteria also had to be met: own a car, have tertiary education or be studying, work in a professional job, live in formal housing in a metropolitan area or pay rent of R4,000 or more.
Since then, the institute has not conducted any further research on the middle class in South Africa.
That said, different studies employing different methods have come up with different figures. For instance, according to a Stellenbosch University study the black middle class totalled nearly 3 million people in 2012, using a cut-off of per capita income of R25,000 per year in 2000 prices.
A 2015 journal article calculated the black middle class to have increased from 7.7 million individuals in 1993 to 10.4 million in 2008 (up 35%), when an after-tax income bracket of R1,400 to R10,000 per capita per month in 2008 prices was applied.
The author, Justin Visagie, wrote: “In sum, the racial composition of the middle class and upper class displayed massive transformation between 1993 and 2008… That said, there is still significant progress to be made.”
Definitions and estimates of the black middle class vary, but the source Louw provided did not support his claim.
Louw cited articles by the Helen Suzman Foundation as the source for his claim, but these did not substantiate his figures.
Africa Check phoned the writer of the articles, economist Charles Simkins, who extracted data from the 2001 and 2011 Census, saying it showed that total black disposable income increased from R16.9 billion in 2001 to R60.4 billion in 2011, a 257% increase. (Simkins said 1996 data on disposable income is unavailable.)
However, researcher and PhD student at Stellenbosch University, Asmus Zoch, said that the census data does not provide an accurate total, because the top income bracket by which a household head can report disposable income is open-ended.
Zoch stressed that researchers should control for both inflation and population increases over long time periods. “This gives a better idea about the wealth changes than [combined] numbers, especially if the total black population is growing,” he told Africa Check.
Converting data from the Project for Statistics on Living Standards and Development, collected in 1993/94, and 2012 data from the National Income Dynamics Study into constant 2000 prices, Zoch calculated that black per capita household income increased from R363 in 1994 to R775 in 2012, a 114% increase.
Africa Check was unable to source data for the specific years mentioned. However, data for a similar timespan is available in a paper called “Trends in South African Income Distribution and Poverty since the Fall of Apartheid” by economists the University of Cape Town’s Southern Africa Labour and Development Research Unit (SALDRU).
Per capita annual personal income in 2000 prices increased by only 93% for black people between 1993 and 2008, from R5,073 to R9,790. (For white people, the increase was 62%, from R46,486 to R75,297.)
Former white areas
Louw provided a slide from a presentation by WHAM! Media, a social investment communications company, as his source.
It showed these figures for the “new black middle class”, taken in turn from the Newspaper Advertising Bureau:
|Polokwane||30% to 64%|
|Midrand||53% to 62%|
|Kempton Park||33% to 50%|
|Centurion||14% to 43%|
|Durban Berea||33% to 53%|
Louw told Africa Check his claim was “extrapolated” from the slide’s figures, but he did not explain how he used five data points to make a national claim.
The Newspaper Advertising Bureau publishes what it calls “South Africa’s largest urban, community level quantitative survey”, called Roots. The most recent edition on their website, Roots 2013, includes demographic and socio-economic data collected from 115 suburbs and towns in South Africa.
Samu Makhatini, research manager at SPARK Media of which the bureau forms part of, told Africa Check that, very much like with the first claim, it would depend on how one defines the black middle class. She said it was possible to cross-tabulate data from their research but she was unable to confirm whether it was carried out correctly without seeing how exactly it was done.
Africa Check also contacted Statistics South Africa to ask if the claim could be verified using their data.
Living conditions survey manager, Patricia Koka, said data published from the Income and Expenditure Surveys and the Living Conditions Survey can be reported at provincial level and other categories like population group, settlement type and sex, but that data cannot be disaggregated into suburbs.
Louw did not provide evidence to support this claim.
A report by the Commission for Employment Equity for 1999 to 2001 showed that 87% of top managers were then white, 6% were African/black, 4% were Indian and 3% coloured.
The commission’s latest report for 2014/15 showed that the share of African/black people in top management positions was 13.6%, a 127% increase. White South Africans held 70% of top management positions in 2014/15.
Important to note though, is submissions from designated employers for 2014/15 only accounted for 7 million employees out of the 20.3 million people in South Africa’s economically active segment.
Louw did not provide a source to support this claim.
In 1994, 67.4% of African/black people were “functionally literate”, Dr Anne Harley, from the University of KwaZulu Natal’s Centre for Adult Education, and other researchers determined from Stats SA’s 1994 October household survey. This refers to people who have completed at least Grade 7 of schooling.
However, Stats SA service delivery statistics manager, Niël Roux, told Africa Check that the “sample and fieldwork execution used in 1994 were much more rudimentary as a result of the shortage of recent census data which could have informed the sample frame that was used”. As a result, Roux said this data was less reliable.
He directed Africa Check to the 1996 census, which estimated that 59.3% of African/black people aged 15 and older were functionally literate then. By last year, it was estimated to have increased to 82.8% – or up 40% since 1996.
Stats SA’s 2015 mid-year population estimates reported that life expectancy increased from 55.3 years in 2006 to 59.7 years in 2011 for all South Africans. In 2015 it stood at 62.5 years.
Manager for demographic analysis at Stats SA, Diego Iturralde, told Africa Check that “life expectancy for race groups isn’t published” by the organisation.
He was however able to extract data which showed that black life expectancy increased from 51.5 years in 2006 to 57.2 years in 2011.
To support this claim, Louw provided a report containing the claim but no source.
South Africa’s department of basic education sent Africa Check a racial breakdown of students at independent schools in South Africa from Pre-Grade R to Matric.
The data showed that in 2014, 65% of students were African/black, 22% were white, 7% were Indian/Asian, 5% were coloured and the race of 2% were not determined.
Louw’s statement was ambiguous. It only became clear that he was referring to the share of degrees awarded per year as we started tracking down the original source of his claim.
This is true – of the 164,423 degrees awarded to South Africans studying at public universities last year, 105,782 (64%) were conferred to African/black students.
Share of ownership
Claim: “Blacks are approaching or have surpassed 50% in almost everything…”
Louw told Africa Check that the source of the statistic was a “media briefing document” from economist Vivian Atud. However, when Africa Check contacted Atud she said that the research she conducted showed that 23.4% of shares were black-owned.
None of the available research on share ownership of the Johannesburg Stock Exchange (JSE) supports Louw’s claim.
Research conducted for the JSE in 2013 estimated that 10% of shares were directly owned by black people (“largely through BEE schemes”) and 13% were held indirectly by black people (“mostly through individuals contributing to pension funds, unit trusts and life policies”). The JSE has not released any subsequent research.
That same year the National Empowerment Fund estimated that black people controlled just 3% of companies on the JSE. Their latest estimate, from May 2017, is still 3%.
Read our guide to the different methodologies that produced these estimates.
2) New companies
Verdict: Unable to verify
In support of this claim, Louw provided a link to the April 2014 archive of ZAeconomist.com, which describes itself as “a blog on the South African economy and its financial markets.”
However, the April 2014 blog posts did not include any information on the percentage of new companies that were black owned.
Data manager at the Companies and Intellectual Property Commission, Marietjie Swart, told Africa Check that 248,902 new companies were registered in 2014. However, she said that the commission does not “record the race of directors of companies”. Because of this, the commission is unable to provide a racial breakdown of company ownership in South Africa.
3) Insurance policies
Verdict: Unable to verify
Louw cited Vivian Atud, who was previously an economist at the Free Market Foundation, as the source of this claim.
Atud told Africa Check that she had conducted research on racial ownership of companies in 2010 but she did not respond to Africa Check’s request for the research.
The South African Insurance Association told Africa Check that most insurers “do not request the race of the policy holder to be disclosed”.
According to South Africa’s 2012 National Development Indicators, social assistance expenditure “increased from R10 billion in 1994 to just under R100 billion in 2011/12 and constitutes 3.4% of GDP”. A South African Reserve Bank report specifies the amount for 2011/12 as R96 billion.
To compensate for inflation, Professor Ingrid Woolard of the School of Economics at UCT, who specialises in researching social protection and poverty and inequality, calculated the increase in real terms.
Using Stats SA’s Consumer Price Index figures, Woolard said that the 2011/12 amount of R96 billion would equate to R34 billion in 1994 terms. Alternatively, the 1994 amount of R10 billion would be equivalent to R29 billion in 2011 prices.
Also read: FACTSHEET: Social grants in South Africa – separating myth from reality
Africa Check was first directed by Louw to page 175 of the Presidency’s Twenty Year Review. But that page noted that the share of all people in South Africa living on less than $2.50 a day fell from 42.4% in 2000 to 29.2% in 2011.
A second source, a presentation by WHAM! Media’s editor-in-chief Paul Pereira, claimed that the percentage of all people living on $2 a day fell from 16% in 1996 to 2.7% in 2009. Pereira told Africa Check there was an error in the slide and that the 2.7% estimate referred to 2011, not 2009.
Gerhard Bijker, country manager for South Africa at IHS, told Africa Check that IHS stopped calculating what percentage of people were living on less than $2 a day when Stats SA published a set of three national poverty lines in 2012.
“Now [Stats SA] has the food poverty line and the upper bound and lower bound [poverty line] and those are the ones that are better to use,” Bijker said.
Africa Check then consulted three further sources, none of which supported Louw’s claim:
© Copyright Africa Check 2019. You may reproduce this piece or content from it for the purpose of reporting and/or discussing news and current events. This is subject to: Crediting Africa Check in the byline, keeping all hyperlinks to the sources used and adding this sentence at the end of your publication: “This report was written by Africa Check, a non-partisan fact-checking organisation. View the original piece on their website", with a link back to this page.