University of Stellenbosch economics professor Servaas van der Berg is the latest academic to attack recruitment firm Adcorp’s employment statistics, describing them as "dangerous fictions".
The Adcorp Employment index is published monthly. Unlike figures from Statistics South Africa (Stats SA), it is not based on surveys and relies upon its own data gathered in interviews with job recruits.
It also uses a controversial methodology to estimate the size of the informal sector to arrive at an estimate of total employment.
› Adcorp job figures fiction - Prof
› Eurozone slips deeper into recession
In an article on Econ3x3, a new online policy forum, Prof van der Berg writes that Adcorp’s data "is so clearly wrong that it does not even warrant paying their methodologies serious attention".
He urges people to consider the "obvious inconsistencies" in Adcorp’s numbers by juxtaposing its unemployment figure - which it put at 1-million people in 2011 - with the number of unemployed graduates, which it estimated to be 600 000 at the time.
This would mean, Prof van der Berg writes, that based on a labour force of 20-million and a graduate population of 1.1-million, unemployment was 5% for the population and 54% for graduates.
In contrast, the last quarterly labour force survey compiled by Stats SA put unemployment at 25.5% and in 2011 found that broad graduate unemployment ( for all tertiary qualifications ) was 340 000.
The Adcorp employment index estimates the informal sector to be much larger than estimates from other organisations. It uses the currency demand method to arrive at its estimate, which takes the amount of cash circulating to estimate the size of the activity not recorded in the formal economy.
The result is a much higher number of employed : 19-million according to Adcorp and 13.5-million according to Stats SA.
This methodology has been criticised in detail by University of Cape Town academics Andrew Kerr and Martin Wittenberg, in two papers written last year.
"The Adcorp employment and unemployment figures are dangerous fictions - and need to be exposed as such," Prof van der Berg writes.
"It is unfortunate that the news media publish them as if they come from a credible source. That feeds at least two contradictory and mistaken beliefs: that unemployment is not a problem of great consequence, and that graduate unemployment is a problem of grandiose proportions.
"Neither of these two conclusions could be further from the truth."
Adcorp economist Loane Sharp said Prof van der Berg’s criticisms were "parochial" while his methodology enjoyed international recognition. He said he would not debate the size of the informal economy - as he had done it before - and won.
Mr Sharp said the apparent anomalies in the number of unemployed graduates compared to the rest of the population was because "there is a problem with Stats SA data".
He believed Prof van der Berg had incorrectly read Stats SA figures on graduate unemployment.
Statistician-general Pali Lehohla, who has been arguing with Adcorp over its employment numbers for years, said on Thursday the company’s objectivity was compromised.
"They are an employment agency. That compromises them severely. To collect statistics, you need to be policy neutral. You can’t measure statistics when you have an interest, especially where the benefits are financial."