At a time when job creation is the focus of economic policy, a colossal fight is brewing over the reliability of employment statistics in South Africa.
The Adcorp Employment Index, released this week, shows that 80 000 jobs were created in January and 103 000 in December, whereas the government’s Quarterly Labour Force Survey shows that 179 000 jobs were created in the fourth quarter of 2011, averaging 60 000 jobs a month. The Adcorp index shows that the greatest number of jobs created in January were in wholesale and retail (13.6%), transport and logistics (8.8%) and government (3.1%).
It raises the question: Which figures are the more credible? Adcorp, an employment services company, has regularly been criticised for its methodology because of its informal sector statistics. Although labour economist Loane Sharp agreed that it was disputed terrain, he said it was largely because of political reasons.
Detractors, however, claim that the figures are simply unreliable. Rudi Dicks of Naledi, Cosatu’s research arm, described the numbers as a “thumb suck”. And in September last year the National Union of Mineworkers put out a statement criticising many of Adcorp’s reports as “nothing but fallacious fabrications of the truth and an ideological warfare meted [out] against the poorest of the poor”. Because part of Adcorp’s business involves the contentious issue of labour broking, Sharp retorted: “To many people on the left with an axe to grind, we are Satan.”
But DataFirst, a research unit at the University of Cape Town, this week released “A Guide to the Adcorp Employment Index” in which research director Andrew Kerr and acting director Martin Wittenberg questioned Adcorp’s methodology.
As Sharp explained to the Mail & Guardian, the model looks at the ratio of notes in circulation to official measures of gross domestic product and expenditure. In the past decade, cash supplies have grown faster than consumer transactions, something unique to South Africa. “The use of cash has increased dramatically and a lot of what is happening in cash is avoidance of tax, illegal activities and payment of wages to people not considered official employees.”
The gap between cash supply and consumer transactions gives an idea of the size of the informal economy. When this is included in the data it reduces the unemployment rate, which, Sharp said, was the core political issue.
But Wittenberg and Kerr said the index used a “crude version of the currency demand method” to estimate the level of unrecorded activity. They said currency demand methods had been criticised by economists at the International Monetary Fund and the Organisation for Economic Cooperation and Development. “One criticism is that it is not clear what the difference between cash supply and gross domestic expenditure actually represents: cash has uses other than in unrecorded activities.”
But Sharp said his methodology was based on an article by Investec economist Brian Kantor that was published in a peer-reviewed journal in 1989 and for which there have been no refutations.
In spite of the dispute about the informal market, Sharp said there was an 85% correlation in Adcorp and Statistics South Africa’s data. “It shows we are basically measuring the same thing.”
But Stats South Africa statistician general Pali Lehohla described Adcorp’s methodology as “ad-hoc rubbish”.
“If there is no general agreement on methods, there can be no general agreement on figures.” Asked whether Stats South Africa figures did not account for some activities in the informal market, Lehohla said: “It is absolute nonsense. There is nothing to add.”
The Adcorp research noted that South Africa’s income gap would close in less than 10 years if the average white person’s income continued to rise by 5.3% a year while the average black person’s income grew by 14.9%. Already, 1.3-million black employees earn as much as, or more than, the average white employee, almost quadrupling from 270 000 in 2000. (See graph)
Sharp attributed the steep increase in black public servants’ incomes to “managerial bloat”: government wages have increased 5.4% a year, but it has used the twin mechanism of promotion and job regrading to increase the income of black public servants. This has pushed up the average remuneration for public-sector workers to 32% higher than private-sector workers.