Labour Force Survey Data is Imperative

The official government statistical agency, Statistics South Africa (SSA), recently suggested that it intends to amend what is arguably the most comprehensive survey of employment and socio-economic status in South Africa – the Labour Force Survey (LFS). This move is unfortunate and inappropriate. The LFS is currently a bi-annual survey that collects data from some 30,000 households of all races and socio-economic status in South Africa. The data collected through the survey forms an invaluable source of information for econometricians and statisticians, which allows policy makers and other interested parties to make informed decisions based on sound economic data that extends across a number of time periods.

The dearth of adequate employment statistics is already a vexing problem for statisticians trying to make meaningful interpretations of the available data. By amending the publication it will make the analysis of time series data extremely cumbersome – if not impossible. Data comparable across time periods provides a better indication of changes in the labour market and allows policy makers to adopt or change existing policies based on the information gathered. Furthermore, by amending this essential publication it renders all past studies and interpretations of the existing survey effectively useless since the intended replacement publication will not be consistent with historical data.

SSA’s continued efforts to improve the quality of its statistics should be commended. However, until the new version of the LFS is comparable across time periods South Africa will no doubt have to rely on the Quarterly Employment Survey (QES) as the main indicator of employment. The QES is however a somewhat watered down version of the LFS, since it focuses exclusively on formal sector employment and non-farm enterprises. The list of registered employers, from which the samples for the QES survey are drawn, has traditionally been dogged with inconsistencies making any meaningful analysis of the QES across extended time periods problematic.

The data collected through the LFS since September 2001 suggests that the number of employed people has been steadily increasing. Over the period September 2001 to September 2006 the cumulative employment gains in the labour market totalled 1.6 million. Yet South Africa continues to suffer from a high rate of unemployment. According to the latest LFS data, approximately 4,391 million working aged adults were officially unemployed. This equates to an official unemployment rate of 25.5%. However, a further 3,217 million did not actively search for work in the month prior to the survey, but indicated they would work if there were jobs available. If we include these discouraged work seekers the total number of unemployed increases to 7,608 million and the unemployment rate shoots to the more realistic rate of 37.3%.

Most countries favour the strict definition of unemployment, since it is assumed that the broad definition overstates the labour supply and that those that are not actively searching for work are simply voluntarily unemployed. The strict version is the measure officially recognised by the International Labour Organisation (ILO). However, the ILO supports appropriate national tests and in a country such as South Africa where there is a large number of unemployed individuals, it is not unreasonable to assume that many individuals stay at home because they simply believe that there is no work available.

The presence of a large body of job seekers, whether discouraged or not, has a dampening effect on wages and the lack of consistent data will interfere with the functioning of the wage setting machinery. By choosing to neglect the publication of data based on the broad definition of unemployment, SSA are attempting to make a dire situation look somewhat more palatable. Instead, it downplays the misfortune of millions of people currently facing the hardships associated with unemployment.

We should not attempt to disguise the reality of South Africa’s unemployment problem. On the contrary, we need to be constantly reminded that there are suffering families out there and that something needs to be done. It is essential for SSA to continue to produce consistent reliable data that reflects a holistic perspective of the labour market so that appropriate policies can be adopted to address the chronic unemployment problem. If the ruling African National Congress is to meet its target of halving unemployment and poverty by 2014, it needs the backing of a solid statistical foundation otherwise any progress made will appear to be the product of statistical manipulation.

Author: Jasson Urbach is an economist with the Free Market Foundation. This article may be republished without prior consent but with acknowledgement to the author. The views expressed in the article are the author’s and are not necessarily shared by the members of the Free Market Foundation.

FMF Feature Article/ 24 April 2007