JABULANI SIKHAKHANE: Question mark over SA’s population and employment data

Economist warns that private datasets offer speed, not representativeness or long-term comparability

Official data sources don’t keep as up to date with changes in the economy as private sources do, the writer says. (123RF/ongcaluck)

By their nature, private sector companies generate a lot of data about economic activity and the behaviour of their customers. This data can offer useful insights into what’s happening in an economy and society at large. But this information must be interpreted with caution, especially where it differs from that gleaned from official sources, including that generated by the national statistical agency.

That’s the message of a recent economic brief, “Public and private labour market data: insights from the government shutdown”, by Claudia Macaluso, a senior research economist at the Federal Reserve Bank of Richmond, one of the 12 regional central banks that make up the Federal Reserve system.

Macaluso’s message has relevance for this country because business people have questioned the accuracy of statistics generated by Stats SA. The relevance of her message is not restricted to the complementarities as well as differences between government and private sector economic data. Her message is also that governments should maintain the integrity of economic measurements by investing in what she calls a “robust public statistical infrastructure”.

Most recently, Alan Knott-Craig, a tech entrepreneur and founder of HeroTel and Fibretime who has a long‑standing interest in township economy development, questioned the country’s population data, arguing that Stats SA’s measure underestimates it by as much as 50%. If true, this would mean the actual South African population is as high as 95-million, versus the official number of 63-million people. Previously, the then CEO of Capitec, Gerrie Fourie, also questioned Stats SA’s measure of employment and unemployment.

Macaluso wrote last week that “large private datasets offer something government statistics cannot: real-time, large-sample observations of economic behaviour”, adding that “because of their size and level of granularity, private data sources often capture detailed economic behaviour that official sources can only approximate in smaller samples”.

As a case study, Macaluso used data from a large job search online platform, which recorded more than 130-million daily job applications for the period January 2020 to November 2025. It also contained employer job postings. The applications contained “precise geolocation data” for both the jobs and the workers and detailed occupational classifications — details that weren’t available from public data sources on vacancies like the job openings and labour turnover survey from the US’s Bureau of Labor Statistics.

However, the advantages of the private data set come with complications. In the case of the online platform data, applicants may have stopped using it in favour of other job search platforms. Therefore, Macaluso says, the data is a partial and selected description of workers’ search behaviour.

She adds that the key difference is that official data sources show the aggregate of the US labour market, while private data sources give only a sliver of that market. Also, the two sources often measure different objects. “Private data sources are generally more responsive to technological change affecting the definitions of economic variables. Public sources guarantee the best comparability across time.”

Official data sources don’t keep as up to date with changes in the economy as private sources do. According to Macalusa, these differences mean private sources can produce timely data, but it may not be comparable across years. She cautions about the use of private sector data as a substitute for official statistics. The private sector can, for example, estimate overall employment from payroll records or job postings but it will find it difficult to measure unemployment.

Macaluso’s comments, not only about the differences between official and private sector data but also about the importance of investing in “robust public statistical infrastructure”, have great relevance for South Africa.

• Sikhakhane, a former spokesperson for the finance minister, National Treasury and South African Reserve Bank, is editor of The Conversation Africa. He writes in his personal capacity.

Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.

Comment icon