About the data
Why do the numbers disagree?
Search for South Africa's population or unemployment rate and you will find more than one answer. Usually none of them is wrong: they measure different things, at different dates, in different ways. This page sets the common ones side by side.
Population: 62.0, 63.5 or 65.5 million?
| Figure | People | Refers to |
|---|---|---|
| People counted in Census 2022A count of every household, adjusted with the Post-Enumeration Survey for people the count missedThe census, held in 2022Stats SA, Census 2022 Provinces at a glance | 62 027 503 | The census, held in 2022 |
| Mid-year population estimateA model that rolls the population forward with estimates of births, deaths and migration30 June 2026Stats SA, Mid-year population estimates 2026 (P0302) | 63 522 380 | 30 June 2026 |
| UN projection, medium variantThe UN's own model, run the same way for every country1 July 2026UN, World Population Prospects 2024 | 65 453 084 | 1 July 2026 |
They refer to different dates. The census describes 2022. The mid-year estimate and the UN figure describe mid-2026, four years of births, deaths and migration later.
They are made differently. Only the census tries to count everyone. Stats SA describes its mid-year estimate like this: “a base population is estimated that is consistent with the known demographic characteristics of the country. The cohort-based population is projected into the future according to the projected components of change.” The UN runs a similar kind of model with its own estimates of births, deaths and migration, and gets 1.9 million more people for 2026.
Estimates are revised every year. Each mid-year release replaces the whole series back to 2002. Stats SA warns: “It would be misleading to compare values and rankings with those of previously published reports, due to revisions and updates of the underlying data and adjustments.” A 2026 figure from the 2025 release and one from the 2026 release are not the same estimate.
Which to use: for official South African purposes, the latest Stats SA mid-year estimate. For comparing South Africa with other countries, the UN series, since it is made the same way for all of them. Quote the source and the date with either.
Source: Stats SA, Census 2022 Provinces at a glance (Report no. 03-01-43), national population, Census 2022. Originally released , revised edition. Downloaded . Checked on SA Data Hub . Original data: Stats SA. Any calculation shown is SA Data Hub's own processing.
Source: Stats SA, Mid-year population estimates 2026 (P0302), Summary, page vii (63,52 million) and Introduction, page 1 (both quotes). Source released . Checked on SA Data Hub . Original data: Stats SA. Any calculation shown is SA Data Hub's own processing.
Source: United Nations, Department of Economic and Social Affairs, Population Division, World Population Prospects 2024, UNdata table "Total population, both sexes combined (thousands)", South Africa, 2026, Medium. Source released . Checked on SA Data Hub .
Unemployment: 33.6% or 43.8%?
Stats SA publishes four measures of people who want more work than they have. The official unemployment rate is the narrowest. The others add people working fewer hours than they want, and people who want work but were not looking or could not start straight away.
| Measure | Q2 2026 |
|---|---|
| LU1: Official unemployment rate“Unemployment rate is the proportion of the labour force that is unemployed.” (Concepts and definitions, page 18) | 33.6% |
| LU2: Unemployment and time-related underemployment“Persons in underemployment (time-related) are employed persons who were willing and available to work additional hours, whose total number of hours actually worked during the reference period were below 35 hours per week.” (Concepts and definitions, page 18) | 36.6% |
| LU3: Unemployment and potential labour force“The potential labour force includes persons who are not in the labour force, but are either available for work but not actively seeking employment or actively seeking employment but not currently available to work.” (Concepts and definitions, page 18) | 43.8% |
| LU4: Composite measure of labour underutilisation“the composite measure of labour underutilisation (LU4) that combines unemployment, time-related underemployment, and potential labour force.” (Highlights of the results, page 2) | 46.3% |
Stats SA added LU2 to LU4 when it revised the survey in the third quarter of 2025. In its words, the report “now includes other labour underutilisation indicators in addition to unemployment, primarily because unemployment is not the only measure of labour underutilisation.” The change did not break the official rate: “headline unemployment figures remain comparable and consistent with previous quarters.”
Many news stories still quote an “expanded” unemployment rate. The two labour force releases SA Data Hub holds from after the change, for the fourth quarter of 2025 and the second quarter of 2026, do not print one. They print LU1 to LU4 instead, so check which measure a story means before comparing it with these.
Which to use: the official rate (LU1) when comparing with past years, since it is the series with the longest comparable history. LU3 or LU4 when the question is how many people want more work than they have.
Source: Statistics South Africa (Stats SA), Quarterly Labour Force Survey, Quarter 2: 2026 (P0211), Table of labour underutilisation indicators, Concepts and definitions page 18, and Introduction page 1 (quotes). Source released . Checked on SA Data Hub . Original data: Stats SA. Any calculation shown is SA Data Hub's own processing.
Found another pair that disagrees?
This page grows as readers send in figures that do not match. Corrections and additions are listed on the updates page, and the methodology page explains how SA Data Hub checks every figure.