Statistics South Africa is vital in providing national and provincial data, oering quarterly employment indicators for the country. However, accurately estimating district (subnational) unemployment rates presents significant challenges due to high sampling variability in the Quarterly Labour Force Survey (QLFS) within smaller areas. To tackle these issues, this study employs a spatiotemporal small area estimation (SAE) technique that integrates both spatial and temporal dimensions. This study uses data from the QLFS (Q1 2022 to Q1 2024) along with auxiliary variables from the 2016 Community Survey to produce robust quarterly estimates of unemployment rates for all 52 district municipalities. Our analysis indicates a statistically robust and consistently observable spatial inequality in unemployment rates between metropolitan and non-metropolitan districts. Specifically, metropolitan districts tend to exhibit lower and more stable unemployment rates, primarily attributed to the
diversity of their economic activities across various sectors. In contrast, non-metropolitan districts experience higher and less stable rates, mainly due to limited industrial diversity and a dependence on seasonal jobs. Our robust methodology also highlights practical implications by identifying specific geographic hotspots of youth unemployment, particularly among the 1524 and 2534 age groups. Additionally, we observe distinct gender disparities, with women facing greater barriers to employment than men. The primary contribution of our study is methodological, as it provides a robust tool for informing evidence-based interventions in local workforce monitoring. Our findings offer valuable evidence for prioritising place-based development initiatives and emphasise the necessity of moving beyond
a one-size-fits-all approach in national policies to effectively address South Africa's deeply entrenched spatial inequities.