Spatial changes in Gauteng’s formal manufacturing jobs: 2014 – 2022

The long term trend for manufacturing jobs in Gauteng is well known: Gauteng has lost significant numbers of manufacturing jobs since 2008 according to The Quarterly Labour Force Survey (Statistics South Africa, 2024). The Quarterly Labour Force Survey has been helpful in analysing employment changes at municipal level, but lacks granular information to understand the spatial shifts. A new dataset called Spatial Economic Activity Data – South Africa (SEAD-SA) (Nell & Visagie, 2023) offers data at the level of planning regions, which are divisions within municipalities. It therefore shows changes in manufacturing employment within Gauteng province at a level of detail that has not been possible before. The data discussed here shows that although there is a net loss in manufacturing employment, not all planning regions are shedding manufacturing jobs – some regions are in fact industrialising.

Manufacturing jobs as a policy objective

Many government strategies including Growing Gauteng Together (GGT) 2030 (Gauteng Provincial Government, 2020) argue that manufacturing jobs are important for promoting economic and employment growth. GGT 2030 captures the response under the heading of Transformation, Modernisation, and Reindustrialisation (TMR). The strategy aims to not only address the decline in employment but to use policy instruments to support job creation and new investments in manufacturing.

These policies respond to what Rodrik calls ‘premature deindustrialization’, where manufacturing investment and employment declines before the sector has matured enough to provide a foundation for the next stage in the country's development (Rodrik, 2015). In South Africa there has been a consistent decline in manufacturing jobs since the mid 2000s, evident in labour market data as well as survey of enterprises (Statistics South Africa, 2023). In a recent book, Owen Crankshaw (2022) explains how Gauteng’s economy has restructured over the last half century away from a Fordist (manufacturing based) economy to a post-Fordist (services based) economy, leaving large numbers of blue collar workers stranded.

The Quarterly Labour Force Survey (QLFS) shows that Gauteng has experienced a large number of manufacturing job losses – with 401 594 jobs lost – between 2008 to 2023. Despite the significant decline in manufacturing jobs, the Gauteng metros and Sedibeng region remain in the top ten districts across the country in terms of manufacturing employment according to a detailed manufacturing study by StatsSA (2023), indicating Gauteng’s continued importance to South Africa’s manufacturing sector.

An important caveat to these calculations is that some jobs may not have been actually lost, but rather transferred out of the manufacturing category in the way employment data is captured. This occurs, for example, where cleaners or security guards working in factories are re-employed by labour brokers. Though their work-location has not shifted, these workers are no longer captured as employed in manufacturing but rather in business services (Tregenna, 2008; Ashman and Newman, 2018).

Using spatial tax data

Recently released spatial tax data allows us to analyse job changes at a smaller scale than was possible with the Labour Force Survey (Robinson, 2022). This dataset utilises tax records from formal businesses reporting data to the South African Revenue Service (SARS). By its nature it does not include those businesses not reporting to SARS, so employment in the informal sector is not captured. Despite these limitations, spatial tax data offers a more spatially granular source of information on economic changes than other freely available sources.

The spatial tax data allows us to break down the manufacturing employment picture to the level of municipal planning regions, thus allowing for a closer look at spatial patterns especially within metropolitan areas. The data mapped in Figures 1 and 2 below shows the number of formal manufacturing jobs in the Gauteng Province in 2014 and 2022.

Figure 1: Number of formal manufacturing jobs in each planning region, 2014

Figure 2: Number of formal manufacturing jobs in each planning region, 2022

Changes in formal manufacturing jobs

The changes in jobs between 2014 and 2022 are presented in Figure 3. Formal manufacturing jobs in 2014 are represented with a teal/blue dot, and those in 2022 with a black dot. These dots are connected with a green line for planning regions that saw an increase in manufacturing jobs, and a red line for those that experienced a decrease.

The data shows that of the 26 planning regions in Gauteng, 13 have experienced manufacturing job losses and 13 have gained jobs. The top 4 regions in 2014 remain unchanged in 2022, and are in Johannesburg and Ekurhuleni.

The data shows that regions in Ekurhuleni (Alberton to Vosloorus, and Kempton Park to Springs) remain the largest areas for formal manufacturing jobs in 2022, but have experienced declines in employment compared to 2014. The data also reveals a significant decline in formal manufacturing jobs in Region 3 of the Tshwane Metro, which includes the area just west of the Pretoria Central Business District. This region has fallen precipitously down the ranks from position 5 in terms of jobs in 2014, to position 13 in 2022.

Figure 3: Change in number of formal manufacturing jobs in Gauteng between 2014 and 2022 in each planning region

These changes between 2014 and 2022 are mapped in Figure 4. The map shows regions of manufacturing job growth (green) and decline (purple) in the province.

Figure 4: Change in the number of formal manufacturing jobs between 2014 and 2022 by planning regions

Understanding job loss areas (purple areas)

The maps indicate areas with job losses in shades of purple. These are:

• Areas adjacent to or in the traditional central business districts (CBD) in Johannesburg (JHB F) and Pretoria (TSH 3): The traditional inner cities of both Johannesburg and Pretoria historically saw close proximity between commercial and manufacturing activities. These areas have experienced significant formal manufacturing job losses. The decline in jobs in Johannesburg Region F (which incorporates the manufacturing belt immediately east and south of the Johannesburg CBD) lost 15 168 jobs, while Pretoria Region 3 lost 26 701 jobs in the same period. Understanding the choices of firms is important here. In the areas adjacent to the Johannesburg and Pretoria CBDs, many manufacturing businesses have moved due to the ageing infrastructure, the collapse in rail transportation reliability and efficiency, and the growth of road-based container transport services. This leads some businesses to favour newer industrial areas closer to arterial roads. Declines in manufacturing jobs are not only due to relocations; many heavy-industry businesses have closed, and, as noted above, some manufacturing jobs have not been lost but have, rather, been reclassified.
• Bronkhorstspruit (TSH 7): The significant decline of manufacturing jobs in Bronkhorstspruit has a historical context. The Ekangala area was historically part of the KwaNdebele homeland, where manufacturing was supported by subsidies provided by the apartheid government as part of its deconcentration strategy. The industrial area, Ekandustria, has declined as subsidies were withdrawn. Efforts to establish new industries in the area have not reversed this broad trend (Peberdy et al., 2017; Todes & Houghton, 2021).
• Vereeniging and Vanderbijlpark (Emfuleni): The area is a critical node for industry in South Africa as some of the country’s largest steel plants (Union Steel Corporation (USCO), and ISCOR) have historically been located there. The decline in manufacturing jobs in the Vereeniging and Vanderbijlpark areas is associated with slow adoption of technology and limited growth in markets (Hlatshwayo, 2014).
• Parts of Ekurhuleni: The decline in manufacturing jobs is not uniform across Ekurhuleni. Towards the edges of Ekurhuleni, towns such as Devon, Nigel and Hiedelberg (EKU E) have experienced relatively sharp declines in manufacturing. Efforts to diversify manufacturing products in these areas have been ongoing, the most notably being the railway manufacturing facility in Nigel (EKU E) (James, 2018). This is observable if one compares the areas in and around Springs, which is one of the anchor points in the maps. The pattern is not a new one with the declining manufacturing base being a concern for the last three decades (See Centre for Development & Enterprise, 1997 for one of the first analyses of policy options in the East Rand, now called Ekurhuleni’).

Job gains (green areas)

The maps indicate areas with job gains in shades of green. These include:

• Centurion (TSH 4) / Midrand (JHB A) / OR Tambo (EKU B) agglomeration: The data suggests that jobs traditionally located in the old manufacturing belts adjacent to inner cities may be shifting spatially to secondary nodes. The maps suggest that an agglomeration of manufacturing jobs along the Centurion and Midrand corridor has developed, with moderate increases in these areas. An important intervention here is the OR Tambo Special Economic Zone (SEZ) (also called Gauteng IDZ).
• There has also seen a concentration of investments in the automotive sector by multinational corporations in the Rosslyn (TSH 1) and Silverton (TSH 3) areas in Tshwane. Government has established the Tshwane Automotive Special Economic Zone (TASEZ) to strengthen investment in these areas.
• Peripheral growth in manufacturing jobs: Mogale City and Midvaal (along the R59 corridor) show moderate increases in manufacturing. The increases are important to note (even if from a small base) as they may represent areas of potential growth in jobs.

Conclusion

Spatialised tax data is important in supporting economic policy making in government, as well as investments by the private sector. Mapping the changes in the number of manufacturing jobs over time, in conjunction with additional economic sector analysis, can help identify areas facing premature deindustrialization. In turn, it allows for the identification of areas of potential growth. In so doing, the data assists in supporting evidence-based policy making at a national, provincial and local levels. This data, together with other data sources indicating broader economic trends and local factors, provides a way to go beyond the headline that we have lost manufacturing jobs in Gauteng, to ask questions of why this has happened and where has this happened. In turn, analysis based on this data can contribute to strategies that support future efforts to arrest job losses through appropriate interventions, and foster reinvestment and employment creation.

References

Ashman, S., & Newman, S. (2018). The evolution of manufacturing in the Gauteng city-region: From DE-industrialization to re-industrialization? In The Changing Space Economy of City-Regions (pp. 131–156). Springer International Publishing.

Centre for Development, & Enterprise. (1997). The East Rand: Can South Africa’s workshops be revived? Centre for Development and Enterprise. https://www.cde.org.za/the-east-rand-can-south-africas-workshops-be-revived/

Crankshaw, O. (2022). Urban Inequality: Theory, Evidence and Method in Johannesburg. Bloomsbury Press.

Gauteng Provincial Government. (2020). Growing Gauteng Together 2030. https://www.gauteng.gov.za/Publications/PublicationDetails/%7B7AB8D1F1-9261-4E7C-A4F4-ED5A0F7D6DA8%7D

Hlatshwayo, M. (2014). NUMSA and solidarity’s responses to technological changes at the ArcelorMittal vanderbijlpark plant: Unions caught on the back foot. Global Labour Journal, 5(3). https://doi.org/10.15173/glj.v5i3.2277

James, N. (2018, October 25). Ramaphosa inaugurates R1bn train manufacturing facility, training centre in Nigel. Creamer Media. https://www.engineeringnews.co.za/article/gibela-2018-10-25

Nell, A., & Visagie, J. (2023). Spatial Tax Panel 2014-2022 [dataset]. Version 3. National Treasury - Cities Support Programme and Human Sciences Research Council [producer and distributor] [dataset]. In Spatial Tax Panel 2014-2022 [dataset]. Version 3. https://spatialtaxdata.org.za/dashboards-landing

Peberdy, S., Harrison, P., & Dinath, Y. (2017). Uneven spaces: core and periphery in the Gauteng City-Region. GCRO Research Report. Gauteng City-Region Observatory. https://doi.org/10.36634/DTLV3720

Robinson, S. (2022). Adventures in City Data: An ethnography. GCRO Occasional Paper. Gauteng City-Region Observatory. https://doi.org/10.36634/RPLR3362

Rodrik, D. (2015). Premature Deindustrialization (No. 20935). National Bureau of Economic Research. https://doi.org/10.3386/w20935

Statistics South Africa. (2023). Manufacturing industry – financial detail, 2021 (Nos. Report-30-02-03 -). Statistics South Africa. https://www.statssa.gov.za/?page_id=1854&PPN=Report-30-02-03

Statistics South Africa. (2024). Quarterly Labour Force Survey (QLFS), 4th Quarter 2023 (Nos. P0211 - 4th Quarter 2023). Statistics South Africa.

Todes, A., & Houghton, J. (2021). Economies and employment in growing and declining urban peripheries in South Africa. Local Economy, 36(5), 391–410.

Tregenna, F. (2008). The contributions of manufacturing and services to employment creation and growth in South Africa. The South African Journal of Economics. Suid-Afrikaanse Tydskrif Vir Ekonomie, 76(s2), S175–S204.

Edits and inputs: Richard Ballard, Graeme Götz and Christian Hamann

Map design: Jennifer Murray

Suggested citation: Hassen, E. K., Naidoo, L., Modiba, M., Mushongera, D. and Labuschagne, H. (2024). Spatial changes in Gauteng’s formal manufacturing jobs: 2014 – 2022. Map of the Month. Gauteng City-Region Observatory. April 2024. https://doi.org/10.36634/LDPH9021