Skip to content

Spatial Development Initiatives and Employment Creation: Will They Work?

Literature › Works

Naude, C.M.; McCoskey, S.K. (2000). Spatial Development Initiatives and Employment Creation: Will They Work?. 2000 Annual Forum, Glenburn Lodge, Muldersdrift, 18-20 September 2000 (Trade and Industrial Policy Strategies / Economic Society forum).
Type conference paper
Scope Post-1994 South African economy
Method mixed
Period studied 1980–1998
Regions South Africa, Gauteng, KwaZulu-Natal, Western Cape, Northern Province, Maputo corridor
Themes manufacturing industrial policy, transition post apartheid, labour markets wages, bantustans removals influx, demography urbanisation, comparative theory
Citation key naude2000

Argument

The paper assesses whether the post-1995 Spatial Development Initiatives (SDIs) can restructure the economy and create jobs, using Krugman-style new economic geography (concentration, core-periphery, transport costs). It finds that SDI projects are very capital-intensive per job and concentrated in declining traditional sectors (agriculture, mining, manufacturing). A magisterial-district regression suggests SDI areas coincide with areas of 1996 employment, though this may be coincidence since the programme was widened over time. The authors argue that government can only influence firm-level investment indirectly, through infrastructure, lower transport costs and education.

Evidence and method

SDI project database (www.sdis.org.za); World Bank sectoral GDP data; Development Bank of Southern Africa provincial GDP/GGP data; Statistics South Africa 1996 census and magisterial-district employment data for 1980, 1991 and 1996 (nine-sector SIC classification); Krugman (1991) and Krugman and Venables (1998) theory; Department of Transport 'Moving South Africa' finding on road transport costs.

Findings

  • SDI projects were identified with large capital value but modest job potential. (p. 4)

    The SDIs referred to above have identified projects to the value of $32,4 billion , creating 86000 jobs.

  • Capital cost per job varies enormously across sectors and is high on average. (p. 5)

    The capital cost per job varies considerably across the sectors, e.g. R1,486 per job in the case of Fishing and Marine products, to R2,674,440 per job in the case of Food & Beverages, with an average of R375,148 overall.

  • South Africa's economy shifted away from industry towards services between 1980 and 1998. (p. 8)

    Table 3 shows that the South African economy has become less reliant upon the traditional sectors of agriculture and manufacturing and more of a service-based economy, with a contribution of 64% from this activity

  • Inland core provinces lost GDP share in 1991-96 while coastal provinces gained, suggesting a shifting core. (p. 9)

    the traditional inland core areas of the economy, e.g. Gauteng and Free State have contributed a declining share of GDP, while the contribution of Northern Province, KwaZulu-Natal and Western Cape have increased.

  • Inland transport costs are high and may explain the shift of industry towards ports. (p. 11)

    The Moving South Africa project of the Department of Transport found that inland road transport costs in South Africa were as much as 30% higher than the equivalent haulage rates in the United States

  • Withdrawal of apartheid-era decentralization incentives was followed by disinvestment from homeland areas. (p. 11)

    Since the withdrawal of these incentives by government, there has been extensive disinvestments from these areas.

  • SDI areas outperformed non-SDI areas in employment growth in 1980-91. (p. 14)

    between 1980 and 1991, employment in the areas now known as SDIs grew by 0.059% per annum on average, versus a contraction in employment in the non-SDI areas of 0.018% per annum.

  • The pattern reversed in 1991-96, with SDI areas shedding jobs. (p. 14)

    SDI areas shed employment opportunities (-0.012% p.a.), while those of non-SDI show employment creation of 0.02% p.a.

  • SDI project employment is concentrated in sectors that have been shedding jobs. (p. 14)

    The Manufacturing sector has also been identified as a source of new employment opportunities (40,106 new jobs or 47% of the total) but this sector has also been contracting over time.

  • The regression finds a positive, significant association between SDI location and 1996 employment, with a caveat about coincidence. (p. 16)

    The results indicate the employment in South Africa as per 1996 and the selection of SDI areas are positively related (0.341). The results are, moreover, statistically significant (t-value of 2.331).

  • Manufacturing employment growth enters the regression negatively and significantly, a worrying sign for industrial job creation. (p. 16)

    manufacturing employment has declined in the period (-0.429) but is significant (t-value of -5.022), while that of mining is positive (0.013) but is not a significant result