Resource Rents, Coercion, and Local Development: Evidence from Post-Apartheid South Africa¶

Literature › Works

Paulo Bastos; Nicolas Bottan (2016). Resource Rents, Coercion, and Local Development: Evidence from Post-Apartheid South Africa. Policy Research Working Paper 7572. World Bank, Development Research Group, Trade and International Integration Team.


Type	working paper
Scope	Post-1994 South African economy
Method	econometric
Period studied	1996–2011
Regions	South Africa, former homelands/bantustans
Themes	bantustans removals influx, mining migrant labour, unions strikes, labour markets wages, inequality poverty unemployment, transition post apartheid
Citation key	`bastos2016`

Argument¶

The paper argues that the interaction between the dismantling of apartheid's coercive labour institutions and the distribution of natural resource rents was an important driver of post-1994 income convergence in former homeland communities. Using 1996 census data it documents large income gaps between communities just inside and just outside former homeland borders, and then shows that 1996-2011 income growth was considerably stronger in homeland communities with higher initial mining employment shares. A unionised international oligopoly (Nash bargaining) model frames this: democracy raised union bargaining power and workers' reservation wages (via removal of mobility and job restrictions), so black mineworkers captured a larger share of mining rents, amplified by the post-2000 commodity price boom. Robustness checks address infrastructure roll-out, market access, migration/demography and a placebo with other sectors.

Evidence and method¶

Community (sub-place) profiles from the 1996 and 2011 population censuses (Statistics South Africa), matched by centroid (~16,000 communities; ~9,583 within 30 km of homeland borders); cartographic data on former homeland boundaries; 1996 census 10% individual sample for wage regressions; 1996 October Household Survey and 2011 Labour Force Survey for unionisation rates; US Geological Survey Minerals Yearbooks; secondary historical accounts (mainly Clark and Worger 2011, Horrell 1978).

Findings¶

Large income gap at the homeland border in 1996, narrowing in relative terms by 2011. (p. 16)

In 1996, the average per capita income just- inside and outside the homelands was 4,526 and 20,651 rand, respectively. At the border the diﬀerence was almost 14,000 rand.
Relative gap fell from about 5 to 3.5 times despite rising absolute gap. (p. 16)

Although the gap at the border increased in absolute terms to around 29,000 rand in 2011, it fell in relative terms from about 5 times to 3.5 times
Main effect size: one SD higher mining employment share in a homeland community associated with higher income growth. (p. 18)

increasing the share of employ- ment in mining by one standard deviation (0.103) is related to a 4.2 percentage point higher total growth rate during the 1996-2011 period.
In 1996 individual data, mining carried a large wage premium and black workers faced a large earnings gap. (p. 19)

we ﬁnd a large wage premium of around 40 percent associated to employment in this industry. At the same time, there is a large income gap against black workers of approximately 50 percent
Black mineworkers initially captured less of the mining rent premium. (p. 19)

We also observe that black mining workers initially beneﬁted less from resource rents than non-blacks.
Historical wage illustration of mining coercion cited from a web source. (p. 9)

in 1972 their wages were about 18 Rands per month, less than the two shillings six pence they had been making in 1902
Black union membership growth before 1994. (p. 10)

By 1984, over half a million black workers were members of organized trade unions. Empowered by a progressively larger membership—reaching about 2.5 million members by the early 1990s
Placebo: only mining shows positive effect; manufacturing and services, though unionised, do not. (p. 23)

Reassuringly, we only ﬁnd positive and signiﬁcant estimates for the mining sector. Both manufacturing and community services have relatively high rates of unionization (over 40% in 1996)
Market access mattered for homeland income growth. (p. 21)

The closer a community is to a major city, the stronger is its growth in income. This eﬀect is even more pronounced for communities located inside homeland areas.
Little evidence that migration explains results. (p. 22)

the estimated coeﬃcients for the initial share of employment in mining and its interaction with the homeland indicator are also not statistically signiﬁcant. These results provide little support to the migration channel.