Estimating the impact of large-scale mining on local communities in sub-Saharan Africa

ABSTRACT

The impact of large-scale mining in sub-Saharan Africa on local livelihoods is controversial. Little consensus exists about whether mining is a ‘blessing’ or a ‘curse’ for affected communities. This article estimates the impact of mine proximity on local socioeconomic outcomes. It does so by linking georeferenced survey data to mine locations, to determine whether living near a mine results in systematically better or worse scores on measures of lived poverty, development infrastructure and access to basic household services. The data (from Afrobarometer Round 6) spans 19 countries, 148 large-scale mines, and 4,796 households. The analysis shows that the average effects of proximity to a mine are modest compared with cross-national and urban-rural differences in living conditions. Mining is not consistently a ‘blessing’ or a ‘curse’ for affected local communities in sub-Saharan Africa.

KEYWORDS:

• Afrobarometer
• large-scale mining
• local communities
• resource curse
• lived poverty

Acknowledgements

Thank you to Mr Jeroen Schmidt for his input regarding spatial analysis, and to Prof. Rod Alence for his guidance during the early stages of this research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data statement

Kindly note that access to Afrobarometer’s low-level geographic data, as used in Estimating the Impact of Large-scale Mining in sub-Saharan Africa, is restricted. Access can be granted to researchers who undertake a personal commitment to use the Afrobarometer geocoded data securely. If users which to access the restricted geographic data, please complete an application form which is available on the Afrobarometer website. Please email your signed application to the Data Management Unit, University of Cape Town [email protected]

Notes

1 Human Development Index

2 Development Infrastructure Index

3 Household Infrastructure Index

4 Lived Poverty Index

5 A mine in the feasibility, advanced exploration or development stage

6 Corporate Social Responsibility

7 Primary sampling unit, which is an official enumeration area (EA) boundary. Coordinates are anonymised, i.e. the geographic location is the centre point within each EA.

8 Mines that fit neither of these definitions were discarded

9 Afrobarometer’s definitions of urban and rural differ from country to country, since countries have different criteria (Emuze & Hauptfleisch 2014). Generally, urban samples are high density large settlements, e.g. Botswana’s urban areas are settlements larger than 5000 people. Malawi’s urban areas are cities with wards. For the Botswana and Malawi samples, semi-urban and peri-urban were recoded to ‘rural’, in accordance with Afrobarometer’s sampling specialists recommendations (Prof. F Kibirige & Prof MG Tsoka).

10 Diagnostic plots, bivariate analysis and robustness checks are available in the online appendix.

11 Figure A.1 in the online appendix.

12 Graph available in online appendix.

13 Please see Figures A.4 – A.6 in the online appendix for the random effects plots for all 3 indicators.

14 Urbanisation is the ‘large-scale transition and centralisation process of resources, industries and population in a certain area’ (Afrobarometer 2014) and usually coincides with a shift from an agrarian to a manufacturing and service-based economy.

15 Environmental Justice Organizations, Liabilities and Trade