Seminar series - “Poverty and ethnicity among black South Africans”
This Wednesday’s seminar will be presented by Carlos Gradín.
“Poverty and ethnicity among black South Africans”
The huge white-black differential in poverty in South Africa, its determinant factors and evolution after the Apartheid have been already studied (e.g. Gradín, forthcoming in Journal of African Studies). However, intra-group differentials among blacks of different ethnicities have been less investigated so far. Given that these groups represent about 80% of the country’s population, and given the increasing role of their growing heterogeneity in shaping the country’s income distribution, a better understanding of these differentials is crucial to ascertain the present and future of wellbeing in South Africa. The aim of this paper is to analyze differences in poverty levels among the main ethnic African groups in South Africa. In this paper, we first document ethnic differences in poverty levels and then study the extent to which they are driven by diverging socioeconomic characteristics across ethnic groups or, otherwise, are the consequence of these characteristics having a different impact on poverty across ethnic groups. For the analysis, we will use a nationally representative sample of private households in South Africa with rich information on households’ living conditions: the National Income Dynamics Study (NIDS, 2008 and 2010 waves) provided by the Southern Africa Labour and Development Research Unit (SALDRU, University of Cape Town). These datasets will be complemented with data from other sources, such as the Project for Statistics on Living Standards and Development (PSLSD 1993), in order to analyze post-Apartheid trends. Ethnicity is approached here by the language spoken at home: IsiZulu (29.1%), IsiXhosa (21.6%), Sepedi (12.2%), Sesotho (10.6%), Setswana (8.5%) or other language (the remaining 18%, including IsiNdebele, SiSwati, Tshivenda, Xitsonga, Afrikaans, English, other language and unknown). Poverty levels will be measured based on income and expenditure, using the usual national poverty thresholds. The main methodology to disentangle the driving factors of ethnic poverty differentials will be Blinder-Oaxaca-type regression-based techniques, adapted for the study of poverty rate differentials (e.g. Yun, Economic Letters, 2004 and related literature). Preliminary results show that per capita income of the two main groups (Xhosa and Zulu) is the lowest among all ethnic groups, around 76-78% of the average for Africans, compared with 135% of Setswana or 128% of “other groups”. Similarly, their poverty rates are highest among all, 62% for Xhosa and 65% for Zulu, using the lower poverty line, compared with 42% of Setswana or 45% of Sesotho. About 68% of the higher poverty rates of Xosha and Zulu, compared with Setswana, are explained by their poorer characteristics, although with some differences. In both cases the most important factors are the demographic differences (especially having more children and dependent adults) explaining around 25% of the differential. Then comes education, that explains a larger share of the differential for Xhosa (22%) than for Zulu (19%), and the labor market participation of household members, which explains a higher proportion for Zulu (19%) than for Xhosa (13%). The overrepresentation of these groups in rural areas also explains a significant 8% in the case of Xhosa (but is not statistically significant for Zulu).
As usual we start at 1pm in Economics' Seminar Room, Level 4, Economics Building, Middle Campus
Reza Daniels, lecturer in the School of Economics and SALDRU associate, was recently placed second in the Cochran-Hansen competition for the best paper on survey research methods submitted by a young statistician from a developing country or transition country. This prize is awarded every two years by the International Association of Survey Statisticians, a division of the International Statistical Institute.
Cally Ardington, an Associate Professor within SALDRU, has recently been made a J-PAL Africa affiliate. Within the sphere of randomised control trials, she is currently involved in an evaluation of a holiday program designed to support Grade 4 learners in the transition to English as the language of teaching and learning in South Africa.