What is the Structure of South African Happiness Equations? Evidence from Quality of Life Surveys

The issue of what determines subjective well-being has been at the centre of a recent flurry of research in the economics field. A necessary part of this understanding is the role relative positions (economic, social, geographic) of economic agents, particularly individuals, play in life (commonly referred to in the literature as rivalry). In this paper, we concentrate on whether the structure of happiness equations of South Africa are the same/similar to those of developed countries. The analysis uses three of the Durban Quality of Life Studies. Firstly these three data series are pooled and a variety of covariates are tested for their significance on happiness. These include age, marital status, employment status, household income and relative household income. Next we estimate yearly cross-sectional models to see if there are consistent findings of what determines happiness across the period considered. Our findings indicate there may be some structural differences between results from the Durban studies and those of international findings. Age appears to play no role in happiness likelihood, nor does marital status. Being unemployed does significantly and negatively effect happiness as does the size of household income, relative household income and whether living in a formal dwelling place. When we distinguish between employment categories we find that being self-employed negatively affects happiness, contradicting findings for developed countries. The authors wish to thank members of the Department of Economics, University of the Witwatersrand, participants at the 10th African Econometric Conference, Nairobi 2005 and International Society of Quality of Life Studies, Grahamstown South Africa 2006 as well as one anonymous referee for helpful comments on earlier drafts of this paper.