No abstract

We present two experiments designed to investigate whether individuals' notions of distributive justice are associated with their relative (within-society) economic status. Each participant played a specially designed four-person dictator game under one of two treatments, under one initial endowments were earned, under the other they were randomly assigned. The first experiment was conducted in Oxford, United Kingdom, the second in Cape Town, South Africa. In both locations we found that relatively well-off individuals make allocations to others that reflect those others' initial endowments more when those endowments were earned rather than random; among relatively poor individuals this was not the case.

This article uses data collected across the four Waves of the National Income Dynamics Study (NIDS) to construct a measure of social cohesion for South Africa. We compare our index to one derived using the Afrobarometer data and find a large degree of consistency in trends in the index and its constituent components over time across the two datasets. However, there is less consistency in the measures once one moves to lower levels of geographic disaggregation. We also find far less variability in the constructed index relying on NIDS panel data as opposed to the repeated cross-sections from Afrobarometer. Having derived the index, we then correlate it with a variety of indicators of social and economic well-being. We show that higher levels of education, per capita income and employment are positively associated with higher social cohesion whilst social cohesion is negatively associated with poverty, service delivery protest and perceptions of crime. In addition, municipal policy and competence are closely associated with higher social cohesion. Whilst this work is exploratory, it is encouraging, and suggests new opportunities for future research to begin to take the link between social cohesion and economic and social development seriously.