Self-centered and non-self-centered inequality aversion matter: Evidence from Uruguay based on an experimental survey

In this paper we provide evidence on the empirical relevance of two notions of inequality aversion that have been explored in the literature: self-centered and non-self-centered inequality aversion. We used a flexible model and designed an experimental survey that allowed us to address jointly both of these notions of inequality aversion and to distinguish their relevance. The survey was administered to a sample of first-year University students in Uruguay. The findings confirm the empirical relevance of both notions of aversion to inequality in a developing country. Most study participants exhibited non-self-centered inequality aversion, while a minority of the individuals in our sample appeared to favor inequality. In general the magnitude of aversion to inequality varied as a function of individuals’ position in the income distribution. self-centered income aversion is influenced by many factors, which is manifested in the fact that the magnitude of its parameters is more heterogeneous in compression to non-self centered aversion. In a minority of individuals, self-centered aversion has zero effect, and they are more willing to pay to reduce non-self-centered inequality. Finally, considering both notions together may help prevent bias in the measurement of inequality aversion.