The atlas of inequality aversion: Theory and empirical evidence from the Luxembourg Income Study Database

In the distributive analysis, the constant relative inequality aversion utility function is a standard tool for ethical judgements of income distributions. The sole parameter ε of this function expresses a society's aversion to inequality. However, the profession has not committed to the range of ε. This paper aims to estimate the parameter ε of the constant relative inequality aversion utility function using datasets available from the Luxembourg Income Study Database. We utilise the method of estimating ε assuming incomes obey the generalised beta distribution of the second kind. The estimator of ε is derived from the mathematical condition of the existence of the social welfare function. We elaborate an 'atlas' of 388 estimates of ε for 55 countries across time. We also verify two hypotheses: 1) The richer the country, the greater the societal inequality aversion; 2) The greater (lower) the inequality aversion, the lower (greater) income inequality. Our data do not confirm the 1st hypothesis. For verifying the 2nd hypothesis, we use the inequality-development relationship augmented by inequality aversion. The 2nd hypothesis is unfalsified in about 90% of country-year cases.