Скачать или смотреть LSE III Event | Inequality measurement for bounded variables

Hosted by the International Inequalities Institute on Tuesday 23 January 2024.

Speaker:
Suman Seth, Leeds University Business School

Chair:
Professor Kirsten Sehnbruch, British Academy Global Professor and Distinguished Policy Fellow at the III

We propose a novel approach to measure inequality for variables with fixed lower- and upper bounds. Inequality assessment with such variables is different from that with variables lacking fixed upper bounds. The maximum-inequality distributions of non-bounded variables always feature every element, but one, equal to their lower bounds, and many existing inequality measures rank them equally. However, the maximum-inequality distributions of bounded variables contain different proportions of elements being equal to the lower bound. We normatively justify a novel axiom requiring maximum-inequality distributions of bounded variables to be ranked equally, irrespective of their means. Our axiomatically characterised indices measure inequality as the observed proportion of the maximum attainable inequality for a given mean. We also characterise a subset of measures that yield consistent inequality comparisons when switching between attainment and shortfall representations. In our empirical illustration, a starkly different picture emerges when traditional inequality indices give way to our approach.