Multidimensional Inequality Indices Research Articles

The Measurement of the Multidimensional Inequality Index in Urban and rural Areas of Iran during the period 1984-2018

The Measurement of the Multidimensional Inequality Index in Urban and rural Areas of Iran during the period 1984-2018

Pandemic and the Crisis of Extreme Inequality in India

The pandemic and the subsequent lockdown created a public health and humanitarian crisis for millions of workers and families in India. Through extensive data analyses of multiple nationally representative sample surveys, we argue that the effects of the pandemic and the associated lockdown have been exacerbated by the pre-existing crisis of extreme inequalities in the country. Multiple dimensions of inequality are explored, such as the disparities in the labour market and in the access to basic amenities. We briefly indicate our most striking findings. 63 percent of the workers and 53 percent of the households in India earned less than Rs 10,000 per month in 2018-19. Substantial gaps in earnings by gender, caste and area of residence persist – on average, female earnings was 63 percent of male earnings, SC earnings was 55 percent of non-ST/SC/OBC earnings and rural earnings was only half of urban earnings in 2018-19. About 905 million people did not have access to piped water, 287 million did not have access to toilets, 127 million lived in rented accommodations, and one-fourth of the population lived in single-room dwellings in 2017-18. The implications of the long-term neglect of the public healthcare system and the disparities in the access to education are discussed. In addition to inequality measures of monetary measures such as household earnings, we construct a multi-dimensional inequality index that combines indicators in housing, water and sanitation. This gives insights on the extent of inequality in non-monetary dimensions. The evidence in this paper clearly identifies that extreme multi-dimensional disparities and disadvantages lie at the heart of the current crisis in India and highlights the need to put redistribution at the centre stage of the development agenda.

Normative properties of multidimensional inequality indices with data-driven dimensional weights: The case of a Gini index

A central problem in constructing multidimensional inequality indices is that of devising weights on the dimensions. There are two different approaches to the problem: the normative and the data‐driven. Indices derived from data‐driven weights have the limitation that they may violate normatively desired properties. This paper asks whether it is possible to obtain normatively acceptable inequality indices from a data‐driven approach. A multidimensional Gini index is derived from an endogenous weighting scheme and is shown to possess a number of desired properties. The existing literature does not seem to contain a Gini index that satisfies all of these properties.

EVOLUTION OF INEQUALITY IN LATIN AMERICA (1980-2014): A MULTIDIMENSIONAL APPROACH BEYOND INCOME

Latin America is the region with the greatest inequality in terms of incomes. Thus, during the last years, the United Nations Development Programme (UNDP) has showed concern about the obstacle that inequality causes in the human development of this region. Given that higher economic growth does not necessarily lead to the achievement of further progress social, development beyond income will be one of the most important UNDP’s messages for its upcoming report for this region to be launched early 2016. In this paper, we propose a new multidimensional approach to study inequality in welfare in terms of the components of the Human Development Index of the UNDP –health, education and income– in the period 1980-2014. For this purpose we use the multidimensional inequality indices proposed by Maasoumi (1986) that can be decomposable into the between- and within-group inequality components.

Joint hypothesis tests for multidimensional inequality indices

An inequality index over p dimensions of well-being is decomposable by attributes if it can expressed as a function of p unidimensional inequality indices and a measure of association between the various dimensions of well-being. We exploit this decomposition framework to derive joint hypothesis tests regarding the sources of multidimensional inequality, and present Monte Carlo evidence on their finite sample behavior.

Axiomatic measurement of multidimensional well-being inequality: Some distributional questions

This paper focuses on two main issues which are crucial for improving on the analysis of multidimensional inequality: the effect of both the dispersion of well-being attributes across individuals and the interaction among attributes on the measurement of multidimensional well-being. To approach these distributional questions we rely on the Atkinson, Kolm, Sen (hereafter AKS) methodology, which defines a multidimensional inequality index consistent with the Pigou–Dalton principle. This index can be decomposed into univariate indexes belonging to the class of AKS indexes, and a residual term accounting for the interaction across dimensions. The empirical application investigates the evolution of inequality in well-being across some EU countries between 1994 and 2001.Since the multidimensional index depends on the values assigned to the parameters, we test the sensitivity of the trend in well-being to the degree of inequality aversion on each dimension. Our empirical results summarize the evolution of inequality for the indicators of well-being considered both separately and jointly, over time and across countries.

Global Inequality: A Multidimensional Perspective

This paper investigates the evolution of global well-being inequality between 1980 and 2010 based on three dimensions: income, health and education. The inequality of each of these dimensions shows a di¤erent pattern over time. To make an overall assessment of the evolution of well-being inequality, I make use of a recently developed multidimensional inequality index which re‡ects the implicit value judgments of the revised Human Development Index. Multidimensional well-being inequality has decreased over the considered period. However, this result is shown to depend crucially on the weighting scheme selected, the aggregation procedure and the transformation of the income dimension.

Statistical inference for multidimensional inequality indices

We use the delta method to derive the large sample distribution of multidimensional inequality indices. We also present a simple method for computing standard errors and obtain explicit formulas in the context of two families of indices.

The Hybrid Multidimensional Index of Inequality

In this paper, we propose a new multidimensional inequality index that satisfies a fundamental set of desired properties. We discus s the case where the social evaluation function of welfare depends simultaneously on unidimensional and multidimensional forms of inequality. We show how this mixed social norm interferes with the most popular axioms conceived specifically for multidimensional indices of inequality. Illust rations of the proposed developments are made using the Cameroonian household surveys, conducted in 2001. It is shown that multidimensional inequality is more pronounced in the Cameroonian semi-urban and rural areas whereas the monetary inequality is more pronounced in urban area.

A Consistent Multidimensional Generalization of the Pigou-Dalton Transfer Principle: An Analysis

This paper explores the implications of using multidimensional majorization criteria to derive inequality measures, without taking into consideration the idea behind the Pigou-Dalton principle, in the sense that if a richer person transfers something of at least one attribute to a poorer person the inequality falls. A new and basic criterion proposed by Fleurbaey and Trannoy (2003) which generalizes this idea to the multidimensional framework is explored, and its logical relationships with the dominance criteria that exist in the literature are analyzed. The paper also surveys the existent multidimensional inequality indices in order to see whether they meet this new criterion.