Volume 8 - Article 11 | Pages 305–358  

Gini coefficient as a life table function: Computation from discrete data, decomposition of differences and empirical examples

By Vladimir Shkolnikov, Evgeny M. Andreev, Alexander Begun

Abstract

This paper presents a toolkit for measuring and analyzing inter-individual inequality in length of life by Gini coefficient. Gini coefficient and four other inequality measures are defined on the length-of-life distribution. Properties of these measures and their empirical testing on mortality data suggest a possibility for different judgements about the direction of changes in the degree of inequality by using different measures. A new computational procedure for the estimation of Gini coefficient from life tables is developed and tested on about four hundred real life tables.
The estimates of Gini coefficient are precise enough even for abridged life tables with the final age group of 85+. New formulae have been developed for the decomposition of differences between Gini coefficients by age and cause of death. A new method for decomposition of age-components into effects of mortality and composition of population by group is developed. Temporal changes in the effects of elimination of causes of death on Gini coefficient are analyzed.
Numerous empirical examples show: Lorenz curves for Sweden, Russia and Bangladesh in 1995, proportional changes in Gini coefficient and four other measures of inequality for the USA in 1950-1995 and for Russia in 1959-2000. Further shown are errors of estimates of Gini coefficient when computed from various types of mortality data of France, Japan, Sweden and the USA in 1900-95, decompositions of the USA-UK difference in life expectancies and Gini coefficients by age and cause of death in 1997.
As well, effects of elimination of major causes of death in the UK in 1951-96 on Gini coefficient, age-specific effects of mortality and educational composition of the Russian population on changes in life expectancy and Gini coefficient between 1979 and 1989. Illustrated as well are variations in life expectancy and Gini coefficient across 32 countries in 1996-1999 and associated changes in life expectancy and Gini coefficient in Japan, Russia, Spain, the USA, and the UK in 1950-1999. Variations in Gini coefficient, with time and across countries, are driven by historical compression of mortality, but also by varying health and social patterns.

Author's Affiliation

Other articles by the same author/authors in Demographic Research

Algorithm for decomposition of differences between aggregate demographic measures and its application to life expectancies, healthy life expectancies, parity-progression ratios and total fertility rates
Volume 7 - Article 14

Components and possible determinants of decrease in Russian mortality in 2004-2010
Volume 28 - Article 32

Estimates of mortality and population changes in England and Wales over the two World Wars
Volume 13 - Article 16

Geographical diversity of cause-of-death patterns and trends in Russia
Volume 12 - Article 13

Life expectancy in two Caucasian countries. How much due to overestimated population?
Volume 5 - Article 7

Two-dimensional contour decomposition: Decomposing mortality differences into initial difference and trend components by age and cause of death
Volume 50 - Article 41

Socio-economic determinants of divorce in Lithuania: Evidence from register-based census-linked data
Volume 33 - Article 30

Average age at death in infancy and infant mortality level: Reconsidering the Coale-Demeny formulas at current levels of low mortality
Volume 33 - Article 13

Official population statistics and the Human Mortality Database estimates of populations aged 80+ in Germany and nine other European countries
Volume 13 - Article 14

Introduction to the Special Collection “Human Mortality over Age, Time, Sex, and Place: The 1st HMD Symposium”
Volume 13 - Article 10

A summary of Special Collection 2: Determinants of Diverging Trends in Mortality
Volume 10 - Article 12

Educational differentials in male mortality in Russia and northern Europe: A comparison of an epidemiological cohort from Moscow and St. Petersburg with the male populations of Helsinki and Oslo
Volume 10 - Article 1

Russian mortality beyond vital statistics: Effects of social status and behaviours on deaths from circulatory disease and external causes - a case-control study of men aged 20-55 years in Urdmurtia, 1998-99
Special Collection 2 - Article 4

Introduction to the Special Collection of papers on "Determinants of diverging trends in mortality"
Special Collection 2 - Article 1

Most recent similar articles in Demographic Research

On the relationship between life expectancy, modal age at death, and the threshold age of the life table entropy
Volume 51 - Article 24    | Keywords: Gompertz law, life expectancy, lifespan variation, longevity, mode, mortality

The role of sex and age in seasonal mortality – the case of Poland
Volume 51 - Article 17    | Keywords: mortality, Poland, seasonality, sex differences

Data errors in mortality estimation: Formal demographic analysis of under-registration, under-enumeration, and age misreporting
Volume 51 - Article 9    | Keywords: age misreporting, data errors, formal demography, mortality

Socio-behavioral factors contributing to recent mortality trends in the United States
Volume 51 - Article 7    | Keywords: despair, health, mortality, National Health Interview Survey (NHIS), smoking, trends

Climate change and health transitions: Evidence from Antananarivo, Madagascar
Volume 51 - Article 6    | Keywords: climate change, health transition, historical demography, infectious diseases, mortality