areppim: information, pure and simple
             HTML 5 Valid CSS Valid RSS

areppim: information, pure and simple


You are here:   areppim >  Glossary of Terms >   Gini Coefficient

Glossary

Gini Coefficient
Lorenz curve plot, indicating the relationship between the cumulative percentage of income and the cumulative percentage of population The Gini coefficient, also known as Gini index or Gini ratio, measures the extent to which income or expenditure is shared among the members of a community. It can be illustrated by means of a Lorenz curve (the thick blue curve in the chart), a graphical device showing cumulative percentage relationships between two variables. The horizontal axis indicates the cumulative percentage of the population. The vertical axis indicates the cumulative percentage of income or expenditure associated with the population on the horizontal axis, starting with the poorest and ending with the richest.
 
The Gini coefficient measures the area A between the Lorenz curve and the theoretical line of perfect equality (the dotted 45-degree line), as a percentage of the total area under the dotted line. As the thick line moves up closer to the dotted line, thus reducing the area A, the poor segments of the population get a better share of the wealth. At the limit, in the case of an exact match between the Lorenz curve and the dotted line, there would be a perfectly egalitarian income distribution, corresponding to a Gini coefficient of 0. Conversely, as the thick line moves down towards the bottom, poor people get less and less, and the wealth accumulates in the hands of the richest. At the limit, a coefficient of 100 would mean that all income is owned by a single individual.
 
Comparisons between coefficients should be handled with caution. In fact, nominal differences may in reality be lower than the margin of error introduced by the data. Bias may derive from various sources:
  • The observed variable may be reported in different terms: disposable income, gross income, expenditure, consumption, earnings, etc.
  • The population coverage may vary: urban, rural, a group of main cities, employees only, all, etc.
  • The unit of analysis may be the person, the family, the household, the tax unit, etc.
  • The survey methodology may differ according to the sources and the years.