next up previous
Next: Tutorial Prerequisites Up: How to Generate a Previous: Meaningful Inputs

Limitations

Lorenz curve limitations:

  1. When comparing two Lorenz curves, it is not possible to determine which distribution has more inequality if the two curves intersect.
  2. The amount of inequality may be misleading. For example when looking at the distribution of income, the amount of inequality could be understated if richer households are able to use their incomes more efficiently than lower income households.
  3. Life cycle effects are ignored. For example an individual's income varies over his lifetime, and this variation is not considered when analyzing inequality with a Lorenz curve.

Gini coefficient limitations:

  1. The Gini coefficient cannot be used when the values of the variable distributed among the population can take on negative values. For example if used to look at income inequality, it must be the case that no one have a negative net wealth.
  2. The measure will give different results if applied to individuals instead of households. For comparison to be meaningful, the populations must be looked at with consistent definitions.

These limitations help to understand the problems caused by the improper use of equality measures. However, they do not render these measurements ineffective. If equality measures are computed in a well-documented and consistent way they can provide a good tool for quantitative comparisons.


next up previous
Next: Tutorial Prerequisites Up: How to Generate a Previous: Meaningful Inputs
info (at) urbansim.org