THE NON‐CONSTANCY OF EQUIVALENCE SCALES

Comparisons of households of differing composition are usually achieved through the use of equivalence scales. It is well known that the choice of scales can have considerable impact on the conclusions drawn from studies of welfare and poverty. There is a considerable literature on the theoretical issues relating to equivalence scales, but applied work on income distribution and related areas almost invariably takes scales to be constant irrespective of income. This paper focuses on the relation of scale to income, by applying theoretical analysis to some simple household types. The conclusion is that scales are not constant and that current practice should be changed.