Eating Like Which 'Joneses?' an Iterative Solution to the Choice of a Poverty Line 'Reference Group'

Abstract

A standard method for calculation poverty lines (e.g. Ravallion, 1994) is not fully specified. The choice of the "reference population" for determining food baskets is left to the decision of the individual analyst. However, the poverty line can be quite sensitive to the real income of the reference group because the "quality" of the food basket - measured as the food expenditures per calorie - rises sharply with income. We propose that the reference group be centered on the poverty line. To address the obvious circularity problem in choosing a reference population at the poverty line to define the poverty line, we use an iterative approach. This iterative method provides a methodological anchor that fixes the reference group.