Abstract
In the analysis of food expenditures, the use of a simple count of household members as an estimate of household size implicitly assumes each household member has the same marginal impact. In this analysis of Mexican food expenditures, endogenously determined adult equivalence scales are estimated in such a way that these marginal impacts are allowed to vary by household member age and gender. The results of a series of hypothesis tests indicate a rejection of the implied null hypothesis of equal marginal expenditure impacts associated with the use of the traditional count based household size variable. This study also rejects the null hypothesis of the equality of adult equivalent scales across the commodities included in this analysis.