Evaluating Expenditure Inequality Using Alternative Social Welfare Functions: A Case Study of Rural India

This paper examines the sensitivity of the normative estimates of inquality to: (i) changes in the planner's 'inquality averesion', (ii) a-priori changes in specification of the individual welfare functions, (iii) incorporation of price variation, (iv) incorporation of relative deprivation, and (v) the type of inquality measure employed, viz 'income inequality' as proposed by Atkinson or 'utility inequality' as proposed by Muellbauer. The empirical results, using time series of expenditure distribution for rural India [October, 1953 to July, 1974], show considerable sensitivty of the inquality magnitudes to the above factors. The paper presents theoretical evidence on the likely impact of price increases on inequality and confirms them with the empirical results. It, also, derives and applies formulae to identify items whose price increases have a relatively large impact on inequality. The paper exploits the link between inequality measurement and welfare/consumer demand theory that is opened up by the normative approach to inequality. However, the extreme sensitivity of the inequality magnitudes and the wide diesagreements between different evaluators, with different sets of prior beleifs and values, on the extent of the inequality in rural India seems to limit the operational usefulness of such measures.