Income redistribution and social choice: A pragmatic approach

Abstract

The conclusions of modern formal welfare analysis can be summarized in two statements. First, although there exists an infinite number of Pareto optimum positions which exhaust all possibilities of improving the welfare of one individual without diminishing the welfare of others, there is no objective criterion for selecting one of these optima as the most preferred or best position.1 Second, if individuals' preferences are to count and other apparently reasonable conditions are to be met, no voting system will be able to achieve a consistent ordering of these op t imum positions, i.e., to pick the most preferred position under all circumstances. This is Arrow's famous Impossibility Theorem.2 These conclusions seem to rule out the possibility of defining a unique welfare maximum and designing public policies to achieve it. Movements along the utility possibility frontier result in some individuals gaining while others lose thus altering the distribution of income and welfare. Ranking alternative positions on the frontier implies comparisions of individuals' utilities or welfare ccontributions which economists deny the capacity to make on behalf of society. Furthermore Arrow's results suggest the individuals would not be able to express their views on these comparisions accurately and consistently through voting.