Ranking intersecting distribution functions

Second?degree dominance has become a widely accepted criterion for ordering distribution functions according to social welfare. However, it provides only partial ordering, and may fail rank distributions that intersect. To intersecting functions, we propose general approach based on rank?dependent theory. This avoids making arbitrary restrictions or parametric assumptions about welfare allows researchers identify the weakest set of needed Our is two complementary sequences nested criteria. The first (second) sequence extends second?degree stochastic by placing more emphasis differences occur in lower (upper) part distribution. characterize separate systems subfamilies functions. us least restrictive preferences give an unambiguous ranking given We also provide axiomatization criteria corresponding show usefulness our using empirical applications; assesses implications changes household income over business cycle, while second performs comparison actual counterfactual outcome from policy experiment.