In 1998, Korpi and Palme proposed a political explanation to account for the greater redistributive success of universalist welfare states. Redistribution, they argued, resulted less from a Robin Hood logic — taking from the rich to give to the poor — than from a broad and egalitarian provision of services and transfers. Hence, the paradox of redistribution: a country redistributed more when it took from all to give to all than when it soaked the rich to help the poor. This account is now challenged by studies failing to see a link between welfare institutions and redistribution. We revisit the question and argue that the original argument was theoretically sound, but inadequately operationalized, because it measured universalism not by the nature of programs, but rather by their outcomes, in distributive terms. Our paper uses new indicators and finds that, indeed, there is still a paradox of redistribution.
