From Charles Karelis. To begin with, I don't think the kinds of income transfers from rich to poor that would significantly raise aggregate utility from consump

From Charles Karelis. To begin with, I don't think the kinds of income transfers from rich to poor that would significantly raise aggregate utility from consumption are in fact politically unfeasible, so I am, as you appreciate, reluctant to get into this what-if area. That said, taxing the poor and transferring the money to the middle class let alone to the rich would carry big negatives according to my approach. We ought to care not only about maximizing aggregate satisfaction from consumption, but about autonomy, including maintaining people's incentives to acquire what they want to consume through their own efforts. Making poor people even poorer will (on my hypothesis) lower (not raise) the marginal utility of their consumption and thereby undermine (not strengthen) their motivation for self-help. If someone with six bee-stings has enough salve to relieve five and the state takes that salve away, this action will probably reduce rather than increase the effort the person will make to get the salve needed to relieve a single sting, because the salve needed to relieve a single sting will make less difference now that the other salve has been confiscated. A fuller answer to your objection than I can give here--including a discussion of the roots of the moral intuition behind your question--is offered in the last chapter of Persistence of Poverty.