Economic Limits on "Rational" Democratic Redistribution

I begin with an economic environment familiar from welfare-and political-economic literatures and show how, with quantitatively reasonable distributions of labor productivity and tax-price-elasticities of taxable income, middle class consumers are (personally) worse off with any negative income tax scheme than they would be with no redistribution at all. This finding has important implications for political-economic theories of redistribution, because it implies that the fully informed median voter cannot be expected to support programs of cash redistribution from rich to poor - such as the negative income tax - merely on the basis of his personal benefits from the program. It also implies that the "rational" median voter model of redistribution is, in the empirically relevant range, inconsistent with a positive correlation between income distribution skewness, or enfranchisement of the poor, and the amount of rich-poor redistribution.