The "popular support for progressive taxation theorem" (Marhuenda and Ortuno-Ort́in, 1995) provides an important formalization of the intuition that a majority of relatively poor voters over rich ones leads to progressive income taxation. Yet the theorem does not provide an equilibrium outcome. In addition, it assumes an overly restrictive domain of tax schedules and no incentive effects of income taxation. This paper shows that none of these assumptions of the theorem can be relaxed completely. Most notably, it is shown that a majority of poor voters does not imply progressive taxation in a more general policy space and that a regressive tax schedule may obtain a majority over a progressive one when individuals' income is endogenous.