Can consumption taxes reduce inequality in developing countries? We combine household expenditure data from 31 countries with theory to shed new light on the redistributive potential and optimal design of consumption taxes. We use the place of purchase of each expenditure to proxy for informal (untaxed) consumption. This enables us to characterize the informality Engel curve: we ﬁnd that the budget share spent in the informal sector steeply declines with income, in all countries. The informal sector thus makes consumption taxes progressive: households in the richest quintile face an effective tax rate that is twice that of the poorest quintile. We extend the standard optimal commodity tax model to allow for informal consumption and calibrate it to our data to study the effect of different tax policies on inequality. Contrary to consensus, we show that consumption taxes are redistributive, lowering inequality by as much as personal income taxes. These effects are primarily driven by the shape of the informality Engel curve. Once informality is taken into account, commonly used redistributive policies, such as reduced rates on necessities, have a limited impact on inequality. In particular, subsidizing food cannot be justiﬁed on equity or efﬁciency grounds in several poor countries.