Does more for the poor mean less for the poor? The politics of tagging

Proposals aimed at improving the welfare of the poor often include indicator targeting, in which non-income characteristics (such as race, gender, or land ownership) that are correlated with income are used to target limited funds to groups likely to include a cincentration of the poor. Previous work shows that efficient use of a fixed budget for poverty reduction requires such targeting, either because agents'income cannot be observed or to reduce distortionary incentives arising from redistributive interventions. Inspite of this, the authors question the political viability of targeting. After constructing a model that is basically an extension of Akerlof's 1978 model of"tagging", they derive three main results: 1) Akerlof's result continues to hold: that, ignoring political considerations, not only will targeting be desirable but recipients of the targeted transfer will receive a greater total transfer than they would if targeting were not possible. 2) A classical social-choice analysis-in which agents vote simultaneously about the level of taxation and the degree of targeting-shows that positive levels of targeted transfers will not exist in equilibrium (an unsurprising finding, given Plott's 1968 theorem). It also shows that a voting equilibrium often will exist with no targeting but with non-zero taxation and redistribution. 3) In a game in which the policymaker chooses the degree of targeting while voters choose the level of taxation, the redistributive efficiency gains from tagging may well fail to outweigh the resulting reduction in funds available for redistribution. These results may be extended readily to account for altruistic agents. The authors stress that even when these results hold, the alternative to targeted transfers - a universally received lump-sum grant financed through a proportional tax - will nonetheless be supported politically and will be quite progressive relative to the pretransfer income distribution.