Groupings and the Gains from Tagging

Abstract

The large literature on “tagging” shows that group specific tax and transfer schedules improve welfare over the case where the government is restricted to a single schedule over the whole population. The central assumption, however, is that the groupings available to the government are given and fixed. But how many and which types of groups should the government choose to tag? This is the question addressed in this paper. Starting with a simple framework and ending with numerical simulations based on data from Finland, we show how groupings should be formed for tagging, and provide a quantitative assessment of how group differences affect the gains from tagging, and of the marginal welfare gains from increasing the number of groups being tagged. We hope that these results are the first steps in a richer analysis of tagging which expands the question of design to the arena of choice over groups being tagged.