1 Preferences for Taxation and Redistribution: An Experimental Study*

Economists debate the reasons why we find progressive taxation and government expenditures, which cause real income to be distributed somewhat more equally after versus before taxes and transfers, in all industrialized countries. Contending possibilities include a shared desire that income be distributed less unequally than market forces alone would dictate, or successful political coalitions to redistribute away from those with higher incomes. We study the relative importance of fairness preferences versus self-interest in determining tax choices in a laboratory decision-making experiment. When choosing as “disinterested observers, ” most of our subjects prefer that there be less inequality among others, and they’re willing to pay something to bring that about. When sure of the effect on their own earnings, subjects ’ experimental tax choices are primarily governed by self-interest, but fairness preferences continue to play a role. The external validity of the experiment is supported by its replication in a non-student subject pool and by the high correlation between tax decisions and political preferences reported in a survey. Results include a correlation between fairness preference and risk-aversion, and a finding that males ’ but not females ’ preferred levels of redistribution are highly responsive to whether differences in pre-tax incomes are due to differences in task performance. Judging by decisions in the “disinterested observer ” condition, subjects ’ aggregate welfare would be highest if post-tax and transfer inequality were less than actual current levels in the United States. * Major funding for this study was provided by the Alex C. Walker Foundation, with additional suppor