We study the tax schemes that maximize social welfare functions built on axioms of responsibility for one’s preferences (the requirement that the social welfare function should treat identically agents with the same wage, independently of their preferences) and poverty reduction. We find zero and negative marginal tax rates on low incomes at the optimum and bunching at the income level of the most hardworking minimum wage households. When preferences are iso-elastic, we derive the optimal tax formula, which we calibrate to the US economy. Our formula approximates the shape of the current US tax function for households with at least one child. This result suggests that a fairness-based approach, and these axioms in particular, can help close the gap between the recommendations of optimal tax theory and actual policies.

我们研究了建立在对个人偏好负责的公理（要求社会福利功能应以相同的工资，而与他们的偏好无关地对待相同的代理人）和减贫上的最大化社会福利功能的税收计划。我们发现低收入的最优边际税率为零和负，而最勤奋的最低工资家庭的收入水平则为零。当偏好具有等弹性时，我们得出最优的税收公式，并根据美国经济进行校准。我们的公式近似为目前有至少一个孩子的家庭的美国税收功能的形式。这一结果表明，基于公平的方法，尤其是这些公理，可以帮助缩小最佳税收理论的建议与实际政策之间的差距。