The taxation of couples

Abstract

We present an analysis of the tax treatment of couples. We first investigate how marriage bonuses and penalties evolved in the US federal income tax since the 1960s. We then develop sufficient statistics for an analysis of Pareto-improving, welfare-improving and political feasible reforms of tax systems that treat singles and couples differently, and interpret the development in the US through the lens of this framework. We also analyze revenue-neutral reforms that depart from the status quo tax treatment of couples in the direction of individual taxation. With such a reform, marginal tax rates are lowered for secondary earners and increased for primary earners. Such reforms are a recurrent theme in the debate about the tax treatment of couples. Again, we derive sufficient statistics formulas that can be used to check whether such reforms would have been Pareto-improving, welfare-improving or politically feasible. We present two main results. First, for the US, with empirically plausible assumptions about the behavioral responses to taxation, moving towards individual taxation for high incomes is Pareto-improving today and would have been Pareto-improving in the last decades. More bluntly, the tax treatment of couples is and was a source of inefficiency in the US tax system. Second, a revenue-neutral reform towards individual taxation creates winners and losers. The winners tend to be couples with a relatively small gap between primary and secondary earnings. The losers tend to be couples with a relatively large gap. Political feasibility requires majority-support, i.e. that the winners outnumber the losers. Over the years, the number of winners from such a reform has grown and only recently passed the majority threshold. Thus, reforms towards individual taxation have not been politically feasible in the past, but have become politically feasible recently. (joint work with Pierre Boyer, Andreas Peichl and Daniel Weishaar)