Thursday, April 16, 2015

Why don't people respond much to marriage penalties and bonuses (insofar as they don't)?

Just a couple of more thoughts about marriage penalties and bonuses in the U.S. federal income tax law, prompted by discussing the issues that were raised by this week's colloquium regarding Larry Zelenak's paper.

First, insofar as marriage rates are observed not to respond much to marriage penalties and bonuses, even in an age when unmarried cohabitation has become far more socially permissible than it used to be, what would be the reason? 

Maybe this sounds a bit too obvious.  It's a big personal life decision, etcetera.  Plus, people often don't know what the marital stakes are, although websites such as try to fill the gap by offering primers.  But a big piece of it may be the following.  Even the members of tightly-knit couples cannot be entirely unmindful of the statements that they may implicitly make towards each other through their words and actions.  Thus, for either prospective spouse to say “No, let’s not get married as it would cost us $4,000 a year,” may unavoidably risk signaling something about his or her personal level of commitment.  That makes it different from, say, mutually agreeing to live where rents are $4,000 a year lower.

Second, as a kind of pedagogical note, as this came up in discussions during the day, one shouldn't necessarily think that a low behavioral response at the marital margin to marriage penalties means that they are an efficient way of raising revenue.  There is still the question of taxable income elasticity given one's marital status.  Thus, even if secondary earners aren't deterred from marriage by marriage penalties, they may be deterred from working by the marginal effect that this has on them.

One also shouldn't be too swift to conclude that net marriage penalties or bonuses at a particular income level must be affecting vertical distribution.  Suppose "the rich" will pay the same overall taxes either way, and that marriage penalties and bonuses only affect the distribution of the burden among them (what I'd call a "horizontal" distribution question).  Then it's just an issue of how to tax the rich, and whom to define as how rich, rather than affecting things vertically overall.  But admittedly, even if this is true in the long run, the short run can be different, as in the recent case where marriage penalties were pretty much ignored in the course of restoring the 39.6% top bracket.

No comments: