Like Byran Caplan and Karl Smith, I think conversational evidence is important to pay attention to. And like Bryan, and unlike Karl, I do hear lots of conversation about the effect of marginal tax rates (MTR), especially on spouse’s labor supply. But the conversation is subtle and needs translating.

When I teach a segment on marginal tax rates and point out that for joint filers, the spouse’s (typically the woman’s) tax rate starts where the husband’s leaves off, we then do some sample calculations. (BTW, the United States is unusual in this respect.) I ask the students (the typical one is a 29 to 35-year-old married Lieutenant or Lieutenant Commander in the Navy) what their marginal tax rate is and they often confuse that with their average tax rate. The latter is pretty low. Give a point to Karl: if marginal tax rates were that well understood, people would be talking about them. They aren’t.

But wait. I can draw out of them that, if their spouse is not working, their marginal tax rate is 15% or 25%. Then I point out that their spouse’s MTR picks up from there, and that they need to include the spouse’s marginal payroll tax, which is 15.3% if self- employed and 7.65% if an employee. (I know that who bears the burden depends not on who is statutorily required to pay it, but, rather, on the elasticities of demand and supply for labor. But they, if their spouse is considering being an employee, are right to take just the 7.65% of the gross pay.) Typically, because they’re military, they have claimed residence in a state that has no income tax and so we don’t add an income tax. We then get to a MTR of somewhere between 22.65% and 40.3%, typically somewhere in the middle of that range.

Then I get a lot of knowing looks. Their intuition told them that it didn’t make sense for their spouse to work outside the home (the typical case for a young family with children) because the after-tax pay just wasn’t high enough to make it worthwhile, especially when, as they pointed out, from that after-tax pay you subtract day-care expenses and expenditures on clothing, more meals out, and more restaurant meals. So their “it just wasn’t worth it,” which they had come to on their own, was evidence that MTRs mattered even if they didn’t know their MTR.