Thursday, October 3, 2013

The Mulligan Marginal Tax Rate

Casey Mulligan has published a truly amazing chart in his latest Wall Street Journal op-ed asserting the following:
The chart nearby shows an index of marginal tax rates for non-elderly household heads and spouses with median earnings potential. The index, a population-weighted average over various ages, occupations, employment decisions (full-time, part-time, multiple jobs, etc.) and family sizes, reflects the extra taxes paid and government benefits forgone as a consequence of working. The 2009-10 peak for marginal tax rates comes from various provisions of the "stimulus" programs in the American Recovery and Reinvestment Act of 2009 and the extension of unemployment benefits to 99 weeks in some states. At the end of 2012, the marginal tax rate index reached its lowest value since 2008: 43.9%. A little over a year later (January 2014), the index will be close to 50%, driven up by the expiration of the payroll tax cut and multiple provisions of the Affordable Care Act.
I have to admit that I have yet to read his NBER paper from which his graph is supposedly taken, but something in all of this looks mighty odd to me. The graph starts in the good old Bush43 days before the Great Recession. I realize that the “1 percent” paid marginal tax rates close to 36% on their Federal income taxes and perhaps a bit extra depending on what state they lived in. But these same folks had a zero marginal tax rate from payroll taxes. Yet, the payroll tax holiday and its expiration change the Mulligan marginal tax rate calculation dramatically. OK, there are a lot of households that were affected by the payroll tax holiday and its expiration even at the margin, but their marginal income tax rate was never anywhere close to 36%. So one has to wonder how few households face anything remotely close to the marginal rates presented in this graph. Given its pro-Republican spin value, Greg Mankiw dutifully linked to it under The Coming Tax Hike supposedly from “The-Not-So-Affordable Care Act” but provided absolutely no commentary or insights. I guess you’ve guessed by now that I’m not buying this Mulligan but it also seems I should go read his NBER paper. Any insights from other economists on whether this chart makes any sense or not would be greatly appreciated.

1 comment:

JW Mason said...

The graph shows the ratio of income of a non-working household to a similar household earning the median wage. In other words, the "tax" here includes not only what we normally think of as taxes, but also the loss of means-tested benefits as your market income increases.

We can debate the relevance of this construct but it is not crazy or wrong in any obvious way.

(I heard him present this work yesterday and had a chance to talk to him about it a bit afterwards. I can tell you, I was very skeptical, but if your reaction is "that's just stupid," you need to think about it more.)