Left Center Left

I like counterintuitive studies. Everyone knows or thinks they know that Massachusetts taxes are high? The Massachusetts Budget and Policy Center runs the numbers and finds out that

Data from the U.S. Census Bureau show that total state and local taxes in Massachusetts amounted to 9.6 percent of personal income in fiscal year 2002 (the latest year for which such information is available for all fifty states). This level of taxes left Massachusetts 38th out of the 50 states (i.e. only 12 states had a smaller state and local tax burden).”

See, if you count not just the income tax rate but look at all taxes (property, sales, excise) - or include local with state taxes - we’re not as tax-and-spend as conservatives would suggest.

Still, something didn’t sit well with me. It was the word “average,” a word that should make alarm bells go off when reading any policy release. Sometimes we do seek to measure the average, but most of the time - in the political realm especially - when we think of “average” income, what we’re really trying to capture is the median. In a population with highly uneven income distribution, the median will vary a lot from the average. Since Massachusetts is richer than other states, it bears to reason that the high end could well make a bigger difference between average and median income, tipping the scales in our favor by increasing the denominator in MBPC’s study.

So I went to the Census site and ran the numbers on my own, this time comparing the per capita tax burden (in which MA ranks eighth) to the median income, not the average income. (Note: I subtracted out corporate taxes to focus on personal taxation, but the results are similar either way.) The result? Taxachusetts is far from being the highest taxer, but based on median income we come in at Number 15, a far cry from the 38th ranking that MBPC has.

Now, MBPC puts out some quality policy studies, and is not merely press release policymaking, but this kind of stunt - and I do believe it was a conscious playing with data - does knock down their credibility a notch or two in my eye.

CORRECTION: Jeff McLynch of the MBPC responds in the comments with a convincing case. Whereas my argument (that the difference between median and average income mattered in tax burden) relied on an assumption that the tax burden is itself not distributed evenly. It’s not, but as this study shows, MA does all right in comparisons of its tax progressivity. I stand corrrected and apologize for imputing bad faith to the MBPC. And I’ll keep my day job.