Tax Advice From Greg Mankiw

Or not. Yesterday Professor Mankiw posted this on his blog:

Tax Fact of The Day

“The U.S. effective corporate tax rate on new investment was 34.6 percent in 2010, which was the highest rate in the OECD and the fifth-highest rate among 83 countries. The average OECD rate was 18.6 percent, and the average rate for 83 countries was 17.7 percent.”

Okay. He chose not to add any commentary, but he’s obviously implying that he believes the rate is too high based solely upon the fact that other countries have lower rates. I suppose that’s a reasonable starting point for a discussion. But his data is extremely misleading. Because the correct way to measure the tax burden on U.S. corporations is to look at what they actually pay, after accounting for all additional provisions in the tax law (i.e., loopholes). The effective tax rate.

So let’s take a look at that. From the New York Times:

So now the tax rate only seems to be unreasonable for a few select industries. What about corporate taxes paid as a percentage of GDP (i.e., the corporate tax burden relative to the size of the economy)? This is probably the best way to compare our effective tax policy with the policies of other OECD countries. Luckily, Ezra Klein had a great post answering this exact question late last year:

Consider this graph, which shows corporate tax rates against the OECD average. As you can see, we’re on the high side:

international_works_2_2.GIF

That’s the evidence for the conservative view. Our rate is about one standard deviation higher than the OECD average. But here’s the thing:

This confirms Mankiw’s data that the U.S. has higher tax rates. Ezra continues:

If you actually look at the amount of money our corporate tax raises, we’re about one standard deviation beneath the OECD average:

international_works_3_2.GIF

The reason is simple enough: Our corporate tax code is complicated. A lot of businesses, like S-corporations and partnerships, don’t pay corporate taxes. Others use loopholes and exemptions and deductions to push their actual rate way down. We could have a simpler code with lower rates that would raise more money.

So U.S. corporations actually pay 1 standard deviation below the OECD average, while U.S. tax rates are 1 standard deviation above the average. There seem to be two take aways from this. First, it’s obvious that we probably need either better tax laws, better tax collectors, or both. And second, maybe we should consider raising the rates (or lowering rates and eliminating loopholes, resulting in a higher effective rate).

Regardless, the conclusion is the opposite from what Professor Mankiw’s “Tax Fact of the Day” would have you imply. If anything, U.S. corporations don’t pay nearly as much as other OECD countries, and the advantage seems to be unfairly skewed by industry. As Ezra says: “We could have a simpler code with lower rates that would raise more money.” And then everyone wins.