Real GDP per Capita and Tax Cuts, Top Marginal Income Tax Rate Edition

by cactus

Real GDP per Capita and Tax Cuts, Top Marginal Income Tax Rate Edition

One often hears that cutting marginal income tax rates, particularly on high individuals, leads to faster economic growth. Let’s dispense with argumentification, opinionizing and pontificatulationizing and graph us some data. Data for this post – top marginal rates from the IRS and real GDP per capita from the Bureau of Economic Analysis.

The first graph shows the annual change in real GDP per capita from one year to the next. I took a few liberties with the graph, namely:
1. I color coded each bar – black means the devil raised taxes, white is for the sweetness and light of a tax cut, and gray means no change to top marginal rates.
2. I included a couple of text boxes. The first shows the average growth in real GDP per capita when you have tax cuts, tax hikes, and no change, and it does so for two periods – 1930 to 2009 and 1952 – 2009.
3. The second text box shows the number of instances of tax cuts, tax hikes, and no change to the tax burden over the two periods.

The graph goes back to 1930 because data on real GDP per capita only goes back to 1929. Here’s what it looks like:

(Graph 1)

Faster economic economic growth occurs in years when you have tax hikes than tax cuts. Wait, that can’t be right. This graph is not showing the Truth! So how do we salvage it? Well, maybe the problem is that it takes awhile for the American public to react to a tax cut. After all, it has to be a huge surprise when a person who has talked about the virtues of tax cuts for thirty years actually (get this!) cuts taxes when he becomes President. I still remember the shock we were all in back when GW cut taxes in 2003 – who could possibly have expected tax cuts from a guy who had cut taxes in 01 and 02 and had been promoting tax cuts as the solution to everything from gout to bad haircuts? So maybe we have to assume that it can take a while for tax cuts to have their glorious effect, allowing us to soar into growthy nirvana.

So the next graph is color coded differently – black means the devil raised taxes, either this year, last year, the year before that, or two, three, or four years before that. White means the same look-back, except this time we’re talking tax cuts. Gray is a situation in which either there’s been no change in the tax rate rate (this year or in the previous four), or both tax cut(s) and tax hike(s) occurred during that period. For instance, from 1939 to 1940, the top rate was raised from 79.0% to 81.1%. It feel to 81.0% in 1941, and then rose to 88% in 1942 and again to 94% in 1944. Because of the tax cut (however infinitesimal) in 1941, 1941 through 1945 are colored gray.

I note that once again, I threw in a couple of text boxes. So here it is:

(Graph 2)

Well, this still cannot possibly be right. Why is the data hiding the True Facts? This is clearly gonna take more digging. So let’s be a bit more systematic and cut to the chase.

The next graph shows the average annual growth rate in real GDP per capita within 0, 1, 2, … and 9 years of a tax cut (with no intervening tax hike) from 1930 to 2009. It also shows the average annual growth rate within 0, 1, 2, … and 9 years of a tax hike (with no intervening tax cut). And of course, it shows the same for periods in which there was no change in the top tax bracket and/or in which there were both tax cuts and tax hikes. Thus, at year 0, growth rates are the one shown in Graph 1. At year 0 to 4, growth rates are the shown in Graph 2. And so on and so forth.

Here’s what that looks like:

(Graph 3)

Well, nine years out and there’s no situation in which tax cuts beat tax hikes.

So one last time to the well… this next graph is similar to Graph 3, but it only includes data for the period from 1952 to 2009.

(Graph 4)

Finally. Some evidence. All of this allows us to state that “carefully selecting data allows one to show that that tax cuts are correlated with rapid growth in the first and second year after the cut, but even that degree of cherry picking indicates that the year of the tax cut, as well as 3 or more years out, growth is faster when taxes are hiked than when they are cut.”

Or we can simply go the Fox News route and say: “Behold the Truth as passed down in the Gospel of St. Ronnie. Tax cuts lead to faster growth.”

I have a few possible explanations for these four graphs. One – the best one – I’m not going to mention since it requires some number crunching to confirm and I simply don’t have the time right now. Maybe in the coming weeks. But here are a few more for the two year positive window on tax cuts:

1. Some of those “going Galt” folks are actually serious. Some of them really do make business decisions based on taxes. But people who make business decisions based on reductions of the income tax don’t know how to run a business. Their initial foray into the business world (or initial expansion) frees up some capital and makes things look good for a while, and then they flop.
2. Similar to 1., except that the business decisions are actually accounting conveniences at first which grow to have real effects a couple years down the road.
3. Those anticipating tax cuts put off converting paper earnings into real taxable earnings until after the tax cuts have gone into effect. Thus, for a few years you have “pent up” profits coming out which disappear after a while.

So why, except for a cherry-picked window, are tax cuts not as good for growth as tax hikes are? A few thoughts:
1. At the margin, given the relative size of the gov’t and private sector, and given that both are made up mostly of very inept and/or corrupt people, the gov’t is not less efficient than the private sector.
2. Less money in the public sector constrains the gov’t at those times when it needs to act and when the private sector won’t or can’t? (E.g., how much less freedom of movement does the gov’t have now than it would had the supluses of the late 90s continued through 2008?)
3. The Megan McArdle hypothesis – if the data shows something different from what we know the truth must be, well, something is wrong with the data and anyone who believes what the data seems to show is craaaazy. Call it a poor man’s version of Maier’s Law, with the entire Austrian school way ahead of the second corollary to that law.

Questions, comments, contributions, donations?
by cactus