Lies, Damn Lies, And The WSJ
A day ago the WSJ published another in its long line of "boy taxes are bad" op-eds. This one uses a chart purporting to display Hauser's "Law", which suggests that Government tax revenue (as a percentage of GDP) is invariant wrt to the top marginal rate.
Sure enough, you can reproduce this at home using data from Bookings Institute and Bureau of Economic Analysis.
But then the author makes two claims:
The data show that the tax yield has been independent of marginal tax
rates over this period, but tax revenue is directly proportional to GDP.
So if we want to increase tax revenue, we need to increase GDP.
What happens if we instead raise tax rates? Economists of all persuasions
accept that a tax rate hike will reduce GDP, in which case Hauser's Law
says it will also lower tax revenue.
Watch the hands: "The data show" and "Economists of all persuasions accept". But why accept that which you can verify? Particularly when you already have the GDP figures in front of you in order to calculate revenue as a percentage of GDP!
Let's look at claim #1 first -- tax yield is invariant of the top marginal rate. Pick a new scale range and we learn: the tax drop in 1982 lowered revenues by 2% although the much larger drop in 1987/88 did not, furthermore after the tax rise in 1993 revenues rose by 2% until the crash through 2003 gutted revenue again. So I'd say the Jury is still out on the invariance.
Claim #2 is that tax hikes reduce GDP (and the unstated converse, that drops increase GDP). Uh oh, bad news for "economists of all persuasions" -- the annual change in GDP is ALSO INVARIANT.
But here I am being a douchebag and choosing axis scales that help support my thesis. So let's be "fair" and pick a more reasonable scale. We learn that the tax drop in 1982 did increase GDP growth -- in 1984 anyhow -- by 1985 it had reverted to the mean. But the tax drops in 1987/88 reduced GDP growth -- in 1990/91 anyhow -- by 1992 it again reverted to the mean. And that terrible tax hike in 1993 -- absolutely no change to GDP growth.
So we learn that the WSJ Op-Ed would like to have it both ways: a 2% variance in revenue isn't interesting while a 2% variance in GDP is. I guess that depends who you're trying to screw.
(All numbers are in Y2k constant dollars (ie: adjusted for inflation). Sorry about the horrible dithering; LJ butchers gifs.)
Sure enough, you can reproduce this at home using data from Bookings Institute and Bureau of Economic Analysis.
But then the author makes two claims:
The data show that the tax yield has been independent of marginal tax
rates over this period, but tax revenue is directly proportional to GDP.
So if we want to increase tax revenue, we need to increase GDP.
What happens if we instead raise tax rates? Economists of all persuasions
accept that a tax rate hike will reduce GDP, in which case Hauser's Law
says it will also lower tax revenue.
Watch the hands: "The data show" and "Economists of all persuasions accept". But why accept that which you can verify? Particularly when you already have the GDP figures in front of you in order to calculate revenue as a percentage of GDP!
Let's look at claim #1 first -- tax yield is invariant of the top marginal rate. Pick a new scale range and we learn: the tax drop in 1982 lowered revenues by 2% although the much larger drop in 1987/88 did not, furthermore after the tax rise in 1993 revenues rose by 2% until the crash through 2003 gutted revenue again. So I'd say the Jury is still out on the invariance.
Claim #2 is that tax hikes reduce GDP (and the unstated converse, that drops increase GDP). Uh oh, bad news for "economists of all persuasions" -- the annual change in GDP is ALSO INVARIANT.
But here I am being a douchebag and choosing axis scales that help support my thesis. So let's be "fair" and pick a more reasonable scale. We learn that the tax drop in 1982 did increase GDP growth -- in 1984 anyhow -- by 1985 it had reverted to the mean. But the tax drops in 1987/88 reduced GDP growth -- in 1990/91 anyhow -- by 1992 it again reverted to the mean. And that terrible tax hike in 1993 -- absolutely no change to GDP growth.
So we learn that the WSJ Op-Ed would like to have it both ways: a 2% variance in revenue isn't interesting while a 2% variance in GDP is. I guess that depends who you're trying to screw.
(All numbers are in Y2k constant dollars (ie: adjusted for inflation). Sorry about the horrible dithering; LJ butchers gifs.)