> I think a max 3.5% decrease in income
It's not a 3.5% decrease in income. It's a 3.5% decrease in income over $5,000,000. The 3.5% rate does not apply to the first five million dollars, only the money after the first five million.
It's not a 3.5% decrease in income. It's a 3.5% decrease in income over $5,000,000. The 3.5% rate does not apply to the first five million dollars, only the money after the first five million.
I think this point is under-appreciated. As the top federal tax rate has dropped over the decades, each additional percentage point drop has given less proportional value to top-rate-payers, while costing proportionally more of remaining federal revenue. In that sense, the lower the tax rate already is, the more expensive a tax cut is to the government. The higher the tax rate already is, the more expensive a tax increase is to taxpayers.
This is why tax cuts under Bush and Trump blew up the federal budget - tax rates were already pretty low, so quite deep cuts to revenue were required to deliver any notable benefit to high-rate taxpayers. It’s hard to make rich people much richer by cutting taxes when they are already keeping most of their income. In the case of the Trump tax cut, rates were already so low it required raising taxes on some to cut for others.
On the other hand, you can go back and find times when after-tax income of top rate payers doubled because the top rate dropped from eg. 91% to 77%, which delivered a lot of relative value to those taxpayers while fitting more easily into the federal budget (although as always, it’s probably more complicated as tax bills make changes to deductions when they change rates).
I just want to point out that after the Trump tax cuts went into effect, revenue increased[1].
[1] https://www.thebalance.com/current-u-s-federal-government-ta...
Note that I used nominal GDP growth because the linked article used nominal tax revenue. To adjust for inflation we subtract the inflation rate from both sides and we’d get the same 4.9% fallback in tax growth relative to GDP growth.
Federal tax receipts as a % of GDP has been largely flat since WW2:
Overall your chart shows the ratio has a range of 19.75% to 14.42% in the past 20 years, tracking changes to the tax code and business cycle, exactly in the way you’d expect. That’s a 36% difference between the min and max ie. not flat.
I think lawmakers supporting tax cuts try to claim that tax revenues are not sensitive to tax rates because it would be convenient if true when trying to pass a tax cut. The facts don’t bear it out, however.
I think the roots of this argument are in the theory of the Laffer Curve: https://en.m.wikipedia.org/wiki/Laffer_curve The Laffer Curve was used to argue for tax cuts when the top marginal rate was 70%. People are still trying to make this argument, but I think they missed the part where it’s supposed to be a curve, so the result of cutting taxes from 70% is not the same as cutting them from 35%.
> I think lawmakers supporting tax cuts try to claim that tax revenues are not sensitive to tax rates because it would be convenient if true when trying to pass a tax cut.
The point of the chart going back to WW2 is to show that it's a little more complicated and we can't draw conclusions either way.
And circling back to the original point of contention, spending is not function of GDP, or in theory it shouldn't be. Spending is a function of population. So if federal tax revenue increases even after a tax cut, but deficit increases, then the culprit isn't necessarily the tax cut, the culprit is the spending increase.
> The point of the chart going back to WW2 is to show that it's a little more complicated and we can't draw conclusions either way.
I don’t think this is correct. Yes, the economy is complicated. No, that doesn’t mean when cutting taxes we can ignore the resulting increase in the budget deficit. A claim that cutting taxes won’t increase the deficit constrains you to make other claims according to the GDP = C + G + I + X equation, and we should demand that lawmakers be specific about those claims.
As I pointed out, the chart is not really flat. The line actually goes up and down.
Great point, but it increased again in 2019. In fact, inflation adjusted tax revenue has been increasing (with minor transient dips) throughout history[1]. You'll definitely find one-off variations, but the trend-line for inflation-adjusted Federal tax revenue is up-and-to-the-right.
> No, that doesn’t mean when cutting taxes we can ignore the resulting increase in the budget deficit...As I pointed out, the chart is not really flat. The line actually goes up and down.
It's complicated because the ups and downs don't perfectly correlate with tax policy. You can find moments where a decrease in tax revenue as a % of GDP coincide with a tax cut, but you can also find moments where a decrease in tax revenue as a % of GDP coincides with a tax hike and vice versa.
It's important because if you zoom out and look at Federal tax brackets over time since WW2, the US's inflation adjusted tax brackets went from 22%-92% in the 1950's[2] to 10%-37% in 2020[3], but Federal tax receipts as a % of GDP is flat / moderately increased over that period.
[1] https://www.taxpolicycenter.org/statistics/federal-receipt-a...
[2] https://taxfoundation.org/us-federal-individual-income-tax-r...
1) increased tax compliance
2) more tax deductions
Overall, you’d need to use a more comprehensive “area under the curve” analysis of every tax to predict the impact of tax policy on total tax revenue, and it would be surprising if it could be predicted based only on the top marginal rate.
A major source of short-term noise is the business cycle, where capital gains and corporate profits fluctuate much faster than GDP, and some automatic stabilizers decrease revenue.
> A major source of short-term noise is the business cycle, where capital gains and corporate profits fluctuate much faster than GDP, and some automatic stabilizers decrease revenue.
Yes, which is why it's tough to suss anything out from the minor variations between consecutive years, and it makes a little more sense to look at a longer time horizon.