June 24, 2006 | Commentary on Taxes
If the American people knew the price of government, they would
purchase a lot less of it. But until the Treasury's fledgling
Dynamic Analysis Division (DAD) came along, government officials
were very successful in keeping that information secret, both from
themselves ("we'd rather not know") and from the voters ("they'd
revolt if they knew").
The secret is beginning to leak out - but only barely. Voters are still being misled and government is still taxing and spending on the false assumptions that $1 spent on a bridge-to-nowhere costs $1 in tax revenue, and that $1 in government tax revenue costs the private economy only $1.
In fact, the cost to the private sector of providing the government an additional $1 in tax revenue is about $2.50, and in some circumstances much more. Even academics on the left now acknowledge that taxes adversely affect economic performance and, therefore, when taxes go up, it is not just the private sector's after-tax income that goes down; its pre-tax income suffers as well.
Thus, when the question is, "How much does it cost the private sector to provide government with another $1 in tax revenue?", the answer is $1 plus the amount by which people's incomes in the future are smaller than they otherwise would be, but for the negative effects that the tax increase has on economic growth.
The amount of future pre-tax income that the private sector loses when government raises taxes varies, depending partly on whether the extra tax is on labor income or capital. Both respond to tax changes: declining when taxes go up and rebounding when taxes are reduced. But capital is particularly sensitive to taxes.
The table below, based on our model, illustrates a range of results when the government attempts to raise an additional $1.
The mainstream story is told by line (c), where the government undertakes to raise $1 in additional tax revenue by an across-the-board rate increase calculated on a static basis to produce that result. It indeed does collect an additional $1 - in a mindless bookkeeping sense - but the economy responds negatively to higher taxes and, as a result, pre-tax incomes (and tax collections) are less than they otherwise would be. On net, the government ends up having collected only $0.68, as shown in column (1). The government "lost" this $0.32 because taxpayers lost $1.07 of pre-tax income that otherwise would have been produced and which would have been taxed at an assumed average marginal rate of 30 percent ($1.07 x .30 = $0.32).
Up to this point in the story, the total cost to the private sector is $1.75 ($0.68 + 1.07), but the government has on net collected only $0.68 in tax. The government can continue to raise tax rates and ultimately net an additional $0.32, as shown in column (5), but that will cost the private sector another $0.50 in lost income, thereby bringing the total cost up to $2.57 per $1 of tax revenue.
The story in line (c) about an across-the-board tax increase on both labor and capital is consistent with recent work by Martin Feldstein at Harvard as well as a recent paper by Gregory Mankiw, the former Bush administration advisor who is also at Harvard. Mankiw also recently performed a groundbreaking analysis of the economic response to a change in taxes solely on capital. His results are consistent with the additional story illustrated in line (d), where the total cost is $4.33 for $1 of additional tax revenue from capital. Mankiw also estimated the economic response from a tax change solely on labor income. Line (a) illustrates the result in the case of a $1 tax increase. The total cost is $1.68.
Line (b) is consistent with the Congressional Budget Office's estimate of the short-term economic response to an across-the-board tax change and line (c) is consistent with CBO's estimate of the long-term response.
By anybody's reasonable estimate, the bottom-line results are clear. The cost of an additional $1 in taxes and spending is much more than $1 - most probably $2 to $3 - and the real burden of taxes on the American people (counting lost income) is much greater than the government admits.
If taxes were both reduced and reformed (so that the drag on economic performance per $1 of tax would be less), the economy would be larger, government would be smaller, and everyone would be better off.
Ernest S. Christian and Gary A. Robbins are, respectively, the executive director and chief economist of the Center For Strategic Tax Reform, a Washington-based think tank, and are also visiting fellows at the Heritage Foundation.
First appeared in the National Review Online