Measuring the Burden of High Taxes

Policy Reports | Taxes

No. 215
Wednesday, July 01, 1998
by Gerald W. Scully

How Taxes Retard Growth

Figure II - Estimated Growth Rates for Various Tax Rates

Figure I shows a gap between potential output and actual output. Assume that this gap is caused by a tax rate that is too high (i.e., higher than the growth-maximizing rate). Then the higher tax is causing a loss of potential output for society as whole. This loss is called a "deadweight loss of taxation."8 One way to measure the burden of this higher tax rate is to divide the gap between potential and actual output by the average amount of taxes collected over the period.9 Since in Figure I the gap between the potential and actual national output grows, the net burden of taxation is increasing over time.

By plugging real-world data into the economic model we've constructed, it is possible to determine the growth-maximizing tax rate, the deadweight loss of the actual rate of taxation and thus the marginal cost of taxation. The marginal cost of taxation is the amount of GDP that is lost as a result of each extra dollar of taxes above the growth-maximizing rate.

The Growth- Maximizing Tax Rate. We use data on the real rate of growth of GDP for the 46-year period from 1950 through 1995 and on federal, state and local taxes as a share of GDP for that period.10 The resulting calculations suggest that:11

  • The estimated growth-maximizing tax rate for the United States is 21 percent of GDP.
  • The overall rate of economic growth that corresponds to that tax rate is 4.8 percent.12
  • Instead of the growth-maximizing tax rate of 21 percent of GDP, however, taxes were 24.2 percent of GDP in 1950 and continued to rise thereafter.
  • The actual annual economic growth rate over the 1950-95 period was 3.4 percent.

In Figure II, the growth rates that correspond to various tax rates are plotted. The illustration clearly shows that the long-term growth rate declines for levels of taxation above 21 percent of GDP.

"If a tax rate of 21 percent of GDP had been in effect, families would have twice as much income today."

The Cost of Higher Taxation. Actual GDP in 1995 (in 1992 dollars) was $6.76 trillion. Absent other influences, if the optimal tax of 21 percent had been in effect throughout the period, real GDP would have been $13.48 trillion - or almost twice the actual figure. Accumulated real GDP over the 46-year time span would have been $261.5 trillion, or $88 trillion more than the actual number of $173.5 trillion. This means the average efficiency loss to the economy from taxation above the optimal level is 34 percent ($88 trillion divided by $261.5 trillion). The accumulated real taxes paid by Americans from 1950 through 1995 totaled $51.45 trillion.

  • On the average, each dollar of tax created a $1.71 deadweight loss of private wealth ($88 trillion divided by $51.45 trillion).13
  • But the marginal cost of taxation is much higher than the average cost,14 with each extra dollar of tax causing a $3.44 loss of GDP.15

"Every additional dollar of tax causes a $3.44 loss of GDP."

If a tax rate of 21 percent had been in effect over the 1950-95 period, the long-term rate of growth would have been about 1.4 percentage points higher. Workers now would be producing (in 1992 dollars) $107,900 in per capita output instead of $54,100. The standard of living would be higher for everyone, the poor as well as the well-off.

Read Article as PDF