Published in The Tennessean, February 14, 2010
Raising tax rates eventually results in smaller tax base
by RICHARD J. GRANT
Ibn Khaldun, the 14th-century historian, observed that, "It should be known that at the beginning of the dynasty, taxation yields a large revenue from small assessments. At the end of the dynasty, taxation yields a small revenue from large assessments."
Throughout history, governments have tried to squeeze more revenue from the people by progressively increasing tax rates. But beyond a certain point, and in the long run, increases in tax rates reduce the growth of the tax base. The amount of tax revenue the government is able to reap will be less than it would have been.
The modern geometric representation of Ibn Khaldun’s observation is called the Laffer Curve (after the Nashville-based economist, Arthur Laffer, who has done a superb job of explaining the benefits of lower marginal tax rates). The curve shows that for any tax base, such as capital gains or income, tax revenue will be zero not only when the tax rate is zero but also when the rate is 100 percent. At a tax rate of 100 percent, the government might capture a windfall, but the next year there will be nothing to capture.
Between zero and 100 percent, revenue will initially rise with rate increases but will eventually peak and then fall. If the government raises tax rates beyond the revenue peak, the resulting revenue will be smaller than what could have been realized at lower tax rates. What's worse, people's incomes and wealth will also be lower than they would have been at the lower tax rate.
Are we now beyond the peak of the Laffer Curve? About 30 years ago, two of my old professors, James Buchanan (who is now a Nobel laureate) and Dwight Lee, explained why governments have an incentive to raise tax rates beyond the revenue peak and have a disincentive to lower them again.
People respond to incentives, but responses take time and planning. When tax rates rise, people shift their activities toward those that are less-taxed. The tax base with the now-higher tax rate will shrink as people try to reduce their tax exposure. They try to maximize their after-tax income or wealth. This means that their pre-tax income and wealth will almost certainly be lower than it would have been otherwise. The sum of tax receipts and after-tax income will be less than the amount of income that would have been produced otherwise.
Why is it likely that governments will go too far? Politicians generally have time horizons that are shorter than those of their citizens. The next election is never more than two to six years away. But the consequences of their actions, whether good or bad, can take much longer to manifest.
All else equal, people do prefer to receive more government services, so politicians have an incentive to promise more. They also have an incentive to borrow, thereby spreading the tax burden into the future. That is why we now have such high government deficits. But there are limits to government's ability to borrow, even if it is only the embarrassment of being the cause of the deficits.
At any point in time, it is possible for government to get a short-term increase in tax revenue by increasing the tax rate. But there are two problems with this. First, long-run total tax receipts will be lower. Second, any attempt to reduce tax rates to more reasonable levels could briefly reduce tax revenue. This latter effect would be especially pronounced if the tax cuts were temporary.
Both the Kennedy and the Reagan tax-rate cuts unambiguously gave us higher revenue. Neither can be blamed for any increase in the budget deficits that followed. The blame lies with increased government spending. The same is true for the Bush tax-rate cuts, though these were temporary, and Bush's successor intends to let the cuts expire on the most dynamic portion of the tax base.
Solution: constitutional limits.
Richard J. Grant is a professor of finance and economics at Lipscomb University and a scholar at the Tennessee Center for Policy Research. His column appears on Sundays. E-mail: firstname.lastname@example.org
Copyright © Richard J Grant 2010