[An updated version of this article can be found at Marginal Tax Rates in the 2nd edition.]
The marginal tax rate is the rate on the last dollar of income earned. This is very different from the average tax rate, which is the total taxes paid as a percentage of total income earned. The seemingly arcane topic of marginal tax rates became the central theme of a revolution in economic policy that swept the globe in the eighties. By the end of the decade, more than fifty nations had significantly reduced their highest marginal tax rates (most of which are shown in table 1). Neither Karl Marx nor John Maynard Keynes had so much influence on so many countries in so little time.
Several economies that seemed on the verge of bankruptcy in the early eighties were suddenly revived once marginal tax rates were reduced. In 1983 to 1984, Turkey's marginal tax rates were slashed: the minimum rate dropped from 40 to 25 percent, the maximum from 75 to 50 percent. Real economic growth jumped to nearly 7 percent in the following four years and to 9 percent in 1990. Like Turkey, South Korea was deep in debt to international banks in 1980, when real output fell 2 percent. Korea subsequently cut tax rates and expanded deductions three times, and economic growth averaged 9.3 percent a year from 1981 to 1989. In the early eighties the African island of Mauritius faced an unemployment rate of 23 percent and massive emigration. Tax rates were cut from 60 percent to 35 percent, and the economy grew by 5.4 percent a year from 1981 through 1987. Egypt, Jamaica, Colombia, Chile, Bolivia, and Mexico had similar experiences after slashing marginal tax rates.
The same pattern was repeated in most major industrial countries. Economic growth in Britain had averaged only 1.2 percent for a dozen years before tax rates were cut in 1984 and 1986. The British economy subsequently grew by 4 percent a year from 1985 to 1989. Economic growth in Japan from 1983 to 1987 had slowed to 3.9 percent—slower than the 4.3 percent growth in the United States. Japan cut higher tax rates by 15 to 20 percent in 1988, and economic growth and investment subsequently boomed. Even in the roaring eighties, economic growth had slipped to around 1.5 percent in Belgium, Austria, and the Netherlands before each country cut marginal tax rates. In the first year or two of tax reform, economic growth jumped to 4 percent in Austria, 4.1 percent in the Netherlands, and 4.3 percent in Belgium. The economies of Canada and West Germany likewise experienced brief booms when tax rates were reduced in 1988 and 1989 respectively, but Canada slipped into recession in early 1990 after reversing course with surtaxes and a new sales tax. Germany likewise added surtaxes and sales tax in mid-1991, with immediate adverse effects on the stock market and the value of its currency.
Despite widespread adoption of such policies, few seem to understand what marginal tax rates are and why they matter. In the United States, for example, it is commonly believed that the Reagan administration "slashed taxes," particularly for "the rich." Actually, real (that is, inflation-adjusted) federal receipts increased by one-third from 1980 to 1990. Moreover, the most affluent 5 percent of all taxpayers paid 45.9 percent of all federal income taxes in 1988—up from 37.6 percent in 1979. Apparent "tax cuts"—from a top marginal rate of 70 percent to 33 percent—became actual tax increases, particularly for "the rich." The explanation for this paradox lies in the critical distinctions between average and marginal tax rates, and between "static" effects right now and "dynamic" effects over years and decades. Dynamic effects include increased intensity and motivation of work effort, more efficient investment, and more innovation and risk taking.
Measuring the taxes that governments collect as a percentage of GNP, for example, is too static. It ignores the destructive effect that steep marginal rates have on both tax collections and GNP. Several African countries attempt to impose marginal tax rates of 60 to 85 percent on people whose income is equivalent to the U.S. poverty line. Yet receipts from such demoralizing income tax systems are usually less than 1 percent of GNP. Productive activity ceases, moves abroad, or vanishes into inefficient little "underground" enterprises. Taxes on sales, imports, payrolls, and profits also reduce the after-tax rewards to added investment and effort, of course. And just as "tax havens" attract foreign investment and immigrants, countries in which the combined marginal impact of taxes is to punish success invariably face "capital flight" and a "brain drain."
In the United States the concept of marginal tax rates is most familiar as tax brackets. Rapid inflation in the seventies pushed many skilled working couples up into the 50 percent tax bracket (then the highest rate on labor income). That did not mean that all of their income was taxed at a 50 percent rate. Instead, the first ten thousand dollars or so might be taxed at a 12 percent rate, the next ten thousand at a higher rate, and so on. Once the 50 percent bracket was reached, though, the federal government really did expect to collect half of any additional earnings. Average federal income taxes—taxes divided by income—have rarely been much more than 25 percent even for the superrich, even when (in the fifties) marginal tax brackets rose as high as 90 percent. By keeping average taxes the same, while reducing marginal tax rates, it is possible to encourage people to earn and report more income. This is a "revenue neutral" tax reform, like the one in 1986.
The marginal tax on added earnings matters because it is easier to earn less than to earn more. To increase income, people have to study more, accept added risks and responsibilities, relocate, work late or take work home, tackle the dangers of starting a new business or investing in one, and so on. People earn more by producing more and better goods and services. If the tax system punishes added income, it must also punish added output—that is, economic growth.
Some economists used to argue that the incentive effect of lower marginal tax rates is ambiguous. Perhaps, they said, people will simply use the "tax cut" to enjoy more leisure, living just as well by working less. This argument again confuses average with marginal tax rates. With a "revenue neutral" cut in marginal tax rates, taxpayers do not automatically receive the increase in after-tax income that is alleged to make them work less. Since average tax rates remain unchanged, the only way to get this added income is to work harder and produce more.
More and more, theoretical and factual research on the sources of both long-term economic growth and short-term disturbances (recessions) has pointed to the level and variation of marginal tax rates. A comparison of sixty-three countries by Reinhard Koester and Roger Kormendi found that "holding average tax rates constant, a 10 percentage point reduction in marginal tax rates would yield a 15.2 percent increase in per capita income for LDCs [less developed countries]."
In 1990 Harvard economist Robert Barro and Paul Romer, then at the University of Chicago, surveyed the latest studies for the year-end report from the National Bureau of Economic Research. "Recent work on growth," they explained, "extends neoclassical markets so that all economic improvement can be traced to actions taken by people who respond to incentives." This approach leads to "very different predictions about how such policy variables as taxes can influence growth....If government taxes or [regulatory] distortions discourage the activity that generates growth, growth will be slower."
What began in the early seventies as a topic that interested only a few quiet specialists in "optimal taxation," and a few noisy "supply-side" economists proposing a remedy for chronic stagflation, has now filtered into several textbooks—such as those written by Robert Barro and by James Gwartney and Richard Stroup. After decades of compulsive tinkering with budgets and money supplies to "manage demand," much of the world has rediscovered an insight as old as economics itself—namely, that cutting marginal tax rates encourages supply.
Alan Reynolds is a senior fellow with the Cato Institute and was formerly director of economic research at the Hudson Institute in Indianapolis.
Barro, Robert. "Taxes and Transfers." In Barro, Macroeconomics. 1987.
Davies, David G. United States Tax Policy. 1986.
Koester, Reinhard B., and Roger C. Kormendi. "Taxation, Aggregate Activity and Economic Growth: Cross Country Evidence on Some Supply-Side Hypotheses." Economic Inquiry 27 (July 1989): 367-86.
McKenzie, Richard B., and Dwight R. Lee. Quicksilver Capital. Chap. 6. 1991.
Reynolds, Alan. "Some International Comparisons of Supply-Side Tax Policy." Cato Journal 5 (Fall 1985): 543-69.