By Robert J. Barro
With the military action in Iraq winding down, the Bush Administration is returning to its tax-cutting proposals. The two main ideas are to accelerate the reductions in marginal income-tax rates and to eliminate the income tax on dividends. I support this plan because it would raise economic growth.
My support has nothing to do with seeing the Bush tax plan as a "stimulus package" that might expand the economy in the short run. This Keynesian approach puts too much emphasis on consumer demand and too little on household and business incentives to work, produce, and invest. By putting the economy on a path of higher growth, the Bush tax program would foster economic expansions similar to those following the Kennedy-Johnson tax cut of 1963-64 and the Reagan cuts of the 1980s.
One concern is that the tax reductions would harm the economy by creating larger budget deficits. A common claim is that the deficits would raise real interest rates and thereby reduce investment. Economic theory tells us that the effects of budget deficits on real interest rates are uncertain and small. Empirical evidence for the U.S. and other countries supports the theory: The numbers show that U.S. budget deficits have played a minor role in determining real interest rates from the late 1950s up to the present.
In economic theory, one key question is whether households feel wealthier when the government runs a budget deficit to pay for a tax cut. Theory argues that if nothing else changes (including the government's purchases of goods and services), the extra future taxes needed to service the larger public debt will offset the current tax cut. Therefore, households are not actually wealthier -- so why should a tax cut increase consumer demand? Households would be wise, instead, to save all of the tax cut so they can afford to pay higher taxes in the future. In that case, we get no effects from a budget deficit on real interest rates. This is the famous "Ricardian equivalence" result (taxes and debt issues have equivalent effects on the economy), associated with the British economist David Ricardo.
Economists have since developed richer theories about the effects of budget deficits and public debt. Some theories assume that people feel wealthier when the government shifts taxes from the present generation toward children and grandchildren. In that case, a tax cut would raise consumer demand, and real interest rates would rise. Other models stress that the economic effects depend on the form of the tax changes -- for example, on whether a tax cut includes a reduction in marginal income-tax rates. If lower tax rates motivate greater work effort and production, real interest rates could fall. The upshot is that budget deficits may affect real interest rates -- but typically by small amounts and in directions that depend on the specifics of a tax program.
Empirically, it is important to assess how real interest rates are determined within an international context. I tend to rely on evidence of the sort I have isolated in a continuing research program begun in 1990 (see "World Real Interest Rates," NBER Macroeconomics Annual, 1990). This work now considers 12 countries: the U.S., Japan, Canada, Australia, and several nations in Western Europe. Here are some of the main findings:
1. Real interest rates depend on cumulated levels of public debt in relation to the gross domestic product, not on current fiscal deficits.
2. Real interest rates in each country depend not so much on that country's debt but rather on overall debt. U.S. real interest rates depend on the ratio of public debt to gross domestic product in the 12-country universe. The U.S. debt matters here, but only to the extent that it contributes to the overall debt.
3. A rise in the 12-country debt-to-GDP ratio raises real interest rates in each country, but by only small amounts. Specifically, if the public debt rose by 1% of GDP in each country, real interest rates would rise by an estimated 0.1%. If debt rose only in the U.S. (which now accounts for 45% of the 12-country GDP), the rise in real interest rates would be about 0.05%.
4. These effects are trivial relative to the fluctuations in real interest rates observed in the U.S. For example, the real yield on 10-year inflation-protected U.S. Treasury securities fell from a peak of 4.4% in 2000 to 2% now. This dramatic shift had nothing to do with changes in public debt levels. In fact, the U.S. government budget shifted over this period from surplus to deficit. (The sharp drop in real interest rates also has little to do with monetary policies but has a lot to do with recessions associated with the end of the technology boom and declines in stock markets.)
The bottom line is that, in evaluating the Bush tax plan, higher real interest rates do not make the top-100 list of economic concerns. We can get on with the business of shaping a tax-reform package that encourages economic growth. The Bush plan is a good start. Robert J. Barro is a professor of economics at Harvard University and a senior fellow of the Hoover Institution (firstname.lastname@example.org).