Rational-expectationist - The Great Deficit Debate

National Review, Jan 27, 1989 by Robert J. Barro

IN RECENT YEARS the U.S. budget deficit has been the dominant popular economic crisis. The supposed harmful effects of these deficits include high real interest rates, low saving, low rates of economic growth, a large current-account deficit, and either a high or a low dollar (depending in some mysterious way on the time period). This crisis scenario has been hard to maintain, given the robust performance of the U.S. economy since late 1982. This performance features high rates of economic growth, declining unemployment, high values of real investment spending, much lower inflation, a sharp decrease in nominal interest rates, a decline in real interest rates since 1984, and a dramatic boom in the stock market until October 1987.

Standard macroeconomic theories predict that budget deficits will crowd out private investment. But, using a broad measure of gross investment that includes purchases of consumer durables, the ratio of real investment expenditure to real GNP averaged 28 per cent from 1984 to 1987-a post-World War 11 high. While less dramatic, even narrower concepts of real gross investment showed a strong performance in recent years.

The standard theory says that budget deficits depress national saving. But the best measure that we have of this saving-the change in the real net worth of householdshas been fairly high, even with allowance for the stockmarket crash, From 1984 to 1987, the ratio of the change in households' real net worth to real GNP (a measure of the national saving rate) averaged more than 11 per cent.

Conventional measures of national saving look low in recent years because they fail to consider the valuation of assets, especially business capital as valued by the stock market and residential real estate as valued by the housing market. Instead of using market-based evaluations of assets, the national-accounts concepts of national saving and net investment rely on essentially arbitrary estimates of depreciation. In recent years, these measures of depreciation have indicated large reductions in the value of capital, just when the markets have signaled the opposite. How can it be that we are providing little for the future by "saving virtually nothing" (as asserted recently by a member of the National Economic Commission), and yet our real national wealth has grown rapidly over time? I submit that the picture of low saving is based on inappropriate concepts of saving, and is simply an inaccurate description of the U.S. economy.

The recent combination of strong investment spending and moderate national saving (appropriately defined) meant that the United States had to rely on foreigners to finance a portion of domestic investment. The reflection of this financing is the current-account deficit. This other deficit is also not cause for concern, especially since the ability to borrow abroad helped to sustain the U.S. investment boom of recent years. In fact, the main legitimate fear about current-account deficits is that government policymakers will take "corrective action."

There is an economic theory that accords better than standard views with the recent U.S. data, as well as with the evidence from other time periods and countries. Economists associate this theory with David Ricardo, the great British economist of the early nineteenth century.

The central point of the Ricardian theory-also stressed frequently by Milton Friedman-is that the cost of government depends on the resources that it uses. Because there is no free lunch, government expenditures have to be paid for by taxes today or taxes tomorrow, but not by taxes never. By running budget deficits or surpluses, the governments can rearrange the time pattern of taxes, butgiven that it is impossible to surprise people systematically by defaulting-it cannot alter the total of taxes that it has to collect. (The term "total" applies here in a presentvalue sense-that is, after adjustment for interest paid or received depending on when expenditures or taxes arise.) Since households care mostly about this total of taxes, and only secondarily about the timing of the tax payments, budget deficits have little effect on households' decisions. In particular, as long as people maintain their beliefs about the prospective levels of government spending, budget deficits do not have major effects on macroeconomic variables, such as consumption and investment, interest rates, and national saving. This result about budget deficits not mattering much fits the facts better than the standard theory that I referred to before.

While U.S. budget deficits do not have the major consequences that are usually suggested, it does not follow that these deficits and the associated quantity of public debt are irrelevant. Budget deficits affect the timing of taxes-say, income taxes-and it is desirable for these levies to behave smoothly, rather than varying erratically from one period to the next. This tax-smoothing argument justifies the longestablished practice of running large budget deficits during recessions and wars. From this perspective it is surprising, and possibly a minor mistake, that U.S. budget deficits were so large during the peacetime boom since 1984. But to the extent that the large deficits were an error, it is also true that much of this error has already been eliminated. When expressed in relation to real GNP, the real federal deficit has declined from 4.4 per cent in 1985 to 2.2 per cent in 1987 and to less than 2 per cent in 1988. (I measured the deficit by the change in privately held federal debt, and I adjusted interest payments for inflation.) The amazing thing is that the amount of ink devoted to budget deficits has not declined in a parallel fashion.