The Concise Encyclopedia of Economics

Federal Debt

by Robert Eisner
About the Author
[Editor's note: this article was written in early 1993. Since then, the debt has grown and a number of other key data discussed in this article have changed.] Everyone talks about the federal debt, but few, literally, know what they are talking about. That is all the more true for the federal deficit, which year after year adds to the total debt outstanding.

Perhaps the first thing to know about the federal debt, some $4 trillion at the end of 1992, is that $1 trillion of it is held by government agencies or government trust funds such as those for Social Security. Excluding that amount leaves the more relevant figure of "gross federal debt held by the public" at $3 trillion.

Even that number is somewhat misleading on two counts. First, because the Federal Reserve banks are technically private corporations, the debt "held by the public" includes the Federal Reserve holdings, which come to almost a quarter of a trillion dollars, although the interest paid on Federal Reserve holdings largely goes right back to the Treasury. Second, "gross" debt does not subtract what the public owes to the federal government or its credit agencies. The net debt—the debt owed by the government to the public exclusive of the Federal Reserve, minus the debt owed by the public to the government—is some 20 percent less. Therefore, the actual net debt was on the order of $2.4 trillion in 1992, or only a little more than half the gross debt.

The debt of U.S. businesses (excluding financial institutions) and households is $7 trillion, far larger than the federal debt. But business debt, it is argued, generally finances income-earning plant and equipment. And individuals borrow largely to purchase homes, which provide implicit income by saving rent that would otherwise be paid. Debt that finances income-earning assets is hardly the same as dead-weight debt that must be serviced out of unrelated income.

The federal government, though, also has real assets: interstate highways, public buildings, and federal land, water, and minerals. All these assets contribute to the national income and hence—indirectly if not directly—to the "earnings" or tax receipts of the government itself. Private businesses compute their net worth by subtracting liabilities from assets. The same should be done for the government. Most unofficial estimates suggest that assets directly owned by the federal government pretty much match the entire federal debt.

Most of the federal debt is owed to Americans. In fact, the share of privately held public debt owned by foreigners fell from 21.2 percent in 1980 to 17.6 percent in 1992. The foreign share of the total gross public debt in 1992 was 12.1 percent. And virtually all of that debt was in dollars, which means that it can be paid off or bought back by the simple device of printing money or, in more sophisticated fashion, open-market operations of the Federal Reserve.

Whether that should be done raises serious issues of economic policy. But if it wishes, the federal government can always create what money it needs to service its debt. In this fundamental sense, then, federal debt is different from private debt or, for that matter, the debt of state and local governments, which do not have the power to create money. Thus the federal government has no reason ever to default on its debt or declare bankruptcy.

The federal debt held by the public differs in another fundamental sense from private debt. For every private creditor there is a debtor who knows he is a debtor. Therefore private debt is, from the standpoint of aggregate wealth or the net worth of the private sector, a wash; the liability of one individual or business is the asset of another. The net debt of the federal government held by the public, however, is an asset of the private sector, of state and local governments, or of the rest of the world. But few people think of themselves as the debtor when the federal government goes into further debt by selling a bond. This means that the bigger the federal debt, the wealthier citizens who own the bonds feel and, hence, the more they are likely to spend. Thus the fundamental importance, for good or for bad, of the federal debt is likely to be its effect on private spending.

If the economy is in a recession because of a lack of effective demand for what can be produced, a bigger federal debt may be useful. The greater wealth it causes in the form of Treasury notes, bills, or bonds gives people less reason to save and, therefore, induces them to consume more. Businesses, which may also feel wealthier with their holdings of Treasury securities, may be expected to produce more to meet the consumer demand and also to invest more in the capacity to meet that demand.

If, however, the economy is already at full employment, with few unused resources, consumers' attempts to spend more as a consequence of their greater financial wealth can only generate higher prices. And there's the rub! Too high a federal debt, or a deficit that increases the debt and, consequently, causes aggregate demand to rise faster than production can be increased to meet that demand, brings on inflation. Then, possibly even worse, actions by the Federal Reserve to combat the inflation will raise interest rates, thus choking off investment in new housing, new factories, and new machinery.

These arguments about the federal debt, its power for good or bad, require two major qualifications. First, the debt must be measured in a correct and relevant fashion. This means, most importantly, that it must be adjusted for inflation. A person who has $101,000 in Treasury securities is not richer than he was a year ago when he held $100,000 if inflation has reduced the value of the dollar by 3 percent. He has, in fact, lost the equivalent of $3,000 in the real value of his securities from what may be called an inflation tax. He is, on balance, $2,000 poorer than he was a year ago. Thus, he is likely to buy fewer goods, not more.

Just as with individuals, so with the federal government. The federal debt must be measured in real terms, and the real deficit must be seen as the change in the real value of the debt. By this measure the apparently huge federal debt actually declined over most of the past half-century. Even now, after a decade of relatively large deficits, the per capita federal debt is still much less than in 1945, when the country and its economy were much smaller. Inflation of only 3 percent reduces a gross debt of $3 trillion by $90 billion. This indicates a real budget deficit $90 billion less than that in official measures.

A good way of judging the size of the federal debt, and hence its likely effect on the economy, is, as for an individual, to take it as a ratio of income. The federal debt reached a peak ratio of 114 percent of GDP after World War II and declined to 26 percent by 1981, before rising again. But even with the subsequent deficits, it was still only 51 percent of GDP in 1992. True "balance" in the budget, it might be suggested, would entail not a zero deficit, but one such that the debt grows at the same percentage rate as GNP, thus keeping the debt-to-GNP ratio constant.

The second qualification is that many economists question whether federal debt is real wealth for the public as a whole. They argue that increases in the federal debt will cause people to expect future increases in taxes in order to service that debt. On the assumption that the present value of the increase in expected future taxes is equal to the increase in the debt, there is no net change in perceived wealth and, hence, no effect of the debt on overall demand for goods and services.

This argument has been severely criticized for claiming too much foresight on the part of individuals and for downplaying serious reservations. Indeed, David Ricardo, the famous early-nineteenth-century economist who first enunciated the idea, was himself skeptical of it. He argued that many people would not worry about future taxes because they might not expect to be alive when taxes were finally raised. The answer that people would still worry about their children and grandchildren is weakened by the facts that some have no children and others don't want to leave money to them anyway. Other factors that weaken the "Ricardian" argument are: lower borrowing costs for the government than the public; the uncertainty that taxes will ever be raised or who will bear them; and the possibility that debt which, in fact, brought higher employment, output, and investment might eventually pay for itself out of higher incomes. In that case tax rates would not have to increase in the future.

Finally, some economists argue that we should include in federal debt the implicit debt from the government's commitment to pay future benefits such as Social Security. If we do so, shouldn't we add assets in the form of the present value of future taxes that might be received? In principle, they are both relevant, but there is a strong argument that in view of the uncertain amounts to be projected for taxes and payments over many decades, it would be better to exclude them from our measures of the debt.

Alternative Measures of Federal Debt
Gross Federal Debt Held by Public1
Year Billions of Dollars Percentage of GDP Per Capita in 1992$

1945 235.2 110.9 15,487
1946 241.9 113.8 12,824
1947 224.3 100.6 10,244
1948 216.3 87.7 9,074
1949 214.3 81.6 8,879
1950 219.0 82.4 8,715
1951 214.3 68.4 8,000
1952 214.8 63.1 7,766
1953 218.4 60.0 7,643
1954 224.5 61.0 7,598
1955 226.6 58.9 7,300
1956 222.2 53.4 6,802
1957 219.3 50.0 6,364
1958 226.3 50.5 6,328
1959 234.7 48.9 6,262
1960 236.8 46.9 6,123
1961 238.4 46.1 6,006
1962 248.0 44.7 6,026
1963 254.0 43.5 6,016
1964 256.8 41.1 5,890
1965 260.8 38.9 5,752
1966 263.7 35.9 5,553
1967 266.6 33.6 5,398
1968 289.5 34.2 5,520
1969 278.1 30.0 4,992
1970 283.2 28.7 4,775
1971 303.0 28.8 4,773
1972 322.4 28.1 4,818
1973 340.9 26.8 4,752
1974 343.7 24.5 4,361
1975 394.7 26.1 4,521
1976 477.4 28.3 5,105
1977 549.1 27.8 5,318
1978 607.1 28.6 5,378
1979 639.8 28.2 5,152
1980 709.3 26.8 5,148
1981 784.8 26.5 5,144
1982 919.2 29.4 5,679
1983 1,131.0 34.1 6,665
1984 1,300.0 35.2 7,268
1985 1,499.4 37.8 8,032
1986 1,736.2 41.2 8,976
1987 1,888.1 42.4 9,372
1988 2,050.2 42.6 9,675
1989 2,189.3 42.3 9,803
1990 2,410.4 44.1 10,227
1End of fiscal years: June 30 to 1976; September 30, 1977-90.
SOURCES: Office of Management and Budget, Budget Baselines, Historical Data, and Alternatives for the Future, January 1993, Table 7.1, p. 346; Economic Report of the President, January 1993, Table B-29, p. 381; Bureau of Economic Analysis, National Income and Product Series Diskette, and Survey of Current Business, January 1993; and author's calculations.

About the Author

Robert Eisner was the William R. Kenan Professor of Economics at Northwestern University and a past president of the American Economic Association. He died in 1998.

Further Reading

Barro, Robert J. "Are Government Bonds Net Wealth?" Journal of Political Economy 82 (1974): 1095-1117.

Barro, Robert J. "The Ricardian Approach to Budget Deficits." Journal of Economic Perspectives 3 (1989): 37-54.

Bernheim, B. D. "A Neoclassical Perspective on Budget Deficits." Journal of Economic Perspectives 3 (1989): 55-72.

Buiter, W. H., and J. Tobin. "Debt Neutrality: A Brief Review of Doctrine and Evidence." In Social Security Versus Private Saving, edited by George M. von Furstenberg. 1979.

Eisner, Robert. How Real Is the Federal Deficit? 1986.

Eisner, Robert. "Budget Deficits: Rhetoric and Reality." Journal of Economic Perspectives 3 (1989): 73-93.

Eisner, Robert, and P. J. Pieper. "A New View of the Federal Debt and Budget Deficits." American Economic Review 74 (1984): 11-29.

Return to top