Edward Renshaw
Professor of Economics
State University of New York at Albany
The data in Table 1.2 suggest that one of the more important impediments to the achievement of a balanced budget in the USA has been the problem of recurring recessions. Since the mild recession of 1960-61 the federal deficit in the national income and product accounts has always soared to a new historic high after a recessionary peak in economic activity and then gradually declined but (until 1996) never enough to achieve a smaller cyclical low before the next peak in business activity.
One of the more interesting points to note in connection with government consumption and investment is the possibility that declines in the real value of federal spending may have helped to tip the US economy into some recessions. In the last century and one-half the United States has never been able to extricate itself from a major war without enduring a wind down recession See /Table 2.1.
In this essay we will examine the bubbly nature of federal purchases and interest sensitive components of real GDP in the vicinity of recessionary peaks in economic activity. The leading indicator quality of more than one bursting bubble might enable policy makers to take actions that will help to prolong business expansions. The data, in any event, are not inconsistent with the possibility that a ridged attempt to quickly achieve a balanced budget might turnout to be counter productive.
The Center for International Business Cycle Research has found that the duration of post World War II expansions in economic activity has been about equal to 15 months plus 1.57 times the monthly lag for a secular rise in the prime rate after months containing an NBER trough in business activity (Lindley Clark 1993). During this period the US has never experienced a recessionary peak in business activity until at least three months after a sustained increase in the prime rate of a third or more from its cyclical low.
Once the prime rate has risen by a third or more it makes sense for the nation's monetary and fiscal authorities to carefully monitor federal expenditure and those interest sensitive components of the national income and product accounts that are likely to be adversely affected by rising interest rates.
In Table 2.2 we endeavor to identify recessionary peaks in business activity before they occur by summing quarterly declines in the real value of consumer durables, producer durable equipment, residential construction and federal purchases of goods and services (which are now classified as consumption expenditures and gross investment).
When the prime rate has risen on a sustained basis by a third or more from its cyclical low and declines in these four components of GDP have summed to .3 percent or more of annualized real GDP in the preceding quarter, the US economy has usually been very close to an actual peak in business activity or at least a one quarter growth recession (as was the case after the first quarter of 1995).
Government downsizing in conjunction with noteworthy reductions in some of the more interest sensitive components of GDP, it would seem, is analogous to a communicable disease that may spread to other parts of the economy.
The longest resolved lead time in Table 2.2 is the ten month lead time associated with the third quarter of 1952 when defense expenditures associated with the Korean war were still increasing at a rapid rate. Production of many goods was seriously disrupted in that quarter by a 53 day strike by the United Steel Workers from June 3 to July 24. All of the other resolved lead times in Table 2.2 fall into the comparatively narrow range of from minus seven to plus two months.
Economic recessions are virtually synonymous with large declines in residential construction and other types of durable goods which sometimes display leading indicator properties. The data in column (4) of Table 2.2 would suggest, however, that note worthy declines in federal purchases may have been among the straws that helped to break the back of some business expansions.
The peaks and troughs identified by the National Bureau of Economic Analysis (NBER) are consistent with the hypothesis that the US economy has never been able to avoid a near term recession after our participation in a major war Table 2.1. The July 1953 peak in business activity can be clearly linked to a contraction of defense spending. Tight money and large reductions in both defense and non defense spending during the fourth quarter of 1959 and especially in the first quarter of 1960 may have also had an important influence in helping to tip the US economy into the mild recession of 1960-61.
The wind down of the Vietnam war and efforts to reduce the size of government may have also helped to trigger two recessions: the mild recession of 1970 and the more severe recession of 1973-75.
The economic recessions of 1980 and 1981-82, when the Fed under the chairmanship of Paul Volcker was allowing interest rates to fluctuate widely in an effort to end the specter of double digit inflation, are not something that can easily be blamed on a contraction of government spending. A concerted effort to pay for a major increase in national defense spending by slashing other types of federal expenditure in third quarter of 1981, however, may have been one of the psychological straws that helped to produce an unexpected recession in July of that year.
There was so much over building and consuming by the end of the longest peace time expansion of the US economy in July 1990 that a recession might have occurred without the 4.8 billion reduction in federal spending during the third quarter of 1990. The Gulf War, which followed the July peak in business activity, was unique in that the weapons were withdrawn from inventory and largely paid for with financial contributions from our allies. Sending large numbers of young men and women to fight a war in a foreign country, however, can create the kind of uncertainty that is very disruptive to the sale of new cars and houses.
During the first three years of the Clinton administsration there was a fairly steady erosion of federal expenditures for consumption and investment which may have helped to slow the annualized growth rate for chain weighted real GDP to only .4 and .7 percent in the first and second quarters of 1995.
It is questionable, however, whether this erosion will be allowed to continue. Reductions in the real value of non defense consumption and investment have a long history of only turning out to be a one or two year phenomenon. The 24.5 percent erosion of the inflation adjusted value of national defense expenditure from 1987 to the third quarter of 1995 has lowered our defense capabilities enough to cause some advocates of a balance budget to propose additional expenditure.
If substantial progress is to be made in the direction of a balanced budget it will almost certainly require greater success at avoiding economic recessions and an unpopular increase in taxes or a slowing down in the growth of transfer payments and welfare entitlements, which are not a part of GDP. Keeping middle- class welfare entitlements from ballooning the deficit and wrecking the economy is one of the most challenging problems that will be faced as the post World War II baby boom approaches retirement. See Longman (1996) for a chilling introduction to this problem.
The first line of defense against a recession that might wreck the goal of a balanced budget is a forward looking monetary policy. From 1947 through the first quarter of 1996 there were 34 quarters when the prime rate charged by banks, which is now closely linked to changes in the federal funds rate, was allowed to decline in the midst of a business expansion. None of these quarters were immediately followed by a recessionary peak in business activity.
The nice thing about the bursting investment bubbles in Table 2.2 is that they have usually provided some advance warning with regard to an impending recession, if the federal government wasn't endeavoring to slash its own expenditure. A better appreciation of the leading indicator properties of some of these bursting bubbles, one would hope, might enable policy makers to take actions that will prolong business expansions and make it easier to achieve that elusive goal of a balanced budget.
In 1986 James Buchanan was awarded a Nobel prize for this many contributions to a relatively new branch of economics. Public choice is more nearly an economic theory of political science than an analysis of how the economy actually operates. It is basically concerned with the behavior of voters, politicians (especially those that are elected to office) and bureaucrats under various kinds of institutional arrangements.
In the public choice literature, there is a presumption that many of these participants in the governmental process may vote or react in their own selfish best interest and sometimes achieve results that are not in the best long run interest of future generations or even the present generation.
One part of the public choice literature is concerned with voting and the outcome of various election processes. Another portion of the literature endeavors to estimate the tax cost of goods and services provided by government and determine the impact of various taxes on the demand for such goods. A number of studies have shown that when local governments are required to raise property taxes, increase user charges, or otherwise pay a higher portion of the cost of building local infrastructure projects, the demand for elaborate public works is substantially reduced.
One of Buchanan's most highly publicized books is Democracy in Deficit. In this book, he and Richard Wagner contend that Keynesian economics has opened up a Pandor's box which has encouraged chronic budget deficits, bloated government expenditure, and resulted in a monetary-fiscal policy mix that is highly inflationary. This book was published in the late 1970s--several years before the Reagan Administration was successful in cutting taxes to the point of creating chronic deficits in the national income and product accounts in the 100 to 200 billion dollar range. At that time, Buchanan and Wagner were already arguing that US budget deficits financed with Treasury debt had become so acceptable to politicians as to reduce the perceived tax cost of additional government expenditure to the point where the only way to save our democratic institutions and protect future generations from debilitating inflation and an excessive debt burden would be to pass a constitutional amendment requiring the Congress to keep the US budget in balance under most circumstances.
The argument in favor of a balanced budget amendment was not so much a contention that deficit spending wouldn't be helpful in lifting an economy out of an economic recession but a fear that it would become so addictive as to eventually create longer run distortions in the use of resources and the distribution of income which would more than outweigh the short run stabilization benefits.
If government deficits are used to finance basic infrastructure, educational and other improvements in the quality of human resources, it is by no means clear that future generations will necessarily be disadvantaged by government debt. Additions to the stock of productive knowledge and increases in the amount of capital available per worker have so far been sufficient to make future generations richer, on the average, than their parents in the US and most other countries.
There is always the possibility, however, that large government budget deficits will begin to crowd out more productive private investments, increase inflationary pressures and eventually result in such high taxes to pay interest on the national debt that many resources get wastefully diverted to the task of avoiding taxes and minimizing the adverse effects of inflation. Public resistance to higher taxes might then create a worst of all worlds where interest on the public debt forces governments to cut back on services, defer maintenance of the public infrastructure and cut back capital expenditures to the point where many productive investments of a public character are never undertaken. The long run economic growth rate might then be less than if government had pursued a policy of balancing its budget.
Clark, Lindley (1993). "A Slump Predictor Clinton Should Love,"The Wall Street Journal, December 28, p. A10.
Lahiri, Kajal and Geoffrey Moore (1991). Leading Economic Indicators: New Approaches and Forecasting Records(New York: Cambridge University Press).
Longman, Phillip (1996). The Return of Thrift: How the Coming Collapse of the Middle-Class Welfare State Will Reawaken Values in America(New York: The Free Press).
------------------Duration in Months------------------------
War time Following Next First Second
Expansion Expansion Expansion Contraction Contraction
Civil War 46 18 34 32 18
World War I 44 10 22 7 18
World War II 80 37 45 8 11
Korean War 45 39 24 10 8
Vietnam War 106 36 58 11 16
Average
Duration 64 28 37 13.6 14.2
The business expansions following major wars have been of below average duration. In the haste to replace warnout equipment and consumer durables production is likely to rise to unsusainable levels and create a problem of accelerating inflation which may encourage the Federal Reserve to tighten credit to the point of tipping the US economy into an early recession.
When the US economy was not involved in a civil war the following recessons have been of below average duration.
Source: Business Cycle Expansions and Contractions estimated by the National Bureau of Economic Research and published in the Survey of Current Business, October 1994, p. C-51.
Decline End of Quarter
Sum as a Lead Time in
Quarter Declines Expressed in Billions of $ % of Real Months from
CD PDE RC FP GDP NBER Peak
(1) (2) (3) (4) (5)n (6)
1948-3 --- -1.8 -2.7 --- -.347 - 2
1952-3 -5.9 -12.6 -.7 --- -1.191 -10
1953-3 --- --- -3.3 -7.7* -.648 2
1957-2 -2.3 -1.0 -2.3 -.4 -.327 - 2
1959-4 -5.5 --- -3.9 -4.2* -.611 - 4
1960-1 --- --- --- -16.4* -.734 - 1
1969-4 -1.1 -1.2 -11.8 -8.8* -.673 0
1973-2 -4.3 --- -12.2 -3.3* -.508 - 5
1973-3 -2.0 --- -9.6 -18.0* -.758 - 2
1979-2 -7.0 -3.5 -4.3 --- -.322 - 7
1979-4 -8.6 -5.1 -9.3 --- -.496 - 1
1981-2 -12.0 --- -7.2 --- -.405 - 1
1981-3 --- --- -14.3 - .3 -.310 2
1989-4 -15.2 -10.3 -4.1 -2.7* -.531 - 7
1990-2 -15.8 -11.0 -11.6 --- -.624 - 1
1990-3 -5.0 --- -12.9 -4.8* -.367 2
1995-1 -13.3 --- -4.4 -8.2* -.387 ?
The quarterly declines in these components of real GDP are measured in 1987 dollars from 1948-58 and in chained 1992 dollars from 1959-95. The dash marks indicate that the component increased instead of declined.
(5)n. The sum of the declines in the components of real GDP in columns (1) through (4) expressed as a percent of annualized real GDP in the preceding quarter. We have omitted all those quarters after a sustained increase in the prime rate amounting to a third or more during a business expansion if the decline sum was less than minus .3 percent of annualized real GDP in the preceding quarter. Quarters containing a recessionary peak in business activity are also ignored if there was no decline in federal purchases.
*Cases where a decline in real GDP amounting $1 billion or more may have been instrumental in helping to tip the US economy into an actual recession or at least a "growth recession".
Go on to Essay 3:
Return to the Introduction