Go to Brad DeLong's Home Page
A balanced-budget amendment would be extremely costly for the U.S. economy. It would have at least four damaging consequences:
First, balancing the budget too rapidly could send the economy into a recession; balancing the budget through spending cuts over 1997-2000 would be likely to raise the short-term unemployment rate by one and a half percentage points.
Second, a balanced-budget amendment would amplify the business cycle. It would intensify the unemployment and economic distress suffered in recessions.
Third, a balanced-budget amendment would further depress our already-anemic public investment effort. America's long-run growth has suffered from the fall in public investment over the past decade; further downward pressure on public investment would further reduce growth.
Fourth, a balanced-budget amendment would put sharp pressure on government analytical agencies to produce numbers that conform to the balanced-budget straight jacket, and reduce the quality of available information
The other side of the ledger is relatively empty. The hope is that a balanced-budget amendment would prevent episodes like 1981, when large tax cuts coupled with spending increases unbalanced the federal budget for more than a decade. But there is little reason to believe that a balanced-budget amendment would prevent such episodes.
The performance of the federal budget over the past decade leaves no doubt that budget discipline is necessary. But current balanced budget proposals principles are at odds with the two principles that should guide fiscal policy--balancing the government's current operating budget over the business cycle so that expenditures to benefit today's Americans are financed by today's taxpayers, and basing government investment decisions on long-term costs and benefits.
Bad for the Economy in the Short Run
The Administration believes that in today's economic situation it would be extremely hazardous to implement a balanced-budget amendment. Rapid implementation of a balanced-budget amendment could trigger a recession.
The deficit reduction efforts already underway are putting a substantial burden on the current recovery. For example, the contraction in government spending on defense has already significantly contributed to the regional recession in California. The rapid contraction in spending and increase in taxes resulting from hasty implementation of an amendment could shrink demand by enough to trigger a nationwide recession.
The figure above shows an estimate of the effects of implementing a balanced-budget amendment by, starting in 1997, imposing sufficient spending cuts to balance the budget by 2000. The estimate assumes that the Federal Reserve manages to reduce short-term interest rates and stimulate the economy by enough to offset half of the contractionary impact of the balanced budget-required cuts. Even so, rapid implementation of the balanced-budget amendment lowers real GDP in 2000 by $110 billion dollars--and reduces the year-2000 level of employment by 2.7 million jobs. The unemployment rate spikes up to 7.6%.
In addition, in the above simulation the recession triggered by the balanced-budget amendment decreases tax revenue and increases spending: more than a hundred billion dollars of additional spending cuts and tax increases are required in fiscal 2000 simply to offset the effects of the induced recession on the budget.
If demand growth over the next five years is significantly stronger than the Administration projects, it may well be possible to undertake additional deficit reduction before the year 2000. But if growth is no faster than or slower than projected, the burden of substantial additional deficit reduction on top of the effort already underway would be more than the economy should bear.
Bad for the Economy in the Long Run
The balanced-budget amendment would do more than run the risk of creating a recession on its implementation. It would have damaging long-run economic consequences as well by eliminating the fiscal "automatic stabilizers" that moderate the depth of recessions.
When the economy enters recession, government revenues automatically fall: as individuals' incomes and firms' profits decline, they owe less in taxes. When the economy enters recession, government spending rises as more people draw on unemployment insurance and other social welfare programs.
The fall in tax collections and the rise in spending both act to boost the economy. They provide a cushion, keeping aggregate demand, and thus employment, from falling as much as they would otherwise.
Some advocates of a balanced-budget amendment attempt to preserve automatic stabilizers, suspending its provisions when a recession is likely or allowing the Congress in a recession to override the presumption that the budget should be balanced.
But most economists believe such attempts to preserve the benefits of automatic stabilizers would be ineffective: recessions emerge with too little warning for mechanical procedures to suspend balanced-budget provisions to be effective in moderating their depth. Thus a balanced-budget amendment would negate the automatic tendency of the federal budget to swing into deficit in recession, thus stimulating the eco-nomy, and into surplus in a boom, thus mod-erating inflation.
A balanced-budget amendment would have made recession periods like 1979-1982 significantly worse: according to one set of estimates, the unemployment rate would have risen 5.4 percentage points (rather than the 3.7 percentage points it rose in actual history).. The 1975 recession would, with a balanced budget amendment, have seen a peak unemployment rate of 10.2 percent-rather than the 8.3 percent of history.
The economy's "automatic stabilizers"--the automatic tendency for the government deficit to swing into deficit and so boost demand in a recession--are a valuable social asset that would be lost under a balanced-budget amendment. The damage done as a result of larger business cycles must be factored into any assessment of the desirability of a balanced budget amendment.
Bad for Public Investment--and Productivity Growth
Public investment in the United States today is at a very low level when measured as a share of GDP. Federal investment expenditures fell by roughly a quarter as a share of GDP over the past twenty years. As a result of this fall in public investment in the 1980s we are now a poorer nation. We lack some $300 billion worth of publicly-provided infrastructure that would have boosted American productivity, had the trends in public investment that prevailed up through the 1970s continued.
The Clinton Administration is committed to boosting public investment. The fiscal 1994 budget sees the Administration close roughly forty percent of the gap between what the GDP share of federal investment spending was in the 1970s and what federal investment had shrunk to during the Reagan and Bush Administrations.
This boost in public investment will pay handsome divi-dends: it will make America a more prosper-ous country, because there are many types of investments that should be undertaken by the government.
A balanced budget amendment would place Administration plans to boost public investment in extreme jeopardy. Such an amendment would force the government to mistake a means--deficit reduction--for an end-boosting America's standard of living.
It is important to recognize that even the most vociferous advocates of deficit reduction see it as a means, and not as an end in itself. It is folly to reduce the deficit by squeezing out public investment, because the main purpose of deficit reduction is to pave the way for more investment in America.
Cutting public investment to make room for more private investment is like running on a treadmill. America needs more of both--not a swap of one for the other.
Pressure on Government Estimates
Passage of a balanced-budget amendment would increase the pressure to resort to budgetary "gimmicks" and to distort the findings of government economic analyses. Pressure on analytical agencies to produce numbers that conform to the balanced-budget straight jacket would intensify.
Such pressure is likely to be strongest to the extent that implementing legislation "rel[ies] on estimates of outlays and receipts." The professional revenue and expenditure estimating organizations of the government would be placed under pressure to adjust their estimates in order to allow Administration and Congressional speakers to assert that current policies met the terms of the amendment.
One guide to the strength of the pressures that already exist in today's climate is provided by the history of budget deficit forecasts and outcomes in the 1980s. The figure shows actual budget deficit outcomes in every fiscal year from 1981 through 1992-along-side forecasts: the first Reagan Administration forecasts made in early 1981, and those forecasts made at the time of budget submission in 1982, 1983, 1985, and 1988.
In almost every case the outyear forecasts reported by OMB drastically understated the budget deficit. In part these understatements can be attributed to the unexpected recession of 1990. But this pattern is apparent long before then--and is in fact most apparent in 1981. Given the strength of the systemic forces that currently work to bias government estimates of spending and revenues, it seems unwise to strengthen them by increasing the incentives to focus on rosy scenarios.
The Other Side of the Ledger
The other side of the ledger is relatively empty. The major hoped-for benefit of a balanced-budget amendment is that it would help prevent disastrous episodes like 1981--when large tax cuts coupled with spending increases unbalanced the federal budget for more than a decade. These extraordinarily large structural deficits have made the U.S. a poorer country--have reduced GDP today by approximately $150 billion a year, and reduced the income of the average family of 4 by approximately $2,400 a year.
But there is depressingly little reason to believe that a balanced-budget amendment would prevent such episodes. To the extent that such episodes are caused by a political reluctance to make hard choices, a balanced-budget amendment will not help: such an amendment does not make hard choices.
To the extent that such episodes are caused by faulty economic theories leading to mistaken forecasts, a balanced-budget amendment would be of little help--recall that when the 1981 tax cut was proposed, it was part of an economic program that the Reagan Administration claimed that would lead to a budget balanced by 1984.
The performance of the federal budget over the past decade leaves no doubt that we need to restore discipline to bring spending and outlays into line.
But the principles that should guide fiscal policy are not those of a balanced budget amendment. The principles that should guide fiscal policy are straightforward:
Balancing the current operating budget over the cycle, so that expenditures to benefit today's Americans are financed by today's consumers, not by our descendants.
Making government investment decisions on the basis of their long-term costs and benefits--not on artificial short-term numerical targets.
A strong case can be made for legislation to reform the budget process to give effect to these principles. But implementation of a balanced-budget amendment would move us further from, not closer to, these principles. Sound principles of fiscal policy are remote from the logic and structure of a balanced-budget amendment.
The Administration's deficit-reduction program has roughly halved the size of the deficit as a share of GDP relative to the baseline.
The figure above shows Treasury Department simulations of the effects of implementing a balanced-budget amendment by imposing spending cuts to balance the budget by 2000. The spending cuts are large enough to send the economy into a significant recession.
In 2000 the unemployment rate, as forecast with a balanced-budget amendment, is up to 7.6%. GDP is lower in 2000 by $110 billion dollars--and employment is lower by 2.7 million jobs.
The balanced-budget amendment would eliminate the fiscal "automatic stabilizers" that moderate recessions.
The figure above shows what the unemployment rate was, and estimates of what the unemployment rate would have been in past recessions had a balanced-budget amendment been in effect--and negated the tendency of the budget to swing into deficit in recession, stimulating the economy.
A balanced-budget amendment would have made recessionary periods like 1979-1982 significantly worse: the unemployment rate would have risen by 5.4 percentage points (rather than the 3.7 percentage points it rose in actual history).
Throughout the 1980s, the Reagan and Bush Administrations claimed that the budget would be balanced in just a few years.
Throughout the 1980s, Reagan-Bush forecasts of future deficits were faulty-both because of optimistic macroeconomic scenarios and because of overruns on entitlements.
To some degree, the analytical process had been corrupted: politically-motivated demands to show the deficit on a downward trend could be made consistent with borrow-and-spend policies only by cooking the books, and creating a "rosy scenario."
A balanced-budget amendment would intensify such pressures. The quality of the analytical work done at OMB and CBO would fall, as the mandated balanced budget would increase the pressure to arrive at more politically-convenient numbers.
Some claim that Reagan-Bush forecasts of deficit reduction turned out to be overoptimistic not because the forecasts were rigged but because of Congressional refusals to enact Reagan-Bush spending cut proposals.
For example, former Reagan Administration political operative Martin Anderson writes that "[t]ime and time again President Reagansent budgets to the U.S. Congress to limit and control federal spending. Time and time again, the Congressrebuffed those plans, and substituted levels of spending that they, and they alone, considered appropriate."
All such claims are wildly, ludicrously false.
As the figure above shows, the bulk of the deficits of the 1980s came about because revenues were less than the level of spending proposed by Presidential budgets.
Congressionally-originated boosts in spending above levels that Reagan and Bush proposed made a very small contribution to the deficits of the 1980s.
November 25, 1993
MEMORANDUM FOR ASSISTANT SECRETARY ALICIA MUNNELL
From: Brad DeLong
Subject: BALANCED-BUDGET AMENDMENT
I had lunch with Sherman Robinson today to celebrate his no longer being an employee of the U.S. government. He brought along Sidney Blumenthal of the New Yorker.
Mr. Blumenthal thinks that the Balanced-Budget Amendment is the next political struggle that is going to catch the Administration by surprise. "Think of NAFTA and August," he said. "While all of you are congratulating yourselves on your political skills like you were doing last August, the members of Congress are sliding into commitments to support the Balanced-Budget Amendment.
"You should start putting out studies of the economic impact of the Balanced-Budget Amendment now," said Mr. Blumenthal. "And if you release them in December the press will write about them, because they need stories and Congress is out of session."
December 5, 1993
MEMORANDUM FOR SECRETARY BENTSEN AND
DEPUTY SECRETARY ALTMAN
FROM: Alan Cohen
Subject: BALANCED BUDGET AMENDMENT OPTIONS
As an alternative to a balanced budget amendment, the Administration might support a requirement that by 2005 the government balance the structural operating deficit. "Structural" means that the budget is in balance when the economy is operating at normal high levels of employment and capacity utilization, but is allowed to go into deficit in recessions. "Operating" means that--as is the case for all private for-profit and non-profit corporations in the United States--investments are classified separately from operating expenditures. Such a proposal helps place fiscal policy on sound principles. It avoids the drawbacks of the current balanced budget amendment proposal.
Currently-discussed balanced-budget amendment proposals are unwise for at least three reasons:
They require balance too quickly; it would be extremely difficult and costly to the nation to reduce spending or to raise taxes by as much as would be needed to meet a 1999 balanced-budget target.
They would destroy the government budget's role as an "automatic stabilizer": recessions would become significantly more painful.
They would put additional downward pressure on already-anemic public investment. The government should finance those investments that have long-term benefits by borrowing, so that the present generation does not bear the entire tax burden of financing such investments.
One possible alternative is that the Administration support either a stand-alone law, or an amendment coupled with implementing legislation, to by 2005 balance the structural operating deficit. "Structural" means that the budget is in balance when the economy is at its normal high levels of employment and capacity utilization, but is allowed to go into deficit in recessions. "Operating" means--as is the case for all private and non-profit corporations in the United States--investments in capital assets are not treated as operating expenditures but classified separately (however, depreciation of capital assets is an operating cost).
Before 2005 the structural operating deficit (currently forecast at some $93 billion for fiscal 1994) would be allowed to be in deficit, but by a decreasing amount as 2005 drew nearer.
Such an alternative would resolve most of the major substantive objections to a balanced-budget amendment.
Objection: An amendment requiring budget balance by 1999 would impose a large burden in additional taxes or reduced government spending on the economy in this decade, and could well trigger a recession.
Solution: Require balance by 2005, which would reduce the associated contractionary shock. The unified budget deficit is projected to be $264 billion in fiscal 1994. The structural operating budget in fiscal 1994 is in deficit by $162 billion--$73 billion of the deficit is because the economy is below normal levels of employment and capacity utilization, and $29 billion of the deficit represents net investment over and above depreciation in capital assets--and according to generally accepted accounting principles is not an operating cost. Thus the alternative proposal requires somewhat less fiscal retrenchment.
Objection: An amendment requiring budget balance year-by-year would force tax increases or spending reductions in recession years, thus making recessions worse. Proposals to allow Congress to vote a suspension of such an amendment in the event of recession are likely to be ineffective because we almost never have enough warning of an approaching recession to accomodate the legislative process.
Solution: Balance the "structural" budget. The Troika would calculate and report to the Congress the amount by which the federal budget is in surplus or deficit because of the deviation of current macroeconomic conditions from normal high-employment high-capacity utilization conditions. Only after adjustment for this "cyclical" component would balance be required--and enforced by sequestration. Thus the normal stabilizing influence of the federal budget on the economy would continue unhindered.
Objection: An amendment requiring budget balance would diminish the government's ability to spend on productive long-term investment. What the country needs is more private and more public investment, not deficit reduction purchased at the price of falling investment in infrastructure.
Solution: The alternative proposal could require balance not of the unified but of the operating budget. The U.S. government would then--like every corporation in America--include as a cost on its income statement not the value of its investments in new capital but the value of depreciation on the government's long-lived assets. The government's ability to invest to increase productivity would not be hobbled by year-to-year requirements, yet long-term financing of investment would be enforced because depreciation would appear as an expenditure on the government's operating budget.
Alan Cohen judges that the shift from the "unified" to the "operating" budget concept would carry political costs that would outweigh its substantive benefits.
Alicia Munnell proposes two possible concepts of "capital": (1) non-defense physical capital investments, plus non-defense research and development; (2) defense and non-defense physical capital and research and development, plus expenditures on employment and training.
One issue left open is the status of the alternative: should it take the form of a law implementing an attached amendment--that is, defining a "balanced budget" to be a "balanced structural operating budget" and setting out estimation and sequestration procedures--or should it be a stand-alone law? There is a strong view that anything other than a law implementing an attached amendment will be dismissed as not serious; there is a countervailing view that it is damaging to change the Constitution without serious substantive need.
The Structural Operating Budget for Fiscal 1994
(Capital Budget Includes Physical Assets and R&D)
$1317.6 Non-Investment Outlays*
$ 137.8 Depreciation Allowances**
$1455.4 Total Operating Outlays
-$204.1 Operating Deficit (-)
-$73.1 Cyclical Deficit Component (-)
-$131.0 Structural Operating Deficit (-)
MEMO: Unified Deficit (-) -$264.0
*Government outlays, excluding outlays for repayment of debt principal, and for defense and non-defense government and government-financed physical investments, and research and development.
**Depreciation on government-owned and -financed defense and non-defense physical investments, and research and development.
The difference between the $264 billion unified deficit in fiscal 1994 and the $131 billion structural operating deficit is made up of two pieces: (a) a $71 billion cyclical deficit component because economic activity is depressed below normal levels, and (b) government gross investments are $60 billion higher than estimated depreciation on government-owned and -financed physical and research and development capital.
Note that in future years the gap between the unified and the structural operating deficit is significantly lower. As the economy is projected to complete its recovery, the cyclical budget deficit component is projected to disappear. And depreciation allowances are projected to grow while government investment sees little change over the next five years.
Go to Brad DeLong's Home Page
Professor of Economics J. Bradford
DeLong, 601 Evans
University of California at Berkeley; Berkeley, CA 94720-3880
(510) 643-4027; phone (510) 642-6615 fax