Behind the Numbers: The Great Surplus Debate

Save It

Alicia H. Munnell

A booming economy, surging tax revenues, and three budget deals (1990, 1993, and 1997) have allowed the administration's Office of Management and Budget (OMB) to project budget balance this year and surpluses thereafter. The Congressional Budget Office (CBO) also projects surpluses, beginning in 2001. No sooner did surpluses appear on budgeteers' spreadsheets than tax cutters, highway builders, and a host of others attempted to claim them. President Clinton countered with his State of the Union call to "Save Social Security First." Although the President flagged the right priority, he missed a striking opportunity. He should have called for separating Social Security from the rest of the budget. The surpluses belong to Social Security; there are no surpluses in the rest of the budget. Taking Social Security out of the unified budget allows Social Security to increase the national savings rate and politically shores up the program


The challenge of balancing the unified budget went very quickly from hard to easy, thanks to strong economic performance and a related but not yet fully understood jump in federal revenues. Future surpluses depend on a continuation of this strong economic performance. Continued economic health depends, in turn, on both continued good luck—for example, the Asian crisis has only minimal impact on the U.S. economy—and the monetary policy decisions of the Federal Reserve. The Fed remains the only game in town, since the march toward surpluses means fiscal policy will be restrictive for the foreseeable future. The economy is in unknown and unexpected territory. No one five years ago would have predicted that unemployment could hover around 4.7 percent while inflation remained around 2 percent. The Fed deserves credit for its willingness to experiment in this new economic environment, and its continued flexibility will be key to a prolonged expansion and budget surpluses.

The 1997 budget deal cut the deficit by $118 billion over 1998-2002 and pushed the budget from red ink into black. The legislation made significant cuts both in the so-called mandatory side of the budget via detailed programmatic changes to Medicare and in the discretionary outlays that include defense and other domestic programs. The budget deal, however, did not spell out how these discretionary programs were to be cut but rather imposed a complicated series of caps. These limit discretionary spending in 2002 to the level of nominal spending in 1999; inflation will pare the purchasing power of that spending by about 12 percent from today's levels. Even if Congress and the administration balk when they see the particulars and trim the discretionary spending cuts to 6 percent, the budget will still be in surplus through 2008, the end of the projection period.

How those short-term changes translate to the long run, however, is more art than science. For example, OMB projections show a "current services" budget in surplus almost through 2060. But these projections are very sensitive to assumptions about the growth in discretionary spending—it grows in line with inflation, grows in line with population, or remains a constant share of GDP—and the future of health care costs. My view is that deficits will probably reemerge as the baby boom retires. The question is what to do with the surpluses in the meantime.


The biggest threat is that the surpluses will be given away either through an assault on the tax code or simply as tax cuts. Although critics bemoan the complexity of the tax code, to date it has been relatively simple for most American taxpayers. Roughly 75 percent of filers use the standard deduction, and most of the remaining 25 percent claim deductions simply for mortgage interest, state and local taxes, and charitable contributions. But this will change. The legislation passed in August 1997 that cut taxes on families with children, small-business owners, and investors required more than 800 changes to the code. Capital gains tax cuts and confusing individual retirement account choices will further complicate the top end of the income scale. For middle-income families the complexity will come from education and child credits, which will require separate calculations and involve complex phase-outs. This increased complexity may increase taxpayer frustration.

H earings before the Senate Finance Committee in late 1997 created the impression that the Internal Revenue Service commonly abuses taxpayers. This is most probably not true. The bipartisan Kerrey-Portman IRS commission recently concluded that there was no systematic abuse of taxpayers. Instead the commission's report noted, "The agency spends significant resources educating personnel to treat taxpayers fairly, and the commission found very few examples of the IRS personnel abusing power."

The four witnesses, who really did have horrible stories, were culled from approximately 1,500 people who had contacted the committee. It's impossible to know how many of them had legitimate complaints, but assume they all did. That's 1,500 compared to 200 million tax returns filed each year and roughly 2 million audits. That's abuse at the rate of .00075 percent—not acceptable, but hardly widespread.

Regardless of the facts, the attack on the IRS has helped fuel the attack on the tax code. Louisiana Republican Representative W. J. (Billy) Tauzin, who joined Texas Republican Dick Armey in an antitax road show last fall, said that the IRS is "drunk with power." He went on to say that "If you've got any agency that's drunk with power, the first and most important thing you do is take the liquor away—the IRS code itself."

Scrapping the Internal Revenue Code has little to recommend it. Moving from progressive rates to a flat rate raises the tax burden drastically on low-income families and reduces it sharply on the rich. [See Joshua Marshall, "A Liberal Tax Revolt."] At the same time, the pressure to return some part of the unified budget surplus in tax cuts will be almost irresistible. Since the budget rules constrain outlays, lawmakers are always tempted to achieve spending objectives through the tax system. Several critics have also been bemoaning the inequities created by the so-called marriage tax. The only hope of permanently protecting the surpluses is to clarify that they belong to the Social Security program and make them unavailable for alternative purposes.

Subscribe to The American Prospect


Until now, separating Social Security from the unified budget has been politically impossible—despite numerous formal attempts—because it would have made the federal deficit look even bigger than it was. With the rosier budget picture, OMB projects that even the non-Social Security part of the budget will be in balance by 2007. Removing all Social Security revenues and expenditures from the budget documents and the annual budget process would focus public discussion, analyses, and budget-balancing efforts thereafter on the non-Social Security section of the budget.

With this change, the Social Security system would become a vehicle to increase national saving and capital accumulation without resorting to the risky and expensive option of individual, private accounts. More saving is not an end in itself but rather a means of achieving more investment, increased productivity growth, and higher levels of income in the future.

To date, increasing saving through accumulations in the Social Security trust funds has produced ambiguous results. Including Social Security in the unified budget has meant that Social Security surpluses have been available to finance deficits in the rest of the budget. Critics contend that the existence of Social Security surpluses encourages either taxes to be lower or non-Social Security spending to be higher than it would have been otherwise. Although little evidence exists to support this contention, an accounting treatment that separates Social Security from the rest of the budget will clarify the extent to which the system is adding to national capital accumulation.

We have tried other means of increasing saving, and they have not worked. When American saving rates dropped dramatically in the 1980s, Congress responded with the development of special tax-favored saving and pension accounts, such as individual retirement accounts, 401 (k) plans, and Keogh plans. A lot of money was accumulated in these accounts, but personal saving declined. My reading of the evidence is that individuals have shifted their saving toward the tax-favored vehicles, but have not increased the total amount of saving. The result has been a loss of federal tax revenues without any compensating response from individuals. Economists agree that reducing budget deficits or accumulating budget surpluses is the most direct and effective way to increase the nation's saving rate.

Social Security has been part of the budget since 1969, when the Commission on Budget Reform advocated the unified budget concept to better measure the effect of budget policy on the economy. But fiscal policy has diminished in importance since the late 1960s, reflecting the observed inability of Congress to make timely decisions, a changed perception of the importance of fiscal policy for stabilization in a world with increasingly integrated capital markets, and the intense focus on balancing the budget.

To politicians who have labored hard during the 1990s to eliminate deficits in the unified budget, shifting Social Security out of the budget will seem to be moving the goalposts. As soon as they declare victory on balancing one budget, they will be faced with a deficit in another budget. Therefore, postponing the shift, perhaps until 2007 or so, would soften resistance. It would also prevent undue pressure to further cut discretionary spending, which has already taken more than its share of hits.

Would it work? Comparisons of the federal government with the states are always tricky, but states have been successful in this endeavor. They accumulate reserves to fund their pension obligations but generally balance their budgets excluding the retirement systems. Their nonretirement budget balance has remained positive, while annual surpluses in their retirement funds have been hovering recently around 1 percent of GDP. Thus, states appear to be adding to national saving through the accumulation of pension reserves. With a commitment to balance the non-Social Security portion of the budget the same should be achievable at the federal level.

The best way to protect the surpluses from unwise tax cuts and to protect the Social Security program from those who want to privatize it is to take Social Security out of the unified budget. In the short run, this move would make clear that there are no surpluses in the non-Social Security part of the budget and would mute calls for tax cuts and unfunded spending increases. For the long run, the accounting change would eliminate the possibility of using surpluses in Social Security to mask deficits in the rest of the budget. Documenting that the Social Security trust funds were an effective saving vehicle would remove a major argument from those who want to privatize Social Security into a system of individual accounts.

Invest It

Dean Baker

W ith the federal budget deficit under control, we can at last focus on the other rising deficit—of public investment. There is a proven correlation between public outlay on physical and human capital and a country's productive capacity. Yet over the last two decades in the United States, public investment in almost all areas has declined precipitously. Since peaking in the late 1970s, federal spending on public investment, measured as a share of total economic output, has fallen by more than a third. If current budget trends hold, it will fall another 35 percent over the next ten years. These investment declines could significantly impede economic growth in the next century.

In orthodox thinking, deficit reduction reduces interest rates and thereby allows for more private-sector investment. This in turn boosts productivity growth and long-term economic growth. The current rosy economy is often attributed to deficit reduction. Certainly, one can cheer aspects of the current economic situation. The unemployment rate has fallen to its lowest level since the early 1970s, and real wages have begun to rise. But while the economy's growth rate has been respectable, it is virtually identical to the path projected before the President's deficit reduction plan was implemented in 1993.

And while the deficit has fallen faster than anyone had anticipated, the impact on real interest rates has been limited. The average real (inflation-adjusted) rate over this period has been 3.9 percent, only slightly higher than the 3.8 percent projected by the Congressional Budget Office (CBO) prior to the passage of the Clinton budget.

Contrary to widespread belief, the investment share of GDP in 1997 was the same as in the last business cycle peak in 1989—10.4 percent, which is far below its 1979 peak. Thus the decline in the deficit did not increase investment share or productivity growth. The rate of productivity growth over the current cycle is nearly the same as the 1980s rate and considerably lower than the rates in both the 1960s and 1970s. While there was strong productivity growth in the first three quarters of 1997, this more rapid growth would have to continue for a considerable time to offset the slow growth earlier in the decade. So deficit reduction has not met expectations in terms of spurring overall growth, private-sector investment, or productivity growth.


Economists agree that what has contributed significantly to economic growth over the last half century is public investment. Some studies suggest that expenditures on public investment actually affect productivity more than private investments do. But even if the impact of public and private investment on productivity is roughly comparable, recent trends are alarming because the decline in public investment has not been offset by increases in the private sector.

This decline in total investment bodes ill for our future. As we enter the twenty-first century, we need to educate and train our workers, upgrade our infrastructure to support new technologies, and give basic science the means to sustain innovation and technological advance. But we may not be able to make these crucial investments.

Human capital. Outlays on education, training, and early childhood needs are sound investments because they increase the economy's productive capacity in the future. But spending on education and training, after rising throughout the 1960s and 1970s, has fallen by about 50 percent since 1976. Spending fell steeply in the early Reagan years, increased slightly under the Bush administration, and then continued to decline under Clinton. This area of spending is projected to experience further cutbacks over the next ten years.

Physical capital. Public expenditures on physical capital provide the basic infrastructure that the economy needs to sustain itself—the roads, bridges, highways, and airports that support the transportation of people and goods. They also include facilities such as water purification systems, which provide clean drinking water to most cities, and sewage treatment facilities, which protect the nation's lakes and rivers. A well-planned system of infrastructure can lead to better use of existing urban areas and nearby suburbs, thereby reducing sprawl and the resulting pollution. Construction of efficient mass transit systems can also provide people with an attractive alternative to driving.

Here, too, spending has fallen. After federal outlay on physical capital peaked at just over 0.8 percent of GDP in 1980, it has since fallen to 0.5 percent of GDP and, if budget constraints on discretionary spending remain in place, may fall to less than 0.3 percent of GDP by 2007. Pollution-control facilities, such as waste-water treatment plants and urban mass transit [see J. W. Mason, "The Buses Don't Stop Here Anymore," TAP, March-April 1998], have suffered the most severe cuts in recent years.

Research and development. In each of the major industries in which the United States is the recognized world leader—aerospace, computers, agriculture, and pharmaceuticals—federal expenditures on research and development have been essential to growth. In the case of both aerospace and computers, research financed by the Defense Department or NASA led to major technological breakthroughs. The research supported by the Department of Agriculture and the system of land grant universities established to promote the spread of modern agricultural practices have been critical to establishing the United States as the world's leading agricultural producer. The government's support for research conducted through the National Institutes of Health has stimulated the pharmaceutical and biotechnical industries. Until recently, more than half of the research and development in pharmaceuticals was paid for by the government.


Source: Office of Management, and Budget, General Accounting Office.

The government's role in supporting basic research is crucial because there is a considerable time lag between scientific discovery and successful commercial applications, and the private sector is generally reluctant to support investments that require such a long payoff period. Furthermore, some discoveries may lead to enormous gains to society but might not provide much profit for an individual firm. Such research requires public support. But federal support has dwindled considerably since the 1960s—and projections show it falling further in the future. In 1966, federal spending on research and development was nearly 1 percent of GDP. On its current trajectory, R&D spending will fall to less than 0.25 percent by 2007.

T he President's 1999 budget does include modest investment in creases: an additional $6.3 billion in research subsidies; tax credits over five years to promote energy conservation; an additional $7 billion to reduce the average class size for young children; and $1 billion a year toward renovating the nation's schools. Though worthwhile, these funding levels are woefully insufficient. And even if all the President's proposals were passed into law, total expenditures on public investment would still decline because of cuts in other areas. The Office of Management and Budget (OMB) projects that, even when the President's entire investment package is included, public investment will increase at the rate of 1.6 percent a year from 1998 to 2003—which is more than one-half of a percentage point less than the OMB's projected 2.2 percent rate of inflation. When investment spending falls in inflation-adjusted dollars, it falls even more rapidly as a share of GDP. Moreover, the President won't get all the investment he has requested; Congress has its own agenda, and some of the funding for the President's proposals depends on uncertain events such as the tobacco settlement. The general trend for public investment is downward, and the only question is how fast it will fall.


To calculate the investment shortfall, we can use 1976, a typical year in postwar fiscal policy, as a baseline. Using this comparison, the shortfall in public investment for 1998 is a whopping $68.2 billion. The largest spending deficit, $26.1 billion, occurs in education and training. The shortfall in spending on physical capital is $21.9 billion, and $18.5 billion in R&D. If projected spending paths hold, the shortfall will rise to $173.1 billion in 2007, at which point we will be spending $55.9 billion less in physical capital, $52 billion less on education and training, and $45.5 billion less on R&D than we were in 1976.

A n even more dramatic shortfall emerges when we compile estimates from government agencies and experts in various fields of the public expenditures needed to meet basic goals, such as maintaining bridges at acceptable standards or providing adequate school facilities for children. For example, an additional $11 billion in annual federal spending will be needed to stabilize the physical condition of the nation's schools. The unmet need for higher education, as measured by the estimated cost of extending the Pell Grant program to all eligible low-income students wanting to attend college, is $10.1 billion. The shortfall in spending on training, as measured by comparing the level of U.S. expenditures as a share of GDP with the average spent on training by the Organization for Economic Cooperation and Development (OECD) countries, is $22.8 billion. (And if you compare U.S. spending on training with comparable spending in Germany, the shortfall is even higher: $32.5 billion.) And to fully fund Head Start, which currently enrolls only one-third of the low-income students who qualify for it, would require an additional $8.6 billion.

The largest estimated needs in the area of physical capital are for expenditures to maintain and improve the nation's highways and bridges and the mass transit capital stock. To fill potholes and re-pave highways, as well as to ensure the structural integrity of the nation's bridges, the government will need to spend $11.4 billion. This is a question not just of basic safety, but of sustaining projected levels of economic growth. Public investment can be used not just to guide economic growth, but to counterbalance suburban sprawl.

The present level of U.S. spending on R&D, measured as a share of GDP, is below the OECD average. To increase civilian research dollars as a share of GDP to Germany's level, for example, the United States would need to invest $6 billion more in basic research; to match Japan's R&D spending relative to GDP, we would need to spend $14 billion more.

Any means of estimating our unmet public investment needs are, by necessity, crude. Even these rough estimations, however, suggest that a substantial shortfall exists. In our deficit-cutting zealotry, we've skimped on public investment for the better part of two decades—and we now find ourselves confronted with a backlog of needs and an inadequate current level of spending to address them. Failure to use the surplus to increase public investment will slow the country's economic growth and jeopardize the health and safety of the population.

Understand It

Robert Eisner

L ast summer, the Wall Street Journal labeled as "Invincible Ignorance" a Republican proposal to run federal budget surpluses. With the surplus at last here, the label is now all too bipartisan, as President Clinton recommends that we reserve every penny of the surplus for Social Security. As tactics, this may be a clever way of heading off regressive Republican tax cuts, but as strategy it just reinforces the conservative fiscal assumptions that continue to paralyze many Democrats.

The government's running a surplus, whether "reserved" for Social Security or anything else, means it takes more in taxes than it gives the public in outlays. That helps neither liberals looking for more public investment nor conservatives who want businesses and households to make their own private spending decisions. Nor does it help the economy or the solvency of Social Security.

A budget surplus reduces the federal debt held by the public, currently holding steady at some $3.8 trillion. The meaningful measure here is the ratio of the debt to our national income or gross domestic product. Measured properly, the federal debt is already declining rapidly relative to GDP. The debt was more than 110 percent of GDP after World War II. It fell last year from 49.6 percent to 47.3 percent. A balanced budget itself, without any surplus, will bring it below 45 percent this year and keep reducing it year after year—just as rapid growth and moderate deficits reduced the World War II debt to about 24 percent at its postwar low in the 1970s.

T he alleged Social Security crisis stems often from uncertain forecasts that the Old Age and Survivors and Disability Insurance (OASDI) trust funds will be running out of money by the year 2029. Using the surpluses to add to the funds, if that is what any in the administration have in mind, would then delay that still distant doom another ten years.

But Social Security checks do not come from the trust funds. They come directly from the United States Treasury. With or without trust funds, our future Social Security benefits are specified by law. They will be paid out of taxes or borrowing, as long as the public supports the program. The OASDI funds are actually no more than accounting entities. The accountants—or these days, the computers—keep track of our payroll taxes and credit them to these accounts. They charge payouts from the Treasury to them. They also credit the accounts with an interest return on whatever positive balances the computers report—currently about 6 percent on balances of some $700 billion.

If increasing balances in the fund accounts seems necessary to reassure the baby boomers that Social Security will be there for them, we can do that easily without budget surpluses. We could readily credit other tax receipts to the OASDI trust funds, just as we do in the case of the Supplementary Medical Insurance trust fund (Medicare part B). For example, we could add credits of income taxes equal to 1.5 percent of taxable incomes to those 12.4 percent payroll tax credits.

Crediting more revenues to the trust fund accounts, however, whether out of a specific income tax credit or out of less certain budget surpluses, would have no effect whatsoever on government spending, taxing, or the debt to the public. Neither would it make any difference for the real issue for Social Security—an aging population.

The bread eaten by those not working must be baked by those working. When the numerous baby boomers retire and no longer produce for themselves, they will require support from the smaller Generation X then working. So if we want to bolster Social Security, the best solution is to increase economic growth rates and increase the wages that finance OASDI payouts.

With even a modest 1 percent per year growth in worker productivity—and our recent robust growth that has turned the deficit to surplus has been greater—we will have some 36 percent more output per worker in the year 2029. That will be enough, even with the predicted 10 percent increase in the burden on the working population, to offer 24 percent additional real income per capita to everyone, young and old. And this amount can be increased even more if we achieve and maintain maximum employment and maximum investment in the skills and productivity of the American people.

In that respect, holding budget surpluses for Social Security is perverse. Not using the surplus means no direct spending or targeted tax credits for old or new programs to invest in education, research, child care, health, infrastructure, or the environment. It also means no general tax cut to offer the public more to spend on its own.

Those advocating running a surplus and "paying down" the debt have one or both of two motivations. Many simply want to reduce the role of government. They fear the alternative of using the surplus for new government programs. Others argue that maintaining a surplus and reducing the debt will increase saving and investment. With less financial wealth and more to pay in taxes, the public will be forced to consume less.

But will less consumption necessarily bring more saving, investment, and growth? One can simply assume this, as did most pre-Keynesian economists as we plunged into the Great Depression of the 1930s. If employment can be assumed to be always full, along with total income and product, less consumption indeed must mean more investment. But suppose lower consumption causes firms to reduce production and lay off workers. Will our national saving rise then? If I do not buy a new car will the automobile manufacturer invest more—or less?

We also hear that with the government buying back its own securities, interest rates will be reduced and that will encourage investment. But economists have found little evidence that the rate of change of the federal debt has had much to do with interest rates. If we wish to stimulate investment with lower interest rates, the path to this runs by Alan Greenspan and his Federal Reserve. They clearly have the power to change interest rates, as anybody in the financial markets can testify. Low interest rates and the contributions of consumption to a prosperous economy are the way to maximize investment.

T he worst part of running surpluses to pay down the debt is its effect on public investment. President Clinton's 1999 budget proposes some $65 billion in new money for child care, Head Start, more teachers and classrooms, basic research, health, infrastructure, the environment, and other programs. That comes to just three-quarters of 1 percent of our projected gross domestic product. For years we have been told we could not have more of this kind of investment in our future because it would increase the deficit. With prospective surpluses, should we not finally provide properly for our children, our grandchildren, and ourselves?

Indeed, I would lay out a more ambitious program. Merely keep the ratio of federal debt to GDP constant. With (nominal) GDP growing at even a modest 5 percent per year this would mean that our current $3.8 trillion debt held by the public could grow by $190 billion. This would be "balance" in the relevant economic sense of the debt staying in the same position relative to our national income. But since the increase in the debt is the deficit, it would mean a deficit of $190 billion—and one growing at that 5 percent per year. I would use this deficit to finance public investment—in the human and physical capital vital to productivity and growth.

This would be the way not merely to "save" Social Security, which is in danger only from those who would nibble away at it or "privatize" it in the way of "reform." It would offer more production and income in the years ahead to all of us, young and old.

You may also like