Skip Navigation

Mr. Chairman and Members of the Committee, I appreciate the opportunity to appear before you to discuss the long-term outlook for the budget and its implications for the U.S. economy. As you know, the federal deficit has dropped substantially from its level in the early 1990s, and as a share of gross domestic product (GDP), it has fallen to a 22-year low.

But this year's good budgetary news should not lull people into complacency; just over the horizon is the retirement of the large baby-boom generation. That retirement will drive up the costs of three important government programs: Social Security (which provides income to retired and disabled workers, their spouses, and others), Medicare (which helps pay the costs of medical care for elderly and disabled people), and Medicaid (which helps finance medical care for certain low-income people, including the elderly). In addition, continued expansion in the volume and intensity of services provided through Medicare and Medicaid is expected to put upward pressure on federal spending for each beneficiary enrolled in those programs. If the resulting budget pressure from both demographics and health spending is not relieved by reducing expenditures or increasing taxes, deficits will mount and seriously erode future economic growth.

Balancing the budget in the near term will help alleviate the pressure and improve the economic outlook. But only eliminating the deficit in the next few years, without changing the underlying growth of spending and revenues, will not ensure that future deficits remain at acceptable levels. Correcting the remaining budgetary imbalances will require some further decisions about taxes and popular government programs.

Those decisions will grow more daunting if action is put off until deficits begin to rise rapidly. Delay increases the amount that the federal government must borrow, which crowds out private investment and pushes up interest rates. And because delay leads to higher federal interest payments, it further expands the size of the budget gap that policy changes must fill.
 

THE CHANGING DEMOGRAPHIC PICTURE OF THE U.S. POPULATION

Some simple demographic facts lie behind concerns about the long-run budgetary situation facing the United States. This country's population is aging. Over the next 35 years, the Social Security Administration estimates, the number of people age 65 and older will double, while the number of people age 20 to 64 will increase by only 20 percent (see Table 1).
 

Some of the demographic changes reflect the welcome fact that people are living longer today. Thanks to improved health care and healthier lifestyles, a growing proportion of the adult population now reaches age 65, and life expectancy at that age has increased by 15 percent since 1970. When Medicare was created in 1965, the average person was expected at birth to live about 70 years. By 1990, that expected lifespan had risen to 75 years, and by 2010, it is projected to increase further to 78. By any standard, those are remarkable gains.

A second factor behind the demographic changes is the baby boom: the large generation of Americans born between 1946 and 1964 (see Figure 1). In 2008, the oldest members of the baby boom will turn 62 and become eligible to claim early retirement benefits under Social Security. That date will end a period of relatively favorable demographics that began with the retirement of the generation born during World War II and the Great Depression, whose relatively small numbers are now providing a respite to Social Security and other entitlement programs for the elderly.
 

FIGURE 1. NUMBER OF BIRTHS IN THE UNITED STATES, CALENDAR YEARS 1910-1994

SOURCE: Congressional Budget Office based on data from the National Center for Health Statistics.

Besides straining entitlement programs, the retirement of the baby boomers will also significantly slow the growth of the labor force. The effect of having such a large group of workers leave the labor force will be accentuated by the fact that the high birth rate during the baby boom was followed by a lower rate (a baby "bust"). As a result, the growth of the labor force will slow to a crawl from 2010 to 2020 and almost to a standstill between 2020 and 2030. That projection stands in stark contrast to the 2 percent annual growth that the labor force recorded from 1960 to 1989, and even to the 1 percent average annual growth rate expected over the next 10 years.

Although the details of such population projections are inherently uncertain, the basic message is unambiguous: with more retirees and little growth in the number of workers, the ratio of retired people to workers (the so-called elderly dependency ratio) will increase significantly in coming decades. In 1960, there were about 20 Social Security beneficiaries for every 100 workers. That ratio has risen to about 30 Social Security beneficiaries for every 100 workers and is expected to swell to about 50 beneficiaries per 100 workers by 2030.
 

THE BUDGETARY IMPLICATIONS OF AN AGING POPULATION AND GROWING HEALTH CARE COSTS

Both the outlay and revenue sides of the federal budget will be strained as the elderly dependency ratio worsens and per-enrollee spending in government health care programs continues to rise. Revenues will be squeezed as the number of people working--and the economy--grows more slowly. At the same time, outlays for government programs that aid the elderly (Social Security, Medicare, and Medicaid) will increase significantly as the number of people eligible to receive benefits from those programs shoots up. In 1996, federal spending for the three programs reached about $630 billion, or 8.4 percent of GDP. But by 2030, when most baby boomers will have retired, the programs are projected to consume 17 percent of GDP--about twice today's percentage.

The projected increase in Social Security spending as a share of GDP results entirely from the surging number of people eligible for benefits, but the growth in Medicare and Medicaid also reflects an increase in spending per beneficiary. Unlike Social Security, whose real spending for each enrollee is set legislatively by a formula that depends on the enrollee's history of wages, traditional Medicare and Medicaid are open-ended entitlement programs that place no dollar limits on the benefits provided to each enrollee. Over most of the programs' history, benefits per enrollee have risen rapidly.

Indeed, the growth in per-enrollee costs is the main reason that federal spending for Medicare and Medicaid, now about three-quarters that for Social Security, is projected to overtake Social Security spending within 10 years. Although outlays for Medicare and Medicaid in fiscal year 1996 were lower than anticipated, the Congressional Budget Office (CBO) expects that growth to pick up again, although at a slower pace than before. The growth in spending per beneficiary reflects an increase in the volume and intensity of services provided through Medicare and Medicaid for each spell of illness. Those factors will continue to increase the burden of federal health costs in the years ahead. Thus, even if the elderly dependency ratio did not climb with the retirement of the baby boom, federal health spending would still be projected to rise faster than gross domestic product and would put increasing pressure on the budget.
 

LONG-TERM PROJECTIONS OF THE BUDGET AND THE ECONOMY

What would happen to deficits and the economy if U.S. budget policy did not change in the face of the impending retirement of the baby boomers? CBO has tried to answer that hypothetical question by projecting future government revenues and expenditures under various economic and demographic assumptions and examining their impact on the federal deficit and economy. CBO reported the results of that analysis in Chapter 4 of The Economic and Budget Outlook: Fiscal Years 1997-2006, published last May. The bottom line of the analysis is that current budget policy is unsustainable. Moreover, attempting to preserve current policy would severely damage the economy in the first quarter of the 21st century.

The lower-than-projected deficit for fiscal year 1996 suggests an improvement in the short-term budgetary outlook, which will be reflected in CBO's upcoming annual report. That improvement will also brighten the long-term picture and delay the emergence of serious trouble for a few years. Although the projections that I am presenting today do not reflect those revisions, CBO's preliminary analysis suggests that the improved short-term outlook would not eliminate the long-term imbalances in the federal budget, and thus, the qualitative conclusions of CBO's May report stand firm.

Budgetary and Economic Assumptions

In conducting its long-term analysis, CBO did not attempt to extend its regular budgetary projections beyond 2007. For one thing, the concept of a current-policy baseline is somewhat ambiguous even for 10-year projections; over a much longer period, that approach could produce misleading results. Instead, CBO simply assumed that spending would grow according to some simple rules for most categories of the budget. CBO also adopted the official long-term projections for Social Security and Medicare made by the trustees of those programs and adjusted the numbers for any differences between CBO's economic assumptions and those of the trustees. CBO also followed the trustees of Medicare's Hospital Insurance Trust Fund in assuming that federal health costs per beneficiary would slow significantly over the next two decades and, after 2020, would grow no faster than overall wage rates--an assumption that is admittedly optimistic. Tax revenues were assumed to remain a roughly stable share of GDP.

Because CBO's analysis focuses on macroeconomic relationships, its long-term projections use the budget categories defined by the national income and product accounts, not the categories of the unified budget, which CBO focuses on in its annual reports.

To assess the effect of long-run budget policies, CBO also had to make assumptions about fundamental forces in the economy over the coming decades. Because the growth of the labor force is expected to slow, CBO's base scenario assumes that annual growth in the total number of hours worked will drop virtually to zero by 2020. The economic projection also depends on the growth of the capital stock, which is affected by the federal budget deficit. But even without factoring in the effects of higher deficits on growth, CBO projects that annual growth of real GDP will slip from 2.1 percent in 2005 to 1.3 percent in 2030 simply because the labor force is expected to grow much more slowly.

Because great uncertainty surrounds both the budgetary and economic assumptions, CBO looked at several different scenarios and tested the sensitivity of its results to changes in the assumptions. In particular, it ran its economic model using a large number of alternative assumptions about population and productivity, reflecting the historical variation of those two key variables. And in a deliberate attempt to be somewhat optimistic about the effects of deficits on the economy, CBO assumed that private savers would offset half of the rise in the deficit and that foreign investors would continue to lend to the United States despite the increasing riskiness of holding U.S. assets as the growth in federal debt escalated. Those two assumptions work to delay the emergence of serious economic problems from growing federal debt.

The specific numbers in CBO's projections are tremendously uncertain and should be viewed cautiously. Even modest changes in the assumptions will affect the projections of the deficit and debt. Because of such uncertainties, CBO does not believe people should focus on the predictions about the specific level of the deficit or debt 30 years from now. Instead, the important message of CBO's simulations is the general trend in those fiscal variables.

Budget Projections Without Economic Feedbacks

Even before considering the interaction between the budget and the economy, the outlook for the deficit is gloomy. Indeed, one seemingly plausible path of revenues and spending would produce a deficit of about 12 percent of GDP by 2030 and would require investors to hold federal debt equal to almost 160 percent of GDP (see Table 2).⁽¹⁾ The deficit has reached levels that high only during major wars, and the debt has never been so large. For those calculations, CBO projected gross domestic product assuming that the growth of the labor supply slows with the retirement of the baby boomers, that capital investment grows with the overall economy, and that total factor productivity grows at historical rates. (The growth of total factor productivity is the growth of output that cannot be accounted for by the growth of capital and labor.)
 

The rise in the deficit stems from an escalation of spending in just four categories of the budget: Social Security, Medicare, Medicaid, and interest on the debt. CBO's projections show Social Security increasing from 5 percent of GDP in 1995 to 7 percent in 2050, outlays for Medicare and Medicaid climbing from 4 percent of GDP in 1995 to 12 percent in 2050, and interest payments on the debt soaring from 3 percent of GDP in 1995 to 14 percent in 2050--even without taking account of the economic feedbacks that push up interest rates. The aging of the population and the growth of per-enrollee health costs explain the rise in spending for the three entitlements programs; the growth in interest costs stems from the additional debt needed to finance higher entitlement spending.

With such massive fiscal changes, it is unreasonable to assume that GDP growth and interest rates will be unaffected. Incorporating economic feedbacks into the projections only makes the outlook worse. Under an array of scenarios with economic feedbacks that assume no change in current budget policy, the debt would increase to historically unprecedented levels in the next four decades (see Figure 2). Moreover, as federal debt pushed up interest rates and lowered the growth of the economy, interest payments would begin to consume an ever larger share of federal spending and eventually grow at an explosive rate. In the end, the total amount of debt held by the public would reach levels that the economy clearly could not support.
 

FIGURE 2. PROJECTIONS OF FEDERAL DEBT AND REAL GNP PER CAPITA, USING THE ASSUMPTIONS OF THE BASE SCENARIO WITH ECONOMIC FEEDBACKS

SOURCE: Congressional Budget Office.

NOTES: Discretionary spending is assumed to grow with inflation after 2006. Simulations I, II, and III are based on alternative assumptions about population and productivity growth. Simulation II is the base scenario, which assumes that the population grows according to the midrange path of the Social Security Administration and that total factor productivity grows at 0.7 percent annually. Simulations I and III are defined so that two-thirds of the 750 alternative simulations fall between them. Thus, the chance of an outcome better than scenario III is about 15 percent; correspondingly, the chance of an outcome worse than scenario I is also about 15 percent.

The projections are truncated when debt held by the public exceeds 300 percent of gross national product (GNP).

The figure reflects the baseline projections that appear in Congressional Budget Office, The Economic and Budget Outlook: Fiscal Years 1997-2006 (May 1996).

Based on the definition of GNP before the January 1996 benchmark revision.

How Rising Deficits Would Affect the Economy

Such rapidly rising debt would have serious consequences for the economy. Federal debt would displace private capital in housing and in business plant and equipment. It would also increase U.S. borrowing from foreigners. Thus, the economy would produce less output, and a larger fraction of that output would have to be paid to foreigners to service the borrowing from abroad. The rising debt would eventually put an end to the long-term growth of real gross national product (GNP) per capita.⁽²⁾

Although CBO's projections show the economy responding smoothly to the rapidly rising debt, those adjustments would probably be much more disorderly. Foreign investors cannot be expected to lend to the United States forever in the face of explosive debt, as CBO assumed in its projections. At some point, foreign lenders would lose confidence in the United States and withdraw their capital. If that happened abruptly, the exchange rate would plummet, interest rates would shoot up, and the economy would drop into severe recession. No one knows when that would occur, but when it did, the United States would have to service its foreign debt at unfavorable terms.

Of course, the above scenario is not a forecast of what will actually happen to the debt and the economy. Instead, it is a conditional projection of what could happen if the United States blindly followed current policies into the 21st century. Surely, policymakers would take the necessary steps to limit the growth of debt before it reached unthinkable levels. But because debt can quickly snowball out of control at levels above 100 percent of GDP, policymakers would need to act well before the debt reached that critical level.
 

THE ECONOMIC BENEFITS OF ACHIEVING A SUSTAINABLE BUDGET POLICY

For any path of spending and revenues to be sustainable, the resulting debt must eventually grow no faster than the economy. One measure of the size of the problem that policymakers face is the amount that revenues would have to rise to keep the debt from exceeding its current percentage of GDP for the foreseeable future. In May, CBO estimated that permanently increasing revenues by between 3 percent and 5 percent of GDP now would achieve that goal, depending on the assumptions about future discretionary spending. That amount would represent a tax hike of 15 percent to 25 percent.

Other policies could also create sustainable budgetary conditions. For instance, a budget that was permanently balanced would freeze the level of federal debt. Thus, as the economy grew, debt would gradually fall as a share of GDP. Because the policy could be maintained forever, it would be sustainable. However, sustainable policies do not require balanced budgets. In principle, the government could keep the budget in deficit forever, as long as the deficit did not grow relative to the economy. Under the assumptions of CBO's long-term projections, if the government stabilized the deficit at about one and one-half percent of GDP, debt would remain at about its current share of GDP--50 percent--indefinitely.

Regardless of what budget strategy lawmakers choose, the economic benefits of achieving any long-term, sustainable budget policy will be substantial. CBO's estimates from May 1996 indicated that permanently balancing the budget could raise real incomes (GNP per capita) in the United States by 12 percent by 2025 and by ever larger percentages thereafter (see Table 3). Moreover, even a policy that just held the debt to 50 percent of GDP (and thus allowed for permanent deficits) would raise real incomes by 10 percent in the next three decades compared with the base scenario. Of course, if the base scenario reflected the recent short-term improvement in the budget outlook, the gains from reaching--and maintaining--a balanced budget would be delayed by a few years, but they would still be substantial.
 

The biggest economic benefits come from moving the budget from an unsustainable track to a sustainable one. A balanced budget provides for higher incomes in the long run than a policy involving permanent deficits, and a policy with budget surpluses boosts long-run incomes even more. But the differences are not large. For example, a balanced budget would raise real incomes by only 2 percent more in 2025 than would a policy that stabilized the debt at its current percentage of GDP, which implies a long-term deficit at about one and one-half percent of GDP (see Table 3).
 

THE BUDGETARY REQUIREMENTS FOR ACHIEVING A SUSTAINABLE POLICY

There are many possible ways to help resolve the policy dilemmas posed by the aging of the population and the growing cost of health programs. One approach incorporated in the Balanced Budget Act that the Congress passed (and the President vetoed) in 1995 would have reduced the growth of nominal Medicare spending and would have converted Medicaid to a block grant administered by the states, thereby ending its status as an automatic federal entitlement. Changes might also be made in the formulas that determine Social Security benefits, such as further raising the normal retirement age and reducing the cost-of-living adjustment. Those approaches would limit the growth of the deficit by restricting entitlements. Of course, the nation might instead decide that the benefits of entitlements are worth paying for and accept higher taxes or cuts in other spending to provide the needed funding.

In any case, achieving a sustainable policy will require significant changes in government spending or revenues. Simply balancing the budget by 2002 with broad-based cuts that reduce the level of spending without reducing its long-term growth rate will not eliminate the long-term imbalances, although it will significantly brighten the outlook. Moreover, the imbalances cannot be corrected simply by cutting discretionary spending. The size of the required budgetary changes would exceed total discretionary spending after 2020. Instead, tough choices will have to be made about taxes or popular entitlement programs.

Achieving Sustainability Through Tax Increases

Lawmakers could correct the projected budgetary imbalance by raising taxes, but the size of the tax increases would be large. To keep the budget balanced through tax increases alone, federal revenues would have to rise from 20 percent of GDP in 1995 to about 28 percent in 2050 (see Figure 3). (That scenario does not describe an immediate increase to a higher constant tax rate such as the one mentioned earlier, but a gradual increase in the tax rate that is sufficient to keep the budget balanced.) Moreover, that estimate probably understates the actual size of the tax increase that would be necessary because it does not account for the adverse impact that rising marginal tax rates could have on incentives to work and save.
 

FIGURE 3. PROJECTIONS OF REVENUES IF TAX INCREASES ARE USED TO ACHIEVE BUDGET GOALS

SOURCE: Congressional Budget Office.

NOTE: The balanced budget path assumes that the budget is balanced by 2002 and remains balanced thereafter. The path with the steady ratio of debt to gross domestic product assumes that the ratio is stabilized at its current level. The projections of the base scenario use the balanced budget economic assumptions. Receipts are as defined in the national income and product accounts.

The figure reflects the baseline projections that appear in Congressional Budget Office, The Economic and Budget Outlook: Fiscal Years 1997-2006 (May 1996).

Keeping the current ratio of debt to GDP steady would at first require smaller tax increases than would balancing the budget, but the additional interest costs would eventually require slightly larger increases. Under the steady debt-to-income strategy, revenues would rise from 20 percent of GDP in 1995 to 29 percent in 2050.

Slowing the Growth of Social Security and Medicare

Another way to solve the long-term deficit problem would be to reduce future spending commitments for Social Security and Medicare. A chapter in CBO's August 1996 report Reducing the Deficit: Spending and Revenue Options examined a wide range of long-term approaches that would prevent Social Security and Medicare spending from growing more rapidly than the economy when the baby boomers become eligible for both programs. Those approaches could be used separately or in combination to achieve substantial savings. CBO will publish a revision of that report in March, including an analysis of the potential effects on the economy of reducing the growth in spending for Social Security and Medicare.

Social Security. To prevent Social Security spending from growing faster than the economy, policymakers would have to substantially curtail the commitments made under current law. CBO's August report examined three main ways of doing that. First, the initial benefits of future Social Security recipients could be reduced below the levels that current law would provide. Across-the-board cuts in initial benefits that were announced well before they took effect could produce substantial savings while still preserving the basic benefit structure of the Social Security system. In principle, workers could offset the cut in their future benefits by either working longer or saving more. However, some people would not be able to make the necessary adjustments and could therefore have much lower income when they stopped working.

Second, the age at which a worker becomes eligible for full retirement benefits--the "normal retirement age"--could be increased. Under legislation enacted in 1983, the normal retirement age is already scheduled to rise from 65 to 67. Some proposals would speed up the transition to 67 and then further increase the age to keep up with future gains in life expectancy. Raising the normal retirement age is, for most purposes, equivalent to cutting initial Social Security benefits, with similar advantages and disadvantages.

Third, future annual cost-of-living adjustments (COLAs) could be reduced. Current law indexes the basic Social Security benefit by the increase in the consumer price index (CPI), beginning when a worker becomes eligible for benefits. Many analysts feel that the CPI overstates increases in the cost of living, although the magnitude of the overstatement and what should be done about it are subject to much debate. For example, the Advisory Commission to Study the Consumer Price Index (also known as the Boskin Commission) recently estimated the size of the upward bias to be about 1 percentage point per year.⁽³⁾ Unlike across-the-board reductions in benefits and increases in the normal retirement age, substantial changes in COLAs would eventually reduce benefits the most for the oldest beneficiaries and those who initially became eligible for Social Security on the basis of disability.

Those and other approaches were considered by the Advisory Council on Social Security in its recent report. As you know, the members of the council were unable to reach a consensus on how to improve the financial status of Social Security. I understand that the chairman and two other members of the council will be appearing before this Committee next week to discuss their competing proposals.

Much of the public discussion about their plans has focused on aspects that involve either requiring workers to invest a certain percentage of their earnings in retirement accounts or investing a portion of the balance in the Social Security trust funds in equities rather than Treasury securities. Ultimately, the success of the approaches presented by the council in preparing the economy for the retirement of the baby boomers rests on the extent to which they would increase national saving. Some of the specific provisions in one or more of the plans would do that by slowing the growth in spending for Social Security. Other provisions could increase national saving if they required workers to save more than would otherwise be the case or if they raised taxes.

Medicare. Three fundamental approaches exist for slowing the growth in federal Medicare spending. The Congress could reduce the number of people eligible for benefits, collect more of the costs from beneficiaries, or restructure Medicare so as to reduce total health care costs per beneficiary.

One way to reduce the number of people eligible for benefits would be to increase the age of eligibility from 65 to 70, for example. That would do little to reduce total health care costs, however, and it would lengthen the period of time during which people who opted for early retirement under Social Security might have difficulty getting private insurance coverage.

Under the second approach, premiums collected from beneficiaries would be increased to cover a substantial part of Medicare's total costs (for both Parts A and B). Nearly all of those collections would represent federal savings since enrollees' premiums cover about 10 percent of Medicare's costs now, and that share will fall steadily after 1998 under current law. This approach would shift costs to beneficiaries rather than constrain the growth in total health care costs. Without any changes to improve the efficiency of the Medicare program, premiums would consume an ever larger share of enrollees' income.

A third approach to slowing the growth of federal Medicare spending would be to restructure the program in such a way as to limit the total health care resources that each Medicare beneficiary uses. One way to do that would be to set up a system of competing health care plans--of which Medicare's traditional fee-for-service sector would be just one--and limit growth in the amount that Medicare contributes to enrollees' premiums. Because enrollees would be responsible for any excess premium amounts (and would receive rebates for plans costing less than Medicare's contribution), they would have financial incentives to be prudent purchasers of health plans. And because plans would be at risk for any costs above their predetermined premium collections, they would have financial incentives to operate efficiently. Control of federal Medicare spending would be assured because the financial risks from higher growth in health care costs would be shifted to health plans and enrollees. However, the effects that those changes would have on total costs for a basic benefit package--and therefore on the costs that beneficiaries would face--are uncertain.
 

THE IMPORTANCE OF ACTING SOON

In solving the long-term budget problem, lawmakers need to address one overarching concern: the issue of timing. The pressures of demographics and rising health costs will become severe in just a few years. Putting off action on the looming budget problem will raise the ultimate cost of constraining the growth of the federal debt. Because delay increases the total amount of outstanding debt, it leads to smaller capital stocks and lower levels of output than would otherwise be possible.

The stakes will get significantly higher when the baby boomers begin to retire at the end of the next decade. At that point, the budget deficit will begin to rise rapidly if no change in policy occurs. Any delay at that time would significantly add to the amount of debt to be serviced and produce higher interest costs. And as interest costs rose, efforts to balance the budget would have to cut more and more deeply into programs or raise taxes even further. Moreover, the accumulation of debt and the diversion of national savings to federal borrowing in the interim would reduce the productive capacity of the economy, thereby cutting into the tax base. Thus, postponing difficult decisions will make the choices that have to be made later even more difficult.

There are other reasons not to delay attacking this problem. If the solution involves changing entitlement programs, concerns for both equity and efficiency suggest that such changes should be committed to well before they are carried out. Entitlement programs for the elderly are generally viewed as long-term commitments between the government and the citizenry, and people have based their behavior on the programs' current provisions. Deciding soon on any future changes in such programs and making gradual changes in spending and tax policies would give people more time to adjust their saving and retirement plans.

Moreover, significant benefits can accrue not just from making decisions about future cuts in the deficit now, but from actually making some cuts. In fact, balancing the budget over the next few years will significantly brighten the budgetary picture in the long run. But if policymakers do not address the factors that will increase the deficit thereafter, those changes alone will not completely eliminate the long-term problem.

The outlook for the economy will, of course, depend on how policymakers reduce the deficit. But that should not obscure the fundamental importance of resolving the long-term budgetary situation. Although alternative deficit reduction packages would affect the economy differently, those differences are much smaller than the benefits that any type of deficit reduction package would bring. Doing nothing about the looming deficits on the horizon, however, is simply not a feasible option.

1. That projection assumes that discretionary spending would grow no faster than inflation after 2006, which implies that its share of GDP would decline significantly over the next 40 years. Assuming instead that discretionary spending in the long-run remains at the same share as in 2006 would produce even larger projected deficits.

2. Unlike GDP, gross national product subtracts the net dividend and interest payments paid to foreigners who invest in the United States; as a result, it is a better measure than GDP of the income available to the U.S. population.

3. Advisory Commission to Study the Consumer Price Index, Toward a More Accurate Measure of the Cost of Living, Final Report to the Senate Finance Committee (December 4, 1996).