This chapter quantifies the fiscal adjustment needed to stabilize debt/GDP over the very long run and also examines the generational imbalance (the difference in net taxes faced by current versus future generations). Both the fiscal and generational imbalances are large: we estimate an adjustment between 7 ¾ and 14 ½ percent of every future year’s GDP to restore sustainability and fiscal equity. A permanent cap on the growth of Medicare spending, along with the 2 ¾ percent of GDP adjustment advocated in the Staff Report, would eliminate 40 percent of the fiscal gap. The needed adjustment would rise if delayed.

Abstract

This chapter quantifies the fiscal adjustment needed to stabilize debt/GDP over the very long run and also examines the generational imbalance (the difference in net taxes faced by current versus future generations). Both the fiscal and generational imbalances are large: we estimate an adjustment between 7 ¾ and 14 ½ percent of every future year’s GDP to restore sustainability and fiscal equity. A permanent cap on the growth of Medicare spending, along with the 2 ¾ percent of GDP adjustment advocated in the Staff Report, would eliminate 40 percent of the fiscal gap. The needed adjustment would rise if delayed.

This chapter quantifies the fiscal adjustment needed to stabilize debt/GDP over the very long run and also examines the generational imbalance (the difference in net taxes faced by current versus future generations). Both the fiscal and generational imbalances are large: we estimate an adjustment between 7 ¾ and 14 ½ percent of every future year’s GDP to restore sustainability and fiscal equity. A permanent cap on the growth of Medicare spending, along with the 2 ¾ percent of GDP adjustment advocated in the Staff Report, would eliminate 40 percent of the fiscal gap. The needed adjustment would rise if delayed.

A. Introduction

1. The United States is facing major fiscal and generational imbalances. The combination of high fiscal deficits, an aging population and rapid growth in government-provided healthcare benefits have put the fiscal accounts on an unsustainable path. Staff and Congressional Budget Office forecasts imply that U.S. debt will rise rapidly relative to GDP in the medium to long term (Figure 1).

Figure 1.
Figure 1.

U.S. Debt in Percent of GDP

(1930–2083) /1

Citation: IMF Staff Country Reports 2010, 248; 10.5089/9781455206759.002.A006

B. Methodology

2. To measure the U.S. fiscal imbalance we compute the “fiscal gap”. Over a finite horizon, it measures the reduction in the deficit required so that the debt-to-GDP ratio in a particular year is the same as today. Over an infinite horizon, it measures the adjustment needed for the government to meet its intertemporal budget constraint, e.g., so that the present value of the excess of future expenditure and current liabilities over future receipts is zero. It has been argued that when fiscal pressures are concentrated in the long run, as in the United States, using the infinite horizon definition is preferable because finite horizon measures of the gap can underestimate the necessary adjustment (see Gokhale and Smetters, 2006).

3. To measure the U.S. generational imbalance we compute a set of generational accounts for all current and future U.S. generations. Generational accounts indicate the net present value amount that current and future generations are projected to pay to the government now and in the future. The accounts can be used to assess the fiscal burden current generations place on future generations, and thus offer a measure of the fiscal adjustments needed to make the fiscal structure generationally equitable (the Appendix offers details on the methodology used to compute the generational accounts).

4. Two main fiscal scenarios are used (Figure 2):

Figure 2.
Figure 2.

U.S. Federal Fiscal Overall (solid) and Primary Deficit (dotted) in Percent of GDP (1980–2083)

Citation: IMF Staff Country Reports 2010, 248; 10.5089/9781455206759.002.A006

  • The staff’s baseline fiscal scenario (hereafter ‘Staff Scenario’), based on the IMF’s staff macroeconomic forecast (see Table 1).

  • An alternative scenario (hereafter ‘Alternative Scenario’) based on the Congressional Budget Office’s (CBO) June 2010 Alternative Long-Term Scenario.

Table 1.

Macroeconomic Assumptions Underlying Budget Projections

article image

5. Both the Staff and Alternative scenarios are based on CBO’s concept of “current policies”, in line with the 2010 CBO Alternative Long-Term Scenario. Both scenarios incorporate the budgetary impact of the Final Healthcare Legislation until 2020 as documented in CBO (2010d). Post 2020, the scenarios make identical assumptions about mandatory spending on health care, namely that several policies enacted in the Final Healthcare Legislation that would restrain growth in spending would not continue in effect (see CBO, 2010e). Both the Staff and Alternative Scenarios incorporate the limit beginning in 2015 on Medicare spending on a per capita basis, to a fixed growth rate, initially set at a mix of general inflation in the economy and inflation in the health sector (in line with CBO, 2010d). However, the scenarios do not incorporate the upper limit on Medicare spending to be set by the IPAB permanently at per capita gross domestic product growth plus one percentage point starting in 2018. Likewise, both scenarios assume that the extra revenues envisaged under a full enactment of the Final Healthcare Legislation will not increase as a share of GDP after 2020. These assumptions have a potentially large effect on the size of the fiscal gap because the full implementation of such policies—according to OMB estimates—could reduce the fiscal gap by some 2 to 3 percentage points of GDP. However, they reflect a view, incorporated in CBO’s Alternative Long-Term Scenario, that changes to the current law are likely to occur or that some provisions of law may be difficult to maintain for a long period. Finally, both scenarios assume that healthcare spending remains stable in terms of per capita GDP after 2083.

6. The main differences between the Staff and the Alternative Scenarios are: (1) the Staff Scenario is based on lower growth and higher real interest rate assumptions between 2011–2015, reflecting the staff’s assumption of a permanent output loss after the crisis, and higher debt financing costs in the medium term; (2) the Staff Scenario assumes, in line with the Administration’s FY2011 budget, that the tax cuts enacted in the Economic Growth and Tax Relief Reconciliation Act of 2001 (EGTRRA), the Jobs and Growth Tax Relief Reconciliation Act of 2003 (JGTRRA) expire only for higher income households, and an extension of some of the tax relief provisions in the American Recovery and Reinvestment Act of 2009 (ARRA, Public Law 111-5) in line with the Administration’s FY2011 budget (the CBO alternative scenario assumes that the EGTRRA and JGTRRA are made permanent).

C. Results

7. The U.S. fiscal gap associated with today’s federal fiscal policy is huge for plausible discount rates (Table 2). Using the same discount rate (3 percent) used by the Trustees of the Social Security Administration (2009) in their own Social Security-specific fiscal gap analysis and by CBO (2010e), and the infinite horizon definition, the U.S. fiscal gap is about 14 percent of the present discounted value of U.S. GDP under the Staff’s Scenario. This implies that closing the fiscal gap requires a permanent annual fiscal adjustment equal to about 14 percent of U.S. GDP, that is to say that fiscal revenues and spending would need to change so that the primary balance predicted under that scenario improves by this amount every year into the indefinite future starting next year.2 Using the Alternative Scenario the fiscal gap increases to about 14½ percent of the present discounted value of GDP (owing to the assumption that tax cuts are made permanent).

Table 2.

U.S. Fiscal Imbalance in Terms of the Present Discounted Value of GDP

article image
Source: Authors calculations.Note: All calculations are obtained taking present values as of fiscal-year-end 2009, and interpreting the policies in the FY 2010 Federal Budget as “current policies”.

8. The fiscal gap under a finite horizon definition, or a larger discount factor, is smaller (but still sizeable) (Tables 2 and 3). Using a 6 percent discount factor reduces the gap to 7¾ to 8½ percent of GDP.3 Targeting a return to the 2008 federal debt-to-GDP ratio by 2083 under the Alternative Scenario, for example, implies a fiscal gap of about 8½ percent of GDP.4

Table 3.

Fiscal Imbalance in Terms of the Present Discounted Value of GDP, 3 Percent Discount Rate

article image

9. The main drivers of the fiscal gap are rising healthcare costs that under current law will boost mandatory spending to above 18 percent of GDP by 2050. 5 Since the federal government has historically collected about 18.4 percent of GDP in tax revenues, this means that mandatory programs may absorb all federal revenues sometime around 2050, or as early as 2026 when the cost of servicing the debt is added. As a result, future entitlement reforms will be necessary to restore fiscal sustainability.

10. The gap remains large even excluding the adverse fiscal effects from the crisis (Table 2). The crisis had a sizeable fiscal impact in deficit terms over 2008, 2009, and 2010. However, its impact is ‘modest when compared to the wave of future liabilities.6

11. The U.S. generational imbalance is also large. Applying “generational accounting” to U.S. data indicates that—under the Staff’s Scenario—unless currently living Americans pay more in net taxes or unless government spending on current generations is curtailed, future Americans will face net tax rates that are about 14½–16½ percentage points of the present discounted value of labor income higher than those facing current newborn Americans under our scenarios (See Table 4).7

Table 4.

Lifetime Net Taxes as a Share of Present Value of Labor Income Under Different Scenarios, 3% Discount Rate

article image

12. Implementing a fiscal adjustment equivalent to 2 ¾ percent of GDP by 2015 as suggested in the Staff Report reduces considerably the fiscal gap (Table 5). Were the adjustment to be followed by a permanent cap on Medicare spending, as mandated under the Final Healthcare Legislation to the Independent Payment Advisory Board and entailing the adoption of a rule that controls the excess growth in healthcare costs from Medicare, this would eliminate 40 percent of the fiscal gap, going a long way in eliminating the country’s fiscal problems (Table 5).

Table 5.

Impact on Fiscal Gap (as % of PDV of GDP) of Fiscal Adjustment by 2015 and of Cap on Medicare

article image

13. The fiscal adjustment would entail significant adjustments in taxes and/or transfers. Under the Staff’s Scenario, for example, the federal government can restore fiscal balance, conditional on the 2¾ percent of GDP fiscal adjustment by 2015 and the cap on Medicare spending by raising all taxes and cutting all transfer payments from 2015 onwards by 18 percent (Table 6). This would raise the U.S. tax revenue-to-GDP ratio to just below Germany’s, while still leaving it below that of many other advanced G-20 countries (Figure 3). A 5- or 10-year delay in the implementation of such residual fiscal adjustment would imply the need of ever larger additional increases in taxes/cut in transfers, equal to 19 and 21 percent, respectively (seeTable 6).

Figure 3.
Figure 3.

Total Revenues and Tax Revenues in Percent of GDP—Advanced G-20 Countries

Citation: IMF Staff Country Reports 2010, 248; 10.5089/9781455206759.002.A006

Sources: IMF, Government Finance Statistics, 2009; International Financial Statistics; and World Economic Outlook.1/ 2006 data for Japan and 2007 data for all other countries.
Table 6.

Additional Percent Increase in Taxes and/or Cut in Transfers Necessary to Close the Fiscal Gap if Adjustment Starts In1,3

article image

Applied to both current and future generations.

Including all other taxes like capital income, excise etc.

Under Staff Scenario, after 2¾ of GDP fiscal adjustment by 2015 and Medicare cap, to close fiscal gap computed using a 3 percent discount rate.

D. Conclusions

14. Sizeable fiscal actions would be needed to close the U.S. fiscal and generational imbalances. Under current policies, the United States federal debt is projected to grow rapidly due to a combination of large budget deficits before and during the crisis, as well as, over the medium term, demographic factors and healthcare inflation. As part of the medium term adjustment, the authorities would need to raise taxes and/or cut transfers substantially to avoid an undesirable escalation of the debt-to-GDP ratio. The longer the wait, the larger the necessary adjustment will be and the greater the burden on future generations.

Appendix 1. Definition of Fiscal and Generational Gaps

A. What is the Fiscal Gap?

The infinite-horizon fiscal gap measures, as a present value, a country’s excess of total expenditures—including those arising from its commitments to spend in the future—over available current and future resources. It is commonly defined as the current federal debt held by the public plus the present value in today’s dollars of all projected federal non-interest spending, minus all projected federal receipts. In symbols:

FGt=PVEtPVRtAt(1)

Where FGt is the fiscal gap at time t, PVEt is the present value of projected expenditures under current policies at the end of period t. PVRt stands for the present value of projected receipts under current policies, and At are assets in hand at the end of period t.

A non-zero fiscal gap implies that the federal government is violating its inter-temporal budget constraint, meaning that it will not be able to finance its expenditures at some point in the future. Independently of solvency considerations, as emphasized on April 27, 2010 by the Director of the Office of Management and Budget Peter Orszag in his testimony before the National Commission on Fiscal Responsibility and Reform, large fiscal gaps from persistent budget deficits lead to a crowding out of private capital. Reducing or eliminating fiscal gaps can thus lead to an increase in capital and an increase in potential growth, a point made, among others by Auerbach and Gale (2010), Reinhart and Rogoff (2009) and Kumar and Woo (2010).

B. What Are Generational Accounts?

Generational accounting—a concept originally developed by Laurence J. Kotlikoff, Alan J. Auerbach, and Jagadeesh Gokhale—answers the hypothetical question: if policy remained as it is for current generations for the rest of their lives, how much would they pay in net taxes and how much would future generations pay? A basic assumption is that there is no default and no free lunch-all net liabilities transferred forward must be paid for eventually. In this sense generational accounts differ from the fiscal gap, which is computed assuming no changes in current policies even if this implies a violation of the intertemporal budget constraint of the government.

Generational accounts indicate the net present value amount that current and future generations are projected to pay to the government now and in the future. The accounts can be used to assess the fiscal burden current generations are placing on future generations, and thus represent an alternative to using the federal budget deficit to gauge intergenerational policy.1 Generational accounts can also be used to calculate the policy changes required for achieving a generationally balanced and therefore sustainable fiscal policy—one that implies equal lifetime net tax rates on today’s newborns and future generations. For further discussion, the reader is referred to Auerbach, Gokhale, and Kotlikoff (1991) or Auerbach and Kotlikoff (1999).

The calculation of generational accounts starts from the government’s intertemporal budget constraint, which implies that the sum of future government consumption spending has to be equal to the sum of all future net taxes (taxes minus transfers all in present value terms) plus current government net wealth. This can be expanded to detail the amount of government consumption and revenues apportioned to current and future generations, where the apportioning is done by summing each generational account across all current generations on one side, and on all future generations, on the other side. Specifically, a generational account is the present value of the remaining lifetime net payments (taxes minus transfers) of the average individual of each generation. In our analysis we distinguish between males and females, and we assume that each individual lives for 100 years. So we have 100 generations (0 to 100) per each gender. Omitting for simplicity gender notation, and following the notation used by Auerbach and Oreopoulos (1999), the government intertemporal budget constraint expressed using generational accounts then becomes:

S=0DNt,ts+S=D+1Nt,ts(1+r)(st)=S=tGs(1+r)(st)Wtg(2)

Where Nt, t-s is the account of the generation born in year t-s, and the index s runs from age 0 to the maximum length of life (year D); Gs is government consumption in year s, Wtg denotes the government net wealth in year t—its assets minus its explicit debt; r is the pre-tax real interest rate. The first term on the left hand side of equation (2) sums together the generational accounts (i.e. the present value of the remaining lifetime net payments) of existing generations. The second term does the same for future generations, with s representing the number of years after year t that the generation is born.

Like more standard versions of the intertemporal budget constraint of the government, equation (2) suggests that intergenerational fiscal policy is a zero sum game: for a given present value of government consumption, lower taxes in present value terms on current generations imply a higher tax burden on future generations, in present value terms.

To compute the first and second term of equation (2) it is necessary to derive individual generational accounts, i.e. present values of lifetime net tax payments per each current and future generation. To do so, in turn, it is necessary to build a set of relative-age profiles for each sex (this is important because the average amount of any tax and transfer can vary greatly by sex as well as by age). Relative-age profiles by sex are derived using micro data from official survey. Below we list the data that we have used to build the profiles used in this analysis.

The profiles are basically distributions of the cumulative incidence of taxes and transfers on all individuals belonging to a particular age cohort. The profiles are “relative” because they are expressed relative to the incidence of taxes and transfers of a 40-year-old male, which acts as a numeraire to ensure profile comparability across age cohorts. The profiles are then transformed into per capita terms using demographic projections and used in conjunction with CBO’s long-term taxes and transfer projections to generate per capita lifetime net tax burdens by age and sex.

Since generational accounts reflect only taxes paid less transfers received, the accounts typically do not impute to particular generations the value of the government’s purchases of goods and services. Therefore, the accounts do not show the full net benefit or burden that any generation receives from government policy as a whole, although they can show a generation’s net benefit or burden from a particular policy change that affects only taxes and transfers. Thus generational accounting tells us which generations will pay for government spending, rather than telling us which generations will benefit from that spending. Another characteristic of generational accounting that should be understood at the outset is that, as its name suggests, it is an accounting exercise that does not incorporate induced behavioral effects or macroeconomics responses of policy changes.5

The generational gap is calculated by assuming that future generations (those born after the base year) pay, in the form of net taxes, all of the government’s bills left unpaid by current generations. This assumption ensures that the difference between generational accounts of the newborn generation and generational accounts of future generations reflects the policy adjustment required to satisfy the government’s intertemporal budget constraint.

To build the relative age/sex profiles of taxes and transfers we have used the following sources and methodologies:

1. Individual Income Taxes, FICA Taxes, Capital Income Taxes, Unemployment Compensation and Child Support, 2007

Source : Current Population Survey, March 2008 Supplement. Data was extracted using Data Ferrett, the “Federal Electronic Research and Review Extraction Tool” (dataferrett.census.gov). Average values by age and sex are provided by Data Ferrett.

2. Food Stamps and General Welfare, 2007

Source : Survey of Income and Program Participation, 2008. Data was extracted using Data Ferrett and average values by age and sex are the final output.

3. Excise Taxes, 2007

Sources : Alcohol and tobacco products use in 2007 by age category are from the U.S. Department of Health and Human Services, Office of Applied Studies, 2007 National Survey on Drug Use & Health: Detailed Tables. Population data comes from the Census Bureau, National Population Projections 1999-2100, middle series data.

Methodology : Department of Health and Human Services tables giving alcohol and tobacco products use per age groups are allocated by age according to the age group average. Consumption per thousand of male and female population at each age is calculated using total population from the Census Bureau to get per capita consumption profile for total population by sex. For age cohort “65 and over”, it was assumed that the consumption happens between 65 and 75 years of age (2/3 from age 65–69 and 1/3 from age 70–74), so that persons 75 and older do not smoke nor consume alcohol.

4. Social Security (OASDI), 2007

Sources : Average benefits and number of beneficiaries data comes from the U.S. Social Security Administration, Office of Retirement and Disability Policy, Annual Statistical Supplement, 2008 (2008 report reflects data for year ending December 2007). Population data comes from U.S. Census Bureau intercensal estimates for July 1, 2007. Census Bureau 5-year age group population counts estimated as 1-year age groups using Beers' Interpolation.

Methodology: OASDI tables giving average benefits and number of different types of beneficiaries by age-sex groups are made into single year series for each sex. All beneficiaries receiving OASDI benefits are added up and multiplied by average annual benefit to get aggregate benefits at each age and sex. Aggregate benefits at each age are divided by total population at each age to get per capita benefit profile for total population by sex.

Medicare and Medicaid, 2003

Sources: Medical Expenditure Panel Survey (MEPS), National Nursing Home Survey (NNHS), National Income and Product Accounts (NIPA), and National Health Expenditure (NHE).

Methodology : Age shapes are estimated using survey or administrative data (MEPS and HHHS - single age shape used for Medicare and Medicaid. Sex profiles are generated applying to the age-profiles the same male/female per capita ratio of the profiles used in Gokhale, Page and Sturrock (1999).

  • Auerbach, A. J., Gokhale, J. and L. J. Kotlikoff, 1991. Generational Accounting: A Meaningful Alternative to Deficit Accounting, NBER Chapters in Tax Policy and the Economy, Volume 5, pages 55-110 National Bureau of Economic Research, Inc.

    • Search Google Scholar
    • Export Citation
  • CBO (2009a), “An Analysis of the President's Budgetary Proposals for Fiscal Year 2010”, Congressional Budget Office, June 2009.

  • CBO (2009b), “Economic and Budget Issue Brief”, Congressional Budget Office, December 2009.

  • CBO (2010a), “The Budget and Economic Outlook: Fiscal Years 2010 to 2020”, Congressional Budget Office, January 2010.

  • CBO (2010b), “Budget and Economic Outlook: Historical Budget Data”, Congressional Budget Office, January 2010.

  • CBO (2010c), “An Analysis of the President's Budgetary Proposals for Fiscal Year 2011”, Congressional Budget Office, March 2010.

  • CBO (2010d), “H.R. 4872, Reconciliation Act of 2010 (Final Health Care Legislation)”, Congressional Budget Office, March 2010.

  • CBO (2010e), “The Long Term Budget Outlook”, Congressional Budget Office, June 2010.

  • Gokhale, J. B. R. Page and J. Sturrock (1999), “Generational Accounts for the United States: An Update”, in Generational Accounting Around the World, A. J. Auerbach, L. J. Kotlikoff and W. Leibfritz (eds).

    • Search Google Scholar
    • Export Citation
  • Gokhale, J. and K. Smetters (2003). “Fiscal and Generational Imbalances: New Budget Measures for New Budget Priorities”, AEI Press, Washington, D.C.

    • Search Google Scholar
    • Export Citation
  • IMF (2009), “The State of Public Finances Cross-Country Fiscal Monitor: November 2009”, IMF Staff Position Note, November 2009.

1

Prepared by Nicoletta Batini, Giovanni Callegari and Julia Guerreiro. Laurence Kotlikoff served as a consultant on this project.

2

Technically, the ratio of the fiscal gap to the present discounted value of GDP shows how much of the gap adjustment can be apportioned to each year from now to infinity to ensure intertemporal solvency.

3

A higher discount rate indicates a low propensity to save now for future consumption, a form of spending impatience, implying that the current government attaches more weight to the welfare of current generations relative to the welfare of future generations. In this sense, the discount rate is different than the cost of financing government borrowing that is embedded in our two fiscal scenarios. In general, the higher the discount rate, the lower the present value of future cash flows—hence a lower fiscal gap.

4

This number is close to the figure (8¾ percent of GDP) derived by CBO (2010) using a 75-year-horizon. The small difference is due to the fact that the CBO calculations employ a variable real interest rate, while Fund staff uses a constant rate throughout.

5

Population aging is also an important driver but far less than the increase in healthcare costs; the increase in healthcare costs is in turn due to various factors, the more important of which is technological change. This factor is summarized in CBO’s “excess growth component” of health-care costs growth (see CBO, 2010).

6

To assess the impact of the crisis, individual and capital income taxes are set at the pre-crisis GDP ratio level for 2009–11. Unemployment compensation and food stamps are set at the pre-crisis GDP ratio for 2009–14. Discretionary spending is reduced in order to exclude fiscal stimulus and above-the line financial sector support above the line. Relatedly, IMF Staff Position Note SPN 2009/13 calculates that the PV of the impact of the financial crisis in only 7½ percent of the PV of age-related fiscal costs.

7

As Table 4 indicates, projected transfers to current generations, particularly health care, have become so substantial as to drive the lifetime net tax payment of current generations negative.

1

However, from a theoretical perspective, the measured deficit need bear no relationship to the underlying intergenerational stance of fiscal policy.

United States: Selected Issues Paper
Author: International Monetary Fund