The annual Budget and Economic Outlook released on Feb. 5 by the Congressional Budget Office (CBO) contained some good news. The deficit is coming down. The debt is relatively stable over the coming 10 years. Revenues are going up and federal health care costs were again revised downward.
On the surface, all this looks reassuring. Unfortunately, the surface is made of thin ice.
As CBO warns, “the projected path of the federal budget remains a significant concern for several reasons.”1 Among those reasons, in CBO’s view:
- The debt-to-GDP ratio remains “very high by historical standards.”
- Deficits and debt will be higher than projected if policymakers alter certain policies scheduled to take effect, including the automatic spending cuts that occur on March 1.
- Caps imposed on annual appropriations, which are assumed in the projections, will be very difficult to achieve in practice.
- The 10-year projection period does not fully reflect upward pressure building on the budget toward the end of that time.
Moreover, in a separate report, CBO laid out three illustrative deficit-reduction paths showing that much more needs to be done to lower the debt-to-GDP ratio over the coming decade. One such illustrative path shows that it would take another $2 trillion of spending cuts and/or tax increases to bring the projected debt-to-GDP ratio in 2023 of 77 percent down to 67 percent. And even this assumes that several questionable benchmarks in the current baseline will be met.
Challenges also loom for the economy. Slow growth is forecast again for this year (1.4 percent of inflation-adjusted “real” GDP) with unemployment remaining above 7.5 percent through 2014, which would be the longest such stretch in 70 years. The economy is projected to pick up to 3.4 percent real growth next year in CBO’s estimate. However, the long-term trend (2019-2023) is for annual growth at 2.2 percent, mainly due to changes in the labor force reflecting the same dynamic that will put upward pressure on the budget -- an aging population.
Thus, while policymakers can claim some credit for beginning to bring the nation’s finances under control, the debt is still on track to grow faster than the economy and the most difficult decisions -- those involving health care, Social Security and tax reform -- remain to be made. That task must be the focus of the coming budget debate.
II. A More Realistic, But More Alarming, Baseline
The main purpose of CBO’s annual Budget and Economic Outlook is to provide policymakers and the public with a baseline projection of the federal budget over the coming decade under current law. CBO also provides an economic forecast showing how fiscal policy interacts with economic trends.
Compared with recent years, the new CBO baseline is more realistic. For that reason, it is also more alarming. The fiscal cliff deal, which permanently extended most of the Bush-era tax cuts and indexed the Alternative Minimum Tax (AMT) to inflation, accounts for most of the change.
Prior baselines had assumed that these tax cuts would expire as provided by law, resulting in a major revenue windfall. It has been clear for some time that neither Democrats nor Republicans were prepared to let that happen, so the fiscal cliff deal essentially confirmed in law what had been an operating assumption. It was nevertheless an expensive change, adding $4.6 trillion to the deficit between 2013 and 2022.
III. Debt and Deficits
Current fiscal policy remains on an unsustainable path. The debt-to-GDP ratio is higher than it has been since 1950, and that was when it was coming down from the height of World War II borrowing. Today, by contrast, the projection is for ever-rising debt.
The new baseline shows the deficit declining from 5.3 percent of GDP this year to a low of 2.4 percent in 2015 before steadily drifting upward to 3.8 percent by 2023. Debt held by the public follows a similar track, going from 76.3 percent of GDP this year to a low of 73.1 percent in 2018, followed by a small but steady increase to 77.0 percent in 2023.
It is important to note that the growth of debt is not caused primarily by war and recession, although those have certainly been a factor in recent years. Instead, the projected growth is largely caused by a structural mismatch between spending promises and current-law revenue that will continue to grow even when the economy fully recovers and war spending subsides.
Two points are particularly significant about the baseline. One is that the debt has settled in at a much higher level than normal. From an average debt level of 39 percent of GDP in the years 1973-2012, we can now expect debt exceeding 70 percent of GDP throughout the next 10 years. Higher debt does not necessarily portend economic calamity, particularly if it is stable. But as CBO points out, “Debt that is high by historical standards and heading higher will have significant consequences for the budget and the economy.”2
These consequences include higher costs of servicing the debt, lower national saving and income than would otherwise be the case, lack of flexibility to respond to new challenges, and eventually the possible loss of investor confidence, leading to a fiscal crisis.
The other significant point about the current baseline is even more problematic: Despite various deficit-reduction efforts that have restrained spending and raised revenues, the debt has not been stabilized as a share of the economy.
That is why the first task of policymakers must be to stabilize the growth of debt so that it is no longer growing faster than the economy. But even if this task is successful, the debt-to-GDP ratio will still be almost double what it has averaged in recent decades.
As noted above, this could have negative consequences for the budget and the economy. So the second task of policymakers must be to ensure that a fiscal plan includes policies that will put the debt-to-GDP ratio on a responsible downward path over time. Implementation must give due consideration to the still fragile economic recovery. Large immediate cuts, such as those scheduled to occur on March 1, as a result of the 2011 Budget Control Act (BCA), would not be wise policy.
This, however, should not be used as an excuse to avoid hard choices on a phased-in fiscal stability plan. As The Concord Coalition and others have often pointed out, it is both desirable and possible to combine near-term economic recovery policies with longer-term deficit reduction. We need both and we don’t have to sacrifice one to achieve the other.
Enacting such a plan would not only improve fiscal flexibility; it would contribute to a growing economy. For example, in CBO’s illustrative deficit-reduction paths, lowering the deficit by $2 trillion over the coming 10 years would improve the Gross National Product (GNP) by nearly one percent by 2023.3
IV. CBO’s Alternative Scenario
The baseline is merely a starting point based on current law. Some of these laws, however, are at odds with recent congressional actions and may prove to be questionable assumptions. For that reason, CBO has prepared an alternative scenario under which some assumptions in the baseline are reversed.
The main assumptions in this alternative are that the sequester from the BCA does not go into effect, that Medicare physician payment cuts (under the Sustainable Growth Rate, or SGR) are postponed -- as they have been for the last several years -- and that various tax breaks (“the extenders”) often extended on an annual basis are extended throughout the projection period. Finally, CBO includes a calculation for the increased debt service (interest payments) that these policies would cause by increasing the deficit. No changes are made to CBO's economic assumptions.
Under this alternative and more realistic scenario, CBO estimates that revenues from 2014 to 2023 would average 18.5 percent of GDP compared to 18.9 percent in the baseline. Outlays would average 22.9 percent compared to 22.1 percent in the baseline.
While these differences may appear minor in any particular year, they add up to a considerable difference over 10 years. Deficits total $9.5 trillion in 2014-2023 in the alternative scenario compared to $6.9 trillion in the baseline. By 2023, debt would reach 87 percent of GDP in the alternative fiscal scenario compared to 77 percent in the baseline.
The alternative scenario closely resembles one of CBO’s illustrative paths in which $2.3 trillion in additional deficits are added to the current-law baseline. The short-term effect in that scenario would be positive for the economy by allowing higher deficits, but by 2023 the cumulative effect of those deficits would reduce GNP by roughly a full percentage point. By implication, this alternative baseline also roughly doubles the amount of deficit reduction that would be needed to put the debt-to-GDP ratio on a clear downward path.
So, to be clear, if policymakers plan to use the official baseline as the starting point for “deficit reduction” measures, they must be willing to:
- Let the roughly $1 trillion of spending cuts scheduled to begin on March 1 take effect, or find equivalent offsets
- Allow Medicare’s doctor reimbursement rate to be cut by roughly 25 percent, or find equivalent offsets, and
- Allow existing tax breaks with a cost of about $1 trillion over 10 years to expire on schedule, or find equivalent offsets.
Any offsets used as substitutes for the above policies would not be available to reduce the baseline deficit. Thus, it may take roughly $2 trillion of offsets simply to get to the starting gate.
The question for policymakers is not an abstract matter of “realistic baselines” but whether they are prepared to do what is necessary to validate the current-law baseline. If not, the alternative scenario gives a good indication of how much deeper the hole is that they need to dig out of, and how much more negative the consequences will be if they fail to do so.
V. The Big Picture on Key Budget Trends
In Fiscal Year 2012, which ended on Sept. 30, the federal government spent $3.5 trillion (22.8 percent of GDP) and took in $2.5 trillion (15.8 percent of GDP), leaving a deficit of just over $1 trillion (7.0 percent of GDP). It was the fourth straight year in which the deficit exceeded $1 trillion.
The situation is projected to improve considerably this year and next, with spending relatively flat (and falling as a share of the economy) while revenues go up. By 2015, spending is projected to decline to 21.6 percent of GDP while revenues hit 19.1 percent of GDP, leading to the smallest deficit of the 10-year projection.
Problems are again apparent, however, in the second five years (2018-2023), when revenues remain relatively flat and spending begins to creep back up. By 2022 and 2023, the deficit is up to 3.8 percent of GDP. The main cost drivers are federal health care programs, Social Security and interest on the debt.
Spending is categorized as either “discretionary,” meaning that is controlled through the annual appropriations process, or “mandatory,” meaning that it runs on autopilot. In 2012, the largest discretionary program was defense ($670 billion). The largest mandatory program was Social Security ($768 billion).
All mandatory spending plus interest on the debt comprise roughly two-thirds of the budget, leaving only one-third to be divided up through the regular budget process. In other words, Congress does not exercise annual control over most of what it spends. As the largest mandatory programs, such as Social Security, Medicare and Medicaid, grow with the aging population, this squeeze on the rest of the budget will become even tighter.
Even today, mandatory spending plus interest on the debt consumes almost all federal revenues. Virtually the entire discretionary side of the budget, including defense, is paid for with borrowed funds.
Economists look at the budget relative to the size of the economy (GDP) because it gives a better sense of what is affordable, rather than looking at dollar figures alone. Over the past 40 years, government spending has averaged about 21 percent of GDP and revenues have tended to equal about 18 percent of GDP. That means the government has tended to run deficits of about 3 percent of GDP.
If the economy is also growing at 3 percent (adjusted for inflation), deficits at this level would be sustainable. But we are a long way from getting the deficit back down to 3 percent of GDP. Moreover, the historic spending pattern is likely to change as Social Security, Medicare and Medicaid grow more expensive with population aging -- even if we assume a full economic recovery. If revenues don’t keep pace, or if other spending is not cut substantially, our deficits will routinely exceed 3 percent of GDP. That means the debt will continue to grow faster than the economy and will eventually become unsustainable.
The most direct and noticeable consequence of the growing debt on the budget is the cost of servicing that debt. We currently spend more than $200 billion a year on interest. However, under current policies that figure is projected to swell to $857 billion within 10 years - more than doubling as a share of the economy. To put this number in context, $857 billion would be more than projected defense spending by the end of the decade or, alternatively, more than all domestic appropriations combined.
Keep in mind also that this only assumes a gradual increase in interest rates from their current rock-bottom levels -- 2 percent on 10-year notes -- back to 5.2 percent by 2018. If we experience a spike in interest rates due to uncertainty about the fiscal outlook or for some other reason, interest costs would go even higher. In other words, we are taking on a massive amount of new debt at a very low cost today but, as with “teaser rates” on consumer loans, the temptation to load up on debt now could come back to bite us when we have to roll over that debt at more traditional interest rates.
VI. Discretionary Spending
For discretionary spending, CBO projects that outlays will decrease from 7.6 percent of GDP in 2013 to 5.5 percent of GDP by 2023. This would represent the lowest level of discretionary spending in over 50 years.
Taken at face value, CBO’s projections appear to show historic progress in cutting discretionary spending. However, it is important to bear in mind that CBO simply reflects current law and that may eventually prove to be politically unrealistic.
For example, under CBO’s baseline assumptions, most appropriations between 2014 and 2021 are assumed to be constrained by the spending caps and automatic spending cuts required by the BCA. Since policymakers are currently thinking about how to respond to the expiring continuing resolution, which provides appropriations, and the scheduled sequester, there is considerable uncertainty involved in these assumptions.
As the budget office points out in its report, deficits “might be larger than in CBO’s baseline projections because holding discretionary spending within the limits required under current law might be difficult.”4
To illustrate the point, if CBO were to assume that appropriations increased at the rate of inflation (a rate of growth itself lower than the historic average) instead of being governed by the BCA provisions, the increase in the deficit would be about $1.5 trillion over 10 years.
Adhering to the spending caps in the BCA would represent a fiscally responsible achievement. Unfortunately, however, no bipartisan consensus has been reached on the specific programs that would need to be cut to meet the targets. If policymakers wish to continue to take credit for the BCA savings, they should clearly explain which programs are on the chopping block. They will also have to resist the temptation to get around the caps by declaring dubious “emergencies,” changing programs from discretionary to mandatory or by using tax credits.
VII. Health Care Costs
One surprising bit of good news in the CBO report is the trend of federal health care expenditures. For the third year in a row, national health expenditures (NHE) grew at a rate substantially slower than normal. In 2011, the latest year for which full data is available, NHE grew by 3.9 percent -- the same growth rate as the economy. Furthermore, NHE has remained constant as a share of the economy since 2009. This represents a dramatic shift, since NHE tends to grow more quickly than the economy, and grew almost twice as fast in 2007 -- the last year prior to the recession.
This slowdown has had a large effect on spending in federal health care programs. The CBO report notes that 2012 Medicare and Medicaid spending was about 5 percent below the amount projected by CBO in March 2010. This slowdown has also led CBO to make a series of rather substantial technical alterations (changes to projections based on modeling changes, not legislative action or economic conditions) to their federal health care projections.
These technical revisions, made in different reports since March 2010, amount to $200 billion in lower spending on Medicare and Medicaid for the year 2020 alone -- totaling 15 percent less in spending on the two programs. Remarkably, the magnitude of these re-estimates exceeds the legislative changes to federal health care spending which are projected to increase in 2020 due to the Affordable Care Act’s (ACA) expansion of health benefits and other health spending by $13 billion.5
Just as remarkable is that if one only looks at the two CBO updates over the last six months, projected 10-year Medicare spending has been revised downward by $306 billion. Projected Medicaid spending has been revised downward by $273 billion (not counting revised estimates of lower Medicaid enrollment due to the Supreme Court’s ruling on Medicaid expansion in the ACA).
Technical re-estimates are an indispensable part of CBO projections and help those concerned with the federal budget understand how real-world changes in the economy and the private sector relate to shifts in federal spending. However, as the CBO points out, not much is known about the causes of the slowdown in health care costs. Of particular concern is the degree to which the recession is the explanatory factor behind the slowdown -- which would make the slowdown only a temporary break in the oppressive march of health care inflation. If that proves to be the case, these budgetary projections could easily be revised back upwards.
On the other hand, the slowdown might be a more lasting phenomenon representing some fundamental changes in the health care sector. In that case, this budget news might represent the leading edge of a permanently improved budgetary outlook.
Yet even under that “better case scenario,” federal outlays for health care will soon be greater than outlays for Social Security -- placing it at the top of all federal budgetary commitments. Over the 10-year budget window, health care spending will experience growth of 1.2 percent of GDP, second fastest only to net interest.
This growth is fueled by an aging population, with 18 million new beneficiaries signing up for Medicare over that period. Additionally, federal health spending will jump in 2014 as the major insurance expansion from the ACA kicks in. Importantly, this subscriber growth is already “baked into the cake,” regardless of how long the slowdown in health care inflation lasts.
That is why, even though the slowdown is undoubtedly good news, those concerned with the nation’s fiscal health cannot just “declare victory” and cease their efforts. Reformers must strongly emphasize the need to adjust federal health benefits and revenues to account for the growing mismatch in the ratio of taxpayers to beneficiaries. On top of that, the difficult work of health care cost control and bringing efficiencies into our health care system must continue in order to, at the very least, lock in whatever progress has been made to this point.
After several years of unusually depressed revenues (below 16 percent of GDP from 2009 to 2012), the new CBO baseline shows revenues soon reaching and even exceeding traditional levels. In the new baseline, that 18 percent of GDP level will be hit in 2014 and from 2015 on, revenues are projected to hover around 19 percent of GDP.
While the higher trend is sometimes portrayed as the result of "tax increases" in the fiscal cliff deal, the net effect of that agreement was to cut taxes relative to what they were scheduled to be. In the August 2012 baseline, which assumed that all of the tax cuts would actually expire on schedule, revenues were projected to be above 21 percent of GDP by 2020.
Nevertheless, the fact that revenue projections have drifted higher than the 40-year average has led some to conclude that further tax increases should be off the table. It should be noted, however, that in the four years of budget surpluses (1998-2001) revenues averaged nearly 20 percent of GDP.
In any event, the revenue level cannot really be decided without also considering what happens with spending. The two cannot be treated as separate propositions having no relationship to each other. If revenues fall too far below spending for too long, the resulting deficits will eventually cancel out whatever positive economic effect there may be from low taxes. In the final analysis, revenues must be sufficient to pay for the cost of government. Debt is not a painless alternative to taxation.
The fact that the revenue baseline is now higher than the 40-year average should not, therefore, be seen as a barrier for further policy changes that would raise new revenues. Policymakers will need to weigh the projected cost increases of federal health care programs and Social Security, which over the next 25 years is mostly driven on autopilot by demographics.
More beneficiaries mean higher spending. It is simple arithmetic. The combination of increased beneficiaries and rising health care costs will have a profound impact on federal spending through the major entitlement programs, pushing it up by five to six percent of GDP by the 2030s.
Keep in mind that the entire defense budget today is about 4.3 percent of GDP. So in the future we are on track to add an annual amount of spending that is the equivalent of more than doubling the defense budget. Or, to look on the tax side, individual income taxes would need to go up by about 75 percent to cover the added cost (today’s individual income taxes equal 7.3 percent of GDP).
And with the non-elderly population growing much more slowly, there will be fewer workers paying into Social Security and Medicare relative to the number of beneficiaries. Already, the respective Social Security and Medicare Part A payroll taxes do not bring in enough money to cover the benefits paid. This gap is projected to grow wider as the huge Baby Boom generation moves steadily into its retirement years, putting a strain on general revenues, which must cover the shortfall.
So given the structural factors at work, the 40-year averages for spending and taxes cannot be taken as relevant benchmarks for the future.
Moreover, some policies traditionally defined as “tax increases” are really “spending cuts.” The current tax code is riddled with "tax expenditures" -- exemptions, deductions, credits, exclusions and preferential rates that function much like entitlement spending. These tax expenditures amount to about $1 trillion annually, more than the entire discretionary federal budget.
As former Treasury Secretary Larry Summers has observed, "There are long-standing privileges in the tax code that perhaps should be thought of as misguided entitlements and … reform of entitlements should also extend to the tax entitlements that benefit many of those who are best off. If we take that approach and we recognize that the idea of expenditure, like the idea of entitlement, is a notion that applies both to what has traditionally been the spending side of the budget and to what has traditionally been the tax side of the budget, I believe we can move forward to make the necessary difficult choices."6
Similarly, Martin Feldstein, who headed the Council of Economic Advisors in the Reagan administration, has said that "much of the spending done by the federal government is done through the tax code, not by writing checks to individuals and companies, but by allowing deductions or credits or exclusions. ... So it ought to be possible for Republicans and Democrats, once we get past the election and the rhetorical barriers that an election imposes … to raise revenue by reducing government spending in the form of tax expenditures."7
The substantive case for eliminating or scaling back tax expenditures is strong. They complicate the tax code, distort economic choices and drain needed revenues.
Moreover, because tax expenditures tend to benefit upper-income households more than middle- or lower-income households, reforms can result in a more progressive system, or at least one that is as progressive as the current system.
Reducing these kinds of tax expenditures actually reduces the size and scope of government in our economy. Thus, it's a base-broadening, revenue-raising, deficit-reducing, yet government-shrinking proposal. It is therefore consistent with the fiscal policy goals of both Democrats and Republicans.
The extent to which large revenue gains can be made by tax expenditure reform is essentially a matter of political will. The largest tax expenditures are not mere “loopholes,” as the public may understand that term. The largest tax expenditures include the exclusion from income of employer-provided health insurance ($132 billion), the home mortgage interest deduction ($70 billion), the Earned Income Tax Credit ($53 billion), charitable contribution deductions ($31 billion) and the Child Credit ($57 billion).8
IX. Beyond the 10-Year Budget Window
The 10-year budget window cuts off at a time when baseline projections, even if achieved, leave the deficit and debt on an upward path due to structural pressures. Looking ahead, the main drivers of federal spending come from the three largest entitlement programs and interest costs, all of which run on autopilot. Interest costs are determined by outstanding debt and interest rates. The entitlement programs are based on formulas written into law. They only change if Congress changes the law.
The trend lines are clear. If left on autopilot, entitlement spending will continue to outpace the growth in the economy and the cost of discretionary programs, including defense, which are projected to steadily shrink as a percentage of GDP. As a practical matter, Congress will have less and less control over the budget as entitlement spending grows in cost. It is the stealth spending spree of automatic entitlement increases that poses the biggest challenge to the fiscal outlook.
Thus the main question of fiscal sustainability is, how do we accommodate the projected explosion of entitlement spending? The question cannot be avoided. That is the real “fiscal cliff.”
The problem is so large that it is unrealistic to think that we can deal with it all by just cutting other spending or by just raising taxes. We must put everything on the table and arrive at a package of reforms that is big enough to stabilize the debt-to-GDP ratio. And we must act soon to prevent interest costs from becoming the biggest “entitlement” of them all.
But even this would not be sufficient unless the policies enacted to achieve that level of deficit reduction are capable of producing growing savings over time. Both size and composition matter.
So too, does timing. Any plan must be phased in so as not to do harm to the fragile economic recovery. As the CBO shows, enacting a such a plan now, with verifiable and growing savings, will insure both room for the economic recovery to continue and prevent the economic harm that would come from larger deficits towards the end of the budget window and beyond.
Perhaps the scariest thing about the new CBO report is that the recent pitched political battles over spending cuts and tax increases have done so little to alter the budget’s basic unsustainable course. Despite tight discretionary spending caps, an unexplained--but welcomed--slowdown in health care costs, and rising revenues, the debt will continue to grow faster than the economy after just a brief pause. That story will not change until political leaders do what is necessary to replace crisis-management budgeting with a strategic framework that deals with the real problems facing the budget and the economy.
1 Congressional Budget Office, The Budget and Economic Outlook: Fiscal Years 2013 to 2023 (February 2013), p. 7.
2 Congressional Budget Office, The Budget and Economic Outlook: Fiscal Years 2013 to 2023 (February 2013), p. 25.
3 Congressional Budget Office, Macroeconomic Effects of Alternative Budgetary Paths (February 2013), p. 8.
4 Congressional Budget Office, The Budget and Economic Outlook: Fiscal Years 2013 to 2023 (February 2013), p. 8.
5 From 2010 to the present, legislative action--most notably as a result of the Affordable Care Act--and an updated economic forecast boosted the estimates for outlays for Medicare and Medicaid by $72 billion and $115 billion for the new subsidies created for people to purchase health insurance under the Affordable Care Act. Congressional Budget Office, The Budget and Economic Outlook: Fiscal Years 2013 to 2023 (February 2013), pp. 56-57.
6 Lawrence Summers, The Challenge of Pro-Growth Tax Reform (addressing a hearing of the Strengthening of America -- Our Children’s Future initiative, New York, N.Y., September 27, 2012),http://concordcoalition.org/files/121002_csis_fiscal_forum_iii_transcript_0.pdf
7 Martin Feldstein, The Challenge of Pro-Growth Tax Reform (addressing a hearing of the Strengthening of America -- Our Children’s Future initiative, New York, N.Y., September 27, 2012),http://concordcoalition.org/files/121002_csis_fiscal_forum_iii_transcript_0.pdf
8 Joint Committee on Taxation, Estimate of Federal Tax Expenditures for Fiscal Years 2012-2017, JCS-1-13, February 1, 2013.