The United States faces a long-run fiscal imbalance because of rapid projected growth in Social Security, Medicare, and Medicaid spending. The policy response to the imbalance will be shaped by four long-term fiscal realities. First, revenue will rise as a share of GDP. Second, entitlement spending will be reduced, relative to current policies. Third, the middle class, broadly defined, will bear much of the burden of addressing the fiscal imbalance. Fourth, consumption taxation is likely to become a significant part of the federal tax system, probably through the partial replacement of the income tax by a value added tax.
It is well known that the United States faces a long-term fiscal imbalance because of the rapid growth of Social Security, Medicare, and Medicaid spending. In this paper, I discuss four long-term fiscal realities that will shape policymakers' response to this imbalance, involving revenue increases, entitlement reductions, the fiscal burden on the middle class and a likely partial movement toward consumption taxation.
1. The United States Faces a Severe Long-Term Fiscal Imbalance
Numerous studies have documented the long-run imbalance between federal revenue and federal spending. Notably, the Congressional Budget Office (CBO)  projects that, under current policies, federal spending on the three major entitlement programs--Social Security, Medicare, and Medicaid--will rise from 8.4 percent of GDP in 2007 to 14.5 percent in 2030, 18.6 percent in 2050, and 25.7 percent in 2082. Other noninterest federal spending declines slightly (from 9.9 percent of GDP in 2007 to 7.6 percent in 2082), but the net result is a surge in total federal spending. In contrast, federal revenue will rise only modestly under current policies, inching up from 18.8 percent of GDP in 2007 to 20.9 percent in 2082. The imbalance between revenue and spending will cause the federal debt to grow explosively.
Over the long haul, Medicare and Medicaid account for nearly all of the entitlement growth; from 2007 to 2082, their share of GDP rises by 15.2 percentage points, but Social Security's share rises by only 2.1 percentage points. In the early part of this period, however, Social Security's relative importance is somewhat greater. From 2007 to 2030, Social Security accounts for 1.8 percentage points of the total 4.1-percentage-point rise in these programs' share of GDP.
The near-term expansion of Social Security partly reflects the retirement of the baby-boomer generation. Over the longer haul, all three programs are affected by the ongoing rise in longevity. Medicare and Medicaid are also affected by a relentless rise in the price of health care, relative to the price of other goods and services. As suggested by the rapid projected growth of the two medical programs, the increase in the relative price of health care is the single most important driver of the long-run entitlement growth.
The key implication of the fiscal imbalance is a reduction in the wealth of future generations. Future generations will face a heavier fiscal burden as they service the large stock of government debt. Unless private saving rises to offset the government borrowing, national saving will fall and future generations will inherit a smaller amount of net wealth. The distribution of the impact across the various future generations depends upon when the fiscal imbalance is addressed. If the policies described by CBO actually remain in place through 2082, generations alive in 2082 and those born thereafter will face a very large decline in wealth. If policies to narrow the fiscal imbalance are adopted before that time, the impact on those generations will be reduced, with part of the burden falling instead on earlier generations.
Although some discussions of the fiscal imbalance emphasize its potential effect on interest rates, any such effects are of secondary importance. In a closed economy, a restricted supply of national saving would be expected to drive up interest rates and reduce domestic investment. Interest rates need not rise, however, if the economy is sufficiently open to international capital flows, as an inflow of foreign savings can then offset the reduction in national saving and avert a decline in domestic investment. Even in that case, however, the decline in national saving still reduces the wealth of future domestic residents. Although the inflow of foreign savings props up the domestic capital stock, part of that capital stock belongs to foreign savers rather than domestic residents, implying that the wealth of the latter group still declines. Although access to foreign savings can ease the macroeconomic adjustment to a decline in national saving, it cannot change the fundamental economic fact that lower saving today leads to lower wealth
Other discussions emphasize disastrous consequences that may result if policymakers wait too long to address the fiscal imbalance: Foreign lenders may eventually be unwilling to buy additional Treasury securities or the government may eventually default on its debt, either explicitly or implicitly through inflation. These outcomes can, should, and probably will be avoided, however, by addressing the fiscal imbalance before such crises arise.
In this paper, I adopt a different emphasis, focusing on developments that are likely to occur over the upcoming decades as policymakers address the fiscal imbalance. In the following four sections, I describe four long-term fiscal realities pertaining to the trajectory of revenue and entitlement spending, the distribution of the fiscal burden, and the shape of the federal tax system.
2. Federal Revenue will Rise as a Share of GDP
The first long-term fiscal reality is that federal revenue will rise above its current share of GDP. Although the revenue share of GDP has fluctuated within a range of 17 to 21 percent in recent decades, the fiscal imbalance will necessitate a movement above that range.
The revenue increase is politically, rather than mathematically, necessary. It would be mathematically (and, for that matter, economically) possible to hold Social Security, Medicare, and Medicaid spending at their current 8.4 percent share of GDP. That, in turn, would make it possible to keep revenue at its current share of GDP. The CBO projections make clear, however, that holding these programs at a fixed share of GDP would require dramatic, and ever-deeper, reductions relative to the benefit levels implied by current policies. By 2082, these programs would be only one-third of the level implied by current policy.
The political feasibility of such a strategy therefore depends upon the public's willingness to support dramatic reductions in entitlement spending (relative to the levels promised by current law) as an alternative to tax increases. For good or ill, such willingness does not exist. Blinder and Krueger [2004, pp. 375-381] surveyed Americans about how to reduce the long-run Social Security deficit. Only 5 percent of the respondents favored relying mainly on benefit cuts, whereas 30 percent supported relying mainly on payroll tax increases. Another 34 percent called for a mix of the two measures.
As discussed in the next section, it will not be possible to rely solely on tax increases to address the fiscal imbalance; entitlement reductions will also be necessary. Nevertheless, the public attitudes found by Blinder and Krueger make clear that it will be possible to secure public support for entitlement reductions only if they are accompanied by tax increases.
3. Entitlement Spending Will be Reduced, Relative to Current Policies
Although federal revenue will rise as a share of GDP, as discussed above, it will not rise to an extent sufficient to finance the Social Security, Medicare, and Medicaid spending growth projected under current policies. Federal revenue would have to almost double as a share of GDP to fully finance the projected growth in these programs over the next 75 years. Such an increase in tax rates would be politically infeasible and economically calamitous. It will therefore be necessary to restrain entitlement growth, relative to current policies.
As noted above, long-run entitlement growth is driven by the rise in the relative price of health care and, to a lesser extent, the increase in longevity. The major entitlement programs will need to be reconfigured to respond in an economically sustainable way to these trends. It is conceptually straightforward, if politically difficult, to adjust the programs for rising longevity. Raising eligibility ages could forestall an increase in the length of time that recipients receive benefits from these programs. Alternatively, the level of benefits could be reduced to offset the longer period over which benefits are received. Diamond and Orszag [2004, pp. xxi-xxiii, 119-120] discuss ways to index Social Security benefits (and tax rates) to life expectancy.
Some analysts argue that the increase in the relative price of health care can and should be halted or dramatically slowed, minimizing the need for entitlement policy to respond to the increase. There is little reason, though, to think that such a halt is desirable or even possible. Such measures as using better information technology, curbing unnecessary treatments, promoting preventive care or limiting medical malpractice lawsuits may modestly reduce the level of costs and may well be desirable. But, such measures cannot change the underlying rate of the cost growth. The cost growth is a longstanding phenomenon driven by fundamental economic forces; from 1960 to the present, the Consumer Price Index (CPI) for medical care has risen 1.76 percent per year faster than the overall CPI. As Gruber [2007, pp. 466-468] and Rosen and Gayer [2008, pp. 200-203] note, the increase in the relative cost of health care has been associated with a dramatic increase in its quality. No economic theory dictates that relative prices or the division of consumer spending between different goods and services should remain constant over time; instead, prices and spending should respond to changes in technology and preferences.
Given that health-care costs will continue to rise, restraining Medicare and Medicaid growth will require that those programs finance a smaller share of national health-care spending, and that the private sector finance a greater share. This shift may ultimately require a broad rethinking of the role of the government in the financing of health care.
4. The Middle Class, Broadly Defined, Will Bear Much of the Fiscal Burden
Regardless of the precise mix of tax increases and entitlement cuts, much of the burden of closing the fiscal gap will fall on households that are commonly understood to be middle class. Only a limited amount of the burden can be imposed on households in the top 5 percent or so of the income distribution and very little can be imposed on those at the bottom. A large part of the burden will therefore fall on those in between.
The definition of "middle class" is subjective. The term could be defined as, say, the middle 60 percent of the income distribution, with the bottom 20 percent being low-income class and the top 20 percent being high-income class. In practice, though, a larger fraction of Americans consider themselves to be either middle or working class. For example, in a 2008 PSRA/Pew survey, 2 percent reported being upper class and 6 percent lower class, with 19 percent upper-middle class, 53 percent middle class and 19 percent lower-middle class. A 2007 Fox News/Opinion Dynamics survey found 8 percent to be upper class, 9 percent lower class, 50 percent middle class and 33 percent working class.
A definition similar to that of the popular understanding has taken hold in tax policy discussions. President Obama has pledged to raise taxes only on households with incomes above $200,000 ($250,000 for married couples); a mere 3 percent of all income tax returns filed in 2006 had adjusted gross income above $200,000. Households with incomes below this level are now viewed as part of the middle class for tax policy purposes.
To the extent that the fiscal imbalance is addressed by reducing entitlement spending, it is evident that the burden will fall on the middle class rather than the wealthy. In 2005, the top 5 percent received only 5 percent of federal cash transfer payments and the bottom 20 percent received 14 percent, with households in between receiving the remaining 81 percent. Entitlement reductions will fall primarily on the middle class.
To the extent that the fiscal imbalance is addressed by tax increases, things are a little more complicated. Because a small set of high-income households earns a large share of national income (a share that has risen in recent decades), it may initially seem that they could bear a large share of the fiscal burden. CBO  reports that households in the top 5 percent of the income distribution earned 32 percent of the nation's before-tax income in 2006. Even after paying 45 percent of the nation's federal taxes, this group retained more than 28 percent of the nation's after-tax income. Since these households had an average income of $564,200 before taxes and $400,400 after taxes, one might think that they could afford to pay more taxes.
In reality, though, only a limited amount of revenue can be raised from this group. To begin, note a subtle, but fundamental, limitation of raising taxes only at the top. A tax increase that applies only to incomes above, say, $250,000 does not apply to the entire income of households with incomes above $250,000; instead, the tax increase necessarily exempts the first $250,000 of each household's income. The average income of all households with incomes above a (high) specified level is usually about double that level; as a result, a tax increase on incomes above a given level applies to about half of the affected households' incomes. A tax increase on households in the top 5 percent would therefore apply not to the full 32 percent of national income that they earn, but to only about 16 percent of national income. Raising even 4 percent of national income in additional revenue from this group would therefore require raising their marginal tax rates by a staggering 25 percentage points above current levels, even if the households did not reduce their income in response to the rate increase. Spending increases of the magnitude featured in CBO's long-run projections cannot possibly be financed solely by taxing this small group.
The conclusion that the fiscal imbalance cannot be addressed solely by taxing households at the top of the income distribution is accepted by economists and others across the political spectrum (for example, see Krugman , The Washington Post , and the economists of varying political viewpoints quoted by O'Toole , Thorndike , and Alarkon and Swanson ). Inevitably, the middle class, broadly defined, will bear much of the fiscal burden.
5. Consumption Taxation is Likely to Become a Significant Part of the Federal Tax System
The rise in federal revenue will place additional strain on the income tax system, which will create pressure to add a consumption tax to the federal tax system. The most likely long-run outcome is the introduction of a value added tax (VAT) alongside the income tax.
The distinctive economic inefficiency of income taxation is that it penalizes saving. (Income taxation also penalizes work, but that is equally true for consumption taxation.) Consider two individuals, Patient and Impatient, each of whom earns $100 of wages today. Impatient wishes to consume only today, whereas Patient wishes to consume only at a distant future date. Savings can be invested to obtain a 100 percent rate of return between now and the future date.
With no taxes, Impatient consumes $100 today, but Patient saves $100, earns $100 interest and consumes $200 in the future. With a 20 percent income tax, Impatient pays $20 tax on his wages and consumes $80, which is 20 percent less than in the no-tax world. Patient pays $20 tax, and saves the remaining $80. She earns $80 interest, on which she must pay $16 tax. Along with the $80 principal, Patient consumes $144, which is 28 percent less than in the no-tax world. Under the income tax, Patient faces a higher tax burden than Impatient, 28 rather than 20 percent, solely because she saves.
Income taxation penalizes saving, and thereby tends to impede capital accumulation. In contrast, a consumption tax is neutral if the tax rate is constant over time. Consider a consumption tax with a 25 percent rate, so that the tax is 20 percent of the combined amount devoted to consumption and taxes. After earning $100 of wages, Impatient consumes $80 and pays $20 tax. Patient saves the entire $100 and owes no tax today. She earns $100 interest, thereby accumulating $200. At the future date, she consumes $160 and pays $40 tax. Both individuals face the same 20 percent tax burden, demonstrating that consumption taxation avoids the penalty on saving. The neutral treatment of saving generally makes consumption taxation economically superior to income taxation.
If the rise in federal revenue discussed above comes solely from higher income taxes, the penalty on saving will be enhanced, magnifying the economic costs of the income tax. One possible response is to try to mitigate the saving penalty by expanding tax-sheltered savings accounts and pension plans. This strategy is not likely to be effective. Such provisions add complexity to the tax system; more than a dozen types of tax-sheltered accounts already exist with different purposes (retirement, health, or education saving) and different complex rules on contributions and withdrawals. Moreover, large savers, who account for the bulk of private saving, often receive no marginal incentive from the accounts, because they either reach the maximum allowable contribution or are disqualified by income-based limitations.
At some point in the future, the need to raise additional revenue without magnifying the tax penalty on saving is likely to lead to the addition of a consumption tax to the federal income tax system. That consumption tax is likely to be some form of VAT. The VAT is similar to a retail sales tax, except that it is collected from firms at all stages of production (manufacturing, wholesaling, retailing, and so on). Each firm is taxed on the value it adds to the product, namely, its sales to consumers and other firms minus its purchases from other firms.
The most probable outcome is the introduction of a VAT alongside the income tax. This arrangement is common for industrialized democracies; all OECD countries other than the United States have both a VAT and an income tax. As Krugman  comments, "Whatever politicians may say now, there's probably a value-added tax in our future." To be sure, this policy change is not likely to occur in the near term.
The adoption of a VAT could forestall the expansion of the income tax that would otherwise occur and could even be used to finance a reduction of the income tax below its current levels. Graetz  and Burman  propose adopting a VAT while reducing the top income tax rate to 25 percent or less. The President's Advisory Panel [2005, pp. 191-205] considered, but rejected, a plan to adopt a VAT and reduce the top income tax rate to 15 percent.
By partially replacing an income tax that penalizes saving with a consumption tax that does not do so, the enactment of a VAT could yield a significantly more efficient tax system. Of course, the gains arise to the extent that the VAT actually serves as a substitute for income taxation rather than as a substitute for spending restraint. The gains would also be diminished if the VAT exempted numerous goods or services or had other inefficient features.
Although the partial replacement of income taxation by consumption taxation would offer substantial efficiency gains, larger efficiency gains would be available from a complete replacement. Complete replacement of the income tax by a conventional VAT would not be politically feasible, however, because such a replacement would shift too much of the tax burden downward to the middle class. Although the response to the fiscal imbalance will, as discussed above, necessarily feature heavier burdens on the middle class, it is unlikely to include a policy that increases middle class tax burdens while lowering burdens on the wealthy.
A complete-replacement consumption tax would therefore have to be more progressive than a conventional VAT. The most promising option is the Bradford X tax, which is a graduated-rate version of the "flat tax" advocated by Hall and Rabushka . The flat tax is a two-level VAT, in which part of the VAT is collected from firms and part from households. Firms are taxed at a flat rate on their VAT base minus their wage payments (business cash flow). Households pay tax on their wages, above an exemption amount, at the same flat rate. Although the exemption amount provides some progressivity, the flat tax is still likely to be politically infeasible. Under an X tax, the tax rate on firms is relatively high and households are taxed on their wages at a graduated rate schedule with the top tax rate equal to the tax rate on firms. By making the rate schedule more graduated, the X tax can be made as progressive as desired.
Because it avoids the distributional problems of the conventional VAT, the X tax could serve as a complete replacement of the individual and corporate income tax, eliminating the federal tax penalty on saving. The President's Advisory Panel developed, but did not recommend, a plan to replace the income tax system with an X tax that featured a 35 percent tax rate on firms (pp. 182-190). One of the two plans that the Panel did recommend would have replaced much of the income tax with a 30 percent X tax while retaining a 15 percent tax on capital income (pp. 151-182).
Despite the conceptual appeal of the X tax, its enactment seems unlikely, a conclusion confirmed by the failure of the Panel's recommendation to draw public or political interest. The X tax is unfamiliar to the public and will not readily be recognized as a consumption tax. (Indeed, although the flat tax has received vastly more attention than the X tax, many people continue to mistakenly think that it is a flat-rate income tax rather than a consumption tax.) Although consumption taxation is likely to become a significant part of the federal tax system, that outcome will probably occur through the enactment of a conventional VAT accompanied by a reduction in the income tax, not through the complete replacement of the income tax.
The long-run fiscal imbalance will play a key role in shaping tax and budget policy in the upcoming decades. The four long-term fiscal realities described above will drive many of the policy decisions. Policies are more likely to be sustainable and beneficial if they accord with these long-term fiscal realities. By this standard, many of the actions taken by President George W. Bush during his eight years in office and many of those now being taken or proposed by President Barack Obama fall short. Middle-class tax cuts have been adopted and more have been proposed, ignoring the need to ultimately raise taxes on the middle class. A new prescription drug benefit has been added to Medicare and initiatives to expand health coverage have been proposed, ignoring the need to ultimately restrain entitlement spending. Vague hopes of cutting health-care costs are being presented as substitutes for entitlement restraint.
Facing some of the fiscal realities outlined in this paper will be painful. The pain will be minimized, however, if they are faced sooner rather than later.
Alan D. Viard is a resident scholar at AEI.
Acknowledgment: The author thanks Jason L. Saving for helpful comments and Amy Roden for research assistance. The views expressed in this paper are solely his own and do not reflect the views of any other person or any institution.
Alarkon, Walter, and Swanson, Ian. 2009. Obama's Tax Reckoning, http://thehill.com/leading-the-news/obamastax-reckoning-2009-04-14.html.
Bankman, Joseph, and Weisbach, David A. 2006. "The Superiority of an Ideal Consumption Tax Over an Ideal Income Tax." Stanford Law Review, 58(5): 1413-56.
Blinder, Alan S., and Krueger, Alan B. 2004. "What Does the Public Know About Economic Policy, and How Does It Know It?" Brookings Papers on Economic Activity, 2004: 327-87.
Bradford, David F. 2004. The X Tax in the World Economy: Going Global with a Simple, Progressive Tax. American Enterprise Institute Press, Washington, DC.
Burman, Leonard E. 2008. "A Blueprint for Tax Reform and Health Reform." Virginia Tax Review, 28(2): 287-323.
Congressional Budget Office, 2007. The Long-Term Budget Outlook, http://www.cbo.gov/ftpdocs/88xx/doc8877/12-13-LTBO.pdf.
Congressional Budget Office, 2008. Historical Effective Tax Rates: 1979 to 2005: Supplement with Additional Data on Sources of Income and High-Income Households, http://www.cbo.gov/ftpdocs/98xx/doc9884/12-23-EffectiveTaxRates_Letter.pdf.
Congressional Budget Office, 2009. Historical Effective Tax Rates: 1979 to 2006, http://www.cbo.gov/ftpdocs/100xx/doc10068/effective_tax_rates_2006.pdf.
Diamond, Peter A., and Orszag, Peter R. 2004. Saving Social Security: A Balanced Approach, revised edn. Brookings Institution Press, Washington, DC.
Fox News/Opinion Dynamics, 2007. FOX News Poll: Feeling Nervous About the Economy, http://www.foxnews.com/projects/pdf/101207_economy_web.pdf.
Graetz, Michael J. 2008. 100 Million Unnecessary Returns: A Simple, Fair, and Competitive Tax Plan for the United States. Yale University Press, New Haven, CT.
Gruber, Jonathan. 2007. Public Finance and Public Policy, 2nd edn. Worth Publishers, New York, NY.
Hall, Robert E., and Rabushka, Alvin. 1995. The Flat Tax, 2nd edn. Hoover Institution Press, Stanford, CA.
Krugman, Paul. 2009. "Climate of Change." New York Times, (February 27): A23.
O'Toole, Chuck. 2009. "Obama's Tax Plan Falls Short, Say Panelists at Tax Analysts Forum." Tax Notes, 123(2):139-42.
Office of Management and Budget, 2009. A new Era of responsibility: Renewing America's promise, http://www.whitehouse.gov/omb/assets/fy2010_new_era/A_New_Era_of_Responsibility2.pdf.
Pew Research Center, 2008. Inside the Middle Class: Bad Times Hit the Good Life, Social and Demographic Trends Report, http://pewsocialtrends.org/assets/pdf/MC-Middle-class-report.pdf.
President's Advisory Panel on Federal Tax Reform, 2005. Simple, Fair, and Pro-Growth: Proposals to Fix America's Tax System, http://govinfo.library.unt.edu/taxreformpanel/final-report/index.html.
Rosen, Harvey S., and Gayer, Ted. 2008. Public Finance, 8th edn. McGraw-Hill/Irwin, New York, NY.
Saez, Emmanuel. 2001. "Using Elasticities to Derive Optimal Income Tax Rates." Review of Economic Studies, 68(2): 205-29.
Thorndike, Joseph J. 2009. "What You Can't See Might Hurt You." Tax Notes, 122(13): 1527-8.
Viard, Alan D. 2008. "Comment on Leonard E. Burman, 'A Blueprint for Tax Reform and Health Reform'." Virginia Tax Review, 28(2): 325-40.
Washington Post, 2009. "Who Pays Taxes: The Super-Rich Can't Plug the Budget Gap on Their Own," (editorial) (April 10): A16.
1. All numbers are taken from CBO's alternative fiscal scenario. The agency's extended baseline scenario shows a somewhat smaller imbalance, but relies on unrealistic assumptions, including the complete expiration of the Bush tax cuts and the unchecked spread of the alternative minimum tax.
2. Rosen and Gayer [2008, pp. 468-471] provide a good textbook discussion of the effects of government debt.
3. I discussed these realities, in much less detail, in Viard [2008, pp. 329-331].
4. For example, President Obama's fiscal 2010 budget outline projects significant reductions in health-care costs by improving efficiency and promoting preventive care, Office of Management and Budget [2009, pp. 9-10, 25-29].
5. For 2006, CBO  classifies a one-person household as being in the middle 60 percent if its income lies between $18,900 and $71,200, with corresponding income values twice as large for four-person households.
6. The poll results are taken from Pew Research Center [2008, p. 10] and Fox News/Opinion Dynamics [2007, p. 4].
7. The shares were computed by the author from the dollar values listed in CBO [2008, Table 5].
8. For mathematically inclined readers, this statement assumes that the upper tail of the income distribution follows a Pareto distribution with parameter 2. As noted by Saez [2001, p. 211] and others, this is a good approximation.
9. Of course, a full comparison of the two tax systems is complex. As Bankman and Weisbach  discuss at length, however, the superiority of consumption taxation holds under a relatively broad set of assumptions.
10. For a description of tax-preferred accounts, see President's Advisory Panel [2005, pp. 91-92].
11. For descriptions of the VAT, see President's Advisory Panel [2005, pp. 37-38], Rosen and Gayer [2008, pp. 487-488], and Gruber [2007, pp. 747-748].
12. For a description of the flat tax and the X tax, see President's Advisory Panel [2005, pp. 39-40]. Bradford (2004) provides a thorough description of the X tax and how it could be implemented in the international context.