It is no longer possible to avoid the need for a new revenue source.

The United States faced a large long-term fiscal imbalance even before the coronavirus pandemic, and the pandemic has made the problem worse. The projected debt buildup threatens long-run economic growth and will burden future generations. Looking ahead to the 2020s, it is no longer possible to avoid the need for a new revenue source. A value-added tax (VAT) is an essential component of a well-designed response to the fiscal challenge.

Any case for the VAT must acknowledge upfront the most significant challenge to its enactment: It would raise taxes on the middle class. Two other policy approaches, tax increases on those with the highest incomes and targeted cuts in entitlement benefits, may appear to allow the fiscal imbalance to be addressed without raising middle-class taxes. But although each option may play a role, their serious limitations prevent them from fully addressing the imbalance. Tax increases on the highest-income households would induce economic distortions that would threaten long-run growth. Entitlement cuts would be severely regressive, imposing much heavier burdens (relative to income) on lower-income households than on higher-income households. Moreover, tax increases on the highest incomes would have limited revenue potential, and significant entitlement cuts would face their own formidable political challenges.

The United States should therefore follow the lead of 160 other countries by adopting a national VAT, which is a consumption tax economically similar to a retail sales tax. The VAT would avoid most of the economic distortions of high-income tax increases, although it would not be as efficient as entitlement cuts. The VAT would also be far less regressive than entitlement cuts would be, although it would not match the progressivity of high-income tax increases. It can therefore be viewed as occupying a middle ground between those two alternatives. Although the VAT may not be anyone's first choice, it could be many people's second choice, as is often true for sensible policy outcomes.

Because budgetary arithmetic makes it likely that a VAT will eventually be adopted, the real question may simply be when to adopt it. We should seek to act sooner rather than later to limit the debt buildup and ease the fiscal burden on future generations.

Confronting the Fiscal Imbalance

In June 2019, long before the pandemic was upon us, the Congressional Budget Office (CBO) projected that the federal government's debt would rise from 78 percent of annual gross domestic product (GDP) in 2019 to 144 percent in 2049 absent legislative changes. Debt would rise over time as population aging and rising health care costs caused spending on Social Security and health care programs to grow more rapidly than revenue would grow.1

The fiscal imbalance has further widened during the recession accompanying the coronavirus pandemic. The recession has eroded tax revenue, and Congress has adopted large spending increases and tax cuts to provide income support and economic stimulus. In September 2020, CBO projected that the debt would grow to 195 percent of annual GDP in 2050.2 The congressional actions to respond to the pandemic were likely necessary, but their costs will have to be addressed.

The CBO has offered a forceful warning about the risks posed by the relentless rise in federal debt. Higher debt would reduce long-run economic growth by crowding out private investment, and it would push up interest payments to foreigners. It would also increase the risk of a crisis in which investors lost confidence in the US government and demanded higher interest rates to hold its debt. Furthermore, higher debt might lead to higher inflation expectations and could reduce Congress' and the president's flexibility to respond to unforeseen events.3

The steps needed to address the fiscal imbalance will become more onerous if action is delayed. CBO estimates that holding the 2050 debt to its 2019 share of GDP would require permanent tax increases or spending cuts equal to 3.6 percent of GDP if they took effect in 2025 but that the required magnitude would rise to 4.4 percent of GDP if action were delayed just to 2030 and to 5.9 percent if action were delayed to 2035.4

A VAT would have little appeal if there were a good way to address the fiscal imbalance without raising middle-class taxes. Unfortunately, the options that avoid middle-class tax increases have limitations that preclude them from providing a full solution.

Little progress can be made by cutting spending on programs other than Social Security and the major health care entitlement programs (Medicare, Medicaid, and the health insurance subsidies created by the Affordable Care Act). CBO projects that other noninterest spending, which includes national defense, other annually appropriated spending, and the smaller entitlement programs, will actually shrink from 9.0 percent of GDP in 2019 to 7.6 percent in 2050 under current law, leaving limited space for additional reductions.5

But what about raising taxes on high-income households? And what about cutting Social Security and the major health care entitlement programs?

In the past few years, Democrats have amplified their calls for high-income tax increases. Democratic presidential nominee Joe Biden has proposed raising ordinary income tax rates, capital gains taxes, and payroll taxes for high-income households and increasing the corporate income tax and the estate and gift tax. Other Democrats have proposed a new annual tax on wealth.

High-income tax increases place fiscal burdens on those with the greatest ability to pay and reduce economic inequality. However, such tax increases would induce economic distortions and threaten to impede the economy's long-run growth.

The individual income tax penalizes saving for future consumption by taxing the returns to such savings. Estate and gift taxes and annual wealth taxes have similar effects. The corporate income tax penalizes businesses that operate in corporate form rather than as partnerships and penalizes corporations that finance investment with equity rather than debt.

In the broader context of the global economy, America's corporate income tax also penalizes corporations that invest and book profits in the United States and operate with a US corporate charter. The income tax system also includes provisions that encourage increased investment in owner-occupied housing and increased consumption of medical care. All these distortions become larger as tax rates increase.

Even if the advantages of higher taxes on the wealthiest are viewed as outweighing their economic distortions, such tax increases have limited revenue potential. Well before the pandemic, commentators who supported high-income tax increases recognized that they would not raise enough revenue to fully address the fiscal imbalance and would have to be accompanied by other measures.6

So what about entitlement cuts? The CBO projects that spending on Social Security and the major health care entitlement programs will surge from 10.2 percent of GDP in 2019 to 15.5 percent in 2050 under current law.7 Restraining these programs' rapid growth could significantly narrow the fiscal imbalance. Also, most entitlement cuts would avoid the distortions that would be induced by high-income tax increases. Indeed, scaling back entitlement programs could reduce the distortions built into these programs, such as incentives for early retirement and increased consumption of medical care. Furthermore, reducing benefits for some future retirees might encourage some workers to save more. Entitlement cuts would therefore likely promote long-run economic growth.

At the same time, however, entitlement cuts would be severely regressive. Because only households with moderate incomes are eligible for Medicaid and health insurance subsidies, cuts in those programs would fall exclusively on those households. Social Security and Medicare cuts might burden all income levels, but lower-income households would bear much heavier burdens, relative to income, than higher-income households would.8 For example, a recent analysis found that across-the-board Social Security benefit cuts would impose burdens on households in the bottom 20 percent of the income distribution approximately 100 times larger, as a share of income, than the burdens on households in the top 1 percent.9 Deep entitlement cuts could also impede some households' access to health care and erode other social protections.

"When addressing the fiscal imbalance, time is not on our side."

Moreover, significant entitlement cuts would face formidable political obstacles. Democrats, who have long opposed Social Security and Medicare cuts, are now proposing to increase Social Security benefits and expand the federal role in paying for health care. Although Republicans have sometimes advocated entitlement cuts in the abstract, they have generally shied away from proposing significant specific cuts. Indeed, Republicans have sometimes accused Democrats of supporting entitlement cuts and claimed that the GOP would block such cuts.10 More recently, Donald Trump vowed not to cut Social Security, Medicare, and Medicaid. With each party pledging to prevent the other from cutting entitlements, it seems unlikely that entitlement cuts sufficient to address the fiscal imbalance will be adopted.

Therefore, although higher taxes on the wealthiest and entitlement cuts may play a role in rebalancing federal finances, their limitations prevent them from offering a full solution to the fiscal imbalance. It is necessary to find a strategy that avoids the serious economic distortions of high-income tax increases and the severe regressivity of entitlement cuts. Any such strategy will involve tax increases on the broad middle class. Although it would be possible to raise payroll taxes or increase income taxes on middle-class households, the VAT is preferable because it is less regressive than the payroll tax and avoids many of the income tax's distortions.11

Understanding the VAT

German businessman Wilhelm von Siemens and American economist Thomas S. Adams developed the VAT in the early 20th century. France adopted the first VAT in 1954. Many countries in Europe and Latin America adopted VATs in the 1960s and 1970s, and many others followed in subsequent decades. The European Union, the International Monetary Fund, and the World Bank have promoted the VAT.12

As of November 1, 2018, the VAT was in place in 168 countries and territories, including every Organisation for Economic Co-operation and Development (OECD) country other than the United States. 13 As of January 1, 2019, the 35 OECD VATs had tax rates ranging from 5 percent to 27 percent, with an average of 19.3 percent. 14 Most countries with VATs also have individual and corporate income taxes.

The VAT is a consumption tax similar to a retail sales tax but is collected in a different manner. A simple example can illustrate the VAT's relationship to the more familiar retail sales tax. Consider an economy with three businesses: a manufacturer, a wholesaler, and a retailer. The manufacturer makes no purchases from the other businesses and sells its output for $500 to a wholesaler. The wholesaler sells its output for $800 to a retailer, who sells a final product to consumers for $1,000. Of the $1,000 value of the final product, $500 is added at the manufacturing stage, $300 at the wholesale stage, and $200 at the retail stage.

A retail sales tax applies to the $1,000 of final sales to consumers, and the retailer remits the entire tax. At a 10 percent rate, the tax is $100.

Under a 10 percent credit-invoice VAT, the government collects the same $100, but a portion of the tax is collected at each stage of production. The manufacturer charges the wholesaler $550, the $500 before-tax price plus $50 tax, and remits the $50 to the government. The wholesaler charges the retailer $880, the $800 before-tax price plus $80 tax.

However, the wholesaler remits only $30 to the government because it receives credit for the $50 tax that it was charged on its purchase from the manufacturer (as shown on the purchase invoice). The retailer charges the consumer $1,100, the $1,000 before-tax price plus $100 tax. However, the retailer remits only $20 tax to the government because it receives credit for the $80 tax that it was charged by the wholesaler. In the end, the government collects $100 of revenue, with the manufacturer remitting $50, the wholesaler $30, and the retailer $20. At each stage, tax equal to 10 percent of value added is remitted. The multistage collection makes the tax easier to enforce than the sales tax, under which the entire $100 revenue is lost if the retailer conceals the sale.

To ensure that the VAT is visible to taxpayers, the full tax charged to the consumer should be listed as a separate item on customer receipts, as is normally done for state and local retail sales taxes. Under that arrangement, the VAT would likely be at least as transparent as employee payroll taxes and individual income tax withholding, which are displayed as line items on paycheck stubs. The VAT would be far more transparent than corporate income taxes and employer payroll taxes.

Any VAT legislation would need to address various design issues, including the transition and the treatment of the financial sector, housing, nonprofit institutions, and state and local governments. Although these issues are not discussed here, solutions are readily available from other countries' VAT experiences.

The VAT has strong efficiency advantages over tax increases on those with higher incomes. Like other consumption taxes, the VAT does not penalize saving. Those who consume today and those who save to consume tomorrow both pay the same tax rate on their consumption when it occurs. Unlike the corporate income tax, the VAT does not distinguish between partnerships and corporations or between debt and equity. The VAT, which applies to consumption of both domestic and imported goods, is also neutral in the global economy. It does not penalize investing or booking profits in the United States or operating with a US corporate charter.

Although a textbook VAT would be neutral among different consumer products, most countries' VATs offer zero or preferential rates for some products, such as food, housing, and medical care, and have special provisions for small businesses. An American VAT would likely include similar provisions. Such features would not detract much from the VAT's efficiency advantage, though, because the income tax system also offers preferences for small businesses, medical care, and owner-occupied housing.

The VAT is also far less regressive than entitlement cuts are. The Urban-Brookings Tax Policy Center (TPC) has estimated that a 5 percent broad-based VAT would reduce after-tax income by 3.9 percent for households in the bottom 20 percent of the income distribution, 3.6 percent for households in the middle 20 percent, and 2.5 percent for households in the top 1 percent.15 Burdens, relative to income, would be only modestly higher on lower-income households rather than vastly higher as under entitlement cuts.

It is easy to overlook the dramatic distributional difference between a VAT and entitlement cuts due to an inconsistent treatment of taxes and spending. It is well understood that a tax is regressive if it imposes larger burdens relative to income on lower-income households. However, some commentators mistakenly view a spending cut as regressive only if it imposes larger dollar burdens on lower-income households.16

Moreover, a VAT would undoubtedly be accompanied by rebates to offset the burden on low-income households. TPC estimated that a 7.7 percent VAT with rebates (which would raise the same revenue as a 5 percent VAT without rebates) would generally be progressive. It would reduce after-tax income by 0.6 percent for the bottom 20 percent, 2.9 percent for the middle 20 percent, and 3.6 percent for the top 1 percent. Furthermore, the VAT would be only one component of the federal tax system. The individual and corporate income taxes would continue to add progressivity to the overall tax system.

Like every tax and spending policy, the VAT has some drawbacks, of course. Although the VAT is more economically efficient than tax increases on higher-income households, it is not as efficient as entitlement cuts. Although the VAT is far less regressive than entitlement cuts, it cannot (even with rebates) match the strong progressivity of tax increases on the wealthy. The VAT in this sense occupies a middle ground between the two approaches to deficit reduction generally preferred by the right and left. As a result, the VAT is more likely to be adopted as part of a bipartisan compromise than by either party alone.

Harvard University economist Lawrence Summers famously said that conservatives oppose the VAT because they think it is a money machine that could fuel government growth and liberals oppose the VAT because they think it is regressive. That suggests, he joked, it could be adopted if liberals decided it was a money machine and conservatives decided it was regressive.17

In isolation, a VAT is clearly unpopular. On April 15, 2010, the Senate voted 85-13 to adopt a nonbinding resolution condemning the VAT. History suggests, however, that this resolution offers little guidance about the VAT's prospects. For example, the Senate voted 98-0 on July 14, 1981, to adopt a nonbinding resolution stating that Social Security benefits should not be subject to income taxes, but partial income taxation of benefits was included in the bipartisan Social Security reform package adopted in April 1983. Overwhelming support for symbolic resolutions that condemn measures in isolation need not preclude their subsequent inclusion in a bipartisan compromise.

Several VAT proposals have entered the policy debate. Bill Gale of the Brookings Institution has proposed a 10 percent VAT as part of a comprehensive deficit-reduction plan. He estimates that the tax would have a net budgetary yield of 2.6 percent of GDP, after providing generous rebates to offset the burden on lower-income households and adjusting certain entitlement benefits.18 Michael Graetz of Columbia Law School has proposed a 12.9 percent VAT as part of a deficit-neutral tax reform plan. 19 Sen. Ben Cardin (D-MD) introduced bills in 2014 and 2016 calling for a 10 percent VAT, and Rep. Jim Renacci (R-OH) introduced a bill in 2016 calling for a 7 percent VAT. All these proposals include rebates or expanded tax credits to offset the burden on lower-income households.

The Way Forward

In view of the budgetary situation, it is likely that a VAT will eventually be enacted to address the fiscal imbalance. The real issue may not be whether to adopt a VAT but when to do so.

Earlier adoption would allow government debt to stabilize at a lower level, which would promote long-run economic growth. Earlier adoption would also reduce the necessary tax rate, as there would be less debt to service. Quicker enactment would also allow the tax to be phased in gradually.

To be sure, early adoption of a VAT would diminish the pressure to address the fiscal imbalance through other means. With the VAT in place, there would be less pressure for tax increases on higher-income people and for entitlement cuts. Strong supporters of those options may want to delay or head off the VAT to ensure maximum scope for their preferred approaches. They should consider the limits of those options, however, and weigh any possible gains against the costs of delay.

When addressing the fiscal imbalance, time is not on our side. To strengthen the economy and ease the fiscal burden on future generations, we should adopt a VAT sooner rather than later. The president who takes office in 2021 should support it.


View Citations Hide Citations

1. Congressional Budget Office, "The 2019 Long-Term Budget Outlook," June 2019, 6, 23-24,

2. Congressional Budget Office, "The 2020 Long-Term Budget Outlook," September 2020,, 6.

3. Congressional Budget Office, "The 2020 Long-Term Budget Outlook," 9-15

4. Congressional Budget Office, "The 2020 Long-Term Budget Outlook," 17

5. Congressional Budget Office, "The 2020 Long-Term Budget Outlook," 28.

6. For example, see New York Times, "Once and Future Taxes," September 3, 2009,; Washington Post, "Who Pays Taxes," April 10, 2009, 04/09/AR2009040903915.html; and Paul Krugman, "Climate of Change," New York Times, February 27, 2009, html.

7. Congressional Budget Office, "The 2020 Long-Term Budget Outlook," 28.

8. Social Security and Medicare means tests that reduce benefits as income rose would be less regressive, but means tests would induce economic distortions by discouraging recipients from increasing their incomes through work and saving. They would also not achieve large budgetary reductions.

9. Sita N. Slavov and Alan D. Viard, "Taxes, Transfers, Progressivity, and Redistribution: Part 3," Tax Notes, November 26, 2018, 1109-21, 1115,

10. For example, Mitt Romney's 2012 presidential campaign attacked President Barack Obama for reducing Medicare provider reimbursements in the Affordable Care Act. Alan D. Viard, "Both Parties Defend Medicare, None the Taxpayer," RealClearMarkets, October 24, 2012, parties_defend_medicare_none_the_taxpayer_99951.html. The Trump campaign recently ran an ad claiming that Joe Biden has repeatedly tried to cut Social Security and Medicare. See Donald J. Trump, "Joe Biden Tried to Cut Social Security and Medicare for Decades!," YouTube, August 9, 2020,

11. In addition to taxing wages as they are consumed, the value-added tax (VAT) taxes consumption out of wealth held when the tax is introduced. The latter feature is a progressive component that is absent from the payroll tax, which applies only to wages.

12. Kathryn James, "Exploring the Origins and Global Rise of VAT," in The VAT Reader: What a Federal Consumption Tax Would Mean for America (Falls Church, VA: Tax Analysts, 2011), 15-22; and William G. Gale, "Raising Revenue with a Progressive ValueAdded Tax," in Tackling the Tax Code: Efficient and Equitable Ways to Raise Revenue, ed. Jay Shambaugh and Ryan Nunn (Washington, DC: Brookings Institution, 2020), 191-236, 201.

13. Organisation for Economic Co-operation and Development, Consumption Tax Trends 2018: VAT/GST and Excise Rates, Trends and Policy Issues, 193, https://www.oecd. org/tax/consumption-tax-trends-19990979.html.

14. Urban-Brookings Tax Policy Center, "OECD Value Added Tax, 1976-2019," August 5, 2019,

15. Eric Toder, Jim Nunns, and Joseph Rosenberg, "Implications of Different Bases for a VAT," Urban-Brookings Tax Policy Center, February 14, 2012, 20-22, 23, 25, These estimates overstate the VAT's regressivity because they ignore the progressive burden imposed on wealth held when the tax is introduced.

16. For discussion of the inconsistent treatment, see C. Eugene Steuerle, "Progressivity and Government Downsizing," Tax Notes, January 15, 1996, 319-20, http:// OpenDocument; and Slavov and Viard, "Taxes, Transfers, Progressivity, and Redistribution," 1116.

17. Jan M. Rosen, "Tax Watch: The Likely New Forms of Taxes," New York Times, December 19, 1988,

18. William G. Gale, Fiscal Therapy: Curing America's Debt Addiction and Investing in the Future (Oxford, UK: Oxford University Press, 2019), 241-53.

19. Michael Graetz presented his original plan in Michael Graetz, 100 Million Unnecessary Returns: A Simple, Fair, and Competitive Tax Plan for the United States (New Haven, CT: Yale University Press, 2008). For the 2013 modification, see James R. Nunns and Joseph Rosenberg, "Updated Tables for 'Using a VAT to Reduce the Income Tax,'" Urban-Brookings Tax Policy Center, November 20, 2013, https://www.taxpolicycenter. org/publications/updated-tables-using-vat-reform-income-tax.