Revenues
Impose a 5 Percent Value-Added Tax
CBO periodically issues a compendium of policy options (called Options for Reducing the Deficit) covering a broad range of issues, as well as separate reports that include options for changing federal tax and spending policies in particular areas. This option appears in one of those publications. The options are derived from many sources and reflect a range of possibilities. For each option, CBO presents an estimate of its effects on the budget but makes no recommendations. Inclusion or exclusion of any particular option does not imply an endorsement or rejection by CBO.
Billions of Dollars | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 | 2026 | 2027 | 2028 | 2019- 2023 |
2019- 2028 |
|
Change in Revenues | |||||||||||||
Apply a 5 percent VAT to a broad base | 0 | 200 | 310 | 320 | 330 | 340 | 360 | 360 | 370 | 380 | 1,160 | 2,970 | |
Phase in a 5 percent VAT to apply to the same broad base | 0 | 40 | 100 | 170 | 240 | 320 | 350 | 360 | 370 | 380 | 550 | 2,330 | |
Apply a 5 percent VAT to a narrow base | 0 | 130 | 200 | 210 | 210 | 220 | 230 | 230 | 240 | 250 | 750 | 1,920 | |
Source: Staff of the Joint Committee on Taxation.
Background
A value-added tax (VAT) is a type of consumption tax that is levied on the incremental increase in value of a good or service at each stage of the supply chain, until the full tax is paid by the final consumer. Although the United States does not have a broad consumption-based tax, federal excise taxes are imposed on the purchase of several goods (gasoline, alcohol, and tobacco products, for example). In addition, most states impose sales taxes, but, unlike a VAT, those are levied on the total value of goods and services sold.
More than 140 countries—including all members of the Organisation for Economic Co-operation and Development (OECD) except for the United States—have adopted VATs. The tax bases and rate structures of VATs differ greatly among countries. Most European countries have implemented VATs with a narrow tax base that excludes certain categories of goods and services, such as food, education, and health care. In Australia and New Zealand, the VAT has a much broader tax base, with exclusions generally limited only to those goods and services for which it is difficult to determine a value. In 2017, the average national VAT rate for OECD countries was 19.2 percent, ranging from 5 percent in Canada to 27 percent in Hungary. All OECD countries that impose a VAT also collect revenues from taxes on individual and corporate income.
In 2017, the personal consumption expenditures of U.S. households amounted to about $13.3 trillion. Two-thirds of that amount was spent on services, and the remaining one-third was spent on goods. Spending on housing and health care services accounted for more than half of the total consumption of services. Spending on nondurable goods—particularly food and beverages sold for consumption off-premises and pharmaceutical and other medical products—accounted for about two-thirds of the total consumption of goods.
Option
This option consists of three alternatives. Each of the alternatives would become effective on January 1, 2020—a year later than most of the other revenue options presented in this volume—to provide the Internal Revenue Service time to set up and administer the tax.
The first alternative would apply a 5 percent VAT to a broad base that would include most goods and services. Certain goods and services would be excluded from the base because their value is difficult to measure. Those include financial services without explicit fees, existing housing services, primary and secondary education, and other services provided by government agencies and nonprofit organizations for a small fee or at no cost. (Existing housing services encompass the monetary rents paid by tenants and rents imputed to owners who reside in their own homes. Although existing housing services would be excluded under this alternative, a tax on the purchase of new residential housing would cover all future consumption of housing services.) Government-reimbursed expenditures for health care—primarily costs paid by Medicare and Medicaid—would also be excluded from the tax base. Accounting for those exclusions, the tax base would encompass approximately 66 percent of household consumption in 2020.
The second alternative would gradually introduce a 5 percent VAT to the same broad base. The VAT would be phased in over five years, starting at 1 percent in 2020 and increasing by 1 percentage point each year.
The third alternative would apply a 5 percent VAT to a narrower base and would, like the first alternative, become fully effective in 2020. In addition to those items excluded under the broad base, the narrow base would exclude certain goods and services that are considered necessary for subsistence or that provide broad social benefits—specifically, new residential housing, food purchased for home consumption, health care, and postsecondary education. Accounting for those exclusions, the tax base would encompass about 42 percent of household consumption in 2020.
Each alternative would employ the "credit-invoice method," which is the most common method used by other countries to administer a VAT. Under that method, at each point in the production process, the total value of a business's sales of a particular product or service would be taxed, and the business would claim a credit for the taxes paid on the purchased inputs—such as materials and equipment—used to make the product or provide the service.
Certain goods and services could be either "zero-rated" (that is, taxed at a rate of zero percent) or exempt from the VAT; in either case, no VAT would be levied on the purchased items. If a purchased item was zero-rated, the seller could still claim a credit for the VAT that had been paid on the production inputs. By contrast, if a purchased item was exempted, the seller would not be able to claim a credit for the VAT paid on the production inputs.
Under all of the alternatives, primary and secondary education and other noncommercial services provided by government or nonprofit organizations for a small fee or at no cost would be zero-rated, and financial services and existing housing services would be exempt from the VAT. In addition, under the third alternative, food purchased for home consumption, new housing services, health care, and postsecondary education would be zero-rated.
Effects on the Budget
The staff of the Joint Committee on Taxation (JCT) estimates that, if implemented, the first alternative would increase federal revenues by $3.0 trillion from 2020 through 2028. The second and third alternatives would raise revenues by $2.3 trillion and $1.9 trillion, respectively, over that same period, according to JCT's estimates. Revenues would be lower under the second alternative than under the first alternative because the 5 percent VAT would be gradually phased in. The revenue estimates for the phase-in period account for reactions to that phase-in by taxpayers—first, shifts in the consumption of some goods to earlier years, when the VAT rate would be lower, and second, higher tax compliance resulting from lower VAT rates. Revenues raised under the third alternative would be lower than under the first alternative because the VAT would apply to a smaller tax base.
The VAT, like an excise tax, would reduce taxable business and individual income. The resulting reduction in income and payroll tax receipts would partially offset the revenues raised by the VAT. The estimates for the option reflect that income and payroll tax offset.
The estimates for this option are uncertain because of uncertainty surrounding future economic activity and taxpayers' responses to a VAT. There is particular uncertainty surrounding taxpayers' compliance with the VAT, which would depend on how it was implemented and might differ from the responses considered here. In addition, there is uncertainty about how consumers would substitute taxed goods and services with those not subject to the tax.
Other Effects
One argument in favor of the option is that it would raise revenues without discouraging saving and investment by taxpayers. In any given period, income can be either consumed or saved. Through exclusions, deductions, and credits, the individual tax system provides incentives that encourage saving, but those types of preferences do not apply to all methods of saving, and they increase the complexity of the tax system. In contrast to a tax levied on income, a VAT applies only to the amount of income consumed and therefore would not discourage private saving or investment in the economy.
A drawback of the option is that it would require the federal government to establish a new system to monitor compliance and collect the tax. As with any new tax, implementing a VAT would impose additional administrative costs on the federal government and additional compliance costs on businesses. Research has shown that at least some countries that have implemented a VAT have devoted significant resources to addressing and enforcing compliance. Because such costs are typically more burdensome for smaller businesses, many countries exempt some small businesses from the VAT.
Another argument against implementing a VAT is that, as specified under all of the alternatives in this option, it would probably be regressive—that is, it would be more burdensome for individuals and families with fewer economic resources than it would be for those with more resources. Because lower-income families generally consume a greater share of their income than higher-income families do, the distributional effects of a VAT would depend on its impact on consumer prices. (Phasing in the VAT, as the second alternative of this option would do, would probably limit the increase in prices from 2020 through 2024.) The regressivity of a VAT, however, depends significantly on the measure of income used to rank families. For example, the burden of a VAT in relation to a measure of lifetime income—which would account for both life-cycle income patterns and temporary fluctuations in annual income—would be less regressive than the burden of a VAT in relation to a measure of annual income, which would not account for those patterns and anomalies.
There are ways to design a VAT—or implement complementary policies—that could ameliorate distributional concerns. One way to make a VAT less regressive would be to exclude certain basic goods and services from the tax base, just as the third alternative of this option does. A VAT with a narrower tax base would be less regressive because low-income individuals and families spend a larger share of their budgets on those basic goods and services than higher-income individuals and families do. (Alternatively, lower rates could be applied to such items.) Those preferences, however, generally would make the VAT more complex and would reduce the revenues it generated. In addition, a VAT with a narrow base would distort economic decisions to a greater degree than would a VAT with a broader base because people could substitute goods or services not subject to the VAT for those that were. Another way to offset the regressive impact of a VAT would be to add exemptions or refundable credits under the federal income tax for low-income individuals and families or to increase the size of existing exemptions or credits. That approach, however, would add to the complexity of the individual income tax and reduce individual income tax revenues, offsetting some of the revenue gains from a VAT.
An alternative approach for raising a broad-based consumption tax would be to impose a national retail sales tax. A national retail sales tax would initially be easier to implement than a VAT. However, it would require the federal government to coordinate tax collection and administration with state and local governments. In addition, there are more incentives to underreport retail sales taxes because they are collected only when the final user of the product makes a purchase, whereas a VAT is collected throughout the entire production chain and reported by both the buyer and the seller until the final stage.