Abstract
This Background Paper on the United States examines the effect of fiscal deficit reduction in the context of the IMF’s multicountry simulation model, on the current account and the real exchange rate. The simulations suggest that, other things being equal, fiscal consolidation will tend to cause the real exchange rate to depreciate in the short term. The paper also estimates a long-term relationship between the real effective exchange rate for the U.S. dollar and a number of variables.
VII. Tax Overhaul Proposals: Replacing Income Tax with Consumption Taxation 1/
The current session of the U.S. Congress has brought forward a number of proposals to overhaul the tax system. At present, the system raises most revenue from taxes on personal income and corporate profits, and the major reform proposals call for a shift to a consumption base and for drastic simplification of the tax code. Proposals by Senators Domenici and Nunn, Senator Specter, and Representative Armey would all eliminate the individual and corporate income taxes, and substitute new taxes on business and individuals. 2/ The Domenici-Nunn proposal would retain graduated tax rates so as to roughly maintain the tax burden across income groups; the Specter and Armey proposals would impose a single “flat” tax rate on a broader tax base and could have major distributional implications.
The arguments for simplifying the tax code relate to economic efficiency and growth. All the proposals would eliminate distortions in the current tax system by taxing all sources of income at the same rate, whereas under the current system, dividends are taxed twice (as corporate and personal income) while fringe benefits to employees are not taxed. In addition, by simplifying the tax code, the proposals are expected to save resources currently devoted to tax compliance and administration.
Proponents of consumption-based taxation argue that by exempting income saved from taxation, a shift to a consumption tax will raise the U.S. saving rate, thereby promoting investment and growth. Both theoretical and empirical studies, however, present mixed conclusions regarding the sensitivity of saving to changes in taxes that affect intertemporal rates of return. Advocates of a “flat tax” with a single rate claim that their proposals can achieve revenue neutrality while lowering marginal rates. The lower marginal rates will, they argue, lead workers to increase their labor supply, thereby increasing incomes and growth. This conclusion, too, is debatable.
This chapter outlines the major reform proposals and evaluates the arguments for reform. The distributional implications of each scheme are explored, using for comparison tax rates at which the reforms would be revenue neutral. The paper then examines the efficiency arguments for reform, including the proposition that consumption-based taxation will increase saving and growth. While the reform proposals strive for the simplest possible new tax code, in practice a reform could introduce new distortions and inequities unless special transitional provisions are included; these transitional issues are also discussed.
1. The flat tax proposal
Representatives Armey and Specter have proposed a single-rate “flat” tax on businesses and individuals, based on the flat tax concept developed by Hall and Rabushka (1985). The flat tax would tax all income exactly once, and tax it at the source. The authors claim that by eliminating most tax deductions and credits and thereby broadening the tax base, revenue neutrality could be achieved with a constant marginal tax rate of either 19 percent (under the Armey proposal, which includes no deductions) or 20 percent (under the Specter proposal, which retains deductions for mortgage interest and charitable contributions). 1/ Progressivity would be achieved by allowing large standard deductions and personal exemptions. A family of four would not pay any tax on income below either $25,500 (Specter) or $34,700 (Armey); after that, the average tax rate would rise gradually with income (Table VII-1).
United States: Average Tax Rates by Taxable Income under Proposed Flat Taxes
United States: Average Tax Rates by Taxable Income under Proposed Flat Taxes
| Taxable Income (dollars) |
Armey’s 19 Percent Flat Tax |
Specter’s 20 Percent Flat Tax |
|---|---|---|
| 24,500 | -- | -- |
| 34,700 | 5 | 6 |
| 45,000 | 9 | 9 |
| 55,000 | 10 | 11 |
| 65,000 | 12 | 12 |
| 75,000 | 13 | 13 |
| 100,000 | 14 | 15 |
| 150,000 | 16 | 17 |
| 200,000 | 17 | 18 |
| 400,000 | 18 | 19 |
| 1,000,000 | 19 | 20 |
United States: Average Tax Rates by Taxable Income under Proposed Flat Taxes
| Taxable Income (dollars) |
Armey’s 19 Percent Flat Tax |
Specter’s 20 Percent Flat Tax |
|---|---|---|
| 24,500 | -- | -- |
| 34,700 | 5 | 6 |
| 45,000 | 9 | 9 |
| 55,000 | 10 | 11 |
| 65,000 | 12 | 12 |
| 75,000 | 13 | 13 |
| 100,000 | 14 | 15 |
| 150,000 | 16 | 17 |
| 200,000 | 17 | 18 |
| 400,000 | 18 | 19 |
| 1,000,000 | 19 | 20 |
The flat tax would eliminate the current double-taxation of dividends and subject all labor and capital income to the same marginal tax rate. Corporate and noncorporate businesses would pay the flat business tax rate on value added defined as: 2/
Total revenue from sales
less material input costs
less wages, salaries, and pensions
less plant and equipment investment.
An unlimited loss-carryforward would apply to businesses that incur net losses. 3/ The business tax would apply only to operations within U.S. borders, so that foreign operations of U.S. companies would not be taxed. 4/ Several other differences from current tax law are significant. Full expensing of physical investment would replace multi-year depreciation schedules, and firms’ financial receipts from holding stocks or bonds would not be taxable. Capital gains on business assets would be taxed, since the full sale price of the asset would be taxed on sale but only the purchase price would have been deducted at time of purchase.
The individual flat tax would apply only to labor income. Individuals’ capital income would not be taxed, since it would already have been taxed once at the business level, thereby eliminating double taxation of distributed capital income. An individual or household would pay a flat rate tax on
Wage, salary, and pension income
less family allowances based on household size.
The individual flat tax is equivalent to a consumption tax since it taxes only labor income, which over a lifetime is equal to consumption. 1/
Estimates made by U.S. Treasury’s Office of Tax Analysis suggest that a flat tax would not be revenue neutral compared to the current individual and corporate income taxes at the rates suggested by its proponents. For example, a rate of 24.9 percent would be required to achieve revenue neutrality with the Armey deductions. Modifying the proposal to retain the earned income tax credit would require raising the rate to 25.8 percent. Exempting certain nonbusiness organizations from the tax base, such as nonprofit institutions and state and local governments, would narrow the tax base and require still higher rates.
2. Saving-exempt Income tax
Senators Nunn and Domenici have proposed converting the personal and business income tax system to a saving-exempt income tax (SEIT). 2/ The SEIT or Unlimited Savings Allowance (USA) tax system would be explicitly a consumption tax because the household tax base would be household income less new saving or investment, and the business tax base would be business cashflow less purchases, including all forms of investment. Like the flat tax proposals, the SEIT proposal would eliminate double taxation of dividend income and would tax all income--wages and salaries, interest, dividends, and capital gains--at the same rate. 1/
The personal SEIT would retain a multi-bracket structure so as to maintain roughly the progressivity of the current tax system; rates would be set so as to ensure that the tax burden paid by each quintile of the income distribution is unchanged. The business SEIT would be levied at a flat rate, but Medicare and Social Security taxes would be deductible against households’ tax liability.
Business taxes would be levied at a flat rate on the business’s value added, on a cashflow basis. The tax base would be
Total revenue from domestic sales
less purchases from other businesses
less purchases of imports.
A business would compute its tax liability as its tax base times the tax rate, then deduct its payments of Social Security and Medicare taxes; a refund would be received in the case of a negative tax liability. 2/
Like the flat tax proposals, the SEIT would allow full expensing of investment in lieu of depreciation and firms’ financial receipts would be excluded from gross income. Unlike the flat tax, businesses could not deduct wage, salary or pension payments but, like the flat tax, employer-provided fringe benefits would be taxable at the business level as part of the gross income tax base. Nonprofit entities would not be subject to the business tax, and partnership income would be taxed only when withdrawn as salaries, dividends or bonuses. 3/
An advantage of the SEIT is that border-adjustments of the tax likely would be permissible under GATT (imports would be taxed at the same rate as the SEIT on importation and exports would be exempt from tax). 4/ Most other industrial countries and many developing countries use border-adjustable or territorial taxes and exempt exports from tax. Proponents of consumption-based taxes argue that switching to such a tax in the United States would level the playing field between domestic and imported goods in the U.S. market by equalizing the amount of tax paid, and would make U.S. exports more competitive in foreign markets by removing the tax from the export price.
The individual SEIT would be levied on an income base with adjustments for saving. Income would be defined as under current tax law to include earned income, pensions and transfer payments, business income, and investment income. Net saving would be fully deducted from income; funds withdrawn from saving and consumed would be included in taxable income. 1/ Taxable income would equal:
Wage, salary, pension and capital income
less net new saving
less family allowances based on household size
less personal exemptions
less mortgage interest, charitable contributions, and exemptible education expenses.
Interest payments on all borrowing (mortgage and other) would be deductible; consumption financed by borrowing would be taxed in the year that the money is borrowed and therefore would affect net saving. 2/ Income from business activities would be taxed on a cashflow basis. Income from capital asset sales would be taxed only if not reinvested. The Nunn-Domenici plan would allow state and local tax payments, charitable donations, and extraordinary medical expenses to be deducted from income. 3/ Individuals also could credit payroll tax payments against their tax liability; the difference would be refundable.
A graduated rate structure and per-person exemptions are included in the Nunn-Domenici proposal to ensure progressivity. For a family of four filing jointly, the combined family allowance and personal exemptions would exclude $17,600 from taxation. Additional itemized deductions would be permitted for mortgage interest paid, charitable contributions, and a new deduction for higher education expenses (up to $2,000 per person per year); all other deductions, including the deduction for state and local income tax, would be eliminated.
Taxable income would then be taxed at three graduated rates. A married couple filing jointly would pay 19 percent on the first $5,400 of their taxable income, 27 percent on income from $5,400 to $40,000, and 40 percent on income above $40,000. The payroll tax credit would reduce the effective marginal rate on labor income up to $60,000 per person by 7.65 percentage points (the employee share of the Social Security and Medicare payroll taxes). In addition, the Nunn-Domenici proposal would expand the earned income tax credit.
3. The equivalence between the flat tax and the consumption tax
A two-period example can be used to show the equivalence of the flat and SEIT (consumption) taxes. Assume individuals live for two periods. In period 1 they work, earning income w, of which they save s and consume the remainder (w-s). In period 2, they do not work or save but consume from their savings. There is only one asset in the economy, which yields a return r; the discount rate is also equal to r. There are no endowments, only labor income, such that individuals have no assets in the beginning of period 1.
The tabulation below demonstrates that the net present value of taxes is equal whether the tax is levied when labor income is earned or when it is consumed.
Tax Revenue under Consumption-based and Flat Taxes
Tax Revenue under Consumption-based and Flat Taxes
| Wages | Consumption | Flat Tax Revenue |
Consumption Tax Revenue |
|
|---|---|---|---|---|
| Period 1 | w | w-s | τw | τ(w-s) |
| Period 2 | -- | s(1+r) | -- | fs(1+r) |
| Present value of tax revenues | τw | τw |
Tax Revenue under Consumption-based and Flat Taxes
| Wages | Consumption | Flat Tax Revenue |
Consumption Tax Revenue |
|
|---|---|---|---|---|
| Period 1 | w | w-s | τw | τ(w-s) |
| Period 2 | -- | s(1+r) | -- | fs(1+r) |
| Present value of tax revenues | τw | τw |
This suggests that the major difference between a flat tax (levied on labor income) and a consumption tax (levied on the difference between total income and saving) is the timing of households tax payments. The flat tax is levied when labor income is earned, and the consumption tax is levied when total income is consumed. However, this equivalence only holds if capital markets are perfect and the private and public rates of return are identical.
The two-period model also can demonstrate that either a flat tax or a SEIT creates an incentive to save relative to an income tax that applies to both capital and labor income. The tabulation below shows that, under a consumption tax, saving increases lifetime consumption more than it does under an income tax.
Consumption and Revenue under Consumption and Income Taxes
Consumption and Revenue under Consumption and Income Taxes
| Consumption Tax | Income Tax | ||||
|---|---|---|---|---|---|
| Wages | Consumption | Revenue | Consumption | Revenue | |
| Period 1 | w | w-s | τ(w-s) | (1-τ)w-s | τw |
| Period 2 | -- | s(1+r) | τs(1+r) | s(1+r(1-τ)) | srτ |
| Present value | w | τw | (1-τ)w-(srτ/(1+r)) | τw+(srτ/(1+r)) | |
Consumption and Revenue under Consumption and Income Taxes
| Consumption Tax | Income Tax | ||||
|---|---|---|---|---|---|
| Wages | Consumption | Revenue | Consumption | Revenue | |
| Period 1 | w | w-s | τ(w-s) | (1-τ)w-s | τw |
| Period 2 | -- | s(1+r) | τs(1+r) | s(1+r(1-τ)) | srτ |
| Present value | w | τw | (1-τ)w-(srτ/(1+r)) | τw+(srτ/(1+r)) | |
The tabulation also shows that an income tax yields more revenue than a consumption tax set at the same rate. Therefore, shifting to a consumption tax basis would require higher average tax rates or a broader base than the present income tax in order to achieve revenue neutrality.
4. Distributional implications of the tax proposals
Since, as demonstrated above, an important difference between the SEIT and the flat tax proposals is the timing of tax payments, their distributional effects will depend on whether individuals are net savers or borrowers. For example, households with high labor incomes relative to consumption (i.e., net savers) will tend to pay taxes under a consumption tax during such a high-saving year than under the flat tax. Over a lifetime, taxes paid under a flat tax will be higher for net savers (i.e., those that leave bequests) than under a consumption tax. Conversely, net borrowers would tend to pay more taxes over a lifetime under a consumption tax than a flat tax.
Since lower-income households would tend to have a higher level of debt relative to income, this might suggest that the consumption tax would tend to be more regressive than the flat tax. However, estimates suggest that the graduated marginal tax rates of the proposed SEIT would tend to preserve the progressivity of the current income tax system, while the proposed flat tax would reduce the progressivity of the tax system.
For example, the flat tax would tend to reduce significantly the aftertax incomes of families with incomes below $20,000 owing to the repeal of the EITC. In addition, the proposed flat tax is regressive since the share of salaries, wages, and pensions in total income tends to fall for higher income households. 1/ Conversely, the Nunn-Domenici proposal would protect low-income households by expanding the refundable earned income tax credit for low-income working families. The three-bracket tax schedule is designed to hold constant the burden of taxes on each quintile of the income distribution.
The tabulation below (see also Table VII-2) contains estimates of the distributional effect of the two proposals, prepared by the Office of Tax Analysis (OTA) of the Department of the Treasury on the assumption that the tax rates were set to ensure revenue neutrality. 1/ OTA (1994) estimates that under the flat tax, the tax burden would decrease only for those families with incomes above $200,000 who typically receive a substantial portion of their income from capital. The tax burden would tend to increase for lower income households suggesting that the flat tax system would be substantially more regressive that the current income tax system. By contrast, the SEIT is estimated to improve the progressivity of the income tax system.
United States: Replace Current Individual and Corporate Income Taxes (including the EITC) with a Flat Rate Tax 1/
(In billions of dollars: unless mentioned otherwise)
Assumes a revenue-neutral rate of 24.9 percent on the basis of 1994 incomes.
Family Economic Income (FED is a broad-based income concept. FEI is constructed by adding to adjusted gross income (AGI) unreported and under-reported income; IRA end Keogh deductions; nontaxable transfer payments, such as Social Security and AFDC; employer-provided fringe benefits; inside build-up on pensions, IRAs, Keoghs, and life insurance; tax-exempt interest; and imputed rent on owner-occupied housing. Capital gains are computed on an accrual basis, adjusted for inflation to the extent that reliable data allow. Inflationary losses of lenders are subtracted and those of borrowers are added. There is also an adjustment for accelerated depreciation of noncorporate businesses. FEI is shown on a family rather than on a tax-return basis. The economic incomes of all members of a family unit are added to arrive at the family’s economic income used in the distributions.
The standard deduction is $24,700 (joint) or $12,350 (single) plus $5,000 for each dependent.
United States: Replace Current Individual and Corporate Income Taxes (including the EITC) with a Flat Rate Tax 1/
(In billions of dollars: unless mentioned otherwise)
| Change in After-Tax Income Under Proposal | Percentage Change In Total Federal Tax | |||||||
|---|---|---|---|---|---|---|---|---|
| Family Economic Income Class 2/ ($) |
After-Tax Income Under Current Law | Repeal Income Tax | 24.9 Percent Tax | Total Change | Percentage Change | |||
| on Wages Over Standard Deduction 3/ | on Employer Fringes | on Business Cash Flow | ||||||
| 0-10 | 79.3 | -1.2 | -0.1 | -0.9 | -0.9 | -3.1 | -3.9 | 48.1 |
| 10-20 | 249.4 | -1.6 | -2.1 | -2.8 | -4.2 | -10.6 | -4.3 | 41.1 |
| 20-30 | 342.6 | 13.7 | -10.9 | -5.1 | -7.1 | -9.5 | -2.8 | 17.3 |
| 30-50 | 728.8 | 61.8 | -39.1 | -11.6 | -17.5 | -6.4 | -0.9 | 4.2 |
| 50-75 | 865.4 | 92.4 | -68.0 | -13.3 | -20.5 | -9.4 | -1.1 | 4.6 |
| 75-100 | 680.2 | 86.5 | -73.3 | -9.6 | -18.0 | -14.4 | -2.1 | 8.2 |
| 100-200 | 902.5 | 148.8 | -110.9 | -9.0 | -40.8 | -11.9 | -1.3 | 4.9 |
| 200 & over | 906.8 | 240.1 | -86.3 | -2.1 | -86.1 | 65.6 | 7.2 | -23.9 |
| Total | 4,711.2 | 641.1 | -391.7 | -54.4 | -195.0 | -- | -- | -- |
Assumes a revenue-neutral rate of 24.9 percent on the basis of 1994 incomes.
Family Economic Income (FED is a broad-based income concept. FEI is constructed by adding to adjusted gross income (AGI) unreported and under-reported income; IRA end Keogh deductions; nontaxable transfer payments, such as Social Security and AFDC; employer-provided fringe benefits; inside build-up on pensions, IRAs, Keoghs, and life insurance; tax-exempt interest; and imputed rent on owner-occupied housing. Capital gains are computed on an accrual basis, adjusted for inflation to the extent that reliable data allow. Inflationary losses of lenders are subtracted and those of borrowers are added. There is also an adjustment for accelerated depreciation of noncorporate businesses. FEI is shown on a family rather than on a tax-return basis. The economic incomes of all members of a family unit are added to arrive at the family’s economic income used in the distributions.
The standard deduction is $24,700 (joint) or $12,350 (single) plus $5,000 for each dependent.
United States: Replace Current Individual and Corporate Income Taxes (including the EITC) with a Flat Rate Tax 1/
(In billions of dollars: unless mentioned otherwise)
| Change in After-Tax Income Under Proposal | Percentage Change In Total Federal Tax | |||||||
|---|---|---|---|---|---|---|---|---|
| Family Economic Income Class 2/ ($) |
After-Tax Income Under Current Law | Repeal Income Tax | 24.9 Percent Tax | Total Change | Percentage Change | |||
| on Wages Over Standard Deduction 3/ | on Employer Fringes | on Business Cash Flow | ||||||
| 0-10 | 79.3 | -1.2 | -0.1 | -0.9 | -0.9 | -3.1 | -3.9 | 48.1 |
| 10-20 | 249.4 | -1.6 | -2.1 | -2.8 | -4.2 | -10.6 | -4.3 | 41.1 |
| 20-30 | 342.6 | 13.7 | -10.9 | -5.1 | -7.1 | -9.5 | -2.8 | 17.3 |
| 30-50 | 728.8 | 61.8 | -39.1 | -11.6 | -17.5 | -6.4 | -0.9 | 4.2 |
| 50-75 | 865.4 | 92.4 | -68.0 | -13.3 | -20.5 | -9.4 | -1.1 | 4.6 |
| 75-100 | 680.2 | 86.5 | -73.3 | -9.6 | -18.0 | -14.4 | -2.1 | 8.2 |
| 100-200 | 902.5 | 148.8 | -110.9 | -9.0 | -40.8 | -11.9 | -1.3 | 4.9 |
| 200 & over | 906.8 | 240.1 | -86.3 | -2.1 | -86.1 | 65.6 | 7.2 | -23.9 |
| Total | 4,711.2 | 641.1 | -391.7 | -54.4 | -195.0 | -- | -- | -- |
Assumes a revenue-neutral rate of 24.9 percent on the basis of 1994 incomes.
Family Economic Income (FED is a broad-based income concept. FEI is constructed by adding to adjusted gross income (AGI) unreported and under-reported income; IRA end Keogh deductions; nontaxable transfer payments, such as Social Security and AFDC; employer-provided fringe benefits; inside build-up on pensions, IRAs, Keoghs, and life insurance; tax-exempt interest; and imputed rent on owner-occupied housing. Capital gains are computed on an accrual basis, adjusted for inflation to the extent that reliable data allow. Inflationary losses of lenders are subtracted and those of borrowers are added. There is also an adjustment for accelerated depreciation of noncorporate businesses. FEI is shown on a family rather than on a tax-return basis. The economic incomes of all members of a family unit are added to arrive at the family’s economic income used in the distributions.
The standard deduction is $24,700 (joint) or $12,350 (single) plus $5,000 for each dependent.
Estimated Change in Tax Burden
Estimated Change in Tax Burden
| Percent Change Relative to Current System |
||
|---|---|---|
| Family income |
Flat
Tax |
Consumption Tax |
| ($,000) | ||
| 0-10 | 48 | -75 |
| 10-20 | 41 | -9 |
| 20-30 | 17 | -14 |
| 30-50 | 4 | -5 |
| 50-75 | 5 | -- |
| 75-100 | 8 | -- |
| 100-200 | 5 | 3 |
| over 200 | -24 | 4 |
Estimated Change in Tax Burden
| Percent Change Relative to Current System |
||
|---|---|---|
| Family income |
Flat
Tax |
Consumption Tax |
| ($,000) | ||
| 0-10 | 48 | -75 |
| 10-20 | 41 | -9 |
| 20-30 | 17 | -14 |
| 30-50 | 4 | -5 |
| 50-75 | 5 | -- |
| 75-100 | 8 | -- |
| 100-200 | 5 | 3 |
| over 200 | -24 | 4 |
A study by the Congressional Budget Office (CBO) examined the sensitivity of this type of calculation to assumptions about the measurement of household saving and the distribution of the corporate income tax burden between labor and capital. In particular, the CBO examined the effect of these assumptions on the rate schedule required under a SEIT to maintain revenue neutrality and an unchanged distribution of the tax burden across income quintiles (see tabulation below).
For example, measuring saving as the difference between income and consumption (the residual approach) yields a more progressive distribution of saving than does measuring saving as the change in a family’s net worth would be steeper if based on residual estimates of saving (see tabulation below). Assuming that the burden of corporate income tax falls entirely on capital income would yield a more progressive distribution of income and would require more steeply graduated tax rates than assuming that half the burden falls on labor income.
The most regressive combination of assumptions (net worth estimation of saving and corporate income taxes falling in part on labor), would require tax rates ranging from 14 to 36 percent. The most progressive assumptions (residual estimation of saving and corporate taxes falling wholly on capital income) would require tax rates ranging from 16 to 55 percent.
Revenue-Neutral and Distributionally-Neutral Tax Rates for the Prototype Saving-Exempt Individual Income Tax 1/
(In percent)
Revenue-Neutral and Distributionally-Neutral Tax Rates for the Prototype Saving-Exempt Individual Income Tax 1/
(In percent)
| Measure of Saving Used in Estimates | |||||
|---|---|---|---|---|---|
| Current Income Tax Rate |
Full Burden of Business Tax on Capital |
Burden Split Between Capital and Labor 2/ |
|||
| Residual | Net Worth | Residual | Net Worth | ||
| First bracket | 15.0 | 16.0 | 13.0 | 16.0 | 14.0 |
| Second bracket | 28.0 | 38.0 | 28.0 | 38.0 | 28.0 |
| Third bracket | 31.0 | 49.0 | 38.5 | 49.0 | 36.0 |
| Fourth bracket | 36.0 | 55.0 | 41.0 | 49.0 | 36.0 |
| Fifth bracket 3/ | 39.6 | 55.0 | 41.0 | 49.0 | 36.0 |
Revenue-Neutral and Distributionally-Neutral Tax Rates for the Prototype Saving-Exempt Individual Income Tax 1/
(In percent)
| Measure of Saving Used in Estimates | |||||
|---|---|---|---|---|---|
| Current Income Tax Rate |
Full Burden of Business Tax on Capital |
Burden Split Between Capital and Labor 2/ |
|||
| Residual | Net Worth | Residual | Net Worth | ||
| First bracket | 15.0 | 16.0 | 13.0 | 16.0 | 14.0 |
| Second bracket | 28.0 | 38.0 | 28.0 | 38.0 | 28.0 |
| Third bracket | 31.0 | 49.0 | 38.5 | 49.0 | 36.0 |
| Fourth bracket | 36.0 | 55.0 | 41.0 | 49.0 | 36.0 |
| Fifth bracket 3/ | 39.6 | 55.0 | 41.0 | 49.0 | 36.0 |
5. Administrative and compliance gains from tax simplification
The cost of compliance with the current income tax system, as well as the cost incurred in reducing one’s tax liability, has been estimated to be as high as $50 billion annually. 4/ It has been argued that an important benefit of the flat tax and the SEIT would be that they would simplify the tax code, reducing the cost of compliance and the incentives for tax evasion. 5/ The two reform proposals would substantially simplify the business tax system since the business tax base would be calculated on a cashflow basis, which would eliminate the need to maintain a set of accrual accounts for tax purposes.
While these proposals have the potential for simplifying the tax system, the extent of simplification would depend crucially on withstanding pressures for introducing deductions or tax credits. Moreover, while the proposed reforms would have the potential for reducing tax evasion by ensuring that different sources of income are taxed at the same rate, the elimination of tax withholding (as suggested under the flat tax) could have the opposite effect. Finally, measures to ease the transition to the new taxes would undoubtedly add to their complexity (see section below).
6. Macroeconomic efficiency issues
It is argued that a consumption-based tax would promote economic efficiency and stimulate output growth by reducing distortions and promoting increased saving and labor supply. An important channel for stimulating saving and investment would be the impact of reform on interest rates. Hall and Rabushka (1995) point out that the current tax system creates a wedge between the return on loans and the cost of borrowing. In particular, the tax on interest income of lenders reduces the after-tax interest rate below the pre-tax rate. If this distortion was removed, the effect would be to shift the supply of credit outward, reducing the pre-tax interest rate and increasing the equilibrium amount of lending and investment. It is argued that the fall in pre-tax interest rates could be substantial since lending comes from corporations and high income individuals who face marginal tax rates of 31 to 40 percent. While gauging the likely effect on interest rates is difficult, Hall has pointed to the fact that municipal bonds yield roughly one-sixth less than comparably-rated taxable bonds as suggestive of the order of magnitude that could be expected.
Other features of a consumption tax also would work to increase investment. For example, the after-tax cost of capital would decline as a result of full expensing of capital purchases, as opposed to depreciation over the life of the asset, as under the current tax system. In addition, eliminating double taxation of dividend income would increase the rate of return to equity-financed capital, which in turn would increase investment demand. Eliminating double dividend taxation should also eliminate the preference for debt over equity financing and remove any bias toward lower-risk investments that can be debt-financed, thereby enhancing the efficient allocation of capital. 1/
The effect of a tax reform on saving, investment, and aggregate output also will depend on the responsiveness of saving to changes in the interest rate. However, there is considerable controversy regarding the interest elasticity of saving; empirical estimates vary widely, and a number of studies suggest that it is relatively low. Nonetheless, studies of the effect of moving to a consumption tax that have relied on a relatively large interest elasticity of saving provide an upper bound of the welfare and output gains that might result. These studies suggest that improvements in welfare from tax reform may be 5-6 percent, and less if the transition costs are taken into account.
Summers (1981) provides relatively strong support for tax reform. Summers determines that replacing a total capital tax of 50 percent and labor tax of 20 percent with a consumption tax designed to generate equal revenue (27.5 percent of GDP) would yield a 75 percent increase in the capital-output ratio and a 54 to 66 percent reduction in the gross interest rate. This would correspond to an increase in steady-state output of between 12 and 18 percent and a welfare gain of 6 to 17 percent (depending on parameter values). Summers cautions, however, that higher steady state consumption is obtained by lowered consumption during the transition to steady state and by increasing the tax burden of those who are elderly when the reform occurs. Summers (1989) argues that convergence to steady state is rapid and that long-term gains outweigh transition costs.
Auerbach et al. (1983), using a 55-period overlapping-generations model, demonstrate that shifting from a proportional income tax to a proportional consumption tax (at a 39.5 percent rate) would raise welfare by about 6 percent in steady state. They also find that shifting from a progressive income tax to a progressive consumption tax, as under the Nunn-Domenici proposal, would yield a 5 percent improvement in welfare and increase the capital/output ratio from 3.04 to 4.38. 1/
However, they note that much of the efficiency gain is due to the transition. A large share of the tax burden is shifted toward the existing assets and inelastic consumption of the elderly, which enables a lower overall tax rate and a lesser tax burden on the young. The increase in their disposable income encourages an increase in saving and investment. The welfare of those aged over 40 at the time of the conversion to the consumption tax would fall by about 1 percent, since the consumption/income ratios of these households would be relatively high.
Boskin (1978) estimated a total interest elasticity of saving (η) between 0.3 and 0.4 using aggregate time-series data for the United States. On the basis of an estimated elasticity of interest rates with respect to the capital income tax of 0.3, Boskin estimates that eliminating capital income taxation would increase the capital/labor ratio by about 15 to 20 percent. However, these estimates suggest that a shift to consumption taxation would increase the saving rate by only 5 to 6 percent. For example, suppose that the initial nominal interest rate was 10 percent and the income tax rate on capital 30 percent, such that the net interest rate equalled 7 percent. Let the post-reform gross interest rate equal 8 percent (all numbers based on Hall and Rabushka (1995)). For an interest elasticity of 0.4 and an initial saving rate of 4.1 (the actual 1994 rate as a share of disposable income), the saving rate would rise only to 4.3 percent.
7. The expected effect of tax reform on labor supply
A further channel through which a conversion of the income tax system to a consumption tax could increase welfare and output is by increasing the labor supply. Hausman (1981) and Hall and Rabushka (1995) have argued that shifting to a proportional tax would enable a reduction in the tax rate on labor income, thereby increasing labor supply, expanding output and consumption. Hausman concluded that existing taxes reduced labor supply by 8 percent on average and induced deadweight loss equal to 28.7 percent of revenue; a proportional tax, he estimated, would reduce average hours worked only one-seventh as much and cause deadweight loss of 7.1 percent of revenue. Moreover, if savings raised the capital/labor ratio, higher nominal wages would increase the incentive to work.
However, if one takes into consideration that the deductibility of state and local taxes will be eliminated under the flat tax proposal, marginal tax rates may rise for many taxpayers. For a family of four, the marginal federal tax rate will equal 25.8 percent on income over $34,700 (using Armey’s deductions and OTA’s revenue-neutral rate estimate), versus a 28 percent marginal rate under the current tax system. Combined with a state income tax rate of up to 9 percent (for the District of Columbia), the marginal tax rate could exceed 35 percent--just under the 36 percent bracket under the current tax system and above the first three brackets (which apply to incomes under $117,950 for single filers and $143,600 for married couples filing jointly). 1/
8. Other macroeconomic Issues
The reform proposals also may affect the economy by affecting trade patterns or the magnitude of business cycles. Both the flat tax and the SEIT could have weaker automatic stabilizer properties than the current income tax system. For example, cyclical fluctuations in revenue from a pure consumption tax or the SEIT would be relatively small because consumption varies less than income. Revenues from a flat tax likely could be highly correlated with the business cycle since the flat tax’s base--business cashflow plus the wage bill--would vary with the cycle. However, since business cashflow is defined using the fully expensed value of investment purchases, which also would be likely to decline in economic downturns, the cycle’s effect on flat tax revenues is uncertain.
Although advocates of the SEIT argue that shifting to a VAT-equivalent, refundable tax will promote exports and reduce imports, this result depends on several assumptions. First, it assumes that corporate income taxes are passed on to export prices and that exporters will reduce prices if the tax on exports falls. Some evidence suggests that additional taxes are reflected in higher prices, but that tax cuts do not result in price cuts; in this case, shifting to a VAT will make imports more expensive, but will not make U.S. exports more competitive abroad. Moreover, a company producing both for export and domestic sale could possibly pass the corporate income tax burden entirely onto domestic sales in order to ensure export competitiveness.
If a shift to a VAT did raise exports and reduce imports, currency appreciation could result, offsetting any impact on the trade balance. Finally, the general equilibrium effects must be considered. If shifting to a VAT did shift the tax burden from exports to domestic sources of income, real incomes may fall, resulting in lower saving; wage increases would likely follow this reduction in real income, and the resulting increase in prices could offset the fall in export prices. 1/
Treatment of foreign production and imports requires further attention under the flat tax proposal. The proposed flat tax would apply only to domestically located business operations, with no border adjustment for imports or exports. This rule would create an incentive for firms to avoid taxation by moving their operations overseas, thereby eliminating jobs and narrowing the tax base. If a corporation moved its production site to a country with a border-adjustable VAT, rather than a corporate income tax, the firm would pay no tax on its production for the U.S. market; if it produced in the United States, it would be taxed on its profits at the flat rate.
9. Transition issues
The introduction of a flat tax or SEIT would introduce a number of transition issues. One problem would be the treatment of existing assets, which had been purchased on the assumption of depreciation allowances (in the case of business assets) or mortgage interest deductibility (in the case of residential assets). An option would be to retain the tax preferences on existing assets, but this would add to the complexity of the tax system, and could create opportunities for evasion. An alternative would be to allow taxpayers to fully expense existing assets at the time of the tax reform. Hall (1995) estimates that to offset the revenue losses from allowing taxpayers to expense all unused depreciation allowances in a single year would require a 1.1 percent increase in the flat tax rate in that year.
The effect of eliminating the deductibility of mortgage interest would likely be to reduce the price of pre-existing homes. While pre-tax interest rates would likely fall, the net effect would also likely be to raise the overall cost of home ownership. Options for the treatment of existing mortgages include allowing partial deduction of existing mortgage interest and requiring that lenders continue to pay taxes on interest income, or to require lenders and borrowers to refinance mortgages and immediately treat the interest according to the new tax law. Note that home ownership would still be favorably treated under either a flat tax or SEIT, as neither would tax the imputed income stream from owner-occupied housing.
An important transitional issue for both the flat tax and the SEIT would be their effect on the lifetime tax burden on the elderly. This segment of the population would be affected adversely since their assets were accumulated using after tax income. Upon the introduction of the consumption tax, the value of these assets would likely fall (for the reasons discussed above). In addition, the future consumption that these assets were intended to finance would now be taxed.
To avoid taxing the principal of existing assets twice (it was previously taxed as earned income), Senator Domenici has proposed two approaches to reducing the burden on existing assets. One would be to value all assets at the time the reform went into effect, then allow individuals to exclude this date-of-enactment value from tax when the assets are sold. Alternatively, individuals could be allowed to deduct this date-of-enactment value from their taxes over the first few years following reform. Either proposal, however, would result in revenue loss, requiring higher tax rates and thereby reducing some of the additional saving the reform is intended to generate.
References
Auerbach, Alan J, Lawrence H. Kotlikoff, and Jonathan Skinner, “The Efficiency Gains from Dynamic Tax Reform,” International Economic Review, Vol. 24, No. 1 (1983), pp. 81–100.
Bartolini, Leonardo, Assaf Razin, and Steve Symansky, “G-7 Fiscal Restructuring in the 1990s: Macroeconomic Effects,” Economic Policy, 20 (1995, forthcoming).
Blanchard, Olivier J., and Stanley Fischer, Lectures on Macroeconomics (Cambridge, Massachusetts: MIT Press, 1989).
Boskin, Michael J., “Taxation, Saving, and the Rate of Interest,” Journal of Political Economy, Vol. 86, No. 2 (1978), pp. S3–S27.
Bosworth, Barry, Gary Burtless, and John Sabelhaus, “The Decline in Saving: Evidence from Household Surveys,” Brookings Papers on Economic Activity (1991:1).
Bradford, David H., “What are Consumption Taxes and Who Pays Them?” Tax Notes (April 18, 1988), pp. 383–391.
Congressional Budget Office, Reducing the Deficit: Spending and Revenue Options (Washington: Government Printing Office, 1994).
Davies, James, France St. Hilaire, and John Whalley, “Some Calculations of Lifetime Tax Incidence,” American Economic Review, Vol. 74, No. 4 (1984), pp. 633–649.
Domenici, Pete V., “The Unamerican Spirit of the Federal Income Tax,” Harvard Journal on Legislation, Vol. 31 (1994), pp. 273–320.
Eisner, Robert., “The Proposed Sales and Wages Tax: Fair, Flat or Foolish?” AEI conference paper: “The Flat Tax: An Alternative to the Current Income Tax” (January 27, 1995).
Garrett, Beth, and Mark Weinberger, “A National Consumption Tax: Why, What and When.” (mimeographed, Offices of Senators Boren and Danforth, 1994).
Hakken, Jon, and Frank Sammartino, “Estimates for a Prototype Saving-Exempt Income Tax,” (memorandum, Congressional Budget Office, March 1994).
Hall, Robert E., and Alvin Rabushka, “The Flat Tax in 1995,” (AEI conference paper: “The Flat Tax: An Alternative to the Current Income Tax,” January 27, 1995).
Hall, Robert E., and Alvin Rabushka, The Flat Tax (Stanford: The Hoover Institution Press, 1985).
Hausman, Jerry A., “Labor Supply” in eds. Henry J. Aaron and Joseph A. Pechman, How Taxes Affect Economic Behavior (Washington: Brookings Institution, 1981).
Joint Committee on Taxation, Discussion of Issues Relating to “Flat” Tax Rate Proposals (April 3, 1995).
Joint Committee on Taxation, Methodology and Issue in Measuring Changes in the Distribution of Tax Burdens (June 14, 1993).
Nunn, Samuel, Congressional testimony, January 10, 1995 as cited in Congressional Record, Vol. 141, No. 16.
Office of Tax Analysis, U.S. Department of the Treasury. “An Analysis of A Flat Rate Consumption Tax.” (mimeographed, 1994).
Poterba, James M., “Lifetime Incidence and the Distributional Burden of Excise Taxes,” American Economic Association Papers and Proceedings (1989), pp. 325–330.
Ramsey, Frank P., “A Mathematical Theory of Saving,” Economic Journal 38, No. 152 (1928), 543–559. (Reprinted in Stiglitz, Joseph E., and Hirofumi Uzawa, eds. Readings in the Modern Theory of Economic Growth (Cambridge, Massachusetts: MIT Press, 1969).
Specter, Arlen, Congressional testimony, March 2, 1995 as cited in Congressional Record, Vol. 141, No. 39.
Summers, Lawrence H., “Capital Taxation and Accumulation in a Life-Cycle Growth Model,” American Economic Review, Vol 4, No. 4. (1981), pp. 533–544.
Summers, Lawrence H., “Tax Policy and a Life Cycle Model.” National Bureau of Economic Research Working Paper No. 302 (1989).
Tait, Alan A., ed. Value-Added Tax: Administrative and Policy Issues (Washington: International Monetary Fund, 1991).
Tait, Alan A., Value-Added Tax: International Practice and Problems (Washington: International Monetary Fund, 1988).
Weidenbaum, Murray, “The Saving-Exempt Income Tax (and the Companion Business Cash-Flow Tax),” Center for the Study of American Business, Policy Study No. 122 (July 1994).
Prepared by Elaine Buckberg.
In addition, Senator Lugar has proposed to replace income taxation with a national sales tax. Representative Gephardt has promoted a separate flat tax proposal, and Congress has appointed Mr. Jack Kemp to chair a commission on tax reform.
For mortgage debt up to $100,000 and charitable contributions up to $2,500.
Currently, corporate profits are taxed at graduated rates and the profits of noncorporate businesses such as partnerships are taxed only when distributed as individual income.
The Alternative Minimum Taxes (AMT) also would be repealed along with the Corporate and Personal Income Taxes. Currently, the AMT for corporations is designed to ensure that all corporations pay tax. A corporation’s AMT liability is determined by recalculating taxable income, making adjustments and adding back certain tax preference items. The corporation then owes tax equal to 20 percent of the AMT base, if that liability exceeds its regular tax liability. An AMT also applies to individuals.
Currently, the income from U.S. corporations’ foreign activities is taxable, although tax due to the host country is deductible from the U.S. tax liability.
This assumes that all saving is ultimately converted to assets (no bequests or debt at death), and that the value of an asset is equal to the income stream it is expected to produce:
where pt represents the asset’s price at t, dt represents the dividend paid at time t, and β is a discount factor. If individuals do not consume all their income and instead leave bequests, the bequest will be taxed when it is consumed ultimately by the beneficiary of the bequest.
See Senate bill No. 722, April 25 congressional testimony of Senators Domenici and Nunn (Congressional Record) and Senator Pete V. Domenici (1994).
Under the SEIT, capital income (interest and dividends) would be taxed at the individual level. Under the flat tax rate, capital income would be taxed at the business level.
The Alternative Minimum Tax would be repealed.
If a non-profit organization is engaged in a business activity unrelated to its tax-exempt status, it would be liable for taxes on the income from that activity under either the business cashflow tax or the Unrelated Business Income Tax (UBIT).
GATT prohibits border adjustments for direct taxes like the corporate income tax or the FICA payroll tax.
The following transactions would count as new saving: deposits to bank accounts; purchases of financial assets and equity; payment of life insurance premia; cash contributions to businesses; purchases of real assets, including homes and major home improvements; contributions to pension and profit-sharing plans; and loan repayments. Mortgage principal and interest payments would be deductible in the year those payments were made. Some form of verification would be required to assess the value of a household’s qualifying assets.
Taxpayers would not need to, itemize their deductions to deduct interest payments under the SEIT.
Medical expenses exceeding 7.5 percent of income would be deductible.
In 1991, wages, salaries, and pensions accounted for roughly 90 percent of adjusted gross income (AGI) for individuals with incomes between $30,000 and $70,000, but only 31 percent of AGI for those with incomes of $1 million or higher. See Eisner (1995), p. 10.
As noted above, OTA estimates that to achieve revenue neutrality, Armey’s flat tax would need to be levied at a 24.9 percent rate; if the earned-income tax credit were retained, a 25.8 percent rate would be required.
Source: Congressional Budget Office.
Assuming half of the burden falls on labor income and half on capital income.
Taxable income in excess of $250,000 is taxed at a 39.6 percent rate under current law.
Domenici (1994), p.284.
Flat tax proponents argue that their system would be so simple that it would be possible to file business or individual tax returns on a postcard.
However, the SEIT would introduce new distortions. The large gap between business and personal income tax rates (11 percent for business income, from 19 to 40 percent for personal income) may create an incentive for tax rate arbitrage so as to avoid taxation. For example, businesses may retain earnings rather than distributing them as dividends.
Auerbach et al. note that lump sum payments could be used to offset the transitional welfare loss to the elderly of moving to a consumption tax.
If leisure is a superior good, the labor supply curve may be backward bending and higher net-of-tax wages would reduce labor supply. Evidence suggests that this is the case: historically, increases in real wages have reduced hours.
See Tait (1991) for a discussion.