CRS Reports on Economic Issues Affecting Across-the-Board Tax Cuts

                         Order Code RL30779

[1] Across-the-board income tax cuts were an important feature of the 1999 comprehensive tax bill that was passed by Congress but vetoed by the President. They were also an important feature of President-Elect Bush's tax proposal. Distributional issues have been the most central to the analysis of across-the-board tax cuts (although issues of growth and simplification are also of concern). Some plans have been described as primarily benefitting the middle class; some plans, or even the same plans, have been criticized as unduly favoring high income taxpayers.

[2] The distributional measure used to characterize a tax cut has important consequences for how the cut is perceived. Absolute measures include tax cut per return and the distribution of the revenue cost: these measures show most tax cuts to favor high income individuals simply because income (and income tax liability) is concentrated in higher income classes.

[3] A variety of relative distributional measures are used, but the measure that indicates the change in income inequality is percentage change in disposable income. Using this measure, recent proposed tax cuts have had very different effects on inequality. A 10% cut in tax rates, for example, increases income inequality, causing disposable income in the top 20% of the population to rise by around 5%, while causing disposable income in the bottom 80% to rise by less than 1%. (Note that outside of a refundable credit, however, the bottom fifth of the distribution would have a very small tax cut from any tax change because tax liability is typically zero in that income category.) A one percentage point cut in rates still increases income in higher brackets by slightly more, but has a much smaller effect on increasing income inequality. To make a tax cut neutral or decrease inequality would require (using rate cuts) a larger percentage point cut in the lower brackets. An expansion of bracket widths would have no effect on the 70% of taxpayers who fall in the lowest brackets; an increase in the standard deduction, however, would benefit these taxpayers.

[4] These findings would be affected if the Alternative Minimum Tax (AMT) were not also reduced, especially in the future when very large fractions of higher income individuals are expected to fall under the AMT. They also depend on whether capital gains taxes are cut.

[5] Tax cuts that favor higher income individuals are also more likely to reduce marginal tax rates, which can have benefits for growth and efficiency. In some cases, however, cutting only individual tax rates can worsen distortions (such as those between corporate and non-corporate investments) or undermine desirable subsidies.

[6] Tax cuts in general do not have a lot of implications for simplification, although they do reduce the incentive to shelter income. They can complicate the tax law by increasing the number of taxpayers subject to the AMT, however. Tax cuts could also be focused on changes that provide simplification (e.g. eliminating phase outs and raising the exemption levels of the AMT). This report will be updated to reflect legislative developments.

                               CONTENTS

[7] Across-the-board income tax cuts were an important feature of the 1999 comprehensive tax bill that was passed by Congress but vetoed by the President. They were also an important feature of President-Elect Bush's tax proposal. In addition, many of the issues associated with across-the-board tax cuts are relevant to evaluating proposals to reduce the marriage penalty, which in the 106th Congress eventually took the form of across-the-board tax cuts for married couples.

[8] Distributional issues have been most central to the analysis of across-the-board tax cuts. Some plans have been described as primarily benefitting the middle class; some plans, or even the same plans, have been criticized as unduly favoring high income taxpayers. The interaction of tax cuts with the Alternative Minimum Tax (AMT) has also been an issue of growing importance.

[9] Despite the focus on distributional issues, there are also issues of efficiency and simplification associated with the enactment of across-the-board tax cuts. Lower tax rates have potential behavioral effects on incentives to work and save and also on currently tax favored activities.

[10] This report examines these economic issues, focusing primarily on distributional issues. The first section provides a general overview of the tax system. The next discusses recent proposals. The third section discusses methods of evaluating alternative types of across-the-board tax cuts; the following section discusses alternative types of tax cuts in light of these distributional analyses. The final section briefly discusses issues of efficiency and simplicity.

BRIEF OVERVIEW OF THE CURRENT TAX STRUCTURE

[11] Before describing tax cut proposals, some highlights of the current system are in order. The current individual income tax system allows personal exemptions (of $2,800 in 2000) and a standard deduction, along with $500 child tax credits that currently exempt the equivalent of $25,217 from income tax for a married couple with two children. Individuals may also itemize deductions for mortgage interest, most state and local taxes, charitable contributions and some other items. Only 30% of returns itemized deductions according to preliminary data for 1998. ¹ Wages and salaries, interest, dividends, capital gains, rents and profits from businesses are taxed.

[12] There are a variety of special tax benefits such as deductions or exemptions for individual retirement accounts and credits for child care and college tuition. An important benefit for lower income individuals is a refundable earned income tax credit (which raises the exempt level for a family of four to about $31,000). Individuals are also subject to an alternative minimum tax (AMT). Many tax benefits are phased out as income rises.

[13] There are separate rate schedules for singles, married couples (joint returns), heads-of-households, and married couples filing separate returns. About 44% of tax returns are joint, and about 43% are single, with most of the remainder (11%) head-of- household. (There is rarely an advantage to filing a married separate return). Standard deductions also vary: they are $7,350 for joint returns, $4,300 for single returns, and $6.350 for head-of-household returns.

[14] About a quarter of all returns had no tax liability in 1998. However, these shares varied across filing types. In 1997, the latest year that disaggregated data are available, 76% of returns had tax liability, representing 86% of joint returns, 77% of single returns and 42% of head-of-household returns. Some of the roughly 25% of returns with no tax liability would be eligible for tax cuts because their lack of tax liability is due to the earned income tax credit which is refundable and should not be taken into account for considering eligibility for rate reductions. Overall, 81% of all returns had tax liability before credits including 88% of joint returns, 79% of single returns, and 65% of head of household. The share eligible for tax cuts would fall between those numbers because some credits are not refundable (e.g. child care credit, dependent care credit).

[15] Tax rates apply to taxable income at 15%, 28%, 31%, 36% and 39.6%. Capital gains rates are lower than ordinary rates. As shown in table 1, tax brackets are widest for married couples and most narrow for single returns, with head of household returns falling in between.

                   TABLE 1: TAX BRACKET WIDTHS, 2000

   TABLE 2: DISTRIBUTION OF TAX RETURNS BY HIGHEST MARGINAL TAX RATE

[17] The distribution of income and income taxes paid is, however, quite different. Returns taxed at rates no higher than 15% account for 70% of tax returns, but only 19% of tax liability, and the 90% of returns that fall in the 15% or 28% bracket account for 45% of tax liability. The 5% of returns falling into the three highest brackets account for 46% of tax liability.

[18] This pattern of the concentration of the population in the lower brackets and the concentration of income tax liability and, to a lesser extent, income in the higher brackets is the crucial underlying reason for the distributional characteristics of alternative tax cut proposals. These patterns are also shown in table 3. Income in the United States has become increasingly concentrated in higher income brackets, so that the top fifth of the population received 57% of the income. (Between 1977 and 1995, according to figures from the Congressional Budget Office, real income in the highest quintile increased by 27% while real income in the lowest quintile decreased by 21%.) It is the concentration of income in these higher levels that is most responsible for the concentration of taxes, although the individual income tax, and to a lesser extent, total federal taxes, are progressive. This progressivity has increased in the lower brackets because of legislation, as well, namely the child credit enacted in 1997 and the expansion in the earned income credit in 1993.

 TABLE 3: DISTRIBUTION OF INCOMES AND TAXES BY POPULATION SHARES, 2000

RECENT PROPOSALS FOR ACROSS-THE-BOARD TAX CUTS

[20] Initial tax cut proposals made in the House in 1999 included a flat 10% across-the-board tax rate cut. That is, the 15% rate would be cut by 1.5 percentage points, the 28% rate by 2.8 percentage points, and so forth. The Senate proposed a quite different form of tax reduction: a reduction of the 15% rate to 14% and an expansion (albeit small) in the 14% bracket. The Conference Committee plan ultimately decided to reduce all tax rates by one percentage point; it also agreed to increase the standard deduction and increase the width of the first rate bracket slightly. Senate Democrats offered a proposal to expand the standard deduction. The proposals also included separate marriage penalty relief provisions that provided general tax cuts for joint returns.

[21] The general rate reductions in the House bill cost more than in the Senate bill and by the year 2009 the annual cost was estimated at that time to be twice as large as that in the Senate bill ($112 billion versus $47 billion) if the expansion of the 14% rate for joint returns is included and $112 billion versus $27 billion if the expansion of the bracket is excluded; the Senate version targeted a larger share of its benefits to married couples. Combined with all marriage penalty relief which also affects the rate structure, the House bill's provisions would have cost $117 billion in 2009 while the Senate bill would have cost $96 billion. Revenue estimates do not allow these distinctions for the Conference agreement which was, for procedural reasons, sunsetted, but in 2008, before the sunset, there was an estimated $57 billion revenue loss from the percentage point rate reduction plus an increase in the 14% bracket by for non-joint returns, that appears to account for about $13 billion of the total. Increases in the standard deduction and the size of the first bracket joint returns to make them twice the size of singles would have added a total of $28 billion, for a total of $85 billion.

[22] These bills also modified the Alternative Minimum (AMT) Tax to allow the use of personal credits to offset AMT tax liability.

[23] President-Elect Bush's tax proposal has been estimated by his staff to cost $1.3 billion over the first nine years (from FY2002-FY2010). He proposed to replace the current structure of taxes with new rate brackets and a 10%, 15%, 25% and 33% rate structure. The child tax credit would be doubled to $1000 and the phase-out of the credit will be increased from $110,000 to $200,000 for married couples and from $75,000 to $200,000 for single parents. The marriage penalty would be reduced by allowing a 10% deduction for the second earner (up to $3000). In addition, the proposal (and others announced subsequently) would eliminate estate and gift taxes, and make a number of other changes in the tax structure.

HOW TO ASSESS PROGRESSIVITY AND DISTRIBUTION

[24] Reports of the distributional effects of tax cuts sometimes appear to depict the same tax change very differently. This difference in how the cut is perceived for distributional purposes arises from the choice of distributional measure. Some of the measures that have been presented include: (1) the share of taxpayers benefitted that fall below an income level, (2) the percentage reduction in taxes paid, (3) the tax cut as a percentage of income (both pre-tax and disposable), (4) the distribution of the tax cut by income class and (5) the average tax cut. The first of these measures is most likely to tend to depict a tax change as favoring lower income individuals relative to higher income ones; the second measure is next most likely, and so forth.

[25] To illustrate this point, consider a 10% across-the-board tax cut (all positive net tax liabilities reduced by 10%). Assuming that the bottom quintile of the distribution does not have tax liability, one could describe this tax cut as one in which three quarters of the beneficiaries have income below $50,000, which might make the cut to appear not to be particularly targeted to high income individuals. Almost any tax cut that is a general one will benefit, in numbers, those outside the high income taxpayers, because high income taxpayers are, by definition, not very numerous. But this description of the tax cut does not tell us anything about how much of a tax cut different groups receive. Table 4 illustrates how such quantitative measures of the types described above would look assuming everyone in each quintile and group has the same average income (an assumption that allows the calculation of measures in the lower brackets where some individuals have negative tax liability because of the earned income tax credit). (Note, however, that this table does not account for a number of other features of the tax law, as discussed in the following section.)

    TABLE 4: ALTERNATIVE MEASURES OF DISTRIBUTION FOR A TEN PERCENT

[27] In discussing these measures that do relate to effects on income inequality, it first is important to distinguish between absolute measures and relative measures. For example, average tax reductions per unit provide information on the absolute size of a tax benefit across the income classes, which is a straightforward measure, and is shown in the last column of table 4. In this example, the second quintile has a tax cut of $23 per person and the highest quintile a cut of $2,841.

[28] Another way of examining this same effect is to compare the distribution of the tax benefit with the distribution of the population, in the next to last column of table 4. If each proportion of the population is getting the same share of the benefit, then the benefits are equal. But the 10% tax cut distributes benefits proportionally to higher income individuals, indicating that incomes are becoming more unequal on an absolute basis. Both of these measures can inform us about how a tax cut is changing income without being misleading, although it is important to remember that existing income and tax payments are more concentrated among high income individuals. Thus there is a tendency for absolute measures to show most across-the-board tax cuts as favoring higher income individuals -- because these individuals have a large proportion of the income and pay an even larger fraction of the income tax. Moreover, unless a tax provision is refundable, it will have little benefit for the bottom fifth of the population.

[29] A different type of measure is a relative one that tries to examine how the tax benefit is changing the overall relative distribution of income in the country -- that is, is it making income shares more equal or less equal? In this case, a tax change that does not alter distribution provides tax benefits to different income classes in proportion to some measure of income. (Higher income individuals would still receive high absolute tax cuts but not higher tax cuts as a percentage of income.) In general, the best method for measuring this type of effect on inequality is to examine the percentage change in disposable (after-tax) income. If the percentage change is equal, then the tax change is not making incomes shares more equal or less equal. If the percentages are higher among higher income individuals the change is making incomes shares less equal. Clearly, the across-the-board proportional tax cut is increasing inequality measured by the relative concept: incomes in the lower brackets are increased by considerably less than 1%, while incomes in the higher brackets are increased by 2% or more.

ALTERNATIVE PROPOSALS

[30] There are many types of across-the board tax cuts. In addition to the proportional rate cut discussed previously, possible tax cuts include equal percentage point reductions in all brackets, larger percentage point reductions in the lower brackets, increases in standard deductions and the width of brackets, increases in credits, and making existing credits refundable or introducing new refundable credits.

[31] Tables 5 and 6 use two measures: absolute change in tax liability and relative change in disposable income, to illustrate the consequences of alternative tax proposals on an illustrative family (married couple with two children). Illustrations are shown for incomes at approximately the 25th, 50th, 75th, 90th, 95th and 99th percentile. Four proposals that are estimated to involve roughly the same magnitude of cost are considered: a 10% cut in rates, a 2.15 percentage point reduction in all rates, a 3.8 percentage point reduction in the lowest (15%) bracket, and an expansion of the 15% bracket through 93% of the existing 28% bracket. Each type of proposal has been considered recently in tax legislation (although the magnitudes were different). The House-passed version of the 1999 tax cut bill contained a 10% cut in rates; the Senate Finance Committee version contained a one percentage point cut in the bottom rate; an amendment on the Senate floor extended the lowest rate bracket, and the final version included a one percentage point reduction in all rates. (Note that the distributional effects of the overall bills, which contained many other provisions, including some broad changes for joint returns, cannot be inferred by comparing the single provisions.)

[32] Table 5 shows the dollar tax cut for each type of proposal. In the very highest percentile, the largest tax cut comes from the 10% across the board tax cut. This result is not surprising, since a 10% tax cut would cut the top rate by 3.96 percentage points, the next lower by 3.6 percentage points, and so forth. The tax cut most favorable to lower income and moderate income individuals (all the way through the 75th percentile) is cutting the bottom bracket rate. An equal percentage point cut in rates falls in between these two. The extension of the 15% bracket provides no benefit for more than half of taxpayers, because most of them do not fall in the 28 percent bracket, but it is most beneficial for high income individuals except the very wealthy.

      TABLE 5: EFFECTS OF ALTERNATIVE TAX CUTS ON TAX LIABILITY,

[33] Table 6 provides a measure of relative equality by examining the percentage change in disposable income. The 10% cut in rates clearly redistributes after tax income to favor high income individuals. The equal percentage point tax cut also redistributes income to favor higher income individuals, although not to the degree of the 10% tax cut. Extending the 15% bracket also tends to make incomes more equal except at extremely high incomes. Only the 3.8 percentage point cut in the lowest rate tends to favor middle and lower income individuals. Note that no change (short of increasing a refundable credit) could have a larger effect on the half-median (25th percentile taxpayer) because their tax liability is eliminated by the rate cut (because of the $1000 in child credits).

      TABLE 6: PERCENTAGE CHANGE IN AFTER TAX INCOME, ALTERNATIVE

[34] What sort of income tax cut would be required to be neutral with respect to the distribution of disposable income? Increases in the standard deduction coupled with an equal percentage increase in rates would probably be most likely to accomplish this effect although individuals that are excluded from the tax would require an increase in a refundable credit (such as the earned income tax credit). Rate cuts that reduce lower bracket rates by larger percentage points that higher bracket rates would also tend to be more neutral across the income classes.

[35] It can also be difficult to evaluate a tax cut that simultaneously changes rates and brackets and does not change them in a proportional way. For example, the proposal of President-Elect Bush would substitute a rates of 10, 15, 25 and 33 for the existing brackets of 15, 28, 31, 36 and 39.6. However, the 10% bracket (cutting the 15% bracket by a third, or five percentage points) would apply to only the first $12,000 (about a fourth) of the existing 15% bracket; remaining income in that bracket would receive no tax cut. The 25% rate would cut the existing 28% rate by 3 percentage points, but the existing 31% rate by 6 percentage points. The 33% rate would cut the 36% bracket by 3% and the 39.6% bracket by 6.6%. Based on the incomes reflected in the table above, the tax cuts would be (beginning with the lowest percentile), $268, $600, $776, $2,017, $4,972, and $14,207. The percentage increases in after-tax income would be 1.1, 1.5, 1.3, 2.2, 4.2 and 4.7%. Basically, the percentage increase at the bottom is smaller because only part of income is taxed and there is a limit to tax liability. The percentage change in after tax income would rise (to around 2% as the new $12,000 bracket is exhausted at around $30,000 of income), then fall again to a bottom of 1.1% for those currently at the top of the existing 15% bracket at around $69,000, then rise again, reaching a peak at the top of the 31% rate bracket at around $200,000 and then falling again and then rising. (Note that this analysis also does not include other features of the Bush proposal.)

[36] The illustrations provided in this section are designed to illustrate the general nature of the distributional effects of various types of tax cuts. These examples are simplified because they do not account for a number of other features. Three of the most important ones are capital gains, the AMT, and the child care credit.

[37] For high income people, a significant fraction of income is received in capital gains, which are taxed at lower rates. If capital gains tax rates are cut along with the rest of the tax rate structure, dollar reductions will be nevertheless be smaller at higher income levels than presented in table 5; they will be even smaller if capital gains tax rates are not cut. In table 6 the percentage changes will be smaller at higher income levels. Omitting this effect does not alter the overall conclusions derived from table 5 regarding inequality, but it does change the magnitudes.

[38] Only a small portion of the population overall pays the minimum tax, although rates are higher in high income classes. However, the coverage of the tax is rising due to the expiring of provisions allowing personal credits to offset the AMT and the continual erosion of the AMT exemption. Currently, 1.3% of taxpayers are projected to fall under the AMT in 2000, although over 15% of tax payers in the $100,000 to $200,000 income class are AMT taxpayers. These shares will increase substantially, to an overall share of 15.7% and as much as 64% in some high income brackets. ² Thus the distributional consequences of any tax cut will depend on whether and how much these rates are cut (or exemptions changed). If AMT rules are not revised, tax cuts for higher income individuals or individuals with certain types of characteristics (families with many children) will be smaller than indicated in the tables.

[39] The final provision is the dependent care credit. This credit increases the exempt level for families who use it and means tax reductions are less likely to benefit lower income individuals. For example, the tax burdens of the 25th percentile taxpayer could be easily eliminated through this provision, which is particularly important for married working couples and heads-of-households. In the latter case, there are many tax returns concentrated at these lower income levels.

EFFICIENCY AND SIMPLICITY

[40] The central issue in across-the-board tax cuts discussed in this paper is the distributional effect, which finds that certain types of across-the-board tax cuts are more likely to redistribute income in certain directions. For some, there is a significant concern about a tax cut that favors higher income individuals.

[41] At the same time, economic theory suggests that tax cuts that tend to favor higher income individuals are more likely to achieve efficiency gains (reduce distortions in behavior the most) and behavioral responses (increases in savings and work effort). These tax cuts favoring higher income individuals have larger reductions in marginal as compared to average rates than do cuts favoring lower income individuals. The marginal tax rate governs the substitution effect, where lower tax rates are expected to increase labor supply and savings; the average tax rate governs the income effect, where lower tax rates tend to decrease labor supply and savings. Thus, as is frequently the case, distributional and efficiency objectives may conflict. Those tax cuts that most favor lower income individuals, such as reducing the lower bracket rate or increasing the standard deduction, tend to have smaller effects, per dollar of revenue loss, on the marginal tax rate, where cuts might increase savings and labor supply.

[42] Despite concerns about behavioral responses, most evidence suggests that labor supply and savings are unlikely to be very responsive to tax cuts. There is some evidence that married women's labor supplies are responsive to tax cuts and some theoretical reasons to believe that a shift to a consumption tax could increase savings, but there are also a number of controversies surrounding these issues. ³

[43] Reductions in marginal tax rates would also have effects on the composition of investment or consumption, some of which may be desirable and some of which may not. For example, present law favors owner-occupied housing as both a consumption and investment good and lower tax rates would reduce that distortion. Present law also favors certain forms of passive investment (e.g. pensions and IRAs) relative to investments not targeted for retirement living, and a reduction in tax rates would reduce that distortion. However, lower individual tax rates would magnify the favorable treatment of unincorporated businesses relative to corporate investments. Tax favoritism can also produce allocations that may improve on the market allocation (such as charitable contributions) and these beneficial re-allocations would be reduced.

[44] Simplicity issues do not always involve a tradeoff. On simplicity grounds, across-the-board rate cuts, which do not add complication, are preferable to many targeted tax provisions that increase tax complexity. There are also a variety of tax cuts that favor either lower or higher income individuals that could simplify the tax law. For example, an increase in the standard deduction would direct benefits toward lower and moderate income individuals, and would also simplify the tax law by increasing the number of individuals who choose not to itemize deductions. For higher income individuals, there are a number of provisions in the tax law that are phased out and that complicate the tax law. The tax law would be simplified if tax cuts to higher income individuals were provided by eliminating these phase-outs. Similarly, allowing personal credits to be used against the AMT and increasing AMT exemption levels would favor higher income individuals but would simplify the tax law.

[45] In any case, reducing rates or expanding income brackets can be done without any important complications, and lower rates themselves reduce incentives to shelter income.

CONCLUSION

[46] This analysis has indicated that different types of across the board tax cuts can have substantially different effects on the distribution of income. Tax proposals that cut rates proportionally tend to redistribute income towards higher income individuals whether one is considering absolute or relative measures. Tax proposals that cut rates by equal percentage points, also redistribute income towards higher income individuals, but by a lesser degree. Proposals that favor lower income individuals include increases in the standard deduction, increases in credits and especially in refundable credits, and cuts in the lower bracket rates. Proposals that expand the rate bracket width provide no benefits to lower and moderate income individuals.

[47] Tax proposals that favor higher income individuals may involve more efficiency and larger behavioral responses, but most evidence suggests these responses are not very large. Across the broad tax cuts do not tend to complicate the tax law, but tax cuts for high income individuals could be used to simplify the tax if they were directed at phase-outs and the AMT rather than rate cuts in higher brackets.

² Robert Rebelein and Jerry Tempalski. Who Pays the Individual AMT? U.S. Department of Treasury, Office of Tax Analysis, OTA Paper 87, June 2000.

³ For a brief survey of the literature, see James P. Zeliak, "Labor Supply and Taxes," and Douglas Bernheim and John Karl Scholz, "Savings and Taxes" in the Encyclopedia of Taxation and Tax Policy, ed. Joseph J. Cordes, Robert D. Ebel and Jane G. Gravelle, Washington, D.C., The Urban Institute, 1999. See also Eric Engen, Jane Gravelle, and Kent Smetters, "Dynamic Tax Models: Why They Do the Things They Do?" National Tax Journal, vol. 50, September 1997, pp. 657-682.