Library / Publications
Federal Income Tax Cuts and Low-Income Families
The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.
Recent large federal budget surpluses have spurred debate on how best to use the money. Some see surpluses as an opportunity to pay down the federal debt, to expand programs to meet such social needs as health care and education, or to secure the future of Social Security. Others propose returning some of the money to taxpayers by cutting federal income taxes, an option championed by President Bush that has widespread support in Congress.
Proponents of tax cuts emphasize that this option would help all families. Many popular options for cutting federal income taxes, however, would provide little benefit to low-income families. This is not surprising, given that these families pay little or no income tax and thus receive little benefit from proposals that only reduce positive income tax liabilities. Although most low-income families do not pay federal income taxes, they do work and pay federal payroll and excise taxes as well as state and local taxes. To benefit these families, income tax cuts must be refundable, that is, available in excess of the families’ current income tax liability.
Study Focus and Purpose
To better understand the impact of various tax reduction proposals on low-income families, we identify the overall tax burdens that low-income families currently face and explain how those tax burdens would change if certain types of federal income tax cuts were enacted. Using detailed household-level data on incomes and taxes, we show how federal income and payroll taxes differ for low-income families and how these families benefit from certain features of the income tax, such as the Earned Income Tax Credit (EITC)—a tax credit applied to the wages of low-income families.
We examine five tax cut options that are representative of recent proposals and compare the benefits to low-income families with those for higher-income taxpayers. Each option is designed to reduce income taxes by approximately the same amount—$34 billion in 1998. Because the analysis is concerned with the effects of income tax cuts on the well-being of different types of families (particularly lower-income families), the analysis focuses on the effect of such cuts on average tax rates rather than marginal tax rates.
We examine the following generic proposals:
For low-income families, the most important element of any tax reduction proposal is whether the tax relief is refundable—that is, available in excess of their current income tax liability. For that reason, the options to increase the EITC or create a new refundable payroll tax credit are most beneficial to low-income working families. Across-the-board cuts in federal income tax rates or increases in the standard deduction and the width of the lowest tax bracket provide very little benefit to low-income families. Increasing the child credit as currently structured also would offer little relief.
Rate Cuts and Other Across-the-Board Tax Cuts: Higher-income families would receive a large share of the benefits from the rate cut option and the option to increase standard deductions and the income range for the lowest tax bracket. Almost 40 percent of the tax cuts from the latter option would go to the 15 percent of people in families with incomes over $100,000, while over 50 percent of the benefits from an across-the-board cut in tax rates would go to those families. While the share of benefits from an across-the-board rate cut exceeds the population share for higher-income families, it is less than their 60 percent share of income taxes and only slightly more than their share of combined income and payroll taxes.
Because higher-income families pay a larger share of their income in taxes, an across-the-board cut in income taxes benefits higher-income families more than lower-income families if those benefits are measured in absolute dollars per family or as a share of income, but not if the benefits are measured as a percentage change in income tax liability. Measuring the changes in tax burdens as the percentage decrease in tax liability can be misleading, because it suggests that low-income families would benefit the most from small tax cuts simply because they pay so little tax in the first place.
Increasing the EITC and a Payroll Tax Credit: The options to increase the EITC or to create a new refundable payroll tax credit would have the biggest impact on low-income families by boosting the amount of their refunds. For the proposals simulated here, 18 percent of the benefits from increasing the EITC would go to families with incomes of $20,000 or less in 1998, while these families would receive 20 percent of the benefits from the option to create a new refundable payroll tax credit. Over 75 percent of the benefits from the EITC option would go to families with incomes of $40,000 or less; these families would receive about 60 percent of the benefits from the payroll tax credit option. About one-fifth of the population had family income of $20,000 or less in 1998, and about 45 percent of the population had income of $40,000 or less.
Increasing the Child Tax Credit: We considered two options to increase the child tax credit—one would increase the credit to $1,300 per child, the other would increase it to $1,200 per child and make it refundable for all families. Both would provide the largest share of benefits to middle- and upper-middle-income families. About 65 percent of the benefits from these options would go to families with incomes between $40,000 and $100,000.
Impacts on Different Family Types: Not all types of families would benefit from the proposals simulated here. In particular, couples and singles without eligible children under age 17 would see no change in their taxes from either proposal to increase the child tax credit. Single adults would benefit the most from more broadly based options, including the rate cuts, the increase in standard deductions and the income range for the 15 percent tax bracket, and most of all from the payroll tax credit. Single parents with children would benefit the most from the option to increase the EITC and the least from the option for a proportional cut in income tax rates.
Other Policy Considerations
Whether or not a tax reduction disproportionately benefits low-income or high-income families is only one criterion by which to judge its merits. The standard principles of good tax policy still apply. Other goals, such as economic efficiency, tax simplicity, and reduction of compliance burdens on taxpayers, are important as well. Proposals aimed at helping low-income families do not necessarily advance these other goals and may detract from them. When benefits are targeted to particular taxpayers, there must be rules determining who is eligible, which in turn increase the complexity of taxes and create opportunities for errors and abuse.
Clearly, fairness does not dictate that all tax relief should be directed to low-income families. However, these families should not be left out of the equation. Congress must first compare tax relief with other options for using the budget surplus, such as paying down the federal debt or addressing the future insolvency of Social Security and Medicare. If a tax cut is the best option, lawmakers must consider the merits of targeting relatively more tax relief to higher-income families who have enjoyed extraordinary prosperity over the past decade when there are so many other families who work and pay taxes, but do not enjoy the same economic security.
The recent "good surprises" of large government budget surpluses have led to proposals to cut both federal and state income taxes. Many states have already enacted significant tax reductions, often forced by budgetary tax limitation rules. At the federal level, Congress recently debated and passed legislation that would have reduced income taxes, but was unable to reach agreement with President Clinton on the appropriate size of those tax cuts. Both major party presidential campaigns proposed substantial reductions in the income tax. With the backing of the new administration and continued widespread support in Congress, enactment of a major tax cut in the coming year is very likely.
Not all income tax cuts are alike, however, and there is reason for concern that many families will not share the benefits from future federal tax reductions, despite claims to the contrary. Some recent proposals, in particular, would fail to provide much tax relief to low-income working families. The reasons are simple:
1. Low-income families pay little federal income taxes. Many of these families have a zero or negative income tax burden because of the Earned Income Tax Credit (EITC). Consequently, even broad-based tax cut proposals that would provide the highest-percentage cuts in income taxes to taxpayers with the lowest incomes would not be especially helpful to low-income families. Large-percentage income tax cuts result in little or no change in a family’s after-tax income when the family has little income tax liability to begin with, and after-tax income, not taxes paid, determine a family’s economic well-being. It takes only a few dollars to give low-income families a high-percentage reduction in their federal personal income taxes.
2. Low-income families do pay federal taxes, but for them, other federal taxes matter much more than income taxes. Most of the federal taxes that these families pay are payroll and excise taxes, not income taxes, and those taxes can amount to a sizable fraction of family incomes. Reducing or eliminating positive income tax liabilities has very limited effects on their overall federal tax burdens.
3. State and local taxes matter. Overall tax burdens on low-income families depend on state and local taxes as well, and including these further diminishes the significance of federal income taxes relative to other types of taxes for these families. Retail sales taxes, in particular, make up about half of state tax revenues and place significant burdens on these families.
In this paper we identify the overall tax burdens that low-income families currently face and show how those tax burdens would change if certain types of federal income tax cuts were enacted. We identify the "low-income" segment of the population and show how their incomes and taxes compare with those of the rest of the population, based on detailed household-level data. In doing so, we show how the impact of federal income and payroll taxes differs for low-income families relative to higher-income families, and how low-income families benefit from certain features of the income tax (such as the EITC).
We then evaluate several generic types of broad-based income tax cuts, including an across-the-board reduction in marginal tax rates, an increase in standard deductions and the income range for tax brackets, higher child credits, an expansion of the earned income tax credit, and a refundable payroll tax credit. We show the importance of refundable tax relief to low-income families—that is, relief available in excess of families’ current income tax liability. We show how much different income tax cuts would benefit low-income families and compare this with the benefits for higher-income taxpayers.
To provide an appropriate comparison among different forms of tax relief, we compare options with approximately equal revenue losses. This comparison recognizes that all tax cuts reduce money available for current and future federal programs, including many programs that benefit low-income families. Therefore, in determining which tax cuts help these families the most, it is important to compare proposals that have equal overall budgetary impacts.
Federal Taxes for Low-Income Families
As a percentage of family income, federal income taxes for low-income families are at a 20-year low (see figure 1). Indeed, for the 20 percent of families with the lowest incomes, federal income taxes including refundable tax credits represent a net transfer of nearly 7 percent of income before tax. Income taxes as a percentage of income are also lower for middle-income families than at any time in the past 20 years, although the differences are not as great as for low-income families. Measured as a percentage of family income, upper-income families now pay about the same income tax as similar families 20 years ago—a slightly higher percentage than in the mid-to-late 1980s.
Two changes have contributed the most to reducing income taxes paid by low-income families. First, increases in the personal exemption and standard deductions enacted in 1986 removed many low-income families from the tax rolls. The level of inflation-adjusted income at which a family of four begins to pay tax is nearly double what it was before 1986 and is significantly above the poverty level of income. Second, several tax bills, especially in 1990 and 1993, greatly increased the earned income tax credit. The EITC is a tax credit applied to wages of low-income families; it phases out as income rises above a certain level. It provides the largest benefits to workers with children, but a small credit is also available to childless workers.1 Because the EITC is specifically designed to assist low-income working families, the recent expansions of the credit have had profound effects on the overall tax burdens faced by such families. About 19 million taxpayers now receive the EITC, claiming nearly $30 billion in credits. About $23 billion of the EITC represents net refunds (credits in excess of positive income tax liabilities).
Increases in the personal exemption and standard deductions also have helped to keep down federal income taxes for middle-income families. Legislation enacted in 1997 provided additional tax relief for middle-income families through the introduction of new deductions and credits. Much of the tax benefits went to families with children—in the form of a new child tax credit. Beginning in 1998, families could claim a credit of $400 ($500 in subsequent years) for each eligible child under the age of 17. The credit phases out for higher-income families with incomes above certain thresholds.2 Credits in excess of income tax liability are not refundable for families with one or two children. The credit is refundable for families with three or more children, but only up to the amount by which their Social Security payroll tax exceeds their EITC.
Because few families qualify for a refundable credit, the child credit provides little tax relief for many low-income families with children. Families with no income tax liability (and fewer than three children) receive no benefit from the credit, while families with low income tax liability receive only partial benefits. A family with two children and income tax liability of $400, for example, would not receive a full $1,000 in credits; $600 would go unused. Because the credit applies against income tax liability before subtraction of the earned income tax credit, however, a family can increase the portion of the EITC that is refundable by taking the child credit. If a family with income tax liability of $400 were also eligible for a $2,100 EITC, without the child credit, $400 of the EITC would be used to offset income tax liability and the family would receive a refund of $1,700. With a child credit of $400 or more, the family could be refunded the entire $2,100.
In 1998, approximately 25 million families claimed full or partial child credits, totaling over $15 billion. This was about 65 percent of all families that filed a tax return and had at least one child under the age of 17. Most of the 35 percent of ineligible families with children had low incomes and no income tax liability, but a smaller number were high-income families with incomes above the threshold levels. Few families received refundable child credits, both because refunds were limited to families with three or more children and because, even for those families, the amount of the refund was limited to the amount of their payroll taxes in excess of the EITC. About 700,000 families received refundable child credits (reported separately as Additional Child Tax Credits) totaling $0.5 billion.
In contrast to the personal income tax, Social Security payroll taxes are higher for families at all income levels than they were 20 years ago. Tax rate increases enacted in 1977 and put in place throughout the 1980s raised payroll taxes for all working families. Increases in the maximum amount of earnings subject to the tax, also enacted in 1977, and the elimination in 1997 of any maximum for earnings subject to the Hospital Insurance (HI) portion of the tax further increased payroll taxes for upper-income families.
Unlike income taxes, payroll taxes are about the same percentage of income at most income levels. Social Security payroll taxes are levied at a single rate paid by both workers and their employers—currently 6.2 percent of earnings for the Old Age, Survivor, and Disability Insurance (OASDI) programs (commonly know as Social Security) and 1.45 percent of earnings for the Hospital Insurance program (Medicare). Because there is a single marginal tax rate (unlike the graduated rates in the income tax) and because there are no exemptions or deductions, taxes as a percentage of income do not rise as income increases. Payroll taxes as a percentage of income are somewhat lower for higher-income families than for lower- and middle-income families because the tax is levied only on earnings, allowing income from investments to escape the tax, and because the OASDI portion of the tax exempts earnings above a maximum level ($76,200 in 2000).
Although payroll taxes make up a significant portion of the total federal taxes that families pay, Social Security payroll taxes for OASDI are sometimes viewed differently from income and HI taxes. Both current payroll taxes and future retirement benefits rise with current earnings. Although all taxes are used to pay for government spending of one type or another, there is no comparable relationship between taxes paid and benefits received for the income tax or the HI portion of the payroll tax. Thus, an alternative way to measure the burden of OASDI taxes would be to count net taxes as the difference between the tax and the expected increase in future Social Security benefits.
A dollar of additional payroll taxes, however, does not generally result in a dollar of additional benefits and produces no benefit increase at all in some cases. The Social Security program is not a defined contribution plan and payroll tax contributions are not credited to an individual’s personal account. Current payroll taxes are used to pay current benefits, much the same as in any spending program. While the benefit formula does give higher benefits to workers with higher lifetime earnings, the rules relating benefits to earnings are complex. How much an additional dollar of earnings raises future retirement benefits depends both on the average monthly earnings of the worker and on whether the worker will receive spouse or survivor benefits. There is no guarantee the benefit formula will remain fixed. Congress is free to change these rules and has done so many times in the past, to both the advantage and disadvantage of future retirees. It may do so again in the future to correct for the imbalance between projected Social Security benefits and taxes.
About three-fourths of U.S. families pay more in payroll taxes than in income taxes.3 Despite the increased burden of payroll taxes, however, many low-income families pay no net tax because they receive an EITC that exceeds the sum of their income and payroll tax liabilities. For example, a four-person family with earnings equal to the federal poverty level would have received a net income tax refund (in excess of all income taxes paid) of about $2,900 in 1999 (table 1). In comparison, that same family would have paid about $1,300 in federal payroll taxes, counting only the employee portion of Social Security (OASDHI) payroll taxes.4 Likewise, a single parent with two children and poverty level earnings would have received an EITC refund of over $3,600 while paying about $1,000 in payroll taxes, for a net transfer of almost $2,600. Combined income and payroll taxes are negative for almost all families with children and poverty level incomes, but single workers with incomes at the poverty level pay positive income taxes in addition to payroll taxes, in spite of the small refundable EITC available to them.
The federal government collects other taxes as well, and determining how much of these other taxes families pay is a thornier task. Corporate income taxes and excise taxes are imposed on businesses, but economists recognize that the burden of these taxes must ultimately fall on households in the form of higher prices or lower incomes. The dominant view in the economics literature is that corporate taxes primarily burden people who own capital, which makes these taxes less important for lower-income families than for others. In contrast, economists generally believe that the burden of excise taxes on specific goods, such as gasoline, alcoholic beverages, and tobacco products, falls disproportionately on low-income families. This occurs for two reasons. First, low-income families spend a larger fraction of their income than do higher-income families. Second, the types of goods that are taxed often make up larger fractions of low-income families’ total expenditures.
The bottom line is that rich and poor people pay different federal taxes. As a share of income, federal income tax rates are much higher for high-income than for low-income families (indeed, many low-income families receive refundable credits rather than pay income tax). In contrast, payroll and excise taxes impose higher burdens relative to income on low-income families.
The State Tax Burden for Low-Income Families
Although state taxes as a share of personal income are less than one-third of federal taxes, low-income families in many states pay more state taxes than federal taxes for two reasons. First, states raise a much larger share of their revenue from expenditure-based taxes than does the federal government. As mentioned above, expenditure-based taxes (such as excise taxes) impose a much larger share of the tax burden on low-income families than income taxes. Second, state income taxes are typically less progressive than federal income taxes.
In 1999, states on average collected about one-third of their tax revenues from personal income taxes, one-third from general sales taxes, and another 15 percent from selective excise taxes on such items as alcohol, tobacco, and gasoline (table 2). Thus, combined taxes on expenditures accounted for nearly one-half of state tax revenues, compared with only about 4 percent of revenues at the federal level. Seven states had no personal income tax, while four states (Massachusetts, New York, Oregon, and Virginia) collected more than half their revenues from the personal income tax. Five states had no general sales tax, but all the states collected some revenue for selective excise taxes, including taxes on tobacco, alcohol, and gasoline.
Because of faster-than-expected revenue growth and expanding budget surpluses, many states have reduced taxes in the past several years. Some have targeted a portion of that tax relief to low-income families.5 Almost all states with an income tax use some combination of personal exemptions, standard deductions, and no-tax floors to exempt a base amount of income from the tax, and many states have increased the combined amount of these exemptions and deductions. More than half the states with an income tax now have some type of low-income tax credit. By 2000, 14 states and the District of Columbia had in place a state earned income tax credit based on the federal credit. The state EITC was refundable in 10 of those jurisdictions. (Eleven of the 14 states and the District had either enacted or expanded their EITC since 1997.6) Some states currently provide further income tax relief to low-income families through tax credits tied to certain expenditures, such as child care, property taxes, and rents.
Despite increased deductions and expanded use of low-income credits, state income taxes for low-income families remain higher than federal income taxes in many states. Tables 3 and 4 show calculations of state income taxes in a number of representative states for a married couple with two children and a single-parent family with two children at various income levels. The calculations include income-based tax credits, such as the EITC and working family credits, but do not include credits based on expenditures. While families with poverty-level incomes either pay no income tax or receive refundable credits in most of the selected states, income taxes are positive for both the married-couple family and for the single-parent family in 4 of the states.7 At 150 percent of the poverty level, taxes in all but 4 of the 11 selected states with income taxes (California, Colorado, Massachusetts, and Minnesota) are greater than zero for the married-couple family and, in all but 5 states (the previous 4 plus New York), for the single-parent family. Either type of family with income equal to 100 or 150 percent of the poverty level would receive a sizeable net rebate from federal income taxes.
Low-income families often pay more state sales and excise taxes than state income tax. Most state sales taxes apply to a broad range of retail purchases. Many states exempt certain "necessities," such as food, clothing, and medicine, as an attempt to reduce sales taxes for low-income families. Although these items generally make up a larger portion of the family budget in low-income households than in others, exemptions are an imprecise way to target sales tax relief to these families. All families spend some portion of their income on exempt items and even though budget shares decline with rising income, the absolute amount spent increases. Some states, including Colorado, Minnesota, and Wisconsin, have provided income-based sales tax credits.
Tables 5 and 6 show calculations of 1999 sales tax payments in the representative states, again for a married couple with two children and a single parent with two children at various income levels. The calculations are based on national estimates of average family expenditures by income and family size taken from the Bureau of Labor Statistics Consumer Expenditure Survey and do not make adjustments for differences in spending across states.8 The calculated amount of tax includes only general sales taxes; it excludes excise tax payments on such items as tobacco, alcohol, and gasoline. The calculated amounts also do not include local sales taxes, which are significant in a number of states.
Sales taxes are higher than income taxes in almost all of the selected states for a couple with income less than 150 percent of the poverty level and for a single parent with income below 200 percent of poverty. There is much less variation in the amount paid in sales taxes by families at various income levels than in the amount paid in income taxes. Even in the selected states that exempt food and clothing, sales taxes are a much larger percentage of family income for the lowest-income families than for families with higher incomes.
Comparing Options for Federal Income Tax Relief
Congress last enacted a significant package of income tax cuts in 1997. Most of those tax reductions were in the form of credits or deductions targeted towards particular families (families with younger children) or towards particular activities (saving for educational expenses or retirement). Burgeoning federal budget surpluses provided an impetus for proposals to further cut federal income taxes, although as the economy slows some tax cut proposals have been recast as economic stimulus packages. The leading proposals thus far have called either for broad-based relief through rate reductions or a further round of targeted credits and deductions, including increases in the child tax credit and the EITC. In addition to income tax reductions, there is considerable support in Congress for reducing federal estate and gift taxes, and some federal excise taxes, such as the tax on telecommunications. There has thus far been little interest in reducing payroll taxes.
We consider several options for reducing federal income taxes that reflect the range of current proposals. There are a great variety of details associated with specific tax relief packages, and we do not attempt to simulate actual plans. We instead select representative options from the types of tax cuts that are characteristic of recent proposals and show how they would affect low-income families. The options include an across-the-board rate cut, an increase in standard deductions and a widening of the income range for tax brackets, an increase in child credits, an increase in the earned income credit, and a new payroll tax credit.
We simulate the effect of each option in tax year 1998. Each option is designed to reduce revenues by approximately the same amount—$34 billion (table 7). This would be about a 2 percent cut in total federal revenues and equal to about 0.4 percent of Gross Domestic Product (GDP). These tax cuts would use up only a fraction of the projected federal budget surplus. In July 2000, for example, the Congressional Budget Office projected that the fiscal year 2001 on-budget federal surplus (the surplus excluding revenues and spending of the Social Security Trust Funds and transactions of the Postal Service) would range between 1.0 and 1.6 percent of GDP. More recent estimates project an even larger surplus.
The six specific options we simulate are:
1. Cut income tax rates by 5 percent. This option would reduce income tax rates on ordinary income by 5 percent but keep the rates on long-term capital gains the same as under current law.9 The option assumes that the tax rates for the alternative minimum tax (AMT) are also cut by 5 percent so that people do not switch over to the AMT (see box).
2. Increase the child credit to $1,300 per child. This option would increase the $400 child credit in 1998 ($500 in later years) to $1,300 per child. The credit would remain partially refundable for families with three or more children and nonrefundable for families with fewer than three children. It would continue to phase out for higher-income taxpayers starting at the income thresholds in current law.
3. Increase the child credit to $1200 per child and make it refundable for all taxpayers. This option would increase the child credit to $1,200 per child and extend refundability to families with fewer than three children. Current rules regarding refundability would still apply. The amount of the credit in excess of a family’s income tax liability would still be refundable only up to the amount by which the family’s Social Security payroll taxes exceeded any EITC. The option would retain the current law credit phase-out for higher-income taxpayers.
4. Increase the standard deduction and extend the income range for the 15 percent bracket by 18 percent. This option would increase standard deductions and lengthen the 15 percent tax bracket for single, married, and head-of-household filers.10
5. Increase the EITC phase-in rate by 25 percent and double the phase-out income threshold. This option would raise the maximum EITC by 25 percent by increasing the phase-in rates by 25 percent. It would also extend the credit to higher-income families by doubling the income threshold at which the credit begins to phase out.11
6. Refundable Social Security payroll tax credit capped at $500. This option would create a new refundable credit equal to the employee share of payroll taxes (7.65 percent of wages) up to a maximum credit of $500. Each working spouse in a couple filing a joint return would be able to claim the credit. Thus, a married couple could receive a credit of up to $1,000. The credit would phase out at a 5 percent rate for couples with adjusted gross income (AGI) in excess of $42,000 and for other taxpayers with AGI in excess of $25,200.
Methodology for Simulating the Options
In order to evaluate the effects of income tax changes on different types of families, we determine the distribution of income and taxes using a model that accounts for the specific socioeconomic characteristics of households. The data in the model come from the March 1999 Current Population Survey (CPS) of the U.S. Census Bureau and are enhanced by imputations from other government databases. The simulation model used is the Urban Institute’s Transfer Income Model (TRIM). TRIM computes income and payroll taxes based on detailed income information from the CPS and on imputations of itemized deductions and other tax variables using Internal Revenue Service (IRS) data from individual income tax returns.
We measure baseline taxes and the change in tax burdens for each proposal in 1998. The current version of TRIM simulates many, but not all, of the provisions enacted in the Taxpayer Relief Act of 1997. In particular, the model does not simulate the HOPE and lifetime learning educational expense credits. We show average tax rates (tax liabilities divided by income before taxes) before and after each reform by income group and family type. Income before tax includes all cash income reported by the family on the CPS, plus realized capital gains imputed from tax return data.14 Even with the imputations from tax return data, the model does not measure the full amount of income reported to the IRS by higher-income taxpayers. Average income and taxes for the highest income group are thus lower than the average for the same group computed solely on the basis of tax return data. Because we measure Social Security payroll taxes as the combined employee and employer share of the tax, we include the employer share of the tax as part of before-tax income.
In addition to showing changes in effective tax rates for each of the proposals, we show the share of the benefits going to different income groups and types of families. We also compare measures of the percentage changes in tax liabilities with the percentage change in after-tax incomes.
Because our analysis is concerned with the effects of income tax cuts on the well-being of different types of families (particularly lower-income families), we focus on the effects on average tax rates rather than marginal tax rates. The marginal tax rate is the tax imposed on the last dollar of income and is important for assessing the effect of taxes on incentives to change work effort, saving, and other behaviors.
Some of the options would reduce marginal tax rates or increase marginal tax subsidies. Cutting statutory tax rates by 5 percent would increase the returns to work and saving, but the change would be slight and would probably have little measurable effect on behavior. The after-tax return to work and saving would increase the most for those in the highest tax bracket, but only by just over 3 percent. For taxpayers in the lowest tax bracket, after-tax returns would increase by less than 1 percent.15 The option to extend the income range for the 15 percent tax bracket would increase after-tax returns by a larger amount for those taxpayers who would otherwise face a marginal tax rate of 28 percent. For them, the after-tax return to additional work effort or saving would rise by 18 percent. The biggest change in work incentives for a substantial number of taxpayers would come from the option to increase the phase-in rate for the EITC by 25 percent. Low-income workers with earnings below the amount at which they would receive the maximum credit would receive a 25 cents higher subsidy for each dollar of additional earnings and thus have more of an incentive to increase their earnings up to the point where they qualified for the maximum EITC. Research has shown that the existing EITC has had a substantial positive effect on hours worked by low-income workers, particularly for single mothers.16
Other options would either increase marginal tax rates or at least extend higher marginal tax rates to more families. Taxpayers’ marginal tax rates depend not only on the statutory tax rate in each tax bracket but also on whether additional income would cause them to lose a portion of their tax credits or deductions. Taxpayers in the phase-out range for the EITC or the child tax credit face higher marginal rates because each additional dollar of income causes them to lose some portion of those credits. The options to raise the maximum amount of the EITC or the child credit would maintain the current phase-out rates but would extend the phase-out over a much wider income range and thus cause more families to face the higher marginal rates from the phase-outs.
Overall tax burdens depend on tax rates on all income, not just the last dollar of income. A policy’s effect on marginal income tax rates is generally not the same as its effect on average tax rates. Credits such as the EITC or the child tax credit that phase out as income increases reduce people’s average tax rates even if they increase their marginal tax rates. The option to extend those credits to more families would increase marginal tax rates for some families at the same time as it lowered their average tax rate. Thus, the higher marginal tax rates would not make workers in these families worse off even though their incentives to work an additional hour or save an additional dollar would be reduced.
How Options Affect Low-Income Families
Results from TRIM for 1998 show that families, on average, paid over 11 percent of their income in federal income taxes and just under another 10 percent in Social Security payroll taxes (table 8). As mentioned above, payroll taxes include both the employee and employer share of the tax. Income taxes for families with incomes under $20,000 were less than zero, on average, indicating that the EITC was greater than positive tax liability for these families, resulting in a net refund. Those refunds were not large enough to offset their payroll taxes, however, and thus combined income and payroll liability for these families was greater than zero, though still a relatively small percentage of their income. Payroll taxes were greater than income taxes for the large majority of families. Average payroll taxes exceeded average income taxes in every income group up through $75,000.
Taxes for low-income families with children were lower than taxes for other low-income families. Couples and singles with children and income under $30,000 on average received net income transfers from the individual income tax, rather than owing any tax. Those transfers were significant in size, amounting to 15 percent or more of family income before taxes for families with incomes under $20,000. Although families in those income groups did pay payroll taxes, the combination of income and payroll taxes still resulted in a net tax refund.
The tax cut options to reduce tax rates, increase the child credit, and increase standard deductions and extend the length of the 15 percent income tax bracket would have very little effect on average income tax rates for low-income families (table 9). This is not surprising given that these families pay little or no income tax and thus receive little benefit from proposals that only reduce positive income tax liabilities. Because of the rules limiting refundability under current law, even the proposal to increase the child credit and extend refundability to families with fewer than three children would yield almost no benefit to the lowest-income families with children. Most low-income families who would have child credits in excess of their income tax liability under the option would also receive an EITC payment that exceeded their payroll tax and thus would not qualify for a refund of the child credit. Of course an alternative option would be to make the child credit refundable without regard to payroll taxes net of EITC, in which case the credit would serve as a flat per-child payment to all families with children, administered through the tax code. Many more low-income families would benefit under that option, but refundability would not be limited to low-income families who work and pay payroll taxes. The refundable child credit option simulated here would have a sizeable effect on middle-income families. Income taxes paid by families with children and incomes between $30,000 and $70,000 would decrease on average by 2 to 3 percent of income.
The proposals to increase the EITC or to create a new refundable payroll tax credit clearly would have the biggest impact on low-income families, boosting the amount of refunds received by those families. Both credits apply only to families with earnings, so the impact on families with children is much greater than the impact on other low-income families and individuals, about 40 to 50 percent of whom are elderly. Families with children and income of $20,000 or less would see an increase in their net refunds of 3 to 4 percent of family income on average, while families with income of $20,000 to $30,000 would see an increase of about 8 percent.
The distribution of benefits from the various tax relief proposals is clearly illustrated by the share of benefits from each proposal going to different income groups (table 10). Over 50 percent of the benefits from an across-the-board cut in tax rates would go to the 11 percent of families with incomes of $100,000 or more. This is somewhat less than their share of baseline income taxes but more than their share of combined income and payroll taxes.
The bulk of benefits (80 percent) from raising the standard deduction and increasing the income range for the 15 percent bracket would go to families with incomes between $50,000 and $200,000. It is not surprising that this proposal tends to benefit middle- and higher-income families. Increasing standard deductions would not benefit low-income taxpayers who already have zero income tax liabilities due to the current standard deductions, personal exemptions, and child credit (but not counting the EITC), nor would it benefit high-income taxpayers who for the most part itemize deductions. Lengthening the income range for the 15 percent bracket would benefit only families who are currently taxed at higher rates. While seemingly aimed at middle-income taxpayers, only about 25 percent of taxpayers, in fact, currently face statutory rates in excess of 15 percent. Furthermore, taxpayers in the 28 percent bracket who have only a small portion of their income taxed at that higher rate would receive only a small benefit from the proposal.17
About 60 percent of the benefits from the options to increase child credits would go to families with incomes between $30,000 and $75,000, with a slightly higher share for families at the lower end of that income range under the option that would make the credit refundable. More than 75 percent of the benefits from increasing the EITC would go to families with incomes of $40,000 or less. On the other hand, about 70 percent of the benefits from a new refundable payroll tax credit would go to these families. Additionally, under this option, a larger share of benefits would go to the very lowest-income families than under any of the other options.
Not all types of families would benefit from the proposals simulated here. In particular, couples and singles without eligible children under age 17 would see no change in their taxes from either proposal to increase child credits (table 11). Single adults would benefit the most from more broadly based options, including the rate cuts, the increase in standard deductions and the income range for the 15 percent tax bracket, and, in particular, the payroll tax credit. Single parents with children would benefit the most from the option to increase the EITC and the least from the option for a proportional cut in income tax rates.
Higher-income families currently pay a larger share of their income in taxes than lower-income families. Therefore, an across-the-board cut in income taxes benefits high-income families relatively more if those benefits are measured in absolute dollars per family or as a share of income, but not if the benefits are measured as a percentage change in income tax liability. Measuring the changes in tax burdens as the percentage decrease in tax liability can be misleading, because it suggests that low-income families would benefit the most from tax cuts simply because they pay so little of the tax to begin with. For example, if the tax liability of a family with $20,000 of income drops from $200 to $100, it experiences a 50 percent tax cut, but only a $100 (or 0.5 percent) increase in after-tax income. In contrast, if a family with $100,000 of income paying $20,000 of tax receives a $4,000 tax cut, it experiences a 20 percent reduction in taxes. However, its after-tax income increases by 5 percent —10 times the percentage increase (and 40 times the absolute increase) of the low-income family.
Measured as the percentage change in income taxes, the option for an across-the-board cut in tax rates appears to be progressive for families with incomes over $20,000—the percentage reduction in taxes declines as income increases (table 12). It is not the same percentage decrease in taxes at all incomes because it is an across-the-board cut in income taxes not total taxes and, as we have seen, the share of income taxes in total taxes varies by income level. It is not even the same percentage decrease in income taxes at all income levels. Families would continue to receive the same dollar amount of credits as before, and thus a 5 percent cut in tax rates results in a more than 5 percent reduction in taxes after credits. Higher-income families would have less than a 5 percent reduction in income taxes because the rate cut does not apply to the tax rate on capital gains.
Measured as the percentage increase in income after taxes, the across-the-board cut in tax rates is not progressive. Benefits from the tax cut increase with family income, with the largest increases for families with the highest incomes (table 13). Likewise, the option to increase standard deductions and extend the income range for the 15 percent tax bracket appears mostly progressive when measured as a percentage reduction in income taxes, but not when measured as a percentage change in income after tax.
Alternative types of broad-based tax cuts differ greatly in how they affect low-income families. Because low-income families pay so little income tax, refundability is the key to whether future income tax reductions will provide any benefits to them. However, those benefits will also depend upon whether or not new or expanded credits are refundable in a meaningful way. For example, extending current rules for refundability of child credits to all taxpayers does little to help low-income families because those rules would effectively preclude refunds for most families.
Consequently, across-the-board cuts in federal income tax rates or increases in the standard deduction and the width of the lowest tax bracket provide very little benefit to low-income families. Increasing the child credit as currently structured also would provide them very little assistance. Among the options we considered, the ones that help low-income families the most are increases in the earned income tax credit and enactment of a new, refundable payroll tax credit.
Whether or not a tax reduction proposal disproportionately benefits low-income families is only one criterion by which to judge its merits. The standard principles of good tax policy still apply. Would the proposal increase the economic efficiency of the tax system by reducing disincentives for work and saving? Would the proposal reduce, or at least not add to, tax complexity and the compliance burden on taxpayers? Proposals targeted to help low-income families do not necessarily also advance these other goals and may detract from them. Income-conditioned tax relief means that benefits must phase-out for families with higher incomes and thus results in higher marginal tax rates over the income phase-out range. Benefits targeted to particular taxpayers also mean that there must be rules determining who is eligible, which increases the complexity of taxes and creates opportunities for errors and abuse.
Fairness does not dictate that all tax relief should be directed to low-income families. However, they should not be left out. It is true that most low-income families, in particular families with children, pay zero or no federal income taxes and many even receive a net income tax refund. Nevertheless, these families do pay federal payroll and excise taxes along with state and local taxes. Just as in evaluating changes to spending programs, Congress must compare tax relief with other options for using surplus revenues, such as paying down the federal debt or addressing the future insolvency of Social Security and Medicare. If a tax cut is the best option, should it be targeted to higher-income people who have enjoyed extraordinary prosperity over the past decade, or to low- and moderate-income families who work and pay taxes but do not enjoy the same economic security?
Source: U.S. Congress, Congressional Budget Office, "Preliminary Estimates of Effective Tax Rates," (http://www.cbo.gov) (September 7, 1999). Urban Institute, 2001.
1. The maximum EITC for tax year 2000 is $2,353 for working families with one eligible child and $3,888 for working families with two or more eligible children. Working families with income in excess of $12,690 receive a reduced credit; they receive no credit if their income exceeds $27,413 (with one child) or $31,152 (with two or more children). Single workers between the ages of 25 and 64 are eligible for a smaller credit of up to $353. Single workers with income in excess of $5,766 receive a reduced credit, and no credit at all if their income exceeds $10,380.
3. See U.S. Congress, Congressional Budget Office, "Estimates of Federal Tax Liabilities for Individuals and Families by Income Category and Family Type for 1995 and 1999," CBO Memorandum, (May 1998); and Andrew Mitrusi and James Poterba, "The Distribution of Payroll and Income Tax Burdens, 1979-1999," National Bureau of Economic Research, Working Paper 7707, (May 2000).
4. Economists generally believe that workers also pay the employer portion of payroll taxes through lower wages. The net income tax refund for this family would exceed payroll taxes even including the employer portion of the tax.
5. For recent developments regarding state tax reductions and how they affect low-income families see: Elaine Maag and Diane Lim Rogers, "The New Federalism and State Tax Policies Toward the Working Poor," The Urban Institute, Assessing the New Federalism Occasional Paper 38, (September 2000); Nicholas Johnson, Robert Zahradnik, and Elizabeth C. McNichol, "State Income Tax Burdens on Low-Income Families in 1999," Center on Budget and Policy Priorities, (March 2000); and Nicholas W. Jenny, "2000 Tax and Budget Summary," State Fiscal Brief, Fiscal Studies Program, The Nelson A. Rockefeller Institute of Government, (September 2000).
7. A married couple or a single parent with two children would either have paid no income tax or received a refundable credit in over half of all states with income taxes in 1999. See Johnson et al. (2000).
8. Bureau of Labor Statistics, Consumer Expenditure Survey 1997-1998, (http://stats.bls.gov/csxcross.htm).
10. Under current rules taxpayers can deduct a standard amount-$7,350 for married couples, $6,450 for heads of households, and $4,400 for single filers in tax year 2000-in computing their taxable income. After subtracting personal exemptions and the higher of the standard or itemized deductions, a 15 percent tax rate applies to the first $43,850 of taxable income for couples, $35,150 for heads of households, and $26,250 for singles. For married couples in 2000, for example, the proposal would raise the standard deduction to $8,650 and the income to which the lowest (15 percent) rate applies to $51,740.
11. For example, in tax year 2000, the EITC is 40 percent of the first $9,390 of earnings for families with two or more eligible children, 34 percent of the first $6,680 of earnings for families with one eligible child, and 7.65 percent of the first $4,460 of earnings for single workers. The option would raise the maximum credit for families and singles by increasing those phase-in rates by 25 percent but keep the income level at which the maximum credit applies the same as under current law. Thus, for example, the phase-in rate for families with two or more eligible children would rise to 50 percent and maximum credit would go up from $3,765 to $4,695-an increase of $940.
Under current rules, families with earnings at or above the level at which the maximum applies but less than $12,260 receive the maximum credit, as do single workers with earnings less than $5,570. The credit is reduced by 21.06 percent of earnings in excess of $12,260 for families with two or more children, 15.98 percent of income in excess of $12,260 for families with one child, and 7.65 percent of income in excess of $5,570 for singles. The option would extend the maximum credit to more families by doubling the income thresholds at which the credit begins to phase out. For example, for families with children, the phase-out range would start at $24,520. Because the rates at which the credit phases out would remain the same as under current law, a phase-out range starting at a higher income level means many families who were not eligible for the credit under current law because their income was too high would receive partial credits. The credit for a family with two children, for example, would not phase out completely until income reached $46,800.
13. For a discussion about how the alternative minimum tax would affect the tax reductions proposed by then Governor Bush see: Al Davis, "Candidate Bush's Tax Plan," Tax Notes, (January 10, 2000), pp. 271-277; Lawrence B. Lindsey, "Governor Bush's Proposal and the Alternative Minimum Tax," Tax Notes, (January 24, 2000), pp.553-556; and Al Davis, "Further Thoughts on the Bush Tax Cut Plan," Tax Notes, (February 7, 2000), pp. 857-859.
15. The percentage change in income after tax is calculated as one minus the new marginal tax rate divided by one minus the old marginal tax rate. For taxpayers in the highest tax bracket this is [(1-.396*.95)/(1-.396)] = 1.0328, or just over a 3 percent increase in the after-tax return.
16. See Bruce D. Meyer and Dan T. Rosenbaum, "Welfare, the Earned Income Tax Credit, and the Labor Supply of Single Mothers," Working Paper, Northwestern University, (September 1999); Nada Eissa and Jeffrey B. Liebman, "Labor Supply Response to the Earned Income Tax Credit," Quarterly Journal of Economics, 112(2), (May 1996); and Jeffrey B. Liebman, "The Impact of the Earned Income Tax Credit on Incentives and Income Distribution," in James M. Poterba, ed., Tax Policy and the Economy, Vol. 12, MIT Press, (1998).
17. For example, the proposal would increase the starting point for the 28 percent bracket from $43,850 to $51,750 of taxable income for taxpayers filing joint returns in 2000. Taxpayers in that filing status with taxable incomes equal to or greater than $51,750 would see their taxes reduced by $1,027, because an additional $7,900 of taxable income would be taxed at 15 percent rather than 28 percent. Taxpayers with taxable income in excess of $43,850 but less than $51,750 would receive a smaller tax cut. There would be no benefit for taxpayers with taxable income of $43,850 or less. Because taxable income is measured after subtraction of personal exemptions and deductions, a married couple with two children and taxable income of $43,850 in 2000 would have at least $62,400 of adjusted gross income.