February 13, 2001
Testimony of Robert Greenstein
Executive Director, Center on Budget and Policy Priorities before the
House Committee on Ways and Means
View PDF version
If you cannot access the file through the link, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.
I appreciate the invitation to testify today. I am Robert Greenstein, executive director of the Center on Budget and Policy Priorities. The Center is a nonprofit policy institute here in Washington that specializes both in fiscal policy and in programs and policies affecting low- and moderate-income families. The Center does not hold (and never has received) a grant or contract from any federal agency.
My testimony today focuses primarily on two aspects of the Bush Administration's tax proposal: the proposal's cost in relation to the available surplus, and how various types of Americans would be affected by the tax cut.
The Plan's Cost in Relation to the Available Surplus
At first glance, the proposal may appear to be readily affordable. If the non-Social Security surplus totals $3.1 trillion over the next 10 years and the tax cut costs $1.6 trillion, the tax cut would consume a little over half of the projected non-Social Security surplus. The problem is that neither of these numbers provides an accurate reflection of the fiscal situation: the proposed tax cut would consume substantially more than $1.6 trillion of projected surpluses, while the available surplus is considerably smaller than $3.1 trillion.
How Much Would the Tax Cut Cost?
Let's look first at the widely cited $1.6 trillion figure. In fact, there is no cost estimate that shows the tax plan to cost $1.6 trillion. The $1.6 trillion figure has been derived as follows.
- Last May, the Joint Tax Committee estimated the cost of the plan at $1.3 trillion over the 10 years from 2001 through 2011.
- This estimate showed no cost in 2001, because the plan wouldn't be in effect yet. Hence, the $1.3 trillion estimate was a nine-year estimate.
- The new budget period is 2002 through 2011. Adding the plan's cost in 2011, based on the assumption that the cost in 2011 is similar to the cost for the 2010 that the Joint Tax Committee reported last May, brings the cost to $1.6 trillion.
But there is no Joint Tax Committee estimate that the plan costs $1.6 trillion. If an estimate were done today of the amount of the projected surpluses that the proposal would consume, the estimate would necessarily be considerably higher most likely about $2.1 trillion. This is the case for two reasons.
- The May 2000 Joint Tax Committee estimate was based on the economic and revenue assumptions in use at that time. The new CBO budget projections assume a larger economy that produces more revenues. When the same tax cuts are applied to a larger economy that produces more tax revenue, they result in a larger dollar tax cut. The change in the economic forecast is expected to add approximately $150 billion to the plan's estimated cost.
- The $1.6 trillion figure does not include the increased interest payments on the debt that would result from using projected surplus funds for the tax cut rather than for paying down debt. These interest costs must be included when determining how much of the projected surplus the tax plan would consume. The tax plan would result in $350 billion to $400 billion in added interest payments on the debt.
- This brings to $2.1 trillion the figure the Joint Tax Committee and CBO would likely estimate to be the amount of the projected surpluses that the plan would consume, if no changes are made in the plan.
Yet even this $2.1 trillion figure is low. It does not include several additional costs.
- The President has said he supports accelerating his proposed tax cuts and making some aspects of the tax cuts retroactive to January 1 of this year. The White House has not specified which aspects of the plan it wishes to see accelerated.
The proposed rate reductions would not take full effect until 2006 under the plan as it now stands. Making the rate cuts fully effective now would add $400 billion to the plan's cost over the 2002-2011 period, including the added interest payments on the debt.
Alternatively, making the doubling of the child credit and the establishment of a new 10 percent bracket fully effective now would add about $265 billion in cost.
- The $2.1 trillion cost is artificially low for a second reason it is based on the assumption that Congress will fail to address key problems in the Alternative Minimum Tax. As this Committee knows, the AMT affected 1.3 million filers in 2000 but will hit more than 15 million filers by 2010 if it is not modified, including large numbers of middle-income families that are not engaged in heavy tax sheltering. The Joint Tax Committee has estimated that if no action is taken and the Bush plan is enacted as it now stands, it will cause an additional 12 million filers to become subject to the AMT by 2010, bringing the number of filers hit by the AMT to an astonishing 27 million. Clearly, this should not occur, and virtually all observers are confident Congress will not let this happen. At his confirmation hearing, Treasury Secretary Paul O'Neill took note of the need to address these problems in the AMT.
This matter directly affects the Bush plan. The Joint Tax Committee estimate of the cost of the Bush plan is artificially low because the Joint Committee had to assume the AMT would not be fixed (since there is no provision in the Bush plan to do so) and that the AMT consequently would cancel out several hundred billion dollars of tax cuts the plan otherwise would provide. Fixing the AMT problem so that the AMT continues to affect roughly the same number of taxpayers as it does today will increase the cost of the Bush plan by another $200 billion to $300 billion. If Congress chooses to address the AMT problem a couple of years at a time (in the manner of the "extenders"), that would not alter the fact that over the course of the decade, this full $200 billion to $300 billion cost still will have to be absorbed.
This means the Bush plan would consume approximately $2.4 trillion in projected surpluses if none of the tax cuts are accelerated (the $2.1 trillion figure cited earlier plus the additional cost the tax plan has once the AMT is fixed) and more than that if the tax cuts are accelerated. Furthermore, these costs do not include:
- The cost of extending the expiring tax credits, which Congress will surely do;
- The cost of various tax reductions that Congress passed last year, such as pension tax legislation and larger marriage-penalty relief than President Bush is proposing;
- The cost of various health-insurance related tax preferences that the Bush Administration is proposing separate and apart from its big tax proposal; or
- Any corporate or capital gains tax cuts.
The Available Surplus
Under the new CBO forecast, the projected surplus outside Social Security is $3.1 trillion over 10 years. However, as analyses issued over the past year by the Brookings Institution, the Concord Coalition, and our Center have noted, there is an important difference between CBO's estimate of the size of the non-Social Security surplus and the amount that actually is available for tax cuts and program initiatives.
In making its surplus estimates, CBO follows certain rules that require it to assume implementation of various tax increases and program reductions that would occur if current law is followed but that would be highly unpopular and that virtually all observers believe will not take place. For example, about 20 popular tax credits and other tax preferences, generally known as the "extenders," are typically renewed for only a few years at a time and are scheduled to expire in the next year or two. There appears to be little question that Congress will extend most or all of them. The CBO projections assume, however, that all of these tax credits will expire. Similarly, the CBO surplus projections assume the law governing the AMT will not be changed and millions of middle-class taxpayers will become subject to the AMT over the coming decade. The increased taxes those taxpayers are assumed to pay are part of CBO's surplus calculation. It is very unlikely, however, that Congress will sit idly by and let the AMT encroach heavily upon the middle class.
Similar issues arise on the spending side. The rules CBO uses lead it to assume that federal payments to farmers will be slashed deeply. In recent years, Congress has provided an average of about $10 billion a year in payments to farmers that are made a year at a time rather than provided under a provision of permanent law. Since these payments are not governed by an ongoing statute, the CBO surplus estimates assume these payments will terminate after 2001. Of course, that won't occur.
Questions arise with regard to discretionary spending as well. The CBO forecast assumes that funding for discretionary programs including the defense budget will simply remain over the coming decade at the 2001 level, adjusted only for inflation. For that to occur, discretionary spending would have to fall in purchasing power per U.S. resident (since the population is rising) despite almost-certain increases in areas such as defense, education, and health research. Since 1987, non-defense discretionary spending has remained constant as a share of GDP which means it has risen by more than the inflation rate and in the last three years, as surpluses have emerged and defense spending has started back up, total discretionary spending has stayed even with GDP. Observers such as Bob Reischauer and Rudy Penner, both distinguished former CBO directors, have warned there is little chance that discretionary spending over the coming decade will remain at current levels, adjusted only for inflation.
If one simply assumes that current policy (as distinguished from current law) is maintained in the areas I've just discussed that is, that the "extenders" are continued, the AMT is fixed so it remains at about current levels in terms of the proportion of taxpayers it affects, payments to farmers remain at current levels and discretionary spending remains at today's level in purchasing power per U.S. resident (which means discretionary spending would have to shrink as a share of GDP) then $700 billion of CBO's projected surplus outside Social Security evaporates. To be prudent, policymakers should take this $700 billion off the table when estimating how much they have available for the cuts and program initiatives, since this $700 billion in unlikely to materialize. If policymakers assume these funds are available, they risk using the same projected surplus dollars twice. Removing this $700 billion reduces the available non-Social Security surplus to $2.4 trillion.
The portion of the projected surplus that would be in the Medicare Hospital Insurance trust fund also needs to be removed from the calculation. Both parties have appropriately said these funds should be set to the side and not used for tax cuts or other program expansions. Some $400 billion of the $3.1 trillion non-Social Security surplus occurs in the Medicaid HI trust fund. Removing these funds yields an estimate that $2.0 trillion over 10 years is available if the projected surpluses fully materialize.
Does this Mean the 107th Congress Should Pass Tax Cuts
and Spending Increases of $2 Trillion?
Two trillion dollars still is a very large amount. But it would be unwise for Congress and the Administration to pass legislation this year that commits all of this amount. These figures are only projections. These large surpluses are not assured.
Any number of events such as slower-than-forecast economic growth or faster-than-expected growth in health care costs could cause the projections to be too high. As CBO has pointed out, its track record shows that its budget projections are subject to considerable error, even in the short term. Since 1981, the CBO forecast for the fifth year out has been off by an average of 3.1 percent of GDP. This means that if CBO's projection for 2006 is off by the average amount that its forecasts for the fifth year have been off in the past, its projection of the surplus in 2006 will be off by $400 billion. A $400-billion overestimate would mean we were again running on-budget deficits.
Furthermore, more than 70 percent of CBO's projected surplus outside Social Security and Medicare would come in the second five years of the 10-year period. Projections made that far in advance are especially uncertain and prone to error. This means that if the full amount of the surplus projected outside Social Security and Medicare is consumed by actions the 107th Congress takes but the projections later prove to have been too optimistic, sizable deficits outside Social Security and Medicare could return.
Another reason that it would be unwise to consume all of the $2.0 trillion in available projected surpluses is that even if the surplus forecast proves correct, acting now to commit all of the available surpluses for the next 10 years will leave no funds for subsequent Congresses to use to address needs that cannot be foreseen today but inevitably will arise. It is inconceivable that no such needs will emerge over the course of the decade. Such needs could be military, international, or domestic. While we cannot know today what these needs will be, we had better plan on some new problems emerging that will have to be addressed.
A final reason that consuming all of the projected $2.0 trillion would be imprudent relates to Social Security and Medicare. If legislation to restore long-term Social Security solvency is to be enacted, a transfer of non-Social Security general revenues from the Treasury to the Social Security Trust Fund (or to private, individual retirement accounts) almost certainly will be required. Without such a transfer, the magnitude of the reductions in retirement benefits that will be required regardless of whether a solvency plan includes individual accounts will almost surely make any plan impossible to pass. As a result, policymakers ought to set aside, or reserve, a portion of the projected non-Social Security surplus funds for this purpose. To be prudent, a minimum of $500 billion over 10 years should be reserved for this purpose. (If 70 percent of the solvency gap is closed by other means, including benefit reductions, and only 30 percent of the gap needs to be closed through additional revenues, $500 billion will be needed from the non-Social Security, non-Medicare surpluses.)
The Bottom Line on the Tax Cut's Affordability
The projected surplus outside Social Security and Medicare is $2.7 trillion over 10 years, of which $700 billion is likely to be noted simply to maintain current policies that command broad support. That leaves $2.0 trillion, some of which should be aside to deal with the all-too-real possibility that the surpluses may not fully materialize and some of which is likely to be needed for Social Security and Medicare solvency legislation. This suggests that substantially less than $2 trillion should be committed now. Moreover, a portion of that amount will be needed for priorities in health care and other areas. Yet the Bush tax cut would consume well over $2 trillion of projected surpluses, or more than 100 percent of what is realistically available. The tax cut is substantially too large.
What About a Trigger?
Before turning to the question of who would benefit from the tax cuts, I would like to comment briefly on the idea of accompanying the tax cuts with a trigger that would stop the next phase of the tax cuts from taking effect if surpluses or the debt failed to reach some specified level. Such an approach may have initial appeal. Closer examination suggests, however, it has a significant risk of being ineffective.
In the 1980s, Congress passed the Gramm-Rudman-Hollings law, which established year-by-year deficit targets and required unpopular actions to occur (specifically, across-the-board program reductions) if the deficit target otherwise would be missed. The law proved largely to be a failure. When the deficit target would be missed by a significant amount, Congress and the Administration resorted to budget gimmicks and creative accounting to make it appear on paper as though the target had been reached, thereby avoiding the unpopular action that otherwise would have been triggered. When that was not sufficient, Congress simply raised the deficit targets. The odds are substantial the same pattern would emerge here.
Moreover, most provisions of the Bush tax cut would phase in fully by the fifth year, 2006. Seventy percent of the surplus that CBO projects outside Social Security, however, would come in the second five years, from 2007 through 2011, and it is the projections for these years that are the most uncertain and subject to large error. A "trigger" would have little effect in undoing the fiscal damage if the budgetary picture in those years was considerably less bright than currently forecast; most of the tax-cut provisions would already be in full effect.
The much more prudent approach would be to enact a smaller tax cut now, place a portion of projected surpluses in a reserve available for neither tax cuts nor spending increases, and wait to see to what extent the projected surpluses materialize. Big tax cuts that will not take effect for several years do not respond to the current economic slowdown, so the danger here is entirely on one side from overdoing it now and fostering fiscal problems down the road. If the type of surpluses that currently are forecast for future years do materialize, Congress can easily enlarge tax cuts and program initiatives at that time. Politically, it is far easier to come back and expand tax cuts than to reverse enacted tax cuts or enforce a trigger that may never be allowed to be pulled.
Who Would Receive the Tax Cuts?
In presenting its tax-cut plan, the White House has placed emphasis on the plan's effects in helping lower-income working families, such as waitresses making $22,000 or $25,000 a year, as well as the beneficial effects that provisions such as the proposed doubling of the child credit would have on such families. Analysis shows, however, that the plan would not do very much for these families and would concentrate an unusually large share of its tax-cut benefits on those on the upper rings of the income ladder.
Effects on Lower-income Families
The plan would provide no assistance to working-poor and near-poor families. The plan only affects families with incomes above about 150 percent of the poverty line. (The exact level at which the plan begins to help families with children varies from about 130 percent to 180 percent of the poverty line, depending on the number of children, the family's filing status, whether the family has child care costs, and other matters.)
The plan bypasses working families with incomes below these levels because it only affects families that owe income tax before the Earned Income Tax Credit is applied. This does not mean that the plan provides a tax cut for everyone who pay taxes. Millions of families owe no income tax but pay substantial amounts of payroll tax, as well as other federal taxes such as taxes on gasoline.
Thus, a single mother with two children who works full time and earns $22,000 pays no income tax but owes $1,234 in payroll tax (net of the EITC). She would receive no tax reduction under the plan. A waitress with two children who earns $25,000 an example the White House frequently cites would receive no tax reduction if she incurs $170 a month in child care costs, not an unusual amount for such a single working parent. The waitress pays $2,325 in payroll tax net of the EITC.
Overall, 12 million families with 24 million children one of every three children in the United States would receive no tax reduction. Some 80 percent of these families have workers. Some 55 percent of African-American children and 56 percent of Hispanic children live in families that would receive nothing from the proposed tax cut.
There are a variety of ways to provide relief to low-income working families. These include improving the Earned Income Tax Credit and extending the child credit to such families by making it refundable for them.(1)
Some other low-income working families would receive small tax reductions. A family of four that earns $26,000 would have its income tax liability eliminated. That family, however, pays only $20 in income tax now. The family's principal tax burden comes from the payroll tax: it pays $2,689 in payroll taxes net of the EITC. Although some proponents of the Administration's tax proposal would describe such a family as receiving a 100 percent tax cut because its income tax liability has been eliminated, the family's overall federal tax burden would be reduced by less than three-quarters of one percent.
Marginal Tax Rates
Some supporters of the Administration's proposal have noted that while such a family might get a modest tax cut in dollar terms, the family's "marginal tax rate" would be reduced by 15 percentage points. Presidential economics advisor Lawrence Lindsey has observed that many families in the $25,000 to $30,000 range face higher marginal tax rates than the wealthiest Americans do. If , however, the Administration's goal is to reduce marginal tax rates on the low-income working families that face the highest rates in the nation surely a worthy goal the plan falls short. Conservative and liberal analysts alike have long recognized that the working families that face the highest marginal tax rates are those with incomes between about $13,000 and about $20,000. For each additional dollar these families earn, they lose up to 21 cents in the Earned Income Tax Credit, 15.3 cents in payroll taxes (including the employer's share), 24 cents to 36 cents in food stamp benefits, and additional amounts if they receive housing assistance or a child care subsidy on a sliding fee scale or are subject to state income taxes. No other Americans in any income bracket have as large a share of each additional dollar they earn "taxed away."
Ways to reduce marginal tax rates for such families are available, well known, and not especially expensive. One can raise the income level at which the Earned Income Tax Credit begins to phase down as earnings rise and/or reduce the rate at which the EITC phases down. Bipartisan Senate legislation that Senators Rockefeller, Jeffords, and Breaux and legislation introduced last year follows such a course, as does legislation that Rep. Cardin introduced in the House. The proposal that Sawhill and Thomas outline in their new Brookings paper to make the child credit partially refundable for low-income working families also would lower marginal rates substantially for such families. The Administration's plan contains no such features.
Furthermore, the Bush plan departs from a bipartisan consensus that formed in Congress over the past two years to reduce marriage tax penalties for low-wage working families, along with middle- and upper-income families. Some of the most serious marriage penalties in the tax code are those that can face low-income working individuals as a result of the way the phase-out of the EITC is designed. Tax bills that Congress passed and President Clinton vetoed in both 1999 and 2000 contained EITC reforms to provide marriage penalty relief for low-income working families. Democratic alternative bills included such provisions as well; this issue had become truly bipartisan. The Bush plan contains no such marriage penalty relief, limiting its relief to families at higher income levels.
The often-cited $1,600 tax reduction for a middle-class family of four also has been subject to some misunderstanding. A large number of individuals and families do not have children or have one child and a substantial share of the families with two or more children owe less than $1,600 in income taxes and hence would receive less than $1,600 in tax reductions. Some 85 percent of households would receive less than $1,600 under the proposal.
I also would note that the $1,600 tax reduction for a family of four with a $50,000 income is not scheduled to occur until 2006 under the plan as now structured. For such a family, the tax reduction is $1,600 in 2006 dollars, or $1,400 in today's dollars. The purchasing power of this tax reduction would fall below $1,400 in today's dollars in years after 2006 because the child credit is not indexed.
Tax Reductions for Higher Income Filers
The average tax reductions for very-high income individuals and families would be quite large. It has been argued that such taxpayers would get a large share of the tax cut because they pay a correspondingly large share of the taxes and that low-income individuals and families would receive the largest percentage tax reductions. Neither of these statements turns out to be correct.
- The best data available on who pays what share of all federal taxes including income, payroll, estate, excise, and other taxes come from a major study conducted by Treasury career staff and released in September 1999. The study shows that the top one percent of families pays 20 percent of all federal taxes.(2)
- The top one percent of families would receive at least 36 percent of the tax cuts under the Administration proposal when the plan is fully in effect. The top one percent also would have the federal taxes it pays reduced by a greater percentage than middle- or low-income households, while low-income households would receive the smallest percentage tax cut of any group. These figures are discussed in the next section of this testimony.
The data presented here on how the benefits of the Bush tax cut would be distributed come from two sources: an analysis by Citizens for Tax Justice, using the Institute for Taxation and Economic Policy (ITEP) model, and the aforementioned Treasury study on how the burdens of various taxes are apportioned among various income categories.
The ITEP model that CTJ uses is a well-respected model developed in substantial part by former staff members of the Joint Tax Committee. CTJ tax distribution analyses, using the ITEP model, have been validated over the years by the fact that they generally have yielded results very similar to those the Treasury Department has produced.
The CTJ analysis of the effect of the Bush plan (when the plan's provisions are fully in effect) finds the bottom 40 percent of families would receive four percent of the tax cuts, with the average tax cut for this group being $115. The bottom 60 percent of families would receive 13 percent of the tax cuts. The 20 percent of families exactly in the middle of the income spectrum would receive eight percent of the tax cuts and receive an average tax reduction of $453.
By contrast, the top one percent of families would receive 43 percent of the tax cuts, and their average cut would amount to $46,000. The top one percent would receive more in tax cuts than the bottom 80 percent of the population.
Some supporters of the Administration's proposal have cited alternative figures from the Joint Tax Committee that are said to show the proportion of the tax cut that would go to the top one percent of families would be significantly smaller. Those figures, however, really do not actually show that to be the case. The JCT figures in question include neither the effects of repealing the estate tax, which accounts for about one-quarter of all tax reductions in the plan when the plan is fully in effect, nor the effects of any provisions in the plan that do not take effect until after 2005. The part of the tax-rate reductions that would not take effect until 2006 is disproportionately beneficial to those in the top brackets. As a result, these JCT figures are not especially useful. The figures Citizens for Tax Justice has produced do not suffer from these omissions.
There has been some debate in the past about the best methodology to use to determine what percentage of the estate tax is paid by people in different income categories and thus what percentage of the benefits from estate tax repeal would accrue to each income group. Under the ITEP model that CTJ uses, 91 percent of the estate tax is estimated to be paid by the top one percent of families. Virtually all of the tax is estimated to be paid by the top five percent of filers. Such results should not be surprising. IRS data show that the estate tax is levied only in the case of two percent of all deaths and that in 1997 half of all estate taxes were paid by the 2,400 largest taxable estates the estates of the wealthiest one of every 1,000 people who died.(3)
To help resolve issues related to how to measure the incidence of the estate tax, the Treasury study issued in September 1999 includes a major analysis of the distribution of the estate tax by income category.(4) Since publication of this study, Treasury has used its results in the distributional analyses it has undertaken.
The Treasury findings on who pays the estate tax are broadly similar, although not identical, to the estimates in the ITEP model, which was constructed before the Treasury study became available. The Treasury study estimates that the top one percent of families pay 64 percent of the estate tax (and thus would get 64 percent of the tax-cut benefits that would result from estate tax repeal), rather than paying 91 percent of the tax as the ITEP model estimates. The Treasury and ITEP figures on the proportion of the estate tax paid by the top five percent of families, however, are quite similar; the Treasury study estimates the top five percent of families pay 91 percent of the estate tax, as compared to 100 percent of the tax under the ITEP model. Under both sets of estimates, the top 20 percent of families pay virtually all of the estate tax, and the tax does not affect the other 80 percent of the population.
Accordingly, another way to estimate the effect of the Bush tax cut on different income groups is to take the CTJ estimate but to modify it by substituting the Treasury estimates on the incidence of the estate tax for the estimates in the ITEP model. Under this approach, the top one percent of the population is estimated to receive 36 percent of the tax cuts under the Bush plan, rather than the 43 percent the CTJ analysis estimates, and to receive an average tax cut of $39,000 rather than $46,000. The top 20 percent of families still is found to receive 71 percent of the tax cut, the same percentage as under the CTJ analysis. Similarly, the bottom 40 percent of families still is found to receive four percent of the tax cut.
Under either approach, the tax cut is found to be tilted heavily toward those with very high incomes and to provide only a modest percentage of its tax-cut benefits to the types of families the White House last week presented as major beneficiaries. Under both approaches, the share of the tax cuts that would go to the top one percent would be roughly double the share of the federal taxes this group pays, and the top one percent would receive more in tax cuts than the bottom 80 percent of the population combined.
Who Would Receive the Largest Percentage Tax Reductions?
White House officials have argued that lower-income families would receive the largest percentage tax reductions. These statements rest on data on the percentage reduction in families' income tax burdens. The most relevant data, however, are those on the percentage reduction in families' overall federal tax burdens. Since low- and moderate-income families pay more in other federal taxes principally the payroll tax than in income taxes and often have very small income tax liabilities, it is possible to eliminate those income tax liabilities without providing a family a substantial tax cut or reducing the family's total federal taxes by a very large percentage.
When the percentage reduction the Bush tax cut would make in total federal tax burdens is examined, a different picture emerges. Under the Bush plan, the top one percent would receive a much larger percentage reduction in the federal taxes they pay than would any other income group. The percentage tax reduction for low-income families would be only about half that which the top one percentage of families would receive. (See Table 1.)
Percentage Reduction in Federal Taxes Under the Bush Tax Plan
ITEP Model (Citizens for Tax Justice)
ITEP Model, but using the Treasury Estimates on Who Pays the Estate Tax
Bottom 20% -5.5% -5.5% Second 20% -6.5% -6.5% Middle 20% -7.3% -7.3% Fourth 20% -7.2% -7.3% Next 15% -6.1% -6.7% Next 4% -4.2% -6.4% Top 1% -13.6% -11.6%
The Child Credit and the 10 Percent Bracket
The figures that show the small percentage of the tax cut that would go to middle- or low-income families may seem surprising given the inclusion in the proposal of the provisions to double the child credit and to create a new 10 percent bracket. These have been presented as proposals designed in substantial part to benefit lower-income working families and help them enter the middle class. In fact, only a modest share of the tax-cut benefits from these two proposals would go to low- or moderate-income families; much larger shares would go to high-income families. And as noted, approximately one-third of children would not benefit from either proposal.
Consider the proposal to raise the child credit from $500 per child to $1,000. This proposal would cut taxes for families with two children that have incomes up to $300,000. Those who would benefit most are filers with incomes in the $110,000 to $250,000 range; they would receive the largest tax cuts under this proposal because the Bush plan not only would double the child credit but also would raise the income level at which the child credit starts to phase down from $110,000 to $200,000 and slow the rate at which it phases out so that families with two children and incomes up to $300,000 would benefit from it. Currently, filers with incomes above $130,000 are ineligible for the credit. For many of these relatively affluent families, the child credit would rise from zero to $1,000 per child. For millions of children in low- and moderate-income working families, by contrast, the child credit would remain at zero or at its current level of $500 per child or would rise to less than $1,000 per child (because their families would have insufficient income tax liability against which to apply the increase in the child credit). Faced with a choice between extending the credit to children in low-income working families that pay payroll taxes but no income tax and extending it to children in families in the $130,000 to $300,000 range, the Administration chose the latter course.(5)
As a consequence, when the increase in the child credit is fully in effect:
- Some 82 percent of the benefits from the child credit proposal would accrue to the 40 percent of families with children with the highest incomes. Only three percent of the benefits from this proposal would accrue to the bottom 40 percent of such families.(6)
- The top 20 percent of families would receive 46 percent of the tax-cut benefits from this proposal, a larger share than any fifth of the population would receive.
The Estate Tax
The feature of the proposal that has the largest effect in making the plan so disproportionately beneficial to those at the top of the economic scale is the proposed repeal of the estate tax. This tax is levied on the estates of only the most affluent two percent of individuals who die. Moreover, in 1997, the 2,400 largest estates the estates of the wealthiest one of every 1,000 people who died bore half of the estate tax. Had there been no tax, the estates of these very wealthy individuals would have received an average tax reduction of $3.5 million each.
Families farms and small businesses do not figure heavily here. Of the approximately 2.3 million people who died in 1998, only 47,500 or about two percent left estates that were taxable. Of those estates, there were just 1,418 or three percent of the taxable estates in which a family business or family farm constituted the majority of the estate. This means that a family business or family farm constitutes the majority of the estate for only six of every 10,000 people who die. Furthermore, a Treasury analysis has found that such estates paid less than one percent of all estate taxes. Relief can be provided to such estates, and other reforms made in the estate tax, for a fraction of the cost of repealing it.
In addition, it is beginning to be recognized that repeal of the estate and gift taxes would open enormous loopholes in other parts of the tax code that could substantially increase the cost of estate tax repeal beyond the levels the JCT has estimated. The matter is explored in a recent issue of the journal Tax Notes by estate tax attorney Jonathan Blattmachr and Hofstra law profession Mitchell Gans. Similarly, a recent New York Times article by David Cay Johnston reports that estate tax attorneys interviewed for the article generally concurred that repeal of the estate and gift tax would spawn major new tax-avoidance strategies.
For example, without the gift tax, a wealthy investor could transfer stock that has appreciated in value by $100 million to an elderly relative, who agreed simply to hold the stock. The elderly relative then would return the stock to the donor when he died, through a provision in his will. The investor would thereby escape capital gains tax entirely on the $100 million profit. This is just one of a number of gaping loopholes that repeal of the estate and gift tax threatens to open in the tax code.
The Bush tax proposal would likely absorb the entire projected non-Social Security surplus that is likely to be available, leaving little margin for error if the surpluses do not materialize fully and squeezing out other priorities that should rank higher than giving tax cuts of this magnitude to those who are at the pinnacle of the income scale, have done the best in recent years, and are least in need of a very large tax cut. Congress can provide relief that still provides significant tax reduction to middle-income families and is more favorable to the lower-income working families the President says he cares about at a far lower cost than the rather extravagant and lopsided Bush proposal.
1. A broader approach would make the child tax credit refundable for all families with children. The bipartisan National Commission on Children recommended such an approach in 1991.
2. Following Treasury usage, "families" includes single people as well as family units. All families are included whether or not any member of the family files an income tax return. The ITEP model uses a similar definition.
3. See Iris Lav and Joel Friedman "Estate Tax Repeal: A Costly Windfall for the Wealthiest Americans," </cms/index.cfm?fa=view&id=314> Center on Budget and Policy Priorities, February 6, 2001.
4. "U.S. Treasury Distributional Analysis Methodology," Department of the Treasury, OTA Paper 85, September 1999.
5. For families with more than two children, the income range would extend even higher than $300,000.
6. Institute for Taxation and Economic Policy, special data run for the Children's Defense Fund.