July 12, 2000
Large Cost of the Roth "Marriage Penalty Relief" Provisions Reflects Poor Targeting
Much of the Benefits Would Go to High-Income Taxpayers
or Those Who Already Receive Marriage Bonuses
by Iris Lav and James Sly
View PDF version of this report.
On June 28, the Senate Finance Committee passed a marriage-tax-penalty relief proposal offered by its chairman, Senator William Roth, that would cost $248 billion over 10 years. The official cost assigned to the bill is considerably less — $55.6 billion — because the legislation will be considered in a form that provides the tax relief only through 2004, to satisfy Senate rules. History shows, however, that legislation of this type rarely is allowed to expire. As a result, the full, permanent cost of the bill should be considered the relevant benchmark.
Although two of the proposal's marriage penalty provisions are focused on middle- or low-income families, the proposal as a whole is poorly targeted and largely benefits couples with higher incomes. The proposal's costliest provision, which accounts for more than half of the package's overall cost when all provisions are in full effect, benefits only taxpayers in the top quarter of the income distribution. In addition, the proposal would provide nearly two-fifths of its benefits to families that already receive marriage bonuses.
Citizens for Tax Justice finds that only 15 percent of the benefits of the Roth proposal would go to low- and middle-income married couples with incomes below $50,000. This group accounts for 45 percent of all married couples. By contrast, the fewer than one-third of married couples that have incomes exceeding $75,000 would receive more than two-thirds of the bill's tax-cut benefits.
The Roth plan contains three principal provisions related to marriage penalties. The most costly of these would reduce the rates at which income is taxed for some married couples. This provision would increase for married couples the income level at which the 15 percent tax bracket ends and the 28 percent bracket begins, and also increase the income level at which the 28 percent bracket ends and the 31 percent bracket begins. The second provision would raise the standard deduction for married couples, setting it at twice the standard deduction for single taxpayers. A third, much smaller provision would increase the earned income tax credit for certain low- and moderate-income married couples with children.
A fourth provision relates to the alternative minimum tax (AMT) and affects both married and single taxpayers; it is not specifically designed to relieve marriage penalties. This provision would permanently extend taxpayers' ability to use personal tax credits, such as the child tax credit and education credits, to offset tax liability under the alternative minimum tax.
The Joint Committee on Taxation estimates that the Roth proposal, without the sunset, would cost $248 billion over 10 years. And the proposal's long-term cost is substantially higher than this. The bill's costly provision that would extend the 15 percent and 28 percent tax brackets would not take full effect until 2008; this slow phase-in markedly reduces the bill's cost in the first 10 years. The Joint Tax Committee estimate shows that when all of the plan's provisions are fully in effect in 2008 through 2010, the bill would cost $40 billion a year.
Once in full effect, the proposal to expand the 15 percent and 28 percent tax bracket itself would cost more than $20 billion a year. This provision would exclusively benefit taxpayers in brackets higher than the current 15 percent bracket; no other taxpayers would be touched by it. Since only the top quarter of taxpayers are in brackets higher than the 15 percent bracket, only those in the top quarter of the income distribution would benefit from the provision.
The bill's tax reductions are not focused on married families that face marriage penalties. Nearly as many families receive marriage bonuses today as receive marriage penalties, and the bill would reduce their taxes as well. The proposal would confer tens of billions of dollars of "marriage penalty tax relief" on millions of married families that already receive marriage bonuses. In fact, only about 40 percent of the $248 billion in tax cut benefits the bill would provide over the next ten years would go for reductions in marriage penalties. A similar proportion of the tax cuts, about 37 percent, would reduce the taxes of families already receiving marriage bonuses. The remainder of the benefits, including portions of the AMT change that would go to taxpayers other than married couples, would neither reduce penalties nor increase bonuses.
Senate Democratic and Administration Proposals
A marriage penalty relief plan that is more targeted on middle-income families and modestly less expensive than the Roth proposal is expected to be offered by Democrats on the Senate floor. This Democratic alternative is identical to an amendment offered by the Finance Committee Democrats during the June 28th mark up of the Roth proposal. This plan would allow married taxpayers with incomes below $150,000 to choose whether to file jointly as a couple or to file a combined return with each spouse taxed as a single filer. The long-term cost of the Democratic alternative appears to be about four-fifths of the long-term cost of the Roth plan. (This provision ignores the cost of the AMT provision of the Roth plan.)
The marriage penalty relief proposals contained in the Administration's fiscal year 2001 budget are significantly less costly than either the Roth proposal or the Senate Democratic alternative. These proposals, which are targeted on low- and middle-income married filers who face marriage tax penalties, would provide substantial marriage penalty relief at about one-fourth the cost of the Roth plan. (This comparison, as well, excludes the cost of the AMT provisions of the Roth plan.) The marriage penalty proposals in the Administration's budget would cost a little more than $50 billion over 10 years.
Democrats Offer More Targeted Plan
Democrats are expected to offer on the Senate floor a modestly less expensive version of marriage penalty relief that is more targeted on married couples that experience marriage penalties under current law.
The Democratic plan would give married couples two different options for filing their taxes. The couples could file jointly, as the vast majority of couples do under current law. Alternatively, couples would have a new option under which a husband and wife could each file as single individuals, although they would file together on the same tax return. Each couple would have the opportunity to make two different tax calculations and pay taxes using the method that resulted in the lowest tax bill. In addition, the proposal would in some circumstances allow each spouse in a family with more than one child to claim a separate Earned Income Credit (for different children), based on that spouse's income; this would effectively double the level of income such a family could have and receive the EITC.
This new option for single filing would begin to be phased out for couples with incomes exceeding $100,000. Couples with incomes exceeding $150,000 would not be eligible to use the option.
The optional separate filing provision would reduce or eliminate marriage penalties for most couples below the $150,000 income limit. It would maintain marriage bonuses for couples that receive such bonuses under current law. In contrast to the Roth plan, however, it would not increase marriage bonuses for couples that already receive them.
The Democratic alternative would cost approximately $21 billion a year when fully in effect in 2004. By comparison, the Republican plan would cost approximately $40 billion a year when fully in effect in the years 2008-2010, of which slightly more than $30 billion a year is attributable to the marriage penalty provisions. (The remainder reflects the costs of the AMT provisions.) When costs for similar years are compared, the fully phased-in cost of the Democratic plan would be about four-fifths of the fully phased-in cost of the Republican bill, excluding its AMT provisions.
The budget surplus projections that the Administration issued on June 26 show a projected non-Social Security surplus under current law of nearly $1.9 trillion over 10 years. While this may make it seem as though the proposed marriage penalty relief could be afforded easily, caution needs to be exercised. The surpluses actually available for tax cuts and programs expansions are considerably smaller than is commonly understood. Furthermore, there is a wide range of priorities competing for the surplus dollars that are available.(1)
The projected surpluses include about $400 billion in Medicare Hospital Insurance (HI) trust fund surpluses that the President, the House of Representatives, and the Senate have agreed should not be used to fund tax cuts or program increases. Excluding these Medicare HI surpluses, the surpluses available to fund tax cuts or program increases amount to less than $1.5 trillion.
That baseline projection, however, does not reflect the full costs of maintaining current policies. For instance, the Administration's baseline projections of the cost of discretionary, or annually appropriated, programs assume that funding for these programs will be maintained at current levels, adjusted only for inflation. The projections do not include an adjustment for growth in the U.S. population, so the projections assume that funding in discretionary programs will fall in purchasing power on a per person basis. Maintaining current service levels for discretionary programs would entail that such spending be maintained in purchasing power on a per capita basis.
Certain legislation that is needed simply to maintain current tax and entitlement policies and that is virtually certain to be enacted also is not reflected in the surplus projections, including legislation to extend an array of expiring tax credits that Congress always extends, legislation to prevent the Alternative Minimum Tax from hitting millions of middle-class taxpayers and raising their taxes, as will occur if the tax laws are not modified, and legislation to provide farm price support payments to farmers beyond those the Freedom to Farm Act provides, as Congress has done each of the past two years. Assuming that legislation in these three areas will be enacted (as it is virtually certain to be) and that the purchasing power of discretionary programs will be maintained at current levels on a per person basis reduces the available non-Social Security, non-Medicare HI surpluses by approximately $600 billion, to less than $900 billion over 10 years.
At least half of this $900 billion is likely to be needed to facilitate reform of Social Security and Medicare that will ensure the long-term solvency of those programs. Since neither party is willing to close the long-term financing gaps in these programs entirely or largely through slicing benefit costs or increasing payroll taxes, a large infusion of revenue from the non-Social Security part of the budget will be necessary. Indeed, nearly all of the major Social Security proposals offered by lawmakers of either party entail the transfer of substantial sums from the non-Social Security budget to the retirement system. Taking this reality into account leaves about $400 billion over 10 years to pay for tax cuts or other program initiatives.
Competing for those funds are other tax cuts, various domestic priorities such as providing a Medicare prescription drug benefit, reducing the number of uninsured Americans, increasing investments in education and research, and reducing child poverty, as well as proposals to raise defense spending. The Senate Finance Committee marriage penalty proposal would eat up more than three-fifths of this $400 billion in a single bill.(2)
Roth Plan Favors Higher-Income Taxpayers
The most expensive provision in the Roth bill would change the tax brackets for married couples. It would raise for couples both the income level at which the 15 percent bracket ends and the 28 percent bracket begins, and the income level at which the 28 percent bracket ends and the 31 percent bracket begins. Joint Tax Committee estimates show this provision would cost nearly $123 billion over the next 10 years even though it does not fully phase in until fiscal year 2008. In the years between 2008 and 2010, it would account for 54 percent of the plan's costs.
Because this provision would raise the income level at which the 15 percent and 28 percent brackets end for married couples, it would benefit only those couples whose incomes exceed the level at which the 15 percent bracket now ends. A couple with two children would need to have income surpassing $62,400 (in 2000 dollars) to benefit.(3) Only one of every four taxpayers, and one of every three married taxpayers, have incomes that place the taxpayers above the point at which the 15 percent bracket currently ends.
Thus, when the provisions of the Roth plan are phased in fully, more than half of its tax cuts would come from a provision that exclusively benefits taxpayers in the top quarter of the income distribution and married couples in the top third of the distribution.
A second provision in the Roth bill would increase the standard deduction for married couples. This approach focuses its tax benefits on middle-income families. Most higher-income families have sufficient expenses to itemize their deductions and do not use the standard deduction. Most low-income working families have no income tax liability and would not benefit. If this provision were effective in 2000, the standard deduction would increase by $1,450, which would generate a $218 tax cut for most couples in the 15 percent tax bracket. This provision would account for a little more than one quarter (27 percent) of the plan's costs over the first 10 years and one-fifth of the plan's annual costs when all provisions of the plan are phased in fully.
The third provision of the Roth plan is an increase in the amount of the earned income tax credit that certain married couples with low earnings can receive. This is the one provision of help to low-income married families. When all of the provisions of the plan are phased in fully, the EITC provision would represent four percent of the plan's annual costs. (This provision would account for six percent of the plan's costs over the first 10 years.)
Low-income married families can face marriage penalties that arise from the structure of the Earned Income Tax Credit. EITC marriage penalties occur when two people with earnings marry and their combined, higher income makes them ineligible for the EITC or places them at a point in the EITC "phase-out range" where they receive a smaller EITC than one or both of them would get if they were still single.
The Roth proposal would reduce EITC marriage penalties by increasing by $2,500 the income level at which the EITC for married families begins to phase down, as well as the income level at which married families cease to qualify for any EITC benefits. For a husband and wife that each work full time at the minimum wage, the Roth proposal would alleviate about 44 percent of their marriage tax penalty.
The plan also contains a fourth provision that is not directly targeted at relieving marriage penalties. This measure would address some of the problems that will result in significant numbers of middle-income families becoming subject to the Alternative Minimum Tax in future years — a situation never intended when the AMT was enacted — by permanently allowing both non-refundable and refundable personal tax credits to offset AMT tax liability.(4) This provision would account for one-quarter of the legislation's total cost when all of the bill's provisions are fully implemented.
Roth Plan Targets Benefits on Higher-Income Taxpayers
The Joint Committee on Taxation has estimated the distributional impact of this proposal on taxpayers in the years 2001 through 2005. For 2005, the JCT found that more than 70 percent of the benefits of this tax proposal would go to tax filers with incomes exceeding $75,000, while only 15 percent of the benefits would go to tax filers with incomes below $50,000. Moreover, these figures understate the extent to which higher-income taxpayers would benefit, because the costly bracket increases that benefit only the top quarter of taxpayers would not be fully in effect until fiscal year 2008. The final year covered by the JCT estimate is 2005.
Some observers note that married taxpayers tend to have higher incomes than other taxpayers, in part because there often is more than one earner in the family. They point out that looking at the distribution of benefits among all taxpayers makes the distribution appear more skewed than it is seen to be if just the effect on married taxpayers is considered. This is not the case, however, with respect to the Roth proposal.
Effects of the Finance Committee
Marriage Penalty Relief Bill
Percent of Total Tax Cut
Figures show the effects of the bill when phased in fully. The income levels in the table are 1999 income levels.
Under the legislation, the changes in the standard deduction and earned-income tax credit for couples would take effect in 2001. The changes in the starting points for the 28% and 31% tax brackets for couples would be phased in starting in 2002 and finishing in 2007. The totals exclude about $0.8 billion in tax cuts for married persons filing separate returns. Changes in the Alternative Minimum Tax, which would maintain the current treatment of tax credits under the AMT, are not included.
Source: Institute on Taxation and Economic Policy Tax Model, March 30, 2000
An analysis by Citizens for Tax Justice shows that even within the universe of married couples, the Roth plan disproportionately benefits those married couples who are at the upper end of the income spectrum. The Citizens for Tax Justice analysis finds that among married couples, those with incomes in excess of $75,000 would garner 68 percent of the benefits of the Roth proposal when the plan is phased in fully. Some 41 percent of the benefits would go to married couples with incomes in excess of $100,000. Only 15 percent of the benefits would go to those with incomes below $50,000. (See Table 1.)
Roth Plan Does Not Focus Its Benefits on Families Facing Marriage Penalties
Three of the proposals in the Roth plan — the standard deduction increase, the tax bracket extensions, and the EITC provision — would provide general tax relief for married couples, rather than marriage penalty relief focused on families that actually face penalties. The fourth provision, allowing tax credits to offset the AMT, is not specifically targeted on married couples.
Under the current tax structure, no one-earner couples face marriage penalties; they generally receive marriage bonuses. The families that face marriage penalties are two-earner families. The Roth plan, however, would reduce tax burdens for one-earner and two-earner married couples alike. As a result, the plan is far more expensive than it needs to be to reduce marriage penalties.
Indeed, nearly two-fifths of the cost of the legislation results from tax reductions that would increase marriage bonuses rather than reducing marriage penalties. Another two-fifths of the cost would reduce marriage penalties. The remaining fifth would not affect marriage penalties and bonuses.
If the "marriage penalty relief" provisions are considered alone, approximately half of the cost of these provisions would go to increase marriage bonuses. When the Treasury Department examined a proposal to expand the standard deduction for married filers and to set the tax brackets for married couples at twice the level for single taxpayers — a plan similar to the Roth proposal — it found that only about half of the resulting tax cuts would go to reduce marriage penalties, with the rest going to increasing marriages bonuses.(5)
Long-Term Cost of Roth Plan
The Roth plan has a $248 billion price tag over ten years, in comparison to the $182 billion cost of the similar marriage penalty relief plan the House passed earlier this year. The major difference relates to the Alternative Minimum Tax. The House bill does not include any provision to allow non-refundable credits to offset the AMT, even though failure to do so would allow the Alternative Minimum Tax in future years to tax back from millions of middle-class taxpayers the tax benefits that the legislation otherwise provides. If one assumes the full cost of the House plan ultimately would include changing the AMT to prevent that from occurring, the full cost of the plan would be considerably higher than $182 billion.(6)
The Roth plan, which includes substantial AMT changes, provides a more accurate view of the total cost. Nevertheless, the Roth plan itself appears to hold hidden costs relating to the AMT. Even under the Roth plan, the alternative minimum tax would prevent some higher-income married taxpayers from enjoying the benefits of the wider tax brackets. If the Roth plan were enacted and the AMT were subsequently modified to address this issue, as would be likely, the changes in the Roth plan would have a larger cost.
Leaving aside the additional AMT issues that might have to be addressed in future years, the Roth plan would rise in cost from $23.3 billion in 2005 to $39.9 billion annually by 2010 (assuming the sunsets do not hold). When the plan was fully in effect, its long-term cost thus would greatly exceed the $248 billion price tag for the first ten years.
1. See James Horney and Robert Greenstein, How Much of the Enlarged Surplus is Available for Tax and Program Initiatives: Available Funds Should be Devoted to Real National Priorities, Center on Budget and Policy Priorities (July 10, 2000).
2. The Finance Committee bill would reduce revenues by an estimated $248 billion over the 2001-2010 period assuming that the changes made by the bill are permanent and do not sunset after 2004. That includes the $45 billion cost of a provision included in the bill that partially ameliorates the growing effect of the alternative minimum tax on middle-income taxpayers. Since the cost of a more complete AMT "fix" is included in the calculation described above that reduces the projected surplus over 10 years to $400 billion, this $45 billion is excluded here for purposes of calculating the effect of the Finance Committee bill on the surplus. The remaining $203 billion reduction in revenues over 10 years, plus a $47 billion increase in interest costs that would result from that reduction in revenues, would lower the non-Social Security surplus by $250 billion. This equals 63 percent of the approximately $400 billion in non-Social Security, non-Medicare HI surpluses for 2001 through 2010 that appear to be available to fund tax cuts and program initiatives other than restoring Social Security and Medicare solvency.
3. The income level at which the 15 percent bracket ends has been widely misreported as being $43,850 for married filers. The $43,850 figure is the level of taxable income at which the 15 percent bracket ends for married filers — that is, the level of income after deductions and personal exemptions are subtracted. The level of adjusted gross income at which filers move from the 15 percent to the 28 percent tax bracket is considerably higher. The lowest level of gross income at which the 15 percent bracket ends for a married family of four is $62,400. (This is the level at which the 15 percent bracket ends for a married family of four that does not itemize deductions.
4. Personal non-refundable credits such as the child credit, dependent care credit, Hope Scholarship and Lifetime Learning credits, and the like would be allowed to reduce or eliminate any minimum tax owed. In addition, personal refundable credits such as the earned income credit would not be reduced by the amount of a taxpayer's minimum tax liability.
5. In 1999, the Treasury department conducted a study analyzing various proposals designed to provide major marriage penalty relief. The study estimated, for each proposal, how much of the "marriage penalty relief" would go to reduce marriage penalties and how much would go to increase marriage bonuses. The study found that 46 percent of the tax reductions resulting from raising the standard deduction for married couples so that it would be twice the deduction for single taxpayers would go to increase marriage bonuses. The study also found that coupling such an increase in the standard deduction with an expansion in the tax brackets so that all of the brackets for married filers were double those for single filers would result in 52 percent of the total tax cut going to increase marriage bonuses.
The Roth plan would increase the standard deduction and the 15 percent and 28 percent tax brackets for married couples so both the deduction and brackets are double the levels for single filers. It also would expand the EITC for married couples. It would not expand the higher tax brackets. The percentage of the Roth tax cut that would go to increase marriage bonuses should be approximately the same or modestly higher than the percentages under the proposals Treasury examined. Expanding the highest tax brackets — a feature of one of the proposals that Treasury examined but that is not part of the Roth plan — would provide larger tax cuts for families that Treasury found to have more marriage penalties relative to marriage bonuses than the rest of the population. This suggests that somewhat more than 50 percent or more of the tax cuts in the Roth package (other than the AMT provisions) would go to increase marriage bonuses rather than reduce marriage penalties.
6. Indeed, the Treasury Department estimates that if the AMT provisions of current law were to remain unchanged, the House marriage penalty bill itself would cause eight million additional families to become subject to the AMT by 2010.