Saver's Credit For Moderate-Income Families Would Fade Away Over Time Under House-Passed Pension Bill
Revised July 17, 2006
The saver’s credit — the only retirement tax cut enacted in 2001 aimed at people with incomes under $50,000 — is scheduled to expire at the end of this year. The House proposed a permanent extension of the saver’s credit as part of its pension bill, and both the House and Senate proposed shorter-term extensions of the saver’s credit as part of their original tax reconciliation bills. All of these proposals, however, would extend this saving incentive in such a way that it would be sharply scaled back over time and eventually would disappear altogether. This would occur because none of these proposals would index the credit’s income or contribution limits for inflation.
The saver’s credit was used by five million households last year. That number will decline markedly, however, if the credit is not indexed to inflation. Without adjustment for inflation, the income levels above which people are ineligible for the credit — $50,000 for married filers and $25,000 for singles — would be permanently frozen. Yet the income levels below which households do not earn enough to owe income tax are adjusted for inflation and consequently rise over time. Because households that do not owe income taxes are effectively ineligible for the credit, over time fewer and fewer families will have incomes in the range within which the credit applies.
Furthermore, for those who do remain eligible, inflation will erode the credit’s value, sometimes dramatically. Under the credit’s current structure, the credit is reduced as a family’s income rises, particularly when a family’s income exceeds $30,000. A couple without children that earns $30,000 and contributes $2,000 to an IRA or 401(k) plan will receive a saver’s tax credit of $1,000 in 2006. If the credit is extended without being indexed, and the couple’s income merely keeps pace with inflation, the tax credit that the couple will receive for making a $2,000 contribution will fall from $1,000 in 2006 to $400 in 2007 — resulting in a $600 tax increase. (See Table 1.)
Selected Tax Years
*Benefit to a married couple with no dependent children that has income that equals $30,000 in 2006 and grows at the rate of inflation in succeeding years. The couple is assumed to make a $2,000 annual retirement contribution.
In sharp contrast to the saver’s credit, the provisions of the 2001 tax-cut law that provide generous benefits to higher-income taxpayers by raising the contribution limits for IRAs and 401(k)s are fully protected from inflation, as a result of being indexed. Moreover, the conferees on the pension bill are considering whether to make permanent these higher contribution limits, including the provisions that index those limits to protect their value over time.
This raises a basic tax equity issue. It is difficult to defend indexing the pension tax preferences that are primarily used by higher-income taxpayers, while failing to index the principal pension incentive for modest-income working families and thereby eroding the incentives available to these families to save for retirement (and raising their taxes). It may be noted that the President’s Advisory Panel on Tax Reform proposed both that the saver’s credit should be indexed for inflation and that the credit should be made available to low-income working households that do not earn enough to owe income tax.
Without Inflation Indexing, Value of Saver’s Credit’s Falls Sharply over Time
The saver’s credit offers significant incentives for low- and middle-income families to boost their retirement savings (see box on page 3). With each passing year, however, fewer and fewer families are eligible for the saver’s credit. As noted above, this occurs because, in addition to its income ceilings, the credit has a de facto income floor. Because the credit is not refundable, households that do not earn enough to incur income tax liabilities cannot use it. For instance, a couple with two dependent children that makes retirement contributions cannot benefit from this credit in 2006 unless its income exceeds $24,100. 
The income levels at which families begin to owe federal income tax areadjusted each year for inflation and so rise over time. As a result, each year fewer families will have incomes that are above the point where they owe income tax but below the $50,000 income limit for the credit. As the income range for eligibility narrows, the number of potential beneficiaries declines.
The lack of inflation indexing also will cause significant erosion in the value of the saver’s credit over time. As noted above, a married couple with no children that earned $30,000 in 2006 and contributed $2,000 to a retirement account will receive a tax credit of $1,000 when it files its 2006 tax return. This essentially lowers the cost to the couple of making a $2,000 retirement contribution to $1,000, providing a powerful incentive for the couple to undertake retirement saving. But if this couple’s income merely keeps pace with inflation, the tax credit that the couple receives for making a $2,000 contribution would plunge to $400 in 2007 and to $200 in 2010. By 2016, the credit the family receives would be worth $161 in today’s dollars (based on CBO’s projected inflation rates).
Saver’s Credit Can Encourage Retirement Saving
The saver’s credit creates an incentive for lower- and middle-income households to contribute to retirement savings accounts by matching some or all of eligible contributions. For those eligible for a full 50 percent credit, a $2,000 contribution to an IRA or 401(k) would yield a $1,000 credit ($2,000 * 50% = $1,000). The $1,000 government credit effectively lowers the cost of the taxpayer’s contribution to $1,000. As a result, the government is essentially matching the taxpayer’s contribution, with each providing $1,000. With a 20 percent credit, the federal government is providing a 25 percent match — providing $400 of a $2,000 contribution that costs the taxpayer $1,600.a
In contrast to the saver’s credit, other current tax incentives for retirement saving give the largest tax breaks to the highest-income families, while giving moderate-income families little or no tax benefits for such saving. These incentives are structured as tax deductions rather than tax credits. As a result, a high-income family in the 35 percent tax bracket gets $350 off its taxes for a $1,000 contribution to a 401(k), while a moderate-income family in the 10 percent tax bracket gets $100 off its taxes for the same size contribution. A family that does not earn enough to owe income tax gets no tax incentive for retirement saving.
In addition, high-income families can afford to save more than families of more modest means. In this way as well, they are in a better position to benefit from the retirement tax breaks in current law than are people of lesser means.
Finally, the available evidence indicates that the retirement tax breaks in current law do little to spur new retirement saving among higher-income households; by and large, these households simply shift existing savings from taxable accounts into retirement accounts to take advantage of the tax benefits.b By contrast, lower-income households are more likely actually to increase their overall savings in response to well-designed retirement tax incentives.c
a See William Gale, Mark Iwry, and Peter Orszag, “The Saver's Credit: Expanding Retirement Savings for Middle- and Lower-Income Americans,” Retirement Security Project, 2005-2, March 2005.
b See Eric Engen and William Gale, “The Effects of 401(k) Plans on Household Wealth: Differences Across Earnings Groups,” National Bureau of Economic Research Working Paper No. 8032, December 2000.
c See Esther Duflo, William Gale, Jeffrey Liebman, Peter Orszag, and Emmanuel Saez, “Savings Incentives for Low- and Middle-Income Families: Evidence from a Field Experiment with H&R Block,” Retirement Security Project 2005-5, May 2005.
The principal reason that this lack of indexing has such a large effect on the credit is that the 50 percent “credit rate” (the percentage of a taxpayer’s retirement contributions that the credit equals) phases down sharply after a family’s income surpasses $30,000. The credit rate falls to 20 percent for couples with income between $30,000 and $32,500, and to 10 percent for couples with income between $32,500 and $50,000. (In other words, for married filers with income between $32,500 and $50,000, the tax credit equals 10 percent — rather than 50 percent — of their eligible retirement contributions.) The effects of just a few years’ inflation can push up a family’s wages enough to cause a sharp reduction in the value of its saver’s credit and, as a result, a significant increase in its tax bill.
Improving the Saver’s Credit
The erosion of the saver’s credit could be avoided if the credit’s income thresholds and contribution limits were indexed for inflation. Adjusting the income thresholds would ensure that as the incomes of low- and moderate-income families rise with inflation, the credit for which they are eligible is not reduced or eliminated. Similarly, indexing the $2,000 contribution limit would ensure that the contribution amount to which the saver’s credit applies is not eroded over time by inflation’s effects.
In addition to indexing the credit’s parameters, other steps could be taken to significantly improve its effectiveness as a savings incentive for low- and moderate-income families. A recent experiment designed by a team of researchers for the Retirement Security Project, and conducted by H&R Block, found that the tax credit they tested produced sharp increases in contributions to retirement accounts by married couples with incomes below $35,000. The credit they designed was simpler and more transparent to tax filers than the saver’s credit, and it also was available to savers who did not earn enough to incur income tax liability. The results of this experiment strongly indicate that the current the saver’s credit could be more effective if it were made simpler, more transparent, and refundable so that the savings incentive would be available to workers who do not earn enough to owe income taxes.
Making the credit refundable could prove particularly important, since it would remove the de facto floor that makes the current credit unavailable (or only partially available) to many or most low-income working families. The Urban Institute-Brookings Institution Tax Policy Center estimates that, of the tax filers with incomes low enough to qualify for the 50 percent credit, only about 14 percent could potentially receive any benefit from the credit if they contributed to a retirement savings account; the remainder would receive no benefit because they would owe no income tax. An even smaller number of filers — 0.1 percent — had incomes in the range that would allow them to receive the full credit if they contributed the maximum amount to a retirement account.
Furthermore, a recent article by two economists at the Joint Committee on Taxation finds that the lack of refundability significantly restricts the benefits the credit provides to those low-income households that are able to claim it. Looking at actual tax returns for 2002, the researchers found that many taxpayers who were otherwise eligible for the 50 percent credit either did not receive a benefit or only received a partial benefit because the credit was not refundable. They also observed that the combination of non-refundability and non-indexed income limits means that the credit’s reach will be even more limited in future years.
In its 2005 report, the President’s Advisory Panel on Tax Reform noted these problems and proposed solutions. The panel called for making the credit refundable. The panel apparently reasoned that a saving incentive intended for low-income households would have a greater impact if it were available to more than one-fifth of its target population. The panel also proposed to fully index the income limits and eligible contribution amounts for inflation and to modify the credit’s structure so its benefits would phase out less abruptly as income increased. 
Some of the potential changes that would improve the saver’s credit, such as making it refundable, apparently are regarded as being outside the issues that the current pension conference is willing to consider. The House and Senate have shown a willingness, however, to reform other expiring tax credits as part of the process of extending them. For example, in their respective tax reconciliation bills from last year, both the House and Senate proposed modifications to a number of expiring provisions, including the research and experimentation credit, in an effort to extend these tax provisions and improve their effectiveness. 
When extending the saver’s credit, Congress should, at the very least, adopt the modest reform of indexing the credit’s income thresholds and its contribution limit. This would ensure that the credit remains an effective savings incentive for moderate-income families, and that those families using the credit do not face higher taxes as a result of the effects of inflation. Such a step would place the saver’s credit on a par (in this regard) with other pension provisions enacted in 2001. To allow pension provisions that primarily benefit higher-income families to be indexed while failing to index the one savings incentive for lower- and moderate-income families would be highly inequitable. It also would run counter to the rhetoric of Congressional leaders who say they want to encourage retirement saving by ordinary Americans and to protect middle-class families from tax increases.
 For a discussion of these provisions, Robert Greenstein and Aviva Aron-Dine, “Pension Bill Conference Report May Make Some 2001 Tax Cuts Permanent Without Offsetting Their Costs Bill Could Set Precedent for Other Unpaid-for Tax-Cut Extensions,” Center on Budget and Policy Priorities, July 17, 2006.
 At lower income levels, either the couple would have no income tax liability for the saver’s credit to offset, or the saver’s credit that the couple could claim would reduce, dollar for dollar, the child tax credit the couple could claim. (An estimated 20 percent of tax filers who claim the saver’s credit do not actually benefit from it, because each dollar of their saver’s credit results in a dollar loss in their child credit. See Esther Duflo, William Gale, Jeffrey Liebman, Peter Orszag, and Emmanuel Saez, “Savings Incentives for Low- and Middle-Income Families: Evidence from a Field Experiment with H&R Block,” Retirement Security Project 2005-5, May 2005.)
If the couple in this example also qualified for other tax credits, such as the dependent care tax credit or education credits, the income level below which the saver’s credit would be of no value to the couple in 2006 would be higher than $24,100.
 Esther Duflo, William Gale, Jeffrey Liebman, Peter Orszag, and Emmanuel Saez, “Savings Incentives for Low- and Middle-Income Families: Evidence from a Field Experiment with H&R Block,” Retirement Security Project 2005-5, May 2005.
 William Gale, Mark Iwry, and Peter Orszag, “The Saver’s Credit: Expanding Retirement Savings fro Middle- and Lower-Income Americans,” The Retirement Security Project, March 2005.
 Gary Koenig and Robert Harvey, “Utilization of the Saver’s Credit: An Analysis of the First Year,” National Tax Journal, December, 2005.
 For a broader discussion of the panel’s recommendations on a range of issues, see Aviva Aron-Dine and Joel Friedman, “Effects of the Tax Reform Panel’s Proposals on Low- and Moderate-Income Households,” Center on Budget and Policy Priorities, February 3, 2006.
 These reforms were part of the reconciliation bills passed by the House and Senate. Although these provisions (along with other expiring provisions) were not included in the final reconciliation agreement that was signed into law, Congress is expected to consider these provision later in this session. See Joint Committee on Taxation, “Comparison Of The Estimated Revenue Effects Of The Tax Provisions Contained In H.R. 4297, As Amended By The Senate, and H.R. 4297, As Passed By The House, Fiscal Years 2006 – 2015,” JCX-10-06, February 9, 2006.
 The 2001act phased in higher contribution limits for both IRAs and 401(k)s, and then indexed these contribution limits to inflation when the maximum limit was reached. The 2001 act did not change income limits. Although the income limits for traditional IRAs are not indexed for inflation, they have been increased under the law every year since 1998. These annual increases in the income limits ended in 2005 for singles but will continue until 2007 for couples. Roth IRAs already are available to couples with incomes up to $160,000. This limit is not indexed, but at such a high income level, the lack of indexing does not limit eligibility for middle-income households. No income limits apply to 401(k)s.