November 23, 2009: A Response to Mathur and Hassett۪s Commentary on Taxes, By Robert Cherry, Broeklundian Professor of Economics, Brooklyn College
With the sunsetting of the 2001 tax cuts, next year۪s Congress must look anew at the various ways the federal system now taxes and subsidizes households with children. In their recent Spotlight commentary, Simplifying Tax Credits for Low-Income People, Kevin Hassett and Aparna Mathur of the American Enterprise Institute rightly criticized the patchwork nature of the child-related subsidies families now receive and reinforced that simplicity and transparency should be a minimal goal of tax reforms. In this commentary, I hope to clarify some of the issues they raised.
Hassett and Mathur began by noting that the cash welfare system “created significant work disincentivesand also discouraged the formation of two-parent families,” but they neglected to pursue how these issues are affected by the shift to “reliance on the tax system as a means of providing assistance to working families.” They mentioned the haphazardness of the current marginal tax rates faced by families with children, especially in the phase-out range of the EITC. They did not indicate, however, that these families will also face the phasing out of other benefits: housing, child care, food stamps, health care, etc.
In a 2005 Future for Children article, Adam Carasso and Eugene Steuerle estimated that the average effective marginal tax rate for single mothers with two children and incomes between $10,000 and $40,000 would be 88.6 percent if they qualified for most of the benefits available. As a result, a large share of employed mothers in this income range will not choose to work additional hours or take a somewhat higher paying job that has fewer personal amenities convenient hours or location because the meager income gains would not be worth it. To reduce the marginal tax rate, the simplified tax credit program could have a low phase-out rate below, say, $30,000, with a higher phase-out rate thereafter.
Hassett and Mathur worry that the present set of policies are not well targeted. It should be made clear, however, that most proposals to simplify the tax system by combining programs, including a 2000 piece by myself and Max Sawicky and a 2001 article by David Ellwood and Jeffrey Liebman, keep benefits to those below the official poverty line virtually the same but provide substantial benefits to near-poor and lower-middle-class families. This impact should be embraced because there is a middle-class parent penalty.
Families with incomes below $20,000 receive substantial child-related benefits, as do those with incomes of $70,000 or more. It is those families in between, particularly those with incomes from $30,000 to $60,000, who have too much income to gain much from means-tested programs and too little income to take advantage of child care credits and the increased value of their dependent allowances. Just as we have seen with the SCHIP expansion, these near-poor and lower-middle-income families have every right to obtain just as much child-related benefits as poor and upper-middle-class families.
Hassett and Mathur did not judge how the current set of policies impacts two-parent formation. These child-related tax benefits provided a substantial financial incentive for welfare recipients to marry a working, childless partner. His wage income would allow the married couple to gain substantial EITC benefits and child credits that were unavailable to single mothers on welfare.
Welfare reform dramatically reduced the number of welfare recipients. These mothers are now members of the paid labor force. Many single mothers who work pay a large financial penalty if they marry a working man, more so if he also has children who qualify for child-related tax benefits. Once they marry, their income will be too high to qualify for the substantial benefits that they individually received before marrying. For example, if a single mother earning $15,000 marries a childless man earning $25,000, the federal EITC of almost $5,000 she was receiving would be virtually eliminated, and undoubtedly she would lose benefits from other means-tested programs, including state EITC and child care programs.
Many observers, including Carasso and Steuerle, Alex Roberts and David Blankenhorn, and Claude Renshaw and Ken Milani believe that there should be tax changes to reduce these financial barriers to marriage among low-wage workers. These changes could include adjusting the EITC slowly to changes in marital status or unlinking some benefits from combined family incomes, at least under certain circumstances for some period of time. Having a transitional period will at least blunt the adverse impact of marriage on family child-related benefits.
I believe that combining a limited number of child-related federal tax benefits the EITC, child credit, and dependent allowance offers the best way to go. Adjustments so that benefits from child care tax credits are better targeted to low-income working families should be handled separately. For many of the poor, child care benefits are primarily gained through access to slots in government-funded child care facilities. For poor parents, their co-payment is negligible, but it becomes substantial for many near-poor families.
As Hassett and Mathur documented, these near-poor families have too little taxable income to benefit from the federal child care tax credit. For example, a single mother with one child and an annual income of $30,000 would receive benefits equal to only about one-quarter of the allowable $3,000 in expenses. This is not enough to make child care facilities affordable, so the vast majority of these families rely instead on informal child care arrangements. At least 28 states, including California and New York, have recognized this difficulty, and most have enacted fairly generous refundable child care tax credits. In these states, this mother could obtain benefits equal to about one-half of the allowable expenses but still face a second hurdle: virtually no child care facility can be obtained for the meager allowable $3,000.
One potential solution is for these near-poor families to use federal dependent care accounts (DPAs). For near-poor families, these accounts are much more valuable than the federal child care credit because by lowering their pretax income, these families qualify for a substantial increase in their EITC. As a result, for our $30,000 family with $5,000 in child care expenses, these accounts offer over $1,000 more in benefits than the federal child care credit. If we added together the federal benefits from a DPA with the state refundable tax credit, our family could obtain total benefits equal to more than three quarters of their $5,000 expenditures in New York and California, as well as a number of other states.
Hassett and Mathur correctly point out that though these accounts raise the allowable expenses for one child to $5,000, they are virtually unused by the near poor. This suggests an informational problem: near-poor families are unaware of the benefits from these accounts, so NGOs and government agencies should get the word out.
These efforts, however, would be premature. Unfortunately, in virtually every state, only expenses used for the federal child care credit qualify for the state refundable credit. Thus, use of a DPA disqualifies the taxpayer from claiming state refundable child care credits. Indeed, only Oregon currently allows expenses used for a DPA to also qualify for their refundable child care credittheir Working Families Tax Credit. Thus, as summarized in a 2009 CPA Journal article, one way to provide more aid to near-poor families would be to encourage states to allow a family۪s child care expenses paid through a DPA to qualify for state refundable child care tax credits.
Combining the universal child-related tax credit policies would increase transparency and simplicity. However, only by lowering its phase-out rate, at least in the income range below $30,000, would it reduce the high marginal tax rate many near-poor working mothers face. Indeed, lowering the marginal tax rate substantially would require policy changes beyond the simplified tax proposals made by Hassett and Mathur and others. Just as important, resolving the marriage penalty problem and the inability of these near-poor working mothers to gain from current child care tax policies also require additional policy changes. Thus, moving to a simplified tax system is helpful but is only one component of tax policy changes to help working-poor and near-poor mothers.
Robert Cherry is Broeklundian Professor of Economics at Brooklyn College