Policy Brief #5

How Do Tax Policies Affect Low Income Workers?
October 2005

Prepared from a paper by Kevin A. Hassett, The American Enterprise Institute and Anne B. Moore, University of California, Berkeley

Download printable version. (PDF)



• Concerns about reducing the tax burden on low-income families have motivated a dramatic reduction in the federal income taxes they pay. However, state and local taxes have changed little. As a result, the entire tax system has not increased in progressivity as much over the past quarter century as one would think if one focused solely on the federal income tax.

• Total taxes paid by low-income families with children have declined over the last quarter century. Lowincome individuals without children have seen little change in their overall tax burden.

• The proportion of taxes paid that is a ributed to non-income taxes has climbed sharply over time (e.g. sales tax, property tax).

• Marginal income tax rates have declined in low income ranges, but have increased or stayed the same in moderate income ranges.


This Policy Brief summarizes a 2005 paper in which the authors, Kevin A. Hassett and Anne B. Moore, assemble a comprehensive analysis of every tax payment made by typical families to government at all levels in the U.S. for the 1979 to 2004 period. They include sales, excise, federal and state income, payroll and property taxes, and even at times state lotteries. The paper documents movements both in marginal and average tax rates and payments, paying particular attention to low-income families and individuals.

The Hassett and Moore paper was presented at a National Poverty Center conference and will appear in an edited volume, forthcoming from the Russell Sage Foundation, "Working and Poor: How Economic and Policy Changes are Affecting Low-Wage Workers."1

Data and methodology

Hassett and Moore analyze taxes paid for three hypothetical low-income families: a single mother with two children, a married couple with two children, and an unmarried adult with no children. The authors assume that the single mother earns $7.00 per hour and, working full-time, earns $14,000 per year. The husband of the married couple works full time at a wage of $8.50 per hour, earning an annual income of $17,000; the wife earns $5.15 per hour, the federal minimum wage since 1997, and makes $10,300 annually, for a total family income of $27,300. The unmarried adult is assumed to have a wage rate of $8.50 per hour, earning $17,000 per year (all in 2004 dollars.)

The authors use two approaches to construct measures of federal and state income taxes paid for the families: one drawn from The National Bureau of Economic Research Tax Simulation Model (TAXSIM) and the other from the Internal Revenue Service's Statistics of Income (SOI).

A major contribution of the paper is the incorporation of state sales taxes and state and local property taxes into the overall analysis of taxes paid by low-income families. For a detailed explanation of the data and methodology read the full National Poverty Center Working Paper version.2

Decline in total direct taxes paid by low-income families

Hassett and Moore find that total direct taxes paid by low-income families have declined significantly, dropping especially sharply since the late 1990s. The paper cites three main factors for the decline. First, federal income taxes have declined sharply for families because of the refundable child tax credit (instituted in 2001 for taxpayers for up to three children), expansions of the Earned Income Credit (EIC), and lower marginal tax rates. Second, over the same period, state sales taxes paid have increased significantly, but not enough to offset the decline in federal taxes paid. Third, payroll (FICA) and property taxes stayed relatively constant throughout the last several decades.

Figure 1 plots the total tax paid across all jurisdictions for the three hypothetical low-income households from 1979 to 2004. The bottom line in the chart is the estimated taxes paid by the single mother with two children with income of $14,000. Her taxes paid increased from 1979 until 1986, peaking at $3,435. Since that time, they have declined steadily. While state and local taxes still impose a burden on low-income single mothers, the total burden across all levels of government drops to a net subsidy of about $1,130 by 2004. This is primarily due to the large increase in the EITC in the early 1990s.

The married couple family of four with income of $27,300 also sees its tax liability drop sharply by the end of the period, from about $6,800 to about $3,400. For this family, the federal income tax results in a tax subsidy, but the taxes paid to other sources are large enough that the net tax paid is positive, despite the large federal income tax net subsidy.

The tax paid by the low-income unmarried adult is remarkably constant over time at around $5,500.

Sharp increase in non-income taxes as a percentage of total tax

A combination of falling federal income taxes and relatively constant state and local taxes contributes to the authors' second finding: non-income taxes as a percentage of total taxes for low-income families have increased sharply over time.

Figure 2 shows the changing relative importance of the various taxes over the past quarter century. It charts the share of taxes other than income and payroll taxes in total taxes. (The chart omits the total tax paid by the single mother with two children, since that tax is now negative.) For the hypothetical married couple with two children, the share of non-income taxes rose dramatically from 23 percent in 1979 to 65 percent by 2004. No such change is evident for the adult without children, whose share of non-income taxes was 20 percent in 1979 and 25 percent in 2004.

One explanation for this pattern may be that concerns about the income taxation of the poor received wide attention in Congress but not in state legislatures. Issues of fairness are more visible for taxes whose incidence is readily observable, but not for taxes - such as sales taxes - with less-obvious distributional effects. Another explanation is that states, which levy most of the non-income taxes, tend to be less redistributive than the federal government.

Increase over time in marginal tax rates for families at moderate income levels

While overall taxes paid have declined notably for families with children, the expansion of the EIC since the early 1990s required a higher phase out rate to offset the higher benefits to families with incomes between about $10,000 and $15,000. As a result, relatively high marginal tax rates apply to families with children with incomes between about $15,000 and $30,000. So while average tax rates have declined for all low-income families, marginal tax rates have, in some income ranges, increased.

Figure 3 presents marginal tax rates for the three hypothetical households in 1980 and 2004, when federal income taxes, the employee and employer share of FICA taxes, and food stamps and AFDC/TANF benefits for the single mother family (and their relevant phase outs) are included. As the figure shows, marginal tax rates have varied enormously over time. At the lowest income levels, marginal rates have decreased tremendously as the EIC phase in rate has become more negative. The marginal tax rate for the single mother family at the lowest income level fell from 2 percent to -23 percent between 1980 and 2004.

At low to moderate levels of income, however, families have seen their average tax rates decrease but their marginal tax rates increase, as a result of the higher EIC phase out rate. For example, at $25,000 of income, the marginal tax rate for the family of four was 29 percent in 1980 but 43 percent in 2004.


Hassett and Moore conclude that while concerns about redistribution have led to dramatic reductions in the federal income tax for low-income families with children, changes elsewhere in the code have not kept pace, and the overall tax system has not increased in progressivity as much as one would think if one focused solely on the income tax.

They call for further research, and note two cautions about their findings. First, they point out that direct measures of taxes paid may be poorly correlated with the actual incidence of the tax. If wages and prices shift because of taxes, the cost (or 'incidence') of the tax may shift to others than those who appear to directly pay the tax. Second, this study made no effort to attribute the corporate income tax to low-income individuals, although increasing tax competition over time may have lead to this tax increasingly being borne by workers.


1 For more information: http://www.npc.umich.edu/news/events/workingpoor/

2 http://www.npc.umich.edu/publications/workingpaper05/paper16/



About the authors

Kevin A. Hassett is a Resident Scholar and Director of Economic Policy Studies at the American Enterprise Institute for Public Policy Research.

American Enterprise Institute
1150 Seventeenth Street, N.W.
Washington, DC 20036
E-mail: khassett@aei.org

Anne B. Moore is a graduate student in the Department of Economics at the University of California at Berkeley.

E-mail: annebmoore@gmail.com


The National Poverty Center’s Policy Brief series summarizes key academic research findings, highlighting implications for policy.

The NPC encourages the dissemination of this publication and grants full reproduction right to any party so long as proper credit is granted the NPC. Sample citation: “Title, National Poverty Center Policy Brief #x”.

Major funding for the National Poverty Center is provided by the Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services.

Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the view of the National Poverty Center or any sponsoring agency.