February 8, 2000 | Center for Data Analysis Report on Taxes
The House and Senate tax-writing committees will try again this year to develop legislation to reform the marriage penalty in the tax code. The challenge is to craft a measure that Congress can pass and the President can sign. Last year, President Bill Clinton vetoed the Taxpayer Refund and Relief Act, which contained $117 billion in marriage penalty relief over the next 10 years. The veto forced about 25 million working families to pay more in income taxes because of the marriage penalty, and it perpetuated the troubling second-earner bias forcing lower earning spouses to see their pay frequently taxed at higher rates than the income of their higher earning spouses.2
This Report contains new estimates of the number of couples in each congressional district and state who pay some form of the marriage penalty in 2000. (See table.)3 These estimates are based on data from the 1999 March Current Population Survey and cover all combinations of the three most frequent reasons that marriage penalties arise:
The marriage penalty stems from the federal government's effort to do three things: 1) tax equal-earning couples at the same rate, 2) tax them at progressive marginal tax rates, and 3) recognize the economic benefits of marriage by requiring married couples to file their taxes on a schedule of tax rates that treats them less favorably than it does single taxpayers. As the Congressional Budget Office notes, "The incompatibility of those three goals...results in continuing tension within the tax code."4 This tension in the tax code harms the pocketbooks of American families and the institution of marriage, and has significant implications for the economic and cultural health of the nation.
The marriage penalty is arguably the most significant of the biases affecting the secondary earner (the spouse with the lower income). As two prominent tax economists have observed, "the basic source of the marriage tax is the fact that key elements of the tax law depend on an individual's family situation, including the rate schedule, the standard deduction, and the earned income tax credit. Hence, the act of getting married per se affects individuals' tax liabilities, even if their work and saving decisions stay the same."5
In most cases, federal income tax laws require that married couples file joint tax returns based on the combined income of husband and wife. When a husband and wife both work, the secondary earner is, in effect, taxed at the top rate of the primary earner. As a consequence, a married couple may pay more taxes than they would if each spouse were taxed as a single wage earner.
According to the Congressional Budget Office, an estimated 42 percent of married couples incurred a marriage penalty in 1996: "more than 21 million married couples paid an average of nearly $1400 in additional taxes in 1996 because they must file jointly."6 Most severely affected by the marriage penalty were couples with a more equal division of income between husband and wife and those who receive EITC benefits. Essentially, Americans with the lowest incomes and families dependent on two wage earners shoulder the largest marriage penalty burdens under the current tax policy.
Consider what happens to two $30,000-a-year wage earners who decide to marry. As a single individual, a $30,000-a-year wage earner would pay $3,000 in taxes. The principle of marriage neutrality would mean that when a $30,000-a-year wage earner marries another $30,000-a-year wage earner, the couple's tax liability should be $6,000. Under the current joint filing schedule, however, this married couple--that now earns a total of $60,000--owes $8,400 in tax per year, a $2,400 penalty for marrying each other.
According to the ideal of marriage
neutrality, tax burdens should not be altered when two people
decide to marry. However, the goal of progressive taxation is
violated under such circumstances.
Progressivity states that a person (or, under today's joint filing requirement, a combination thereof) who has twice the income of another would pay more than twice in taxes. The current tax system sides with the ideal of progressive taxation and punishes hardworking Americans.
The marriage penalty can have significantly negative economic implications for the country as a whole. Not only does this feature of the tax system stand as a likely obstacle to marriage, it can actually discourage a spouse from entering the workforce. Edward McCaffery, a law professor at the University of Southern California has said that: "By adding together husband and wife under the rate schedule, tax laws both encourage families to identify a primary and secondary worker and then place an extra burden on the secondary worker because her wages come on top of the primary earner's. The secondary earner is on the margin."7
As the American family realizes lower income levels, the nation realizes lower economic output. From a strictly economic standpoint, the fact that potential workers would avoid the labor force as a result of peculiarities in the tax code is a clear sign of a failure to maximize eligible resources. As a result, the nation as a whole fails to reach its economic potential, which is demonstrated by decreased earnings, output, and international competitiveness.
This analysis estimates the number of married taxpayers who are affected by marriage penalties in the tax code that stem from the standard deduction, taxable income thresholds for marginal tax rates, and the phase-out structure of the Earned Income Tax Credit (EITC). The vast majority of couples that we estimate to be affected by a marriage penalty are two-earner families in which the secondary earner's wages are a significant portion of a family's income and taxed at a higher marginal rate. Readers should note that these incidence estimates cannot be directly employed to estimate the dollar amount of penalty per family due to differences in the marginal rate structure and the EITC.
We employed the March 1999 Current Population Survey, which contains 1998 income and demographic data, to determine the number and type of families that suffer the marriage penalty in each state.8 We then used the most recent IRS Public Use File to determine the percentage of families in each tax bracket that typically use the standard deduction instead of itemizing their taxes.9
We assumed that married taxpayers will incur a marriage penalty if they had two earners in a tax bracket that is higher than 15 percent or if they were likely to use the standard deduction in the 15 percent bracket. We also assumed that married families receiving the EITC suffer a penalty due to the steeper phase-out of the credit for married taxpayers than for single taxpayers. Eligibility for the EITC is determined on the qualifying income of the taxpayers. The phase-out range for married taxpayers is less than twice that of two single taxpayers. Readers should note that the number of families who do not currently receive the EITC but would if they were both single was not used in this analysis.
The estimated percentage of families suffering a penalty at the congressional district level was derived from the 1990 Census updated to the 104th Congress. The percentage of families likely to suffer a penalty was held constant for families in 1998. The overall number of families affected in 1998 is based on U.S. Treasury estimates of 25 million.10
Tables Number of Married Couples Affected by the Marriage Penalty by Congressional District and State
Note: Estimates may
not sum due to rounding. Population data based on 1990 census;
congressional district boundaries based on 104th Congress.
Sources: Heritage calculations based on Census Bureau and Internal Revenue Service Data.
2. U.S. Treasury Department estimates cited in Curt Anderson, "GOP Offers Bigger Marriage Tax Cut," Associated Press wire story, January 31, 2000. Also see House Ways and Means Committee press release, "Archer Announces Marriage Tax Penalty Relief Act of 2000," January 31, 2000.
10. Anderson, "GOP Offers Bigger Marriage Tax Cut." For a complete description of how estimates of married families are calculated by congressional district, see Gareth G. Davis and Philippe J. Lacoude, What Social Security Will Pay: Rates of Return by Congressional District(Washington, D.C.: The Heritage Foundation, 2000), pp. 159-168.