To be sure, it's true that most married couples, especially higher-income ones, could cut their taxes substantially by divorcing. But if Congress were to
|
Effects of Eliminating the Marriage Penalty Perceptions and Reality |
||||||
| Married Income Group | Income Range | Average Income | The "Penalty" You See | The Net Tax Cut You'd Get | ||
| Lowest 20% | Less Than $26,000 | $ 17,300 | $ +45 | $ 36 | ||
| Second 20% | $26,000 42,000 | 33,700 | +149 | 116 | ||
| Middle 20% | $42,000 60,000 | 50,100 | +151 | 117 | ||
| Fourth 20% | $60,000 85,000 | 70,500 | +689 | 400 | ||
| Next 15% | $85,000 163,000 | 110,500 | +1,142 | 493 | ||
| Next 4% | $163,000 409,000 | 236,400 | +2,878 | 659 | ||
| Top 1% | 1,086,000 | +11,858 | 613 | |||
| Notes: Penalties are the higher income taxes that couples pay compared to what they would owe if they divorced with an equal-income-split separation agreement. Net tax cuts reflect elimination of the marriage penalty in a revenue-neutral way. (Unmarried taxpayers would pay higher taxes, typically $67 more each.) Citizens for Tax Justice, January 1998 | ||||||
Suppose, for instance, that the cost of marriage penalty relief is covered by simply imposing a 5 percent surtax on all taxpayers. Then what starts off as a $34 billion a year gross tax cut for marrieds ends up as only about a $8 billion net cut-- accompanied by a $8 billion tax increase on unmarried taxpayers. When all the dust had cleared, the typical married couple making $50,000 a year would pay about $120 less in taxes, while the typical single taxpayer would pay about $70 a year more. A more sophisticated approach, shown in the accompanying table and chart, might make the total dollar shift between marrieds and singles a bit larger-- but still would end up with similarly small tax cuts for typical couples.
Although the "tax on marriage" is in reality much less significant than often claimed, it is annoying and unfair. But despite the fact that the marriage penalty is almost universally deplored, there is substantial disagreement over both whether it can be eliminated, and, if so, how to go about it. Some people argue that intractable problems in balancing the interests of married and unmarried individuals prohibit a universally fair resolution of the competing interests, and that the best we can do is to adopt some gimmick involving special deductions or credits for two-job couples. Others think the "solution" is to abandon the tax equity goal of treating couples with the same total income alike.
In fact, however, there is one approach which could completely wipe out the marriage penalty: simply tax married couples as if each spouse were a single person earning half the total family income. This "income splitting" approach would not be a radical step-- it was the law in the U.S. from 1948 until it was repealed in 1969.
Income splitting makes the reasonable assumption that married partners share the total family income about equally, and that each individual spouse therefore has about the same ability to pay taxes as a single person with half the family income. Current law, on the other hand, randomly assumes income sharing splits of between 60-40 and 80-20 in most cases, an approach that is hard to defend on any logical or empirical basis.
Undoing what was done in 1969-- and returning to the pre-1969 system of "income splitting" for married couples-- combines several compelling advantages: (1) It is the only system that completely eliminates the tax on marriage. (2) Unlike other proposals, it preserves the important principle that couples with equivalent incomes should pay the same amount in taxes. (3) It is by far the easiest system to administer. And (4) perhaps most important, it has the firmest underpinning in tax policy principles, combining respect for married and unmarried persons as individuals with conformance to the most commonly recognized pattern of marital income sharing.
| Given all these advantages, what stands in the way of returning to income splitting?
Certainly, distributional issues are not a significant concern. The current overall distribution of the tax burden can be easily maintained in the context of income-splitting (see table). The most significant political opposition to such a step would likely come from the Treasury and from advocates for giving preferred status to single persons. In addition, some economists maintain that the
tax on marriage, at least as it applies to one-earner couples, is justified by so-called "imputed
income" theories. And finally, there are reasonable arguments to be made in support of retaining
the tax on married created by the different standard deductions for couples versus two singles.
In the past, the Treasury has had problems with income splitting, largely due to the institutional angle from which its tax policy experts approach history. When the income tax was established, the framers did not confront issues like the tax treatment of the family. The Revenue Act of 1916 simply applied to "the entire net income...received by every individual," offering no suggestion as to how "income" should be apportioned among family members with potential claims to it. The issue was soon confronted in the courts, however, and the IRS succeeded in having its position-- designed to maximize revenue but without any basis in considered tax policy-- adopted. So the rules developed that wage income was taxable to the person who performed the services, and property income to the person owning (or having the best claim to) the property. Within these constraints, taxpayers worked on schemes, involving trusts, family partnerships, and the like, to have income attributed to other family members. Some of these attempts were successful, and all of them were attacked as "loopholes" by the Treasury and the IRS.
The biggest such "loophole," from the Treasury's viewpoint, stemmed from the fact that the courts, carried by the logic of their "property interest" rule, felt compelled to allow full marital income splitting in community property states, where, it could not be gainsaid, the income of either spouse was, legally and without the use of any tax avoidance device, attributable equally to each member of the couple. Thus, as a quite nonsensical consequence of the property interest rule, taxes on married couples tended to be higher in the common law jurisdictions. When tax rates shot up to finance World War II and stayed up ever after, the income tax changed from a class tax on the relatively wealthy to a mass tax on almost everyone. And numerous states began to move toward adopting community property laws to protect their married citizens from part of the increased federal exactions. In response, Congress voted in 1948 to allow income splitting for all married couples, wherever they happened to reside. Although Ways and Means Democrats (for example) congratulated themselves on convincing the Republican majority to make the change, based on "considerations of equity and the elimination of discriminations in the tax system," Treasury saw the congressional action as a capitulation to the income-splitting "loophole." And its opposition to the concept of income splitting has remained, even in the face of rather compelling arguments that such an approach is actually theoretically correct. As noted earlier, the tax on marriage was created in 1969 when Congress created a new tax schedule to provide a rate advantage to higher income single persons. Prior to that time, there had been basically only one rate schedule in the Internal Revenue Code, applicable to "the taxable income of every individual." In determining the taxable income of married individuals, the code had provided since 1948 that each partner in a couple was to be treated as enjoying half the joint income. Once this split was made, married people were taxed on their individual incomes using the same rate table employed by single persons. The singles rate table was adopted in 1969 in response to protests that the system of allocating income between spouses discriminated against singles. This claim of discrimination grew out of a comparison between the tax paid by a married couple and the tax on a single person with the same income. Since the tax on the single person was higher than the sum of the taxes on the two married persons, the tax system was said to impose a "tax on remaining single." This complaint must be distinguished from the "tax on remaining unmarried" discussed above, where the comparison was between a married couple and two unmarried persons. The alleged "tax on remaining single" is, of course, the normal consequence of a progressive income tax system. With progressive tax rates, a person with $60,000 of income should generally pay more in tax than two persons each with $30,000. Under income splitting, a husband and wife with family income of $60,000 are treated as each taxable on their share of the income pool, in this case, on $30,000 each. The total tax on the couple is therefore less than the tax on a single person with an income of $60,000. Those complaining about a "tax on remaining single" apparently considered a married couple as one taxpayer instead of two, presumably treating one spouse as the real taxpayer and the other as the "tax shelter." The mind set which produced this offensive categorization was, unfortunately, rather common even in supposed "enlightened" circles until very recently. Thus, in 1964, liberal tax reformed Phillip Stern, in his best-selling and influential The Great Treasury Raid, could refer without a twinge of embarrassment to the worth of "the little woman" as an "asset" of great benefit to "the husband, in making out his tax return." Such an approach may be encouraged by the rather unwise phraseology of Form 1040, which has persons filing a joint return list one marital partner as the taxpayer and the other as the spouse. In reality, of course, both are taxpayers (and both are spouses), even if one of the partners provides the source of the entire family income. To ignore the existence of one of the marital partners is to pretend that two persons with a given total income are, by the fact of marriage, in the same economic position as one person with the same total income-a quite nonsensical assumption. Whether this outlook is ultimately based on outrageous sexism, unabashed self-interest, or mere logical error is unclear. But it is clear that the philosophical position which formed the basis for the 1969 attack on income splitting was bankrupt both then and now, and Congress ought to admit to its mistake. Many singles advocates are willing to admit that two cannot live as cheaply as one, but they still maintain that economies of scale typically enjoyed by married persons make them better able to pay taxes than are singles. Thus, a married couple needs only one bed, one refrigerator, and one house. Meals for two are cheaper to prepare per person than meals for one. And so on. Of course, this argument presumes that single people live alone, which means that the preferential singles rates should not be available to unmarried couples or singles with roommates-a large percentage, perhaps a majority, or single people. Ultimately, the economies of scale argument fails because it proves too much. We simply do not want a tax system-or an Internal Revenue Service-which inquires so deeply into how we structure our personal affairs. Rather, a system which taxes our measured economic incomes, leaving to individuals the decision as to how to spend their money and structure their living arrangements, is far preferable. One additional argument frequently put forward in an attempt to justify preferential tax treatment for singles and/or two-job couples--as opposed to one-earner couples--involves the concept of "imputed income." One of the unfortunate failures of our income tax system, some economists are fond of lamenting, is its inability to take account of the value of self-performed services or leisure. But, it is argued, we can mitigate this pernicious deficiency by taxing those classes of people likely to have excess shares of such "imputed income" more heavily than those likely to have less of it. The concept of "imputed income" is elastic enough to include virtually everything people do for themselves, from reading a book to fixing the car, from washing the dishes to raising one's children, from chewing food to enjoying the sunset. Attempts to limit the scope of the idea in a principled fashion to activities relevant for tax purposes have not had great logical success. Focusing exclusively on items that can be purchased in the marketplace might cut out sleeping and chewing, but would still leave shaving, grooming, and gardening. De minimis rules still retain items frequently performed-- a year's worth of shaves would cost upwards of $700 if purchased. Somewhat suspiciously, the proponents for making tax adjustments for assumed levels of "imputed income" have chosen to focus particularly on activities thought to be performed by stay-at-home spouses, primarily wives. As one paper in this field put it, "Because the housework and leisure of a non-working spouse is excluded from taxable income, the taxable income of one-earner couples is understated relative to that of two-earner couples . . . [and] single persons." One can marvel at the elegant way in which this statement finesses one critical problem: the lack of evidence about "housework" levels as between one job couples on the one hand, and two-earner couples and singles on the other. Even if-- as may well be true-- many, or most, two-job couples and singles are just as neat as their one-job neighbors, and even if many, or most two-job couples and singles are at least as handy with cars, electrical work, and gourmet cooking as their "traditional family" counterparts, the one-job couples clearly have more time to engage in such activities. And if they waste this opportunity by watching TV or smelling the daffodils, why, so much the worse for them. We can still tax their extra leisure-or their potential imputed income. Now this raises some serious questions on its own--treating leisure as taxable would place a heavy burden on small children, retired persons, students, and the unemployed. And the federal income tax has never been considered as a levy on "potential income," although many economists might wish it were so. When one actually confronts the "practical" approached advocated for taxing "imputed income, one's fears that this frightening concept might have some relevance to tax policy are mercifully laid to rest. If anything, the schemes seem even more ludicrous in application. One system, actually part of the tax code from 1981 to 1986, was to provide a special deduction or credit for two-job couples (why singles are excluded is not completely clear), based upon a percentage of the earnings of the lower-income spouse. This plan assumes that dollar earnings are a reasonable proxy for leisure time foregone (leisure time, it must be remembered, being a reasonable proxy for "potential imputed income"), a proposition which minimum wage workers might question in light of, for example, hourly attorney fees. Another approach is to provide a deduction (or credit) for purchases of certain kinds of services which, it is assumed, one-job couples provide for themselves. This was the rationale when Congress adopted the credit for child care expneses that remains in the law today. But if the purpose of the credit is to make an allowance for the failure of the tax system to tax the "imputed income" from taking care of one's own children, then the credit is exceedingly unfair to those without children (who therefore perform no child care services, and have no "imputed income" therefrom). If self-performed child care services are to be treated as "income," the fairest system would seem to be to provide a large deduction for taxpayers without children, and a somewhat smaller one for those with children who purchase child care services (and presumably perform some such services themselves as well). To be sure, the child care credit has also been defended as an allowance for a necessary business expense of working parents. But if this justification makes sense, then the provision should be a deduction rather than a credit. In fact, the child care credit is really a tax expenditure in search of a fully logical rationale. Similar analyses can be made of other "imputed income" tax proposals, including differential rates. The fundamental point is that the "imputed income" argument for higher taxes on one-job couples is so arbitrary in its choices of what to tax and so illogical in its application, that it should not be seriously considered in writing the tax laws. As noted above, differences in the standard deduction are the primary cause of the tax on marriage for average and lower-income couples (excluding the massive marriage penalties created by the earned-income tax credit rules, which are beyond the scope of this essay). This discrepancy could be eliminated by making the married standard deduction exactly twice that for singles, but such a step would conflict with one of the stated rationales for the standard deduction-- to assure that the poor pay no income tax. The poverty level is a welfare term and is officially defined in terms of marital status. The current standard deduction, in conjunction with exemptions and credits, results in a no-tax level of about 15% above the poverty level for a childless couple, and about 6% below the poverty level for a single person. If a marriage neutral standard deduction-- say $4,000 per person-- were established, this would reduce the no-tax level to only 92% of the poverty line for singles, while increasing it to 123% for childless married couples. Whether this would necessary be a bad result is debatable. The assumption that married persons have a monopoly on cost-saving arrangements is clearly not true, since single persons can also arrange to pool their resources and share costs. And, in any case, it may be more important for the tax system to achieve neutrality than to foster welfare goals. Nevertheless, it does not seem likely that Congress would be willing to abandon the poverty-related aspect of the standard deduction. It explicitly confronted the problem in the 1977 Tax Act and decided that the tax on marriage created by the standard deduction was a necessary evil. This conclusion is certainly a reasonable one. But it certainly has no relevance to the more significant tax on marriage created by the preferential tax rates for singles. On Capitol Hill, one marriage tax "solution" that has gained significant support is a perennial bad idea: "voluntary separate filing." Introduced by Rep. David McIntosh (R-Ind.), this plan would give married people the option of ignoring their marriage contract and filing as singles when this produces a tax advantage. If, as has been already pointed out, Treasury's judgment about possible solutions to the marriage tax may be somewhat overwhelmed by its historical perspective, supporters of the McIntosh proposal have no such problem. Instead they seem intent on returning the tax system to the pre-1948 regime, apparently in blissful ignorance of the multitudinous problems which existed in these "good old days." Beneath the surface appeal of their voluntary separate filing approach lurk administrative nightmares and clear and gross unfairness, as ought to be apparent to anyone who analyzes our pre-1948 experience. Separate filing supporters are willing to jettison the principle, now well-established, that couples with equivalent incomes ought to pay the same tax bills. Thus, under their optional separate filing rule, a couple with an 85%-15% earnings split would pay 9 percent more in taxes than a 50-50 couple at the $30,000 or $50,000 income level, 13 percent more at the $60,000 level, and 25 percent more at the $70-$80,000 level. Furthermore, couples with substantial shares of their incomes from property (such as interest and dividends) could use all the devices perfected in the pre-'48 era (and all the new ones which sharp tax lawyers could easily create) to maximize their tax savings at the expense of wage income couples. Finally, the "tax on marriage" would still remain for most couples: It would still pay to obtain a divorce with an alimony agreement to split income. As people have discovered when separate filing was proposed in the past, it is hard to conceive of a worse approach to the marriage penalty problem than the McIntosh plan. The potential revenue significance of the "tax on marriage" underscores the fact that it is not to be trifled with. McIntosh, for example, proposed no offset for his $18 billion a year separate-filing plan other than unspecified reductions in government services or a larger national debt. As noted earlier, however, the $34 billion that would be "lost" to the Treasury from truly eliminating the marriage tax (by returning to income splitting) could be made up by increasing other revenues, probably individual income taxes generally. That brings us back to a more manageable bottom line: an average tax cut of $117 for the typical married couple and an average tax increase of $67 for the typical single. Furthermore, if the marriage penalty were eliminated as part of a program of real tax reform, the shifts in burden could be even less, since tax preferences tend to benefit married people more than singles. Rep. Dick Gephardt (D-Mo.), for example, has introduced a tax overhaul plan that closes loopholes, puts three-quarters of all taxpayers in the 10 percent or less tax bracket and by the way, eliminates the marriage penalty by returning to full income splitting. Because of his reforms, Gephardt is able to reduce personal income taxes for all but the very highest-income people, and yet raise just as much in total revenue as does current law. Congress will be debating a number of ways to change our tax system over the next few year. Some of the proposals-- such as a high-rate national sales tax or a flat-rate wage tax-- are technically, fiscally and distributionally disastrous. Others, such as the Gephardt plan, have more promise. But whatever happens, the debate could provide a good opportunity for Congress to eliminate the unnecessary marriage tax inequity from our tax laws. IMPORTANT CAVEAT: This article does not address the very large disparities between married and unmarried couples that can be created by the earned-income tax credit-- which can approach $6,000 a year in some cases for a couple making $25,000. More information on the EITC marriage tax issues can be obtained from Citizens for Tax Justice, 202/626-3780. Scoring your answers to the quiz according to the following scale will give a rough gauge of your attitude toward what, if anything, ought to be done about the marriage tax. A high "plus" score means you'd probably favor elimination of the marriage penalty for all couples, by adopting a joint filing rule with half of a couple's total income attributed to each partner ("income splitting"). A very negative score means you'd probably prefer a system of individual filing, with no spousal attribution rules. This would reduce the tax on marriage for many two-job couples, but would also mean that couples with the same total incomes could pay very different total tax bills. If your score was in the middle, you may be satisfied with the present system or, at most, would like to see a special deduction or credit for some two-job couples.
INTERPRETING THE RESULTS
People & Taxes, Vol. VIII, No.3, May 1980  Back To Reports
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||