Democrats control Congress, and one of their top priorities is reducing the number of taxpayers afflicted by the alternative minimum tax (AMT), probably by raising other taxes. But they should start paying attention to some imaginative thinkers on the left who are proposing making the Earned Income Tax Credit (EITC)—this generation’s most important new means of helping the poor—more generous, and eliminating a wide range of “categorical” poverty programs, from housing vouchers to food stamps. Democrats could combine this sweeping change with another one: using the billions that would be saved in the categorical programs’ administrative costs to help fix the AMT problem. In essence, this would be a grand compromise between social liberals and fiscal conservatives, shrinking government while still using it to reduce inequality.

Congress passed the AMT in 1969 to force a handful of the extremely wealthy to pay taxes, but this year it’s on track to hit more than 23 million upper-middle-class households. New York’s Charles Rangel, chairman of the House Ways and Means Committee, has hinted that this summer we may see a formal proposal for dealing with the AMT, in all likelihood by raising other income tax rates or taking a bigger bite out of hedge-fund proceeds.

But the EITC presents a better way to reduce the AMT’s impact. Expansion of the EITC—essentially a cash supplement to wages for the working poor—was one of President Clinton’s signature accomplishments, the key instrument that he chose to “make work pay.” Its more than $40 billion in annual payments to low-income households already surpasses the costs of other major antipoverty programs, such as food stamps ($35 billion), housing vouchers ($16 billion), and Temporary Assistance to Needy Families, or welfare (just over $17 billion). In effect, the EITC has made the Internal Revenue Service the country’s most important antipoverty agency.

Here’s how it works. The EITC matches each dollar in earnings with up to 30 cents. A single parent with two children earning up to $14,800 qualifies for the maximum payment of $4,536, for example. But as a household’s earnings grow, the government reduces the EITC rapidly, cutting it in half by the time a household earns $25,000 and eliminating it altogether once earnings reach just $36,000.

Hidden in those figures is what Harvard economist Jeffrey Liebman, who has studied the EITC extensively, calls “one of the largest marriage penalties” in the tax code. “If a custodial parent of two children with zero earnings—that is, relying on welfare—marries a single man with $12,000 in earnings, the couple will gain nearly $4,000 in child tax benefits, all from the Earned Income Tax Credit,” Liebman and his colleague David Ellwood wrote in 2001. “But if the mother were already earning $12,000 and she married the same man earning $12,000, child benefits would fall by roughly $1,200 and serve as a marriage penalty. At $24,000, the child benefits are smaller than at $12,000.” And the EITC’s marriage penalty has grown even larger since 2001. This system is counterproductive, to say the least, since it discourages both marriage—the key engine of uplift for the poor and their children—and work.

Ellwood and Liebman, who both served in the Clinton administration and are among the Democratic Party’s most creative public intellectuals, have explored the idea of removing the marriage penalty by extending the income range that the EITC benefits. Their system would also increase the maximum benefit and reduce it much more gradually, as households’ earnings rose, than at present. Such a plan, they observe, would not only boost aid to the poorest but also improve the lot of lower-middle-class families who, as they work their way up from poverty, find themselves hit by a suddenly declining EITC. And innovative ideas about the EITC aren’t confined to academe. In New York, the Bloomberg administration, no longer Republican even in name, may soon offer its own version of an EITC expansion, realizing that the current program provides a disincentive for low-income fathers to marry the mothers of their children. But expanding the EITC would be expensive: Ellwood’s and Liebman’s various proposals bear price tags ranging from $15 billion to $53 billion.

That’s where a grand social-policy compromise could help. We could pay for the newly expanded EITC by scrapping “categorical” antipoverty programs, in a plan similar to the one that Charles Murray recently proposed in his book In Our Hands—though Murray’s proposal is far more ambitious, replacing Social Security as well as antipoverty programs. “Cashing out” food stamps, housing vouchers, assistance for heating costs, and the Women, Infants, and Children program for new mothers—just some of the more prominent categorical programs—would yield $50 billion in program spending and, crucially, at least $5.5 billion in administrative costs. We could redirect the program spending into the pockets of the working poor through an EITC increase—and use the administrative savings to help reduce the AMT for those of middle income. It won’t be enough to fix the AMT entirely, of course. But if Democrats want to pay for most of an AMT reduction by raising marginal tax rates and hiking the tax rates on hedge-fund proceeds, it’s only fair to ask the antipoverty programs that those taxes make possible to chip in by becoming more efficient.

Elements of such a grand bargain are already being floated—again, by creative Democrats. David Riemer—a former top aide to a Democratic governor of Wisconsin and a Democratic mayor of Milwaukee, as well as a onetime aide to Senator Edward Kennedy, and currently the head of an organization dedicated to expanding health-care access—has pushed for a new, cash-based approach to fighting poverty. The cost of such a system would be “more than offset by the elimination of dozens of symptom-treating programs in the area of nutrition and housing,” he says.

Further, cashing out categorical programs would minimize the market distortions that they create. The Section 8 housing-voucher program, for instance, encourages low-income households to cluster in those neighborhoods where landlords are willing to accept them. More broadly, getting rid of categorical programs would pass responsibility for life decisions from the government to low-income households themselves. Why, after all, should we tell the poor how to spend the assistance that we give them? Why not, instead, simply augment their incomes directly—tying the assistance to a work requirement, but otherwise trusting them to know what’s best for them? Let’s take steps to ensure, as Clinton put it, that those who work are not poor. That we can do so in a way that helps provide middle-class tax relief makes the prospect all the more compelling.


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next