Commentary | Budget Taxes and Public Investment

Lessons—How Tax Code Worsens Education Gap

Share this page:

These pieces originally appeared as a weekly column entitled “Lessons” in The New York Times between 1999 and 2003.

[ THIS ARTICLE FIRST APPEARED IN THE NEW YORK TIMES ON APRIL 25, 2001 ]

How tax code worsens education gap

By  Richard Rothstein

CONGRESS will soon debate the government’s biggest education program, Title I, which has origins in President Lyndon B. Johnson’s war on poverty and sends nearly $9 billion a year to schools with low-income children.

While some dismiss Title I as a failure, no one disputes its intent to aid needy children. Yet few recognize that over all, the federal government exacerbates inequality in education, giving more money to districts with affluent children than to those with poor ones.

It does so with a tax system that subsidizes school spending in home- owning communities, many of them upper middle class or even wealthy. Homeowners who itemize deductions reduce their federal income taxes by a portion of their property tax payments. A family in the 28 percent bracket that pays $1,000 in local property taxes for public schools can deduct that payment on its income tax returns, reducing its income tax bill by $280. Of the $1,000 going to schools, the family pays only $720 out of its earnings. The federal government contributes the $280 balance.

Economists term these subsidies “tax expenditures,” because they have the same effect as direct government spending. Yet the federal education budget highlights only direct outlays, perhaps because tax expenditures would be politically indefensible if widely publicized.

The property tax subsidy aids affluent families more than lower-income ones. It helps only those who itemize deductions, and itemizers have higher incomes on average than taxpayers who take the standard deduction. Nearly all families with annual incomes of $100,000 itemize, as against fewer than a third of families with incomes of $35,000.

And because the subsidy is tied to a family’s tax bracket, even among itemizers the subsidy grows as income rises. Families in the 28 percent bracket get a $280 subsidy for each $1,000 in property taxes, but those in the 15 percent bracket get only $150.

Dr. Susanna Loeb, a Stanford University economist, notes that this system spurs school spending in wealthy communities, both in total dollars and relative to spending in less wealthy districts. When larger shares of property taxes are underwritten by the federal government, families become more willing to raise levies for better schools. Districts in wealthier communities can raise property taxes more easily, knowing that Washington picks up more of the tab.

There are some offsetting factors. One is the alternative minimum tax, paid by those who claim so many tax breaks that they would otherwise pay little or nothing in income taxes; this effectively reduces the property tax subsidy. On the other hand, many other, less affluent taxpayers do not itemize deductions at all, mostly out of ignorance. A community’s schools get no benefit if its residents are lower-middle-income homeowners who take the standard deduction instead of itemizing.

Another countervailing factor is state income taxes, also deductible on federal forms. If a state uses its income tax revenue to equalize school spending, the federal system helps it do so. But this effect is limited. A homogeneous affluent community can more easily respond to federal tax incentives by voting to increase its property levy than a state as a whole can respond by increasing its income tax rates.

On balance, direct federal education outlays are mostly for poor children, while indirect spending mostly benefits the affluent. And federal tax expenditures for schools exceed direct spending.

Dr. Loeb has calculated federal per-student education spending for 1989. (Calculations for recent years must await data from the 2000 census.) She found that federally stimulated inequality occurs both among and within states.

In New Jersey, federal tax expenditures were $1,257 per student, but direct spending was only $237. In Alabama, tax expenditures were $165, while direct spending was $371.

Among districts within states, the differences were just as stark. Because tax expenditures are so high in wealthier districts, Princeton, N.J., got $2,399 in total per-student federal aid. But Camden, despite high Title I grants, got only $1,140.

Other tax expenditures increase inequality further. For example, the mortgage interest deduction also subsidizes homeowners’ costs, lifting property values. This, in turn, disproportionately adds to the income of wealthy school districts, because tax rates are a percentage of assessments.

Politically, it is hard to imagine that either Democrats or Republicans will meddle with these upper- middle-class tax benefits, or appropriate enough Title I aid to outweigh them. But there is something perverse about both parties’ proclaiming that they wish to leave no child behind, when the federal government plays so big a role in pushing affluent children farther ahead.

Return to the Education Column Archive


See related work on Budget Taxes and Public Investment | Education

See more work by Richard Rothstein