Originally published in The Wall Street Journal

July 20, 2010

The 'Tax Expenditure' Solution for Our National Debt

By MARTIN FELDSTEIN

(PDF Version)

When it comes to spending cuts, Congress is looking in the wrong place. Most federal nondefense spending, other than Social Security and Medicare, is now done through special tax rules rather than by direct cash outlays. The rules are used to subsidize a wide range of spending including education, child care, health insurance, and a myriad of other congressional favorites.

These tax rules—because they result in the loss of revenue that would otherwise be collected by the government—are equivalent to direct government expenditures. That's why tax and budget experts refer to them as "tax expenditures." This year tax expenditures will raise the federal deficit by about $1 trillion, according to estimates by the congressional Joint Committee on Taxation. If Congress is serious about cutting government spending, it has to go after many of them.

For example, the Joint Tax Committee identified more than a dozen tax-based programs that subsidize education and training. These include small ones like the Coverdell education saving accounts (with a 2010 tax expenditure cost of $100 million) and much larger ones like the various tax credits for tuition (costing $11.7 billion). The hundreds of other tax expenditures include a $500 million annual subsidy for the rehabilitation of historic structures and a $4 billion annual subsidy of employer-paid transportation benefits.

President Obama's recent plan to expand the existing dependent care tax credit is a good example of how the welfare state grows through the tax code. At the same time he proposed a three-year freeze on all nondefense discretionary programs, Mr. Obama disingenuously called for an increase in the $3 billion tax credit for taxpayers who pay someone to look after their children or their aged parents while they go to work.

Neither party has focused on controlling this kind of spending. Democrats are reluctant to cut such programs, because once built into the tax law they don't have to be reauthorized each year, but remain on the books unless they are repealed. Income limits on the taxpayers who can take these deductions or tax credits allow Congress to target the benefits to lower-income groups. Moreover, many tax expenditures are refundable, so the government sends the individual a check for the benefit even if he owes no tax. Democrats can thus cleverly avoid the traditional accusation of being the party of "tax and spend."

Republicans also are reluctant to cut these tax perks, because they regard the additional revenue collected by the federal government as a "tax increase"—even though the increased revenue is really the effect of a de facto spending cut. A Republican who would vote to cut or eliminate an ordinary spending program therefore won't do so if it is packaged as a tax benefit.

But eliminating tax expenditures does not increase marginal tax rates or reduce the reward for saving, investment or risk-taking. It would also increase overall economic efficiency by removing incentives that distort private spending decisions. And eliminating or consolidating the large number of overlapping tax-based subsidies would also greatly simplify tax filing. In short, cutting tax expenditures is not at all like other ways of raising revenue.

If tax expenditures are not cut, taxes on households and businesses will have to rise to prevent an explosion of the national debt, which is now projected to increase to 90% of GDP by 2020 from today's 63%. When benefits for Social Security and Medicare are set aside, the rest of the outlay side of the budget is too small—7.5% of GDP—to provide much scope for reducing annual budget deficits that are now projected to average 5% of GDP for the rest of this decade. In contrast, total tax expenditures are now 6.4% of GDP.

Not every type of tax expenditure should be cut. Some provide good incentives while others increase the fairness of the tax system. But they can be reduced by one-third or more.

Tax expenditures have been cut before on a large scale. President Ronald Reagan's 1986 tax reform reduced tax expenditures to 6% of GDP (from 9%), the level at which they remain today. Cutting them another 2% of GDP would reduce the national debt in 2020 by some $4 trillion, bringing the projected debt down to 72% of GDP from 90%.

In 1986, the cuts in tax expenditures were made politically attractive by combining them with tax rate reductions. Although such rate cuts cannot be afforded now, there is currently widespread public agreement that the deficit must be reduced, and a growing acceptance that cuts in government spending are the way to do it.

Cutting tax expenditures is really the best way to reduce government spending. And to be politically acceptable, the cuts in tax expenditures must be widespread, requiring most taxpayers to give up something so that the fiscal deficits can decline.

While some of the dozens of small tax perks should be eliminated all at once, others should be reduced gradually in order to avoid economic disruptions. Some of the biggest ones, like the deduction on federal tax returns for local property taxes (projected to cost the federal government $25 billion in the coming fiscal year) might be reduced but not completely eliminated.

One simple approach would be to reduce such tax deductions by 10% in the first year, 20% in the second, and so on until the deduction is cut to 50% of its current size. That same kind of gradual and partial phase-down could also be applied to some of the employer-provided benefits such as life insurance premiums and travel costs that are now excluded from taxable income.

The enormous projected fiscal deficits are a threat to our economic future and our national security. The American public wants to reduce those deficits by cutting government spending. A major reduction of the spending that is built into our tax code is the best way to achieve that.

Mr. Feldstein, chairman of the Council of Economic Advisers under President Reagan, is a professor at Harvard and a member of The Wall Street Journal's board of contributors.