"Economic Perspective" column reprinted with permission.
Copyright 2001 TAX ANALYSTS
The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.
Traditional tax policy analysts often object when expenditure goals are pursued through the tax system rather than through outlay programs. Tax legislation in 2001, however, has brought to light the difficulty with trying to deal with only one side of the budget at a time. For many policy objectives, consistent thinking and action is only possible by looking at both sides of the budget. Two prime examples played out in the recent legislation concerned progressivity and marginal tax rates. Here we will examine the progressivity issue; part two will deal with marginal tax rates.
The discussion of progressivity is hampered by the simple fact that we tend to use inconsistent measures on the tax and expenditure sides of the budget. From the tax side, we tend to measure progressivity by the rate of tax. For example: Is the system proportional (an average rate measure), less than proportional, or more than proportional? Alternatively: Does the marginal rate of tax increase or decrease as income increases? On the expenditure side the tendency has been to measure progressivity by whether more dollars are given to higher or lower income people. For example, a bill that gives more to the rich than to the poor is judged to be regressive—as measured by absolute dollars.
These two standards are inconsistent. We see this when we look at social security. Using the definitions above, the social security tax is regressive because the marginal rate is essentially the same up to a given earnings level, then falls to zero. Expenditures under the system, in turn, are regressive because higher earners get more dollars of benefits on average than do lower earners. Yet the social security system as a whole was meant to be progressive. In other words, on an overall basis the system aims to provide a net redistribution from richer to poorer people—a redistribution that doesn't appear possible when taxes and expenditures are both labeled regressive by inconsistent measures. (I skip over the issue here of whether this redistribution is actually achieved because of other factors such as higher mortality rates for those with lesser earnings.)
Now consider the recent income tax cut. The administration proposed an income tax cut that gave larger percentage cuts to those paying lower amounts of tax than it did to most taxpayers at higher income levels. In the 1970s people would have considered that type of proposal modestly progressive by the traditional tax definition. The complication, of course, is from a broader budgetary perspective; cutting taxes means something has to give on the expenditure side. Measuring across both sides of the budget, the change may not be progressive at all.
I once suggested in this Tax Notes column that in most cases larger government—in particular, modern government dominated by social expenditures—is almost always on net more progressive and redistributive toward lower income classes. Even a regressive tax—regressive by the traditional tax definition—is likely to take more dollars from the rich than the poor, while the expenditures financed by that tax are unlikely to give back to the rich the same disproportionate amount it took from them.
Take the simple case of a proportional tax and an expenditure that are exactly the same for each person. On net there is redistribution. Cut back on the tax in a proportionate or even disproportionate way that favors those with less income, and the amount of redistribution is still liable to be reduced.
By this type of calculus, almost any attempt to create a smaller government is going to produce less net redistribution (I don't address the efficiency and growth issues here, which often are the primary justification for either less or more government action). With a consistent measure of progressivity, almost any cutback in government can be attacked as "regressive," whether the reason is that government has become bloated, that a society wants to move away from communism or socialism, that revenues increased unexpectedly beyond what was necessary, that inefficiency rises exponentially as government becomes larger, or that such a move would be growth-enhancing.
How did this issue play out in 2001? Partly, because we have gone a long way toward removing lower income people from the tax rolls; it was hard to have an income tax cut that did much for those at the bottom of the income scale. But consider it more closely. The argument really is not so much that the government should give lower income persons a disproportionate share of an income tax cut. The argument, instead, is that lower income individuals deserve more expenditures.
The 2001 tax bill resolved some of the distributional issues by doing just that—adding in expenditures for households at moderate income levels. For the most part, expanded earned income tax credits and refundable portions of child credits are not income tax cuts at all. Even the budget counts the refundable portion of these credits as expenditures (although administered by the IRS). The issue could not be resolved or even compromised within the tax system alone. Mixing and matching across both sides of the budget were required.