TCS Daily

Reality Economics

By W. James Antle - May 27, 2003 12:00 AM

While the passage of a tax plan that at least temporarily ends the double-taxation of dividends is a victory of sorts, the final package could have been stronger and the uncertainty of changing dividends tax policy could have been eliminated. Yet another tax cut has been reduced in size largely due to costs projected using static revenue scoring.

Such developments are why proponents of low-tax, free-market economic policies favor jettisoning the static method of scoring proposed tax cuts in favor of dynamic or "reality-based" estimates. An entire tax debate was bogged down in such essentially imaginary numbers as $726 billion, $600 billion and $350 billion. The case against static scoring is powerful. This method falsely assumes that taxpayers do not change their behavior in response to changes in tax rates, a position at variance with such basic economic concepts as supply and demand. Greater rewards for work, innovation, and risk-taking produce more of the same.

Consequently, static estimates have routinely overstated the reduction in revenues that follows tax cuts. A case in point was the 1997 cut in the capital gains tax rate from 28 percent to 20 percent. It was scored as a revenue loser under static analysis, but in fact receipts from this levy doubled within four years. Many sound tax cut proposals have never even been given a chance to prove these projections wrong; inflated estimates of revenue loss have given lawmakers "sticker shock" and prevented their enactment. Fear that lower marginal tax rates would reduce revenues by the amount shown in static estimates without any long-term payoff has been a serious obstacle to aligning federal tax policy with economic reality.

But can reality-based scoring effectively replace the conventional, if inaccurate, assumptions that block pro-growth tax initiatives? There are several problems that hinder the adoption of dynamic analysis as a widely accepted method of scoring tax policy changes.

First, there does not seem to be any consensus as to what the dynamic model would look like. How much will a given reduction in marginal tax rates increase economic growth? What level of economic growth will provide the right degree of "feedback effect," offsetting revenues that would be presumed lost by direct cost estimates? Is there even a consensus as to what the prohibitive range on the Laffer Curve is?

If there is no consensus on the above questions among proponents of dynamic analysis, how can the resulting estimates be persuasive to those who don't share our assumptions about lower marginal rates and economic growth? Will dynamic estimates simply be viewed as the conservative/market liberal equivalent of the "fuzzy math" employed to discredit tax cuts? We don't trust the estimates of left-liberal economists, for good reason. But to what extent will we really advance tax policy if the wider public doesn't accept ours?

As is the case with all budget projections, there is also the risk of political manipulation. This is probably to some extent unavoidable when it comes to projections made by specific administrations or congressional groups. But the adoption of agreed-upon standards for dynamic analysis would mitigate this practice by research institutes, the Congressional Budget Office and nonpartisan entities. Static analysis is used to prevent the enactment of tax cuts by presenting them as too costly; dynamic analysis could potentially be abused to offer "rosy scenarios" that understate the revenue loss of proposed tax cuts. Contrary to popular belief, supply-siders do not believe that all tax cuts will increase revenues. Tax rebates and credits are typically sure revenue losers. When marginal tax rates are lowered to a certain point, further cuts will lose revenues. A realistic dynamic model - true reality-based scoring - would need to incorporate these facts.

Dynamic scoring also fails to address a key fallacy promoted by tax-cut opponents: that any reduction in federal revenues is necessarily a bad thing. By contrast, Milton Friedman has argued that the most important reason to cut taxes is to deprive the government of the money it needs for further spending increases. There appears to be a bipartisan consensus that most federal spending is sacrosanct and that the overall level of expenditures should never be cut. Yet the problem of tax rates that are counterproductive from an economic growth perspective fundamentally derives from a ceaselessly expanding government.

Even with these potential problems, dynamic scoring is conceptually better than the unsound assumptions utilized in static estimates. Analysis that leads to more realistic conclusions about the revenue effects of tax rate changes, and thus fosters more productive debates about shaping tax policy is an objectively good thing.

TCS Daily Archives