The appointment of Keith Hall as director of the Congressional Budget Office coincides with the adoption by Congress of a rule change that requires “dynamic scoring” of proposed tax law changes.
Hall is chief economist at the U.S. International Trade Commission, served as head of the President’s Council of Economic Advisers under George W. Bush and was Commissioner of the U.S. Bureau of Labor Statistics from 2008 to 2012. In April 2015, he will replace CBO Director Douglas Elmendorf, who has served since 2009.
The appointment of Hall appears to signify an intention, going forward, for the CBO to adhere to the spirit, as well as the letter, of the new congressional mandate. He should not be distracted from this effort by politically-driven griping from economists who should know better than to question the congressional intent behind that mandate.
The only question that Hall or any competent economist might ask in considering the question of dynamic scoring is, “Why would we do it any other way?” Dynamic scoring means measuring the impact on tax revenues of a change in tax law by taking into account how that change will affect the base on which the tax is imposed. Because changing a tax law will always change the economic activity on which the tax is imposed, it would be nonsensical to assume the tax base will remain fixed under a new law. Yet that is exactly what the proponents of “static scoring” want to assume. And it is static scoring that dynamic scoring is intended to replace.
Suppose that someone committed to the idea of static scoring thinks income taxes are too low. If the combined federal and state income tax applicable to the top federal tax bracket is 50 percent (as it is in some states), then, under the canons of static scoring, we might as well double the rate to double the amount of revenue collected from that tax bracket. Dynamic scoring would produce the obvious conclusion that the amount of revenue collected would go to zero, inasmuch as no one will bother to earn (or report) any income that is taxed at 100 percent.
Actual proposed tax changes are seldom so drastic (even though the top federal rate was 91 percent when Kennedy took office and 70 percent for Reagan). But the principle always holds: If the government increases the tax rate 10 percent, revenues will rise less than 10 percent because the base will shrink in response to the higher rate. Similarly, if the government cuts the tax rate 10 percent, revenues will fall less than 10 percent since the base will expand. Anyone who took Economics 101 had to understand that principle in order to get a passing grade.
Oddly, some of the nation’s top economists are raising concerns about that principle. For example, Alan Blinder complained in the Wall Street Journal that dynamic scoring of tax increases will fail to account for the benefits that might redound from the new spending a tax increase would finance. Blinder should realize that the only question here is whether the new spending is justified, given the inevitable shrinkage in economic activity the new taxes would bring about. Gregory Mankiw refers to dynamic scoring as “opening a can of worms,” raising the suspicion he would be more comfortable with static scoring — even though, by his own admission, static scoring makes no sense.
If we were to psychoanalyze economists willing to take issue with dynamic scoring, we might suggest their subconscious worry is that, in order to measure the effects of tax increases on tax revenues, it is necessary to understand how those increases exert disincentive effects on economic activity. Dynamic scoring has the collateral effect of revealing how every tax increase, no matter how well intended, shrinks the base on which it is imposed. If that base is income, then we get less income and less production. By exposing this negative effect, dynamic scoring forces lawmakers to weigh the bad against the good in considering new spending programs and the inevitable tax increases they require.
How will Keith Hall stand up against economists who would rather not recognize the bad effects of additional taxes? There is every reason to be hopeful. Hall worked in the administration of George W. Bush and at the Mercatus Center at George Mason University. Both the Bush administration and Mercatus understand the principle behind dynamic scoring. Hall’s own research has focused on concerns about declining labor-force participation and the disincentive effects of safety-net legislation, minimum wage laws and Obamacare. Like tax hikes, these policies have negative dynamic effects.
The Beacon Hill Institute and the National Center for Policy Analysis’ Tax Analysis Center are particularly interested in what Hall has to say about macroeconomic modeling. He provides some clues in a comment he submitted to the Environmental Protection Agency on “The Role of Economy-Wide Modeling in US EPA Analysis of Air Regulations.” There, for example, he says that “economywide modeling will need to be adapted to educate the EPA and the public on the likely number and types of workers displaced if costs increase due to regulation.” He also says that “the use of simulation models of any type . . . is more an exercise of applied economic theory than an evaluation of empirical evidence,” for which reason “transparency is extremely important.”
We agree completely. The best way to model the effects of changes in tax law or environmental rules is to construct a simulation model that takes as a given what the evidence has to say about how economic agents (buyers and sellers) respond to the incentives and disincentives created by those changes. We hope and expect that Director Hall will provide transparency in modeling tax law changes. He needs to keep firmly in mind the principle that any model worth considering recognizes tax increases exert negative effects, and tax decreases exert positive effects, on economic activity.
David G. Tuerck is a senior fellow with the National Center for Policy Analysis and executive director of the Beacon Hill Institute and professor of economics at Suffolk University in Boston.