Tax Analysts Blog

Your Quick Guide to Dynamic Scoring in the Next Congress

Posted on Nov 10, 2014

We've been here before. A midterm election has handed Republicans control of both chambers of Congress, and at the top of their to-do list is getting the staff of the Joint Committee on Taxation to adopt dynamic scoring. It happened at the beginning of the 104th Congress in 1995, and all indications are that it will happen again at the beginning of the 114th in January.

Dynamic scoring refers to a way of doing revenue estimates that is different from conventional scoring done by the staff of the JCT. Currently, the JCT staff--the official scorekeeper of changes in federal tax law--assumes tax legislation does not affect the size of the economy. But if a tax cut or tax reform increases economic growth, that growth will reduce the revenue loss from any tax change because a larger economy produces more revenue.

Conservative tax reformers are hoping that favorable dynamic scoring will make tax reform easier to pass. If the net revenue loss from a rate cut is estimated to be lower, they will have to eliminate fewer tax breaks to pay for those lower rates.

An amendment to the Senate Democrats' budget offered in March 2013 by Senate Finance Committee member Rob Portman, R-Ohio, would have required the JCT to provide dynamic estimates. In April 2014 the House passed a bill (H.R. 1874) that would have required the JCT to provide dynamic estimates. And in May of 2014 Portman introduced a bill (S. 2371) that similarly would require the JCT to provide a dynamic score. Rep. Paul Ryan, R-Wis., who is likely to be the next chair of the Ways and Means Committee, will request that the JCT include macroeconomic effects in its estimates of tax bills and present a single dynamic revenue estimate.

Non-economists are quickly overwhelmed by the morass of complex economic issues that arise when dynamic scoring comes up for discussion. But you don't need semesters of graduate macroeconomics under your belt to understand how dynamic scoring can change the way Congress makes tax law. To monitor developments in the upcoming debate, just keep the following four points in mind.

(1) As long as any new JCT dynamic revenue estimates are not integrated into the official scorekeeping used for the congressional budget process, and they only provide additional information for lawmakers and the public, they will have little impact on the passage of tax legislation.

This is by far the most critical point, and it does not even directly involve economics. It is a matter of procedure. If the JCT's dynamic estimates just tell us that so-and-so's policy will create some number of new jobs, that's useful background for the policy debate. We could get the same type of analysis from think tanks and interest groups. But the JCT staff is not just another think tank. It has a hard-won reputation for economic analysis of the highest quality and for nonpartisanship. So whatever the JCT says about the growth effects of proposed legislation would become the most quoted and most authoritative economic analysis of that proposal.

But in the end, any economic analysis -- no matter how technically proficient or politically unbiased -- is still only feedstock for talking points in a congressional debate in which most everybody has already made up his mind. On the other hand, no legislator can ignore official scoring of a tax bill. Although an estimate may be highly uncertain, once a proposal is assigned a quantitative magnitude by the JCT, that magnitude determines how much precious space it occupies inside all-controlling budget targets.

(2) If the JCT attaches greater importance to a new class of models that puts more emphasis on economic theory than observed empirical relationships, the estimated positive macroeconomic feedback effect could be much larger than if more conventional models are used.

The JCT uses two models to do its dynamic scoring: the MEG (macroeconomic equilibrium growth) model and an OLG (overlapping generations) model. The first type is a more sophisticated version of macroeconomic forecasting model that economists have been using for over 50 years. For the difficult task of predicting the economic future, these models use a combination of theory and empirically observed relationships -- whatever works best at improving accuracy.

The OLG model uses a combination of theory and empiricism, but it belongs to a class of models in which there is more emphasis on consistency with theory and with the assumptions economists routinely make about the way they believe the economy should work. Models like the MEG are used much more than OLG models by businesses and governments that need forecasting services. A strong case could be made that OLG results should be entirely excluded from official estimates given the underlying assumptions' poor correspondence to reality.

Because OLG models produce much larger—often three or four times larger--dynamic effects than the more widely used conventional models, the amount of weight the JCT gives to its OLG estimates for any single estimate will significantly change the results.

(3) To the extent Congress adopts macroeconomic scoring of tax bills, consistency requires it to similarly score other legislation that affects economic growth. This nontax legislation would include immigration reform, highway bills that increase infrastructure spending, and any bill that significantly changes the federal deficit from levels projected in the current-law baseline.

Nontax legislation clearly has large economic effects to warrant the same dynamic scoring treatment as tax legislation. Otherwise, many strange and unexpected situations could arise. For example, if only tax changes are dynamically scored, and assuming the economy is at full employment so there is no stimulus effect, it is entirely possible that a deficit reduction package of tax increases could get a better score than an equal cut in spending. That's because the positive macroeconomic effects of deficit reduction would only be scored for the tax increase.

(4) Even if it becomes standard practice on Capitol Hill, dynamic scoring may not significantly ease the passage of tax legislation because the dynamic effects often are small and sometimes negative.

Despite all the post-election happy talk about common ground, enacting comprehensive revenue-neutral tax reform has close to zero chance of enactment anytime in the foreseeable future, as evidenced by the chilly reception that House Ways and Means Committee Chair Dave Camp's draft tax reform proposal received earlier this year. Even business-only tax reform, which would encounter one fewer hurdle than across-the-board reform, faces extremely long odds. Given this gloomy outlook, conservative tax reformers are hoping that favorable dynamic scoring can break the logjam. But it is by no means certain that the hoped-for estimates will materialize.

It is possible that dynamic estimates will show that revenue-neutral tax reform hurts rather than helps economic growth. How can this be? You don't have to look far to find an example. Under a plain vanilla, revenue-neutral corporate tax reform in which rate reduction is paid for with less accelerated depreciation, the cost of capital for new investment is increased so capital formation is estimated to shrink. This happens because depreciation changes only apply to the taxation of new capital, while much of the benefit of rate cuts applies to income generated by existing capital. Cutting taxes on capital already in place does little to promote investment in these models.

Of course, there are many variations of tax reform that can produce positive estimates of economic growth. But it is important to keep in mind that most realistic packages will be a combination of proposals -- some that increase growth, some that reduce growth -- and it should not be a surprise if the net effect is small.

This post is based on an article in the November 10 issue of Tax Notes.

Read Comments (1)

edmund dantesNov 10, 2014

Has anyone ever checked on the accuracy of these JCT revenue projections? That
is, compared the 10-year revenue projection of a given piece of legislation to
the actual experience ten years later? I know, for example, that cuts in
capital gains tax rates have always been scored as losing revenue, and it turns
out that they have typically gained revenue, sometimes large amounts of
revenue. Have JCT projections been within, say, 50% of actual experience?
Which way do their errors tend to go?

I have also frequently been struck by projections that seem only loosely
connected to reality. For example, in 2007 the kiddie tax was extended to
cover those up to age 24 if they are students, essentially college students.
JCT scored that change as raising $150 million per year for ten years. That
implies every year college students will be selling appreciated assets with at
least $1 billion of taxable gain (the top tax rate was then 15%). And they
would do this every year. Really, college students just don't have portfolios
of that size.

I'd love to see much more transparency and accountability in this area. Before
tax legislation is passed, the JCT pronouncements are treated like the Word of
God; after passage they are mostly just forgotten.

Submit comment

Tax Analysts reserves the right to approve or reject any comments received here. Only comments of a substantive nature will be posted online.

By submitting this form, you accept our privacy policy.


All views expressed on these blogs are those of their individual authors and do not necessarily represent the views of Tax Analysts. Further, Tax Analysts makes no representation concerning the views expressed and does not guarantee the source, originality, accuracy, completeness or reliability of any statement, fact, information, data, finding, interpretation, or opinion presented. Tax Analysts particularly makes no representation concerning anything found on external links connected to this site.