If You Are Worried About the National Debt, Be Wary of “Dynamic Scoring”

Wonder Where the Numbers Come From?

Wonder Where the Numbers Come From?

Republican Congressmen led by House Appropriations Chair, Paul Ryan, are going to direct congressional staff to change the way they estimate the revenue impact of proposed changes to the tax code. They want the Congressional Budget Office (CBO) and the Joint Committee on Taxation (JCT) to use “dynamic scoring” to estimate the cost of tax policy changes. The method reflects the belief, not without some merit, that tax cuts can change behavior in ways that boost economic activity and generate new revenue, thus paying for themselves without contributing to deficits.  You can see why this scoring approach might be attractive!

The story did make the front page of the LA Times business section. But because it’s an arcane subject it’s been flying under the radar.

It’s not entirely clear how the Ryan proposal would affect the scoring of spending bills. Some government spending, especially in education and transportation, is demonstrably positive for the economy.  Such measures too can also be scored as “paying for themselves” and being debt neutral.  Dynamic scoring is thus a double edged sword for fiscal conservatives.

The dynamic scoring approach, championed now by “conservatives,” ironically, would make it harder to bring down the national debt; but easier to pass tax cuts. It would also make it easier to pass spending bills, which is, of course, not Mr. Ryan’s intention.

The “scoring” of tax (and spending) bills is a nerdy and technical enterprise, performed often by PhD economists and MBAs,  but it has serious consequences for the nation’s short and long term fiscal outlook—both perception and reality. The contrasting approach to dynamic scoring is “static scoring.” These are not either/or options; they are on a continuum.

Ideally, government needs an approach to scoring taxes (and spending) which is:

• Transparent
• Replicable
• Explainable (to people without MBAs or doctorates in economics)
• Fiscally prudent
• Not easy to manipulate for political purposes

The Ryan dynamic scoring proposal falls short on all these accounts. Static scoring, the traditional approach, scores much better.

The Ryan proposal is also breath-taking in its hypocrisy, because its proponents talk about using “sophisticated advances in economic science” (new and better economic models) that make dynamic scoring more reliable.  So, is “economic science” (and its modeling) more reliable than, say, “climate science” (and its modeling infrastructure)?  Really?

On the spending side, the CBO, under pressure from both Republicans and Democrats, used a form of scoring with some dynamic elements, to estimate the impact of the Affordable Care Act (ACA).  Thus, the CBO says Obamacare will slow down the overall rate of growth in health care costs and make it easier to reign-in the national debt. Some of that has actually happened since the measure passed, though it’s not clear how much of it is due to the ACA. This made Democrats happy and made it easier to pass the ACA in 2010.

The CBO also put its toe in the dynamic scoring waters by assuming the ACA would cause a number of people to exit the labor market because they would no longer be dependent on employer provided health insurance.  This CBO analysis made it easier for Republicans to argue that the ACA would shrink the U.S. economy. (Funny, how only part of the CBO’s story about ACA seems to have made it into the consciousness of Americans).

You can see the can of worms being opened here. There is, of course,  some merit to identifying the “full” impacts of tax (or spending) bills, but trying to score the budget this way is not only dicey, but it’s not transparent, replicable, fiscally prudent, or readily explainable; and it’s ripe for political manipulation.

Dynamic scoring is not quite the equivalent of “voo-doo economics,” like its fiercest critics say. A cut in the gasoline tax (up to a point) will likely induce more economic activity across the economy. Even a Marxist economist (with a real PhD) would probably own up to that. But the more dynamic you get in scoring, the closer it gets to voo-doo. That’s because even the best economists and models can’t predict these secondary and tertiary and Nth order impacts very accurately.

Not only that, but dynamic scoring doesn’t typically reflect the spending cuts that would accompany a tax cut (if one wanted to avoid increasing the debt). A gasoline tax cut would likely reduce the amount of money available to build highways, bridges and other transportation infrastructure. That could have a very large negative impact on the economy, maybe not immediately, but down the (better paved) road. Typically, dynamic scoring models either don’t measure that at all, or do it inaccurately.

Besides, there is always room for fancier modeling and research as a supplement to official scoring, where economists can legitimately talk about broader effects without affecting the official balance sheets,

If you want a better understanding of dynamic (versus static) scoring, read on.

The polar opposite of “dynamic scoring  is “static scoring.”   it’s not entirely clear as to where one ends and the other begins, but here is an effort to explain the differences more concretely, using a hypothetical proposal to cut the federal gas tax.

If Congress passed a bill lowering the federal gas tax by 6 cents per gallon (cpg), a strictly static approach to scoring would first note that 6 cents is about one-third of the current federal gas tax (18.4 cpg); and that the total annual revenue from the tax is about $30 billion. All of that is verifiable.    So a 6 cpg tax reduction, looking at it simply and statically, would reduce revenue by about one-third, or $10 billion a year.  QED!

In practice, “static scoring” would also likely include a forecast of how many more gallons of gas might be bought if the tax (and thus the price of gas) was lowered. With this wrinkle, the estimated revenue loss would be calculated at less than a third, maybe just 30%, which means the total hit on federal revenue would be just $9 billion.

Dynamic scoring on the other hand, would go much further,  estimating many of the less obvious (but also a lot more uncertain) effects of the tax cut. Consumers would keep more of their income (due to the lower tax) and spend or invest some of it, expanding the national economy.  That would generate more corporate income tax revenue.  Profits of the airline and trucking industries, and other sectors that use a lot of fuel, would grow the most, and generate more federal revenue.

So, with dynamic scoring, the one third reduction in the gas tax, might be expected to lower total federal revenue by much less than $10 billion. Maybe by only $5 billion.  Or maybe by zero. If the federal gas tax was eliminated altogether, some economic models would predict an increase in federal revenue.

Again, you can see why dynamic scoring is attractive, arguably more sophisticated than the static approach…..and also dangerous for budgeting.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s