Event Recap: A Discussion on Dynamic Scoring

The Committee for a Responsible Federal Budget hosted a policy discussion this past Tuesday on dynamic scoring. CRFB President Maya MacGuineas opened the event by noting that dynamic scoring is an issue that will receive considerable attention over the coming months and could have an impact on fiscal policy decisions. Speakers offered their perspectives on the merits and challenges of using dynamic estimates in the legislative and budget process. Senator Rob Portman (R-OH) and Representative Chris Van Hollen (D-MD) offered remarks on their opinions and perspectives on dynamic scoring. A panel of dynamic scoring experts followed, moderated by CRFB President Maya MacGuineas. See CRFB's paper on dynamic scoring for a detailed discussion or our updated 2-page summary.

Sen. Portman spoke in favor of CBO and JCT providing dynamic estimates of bills. He argued that, at the very least, estimates should be done to inform staff and lawmakers how bills will affect the economy. He spoke about the bipartisan support for his bill, which passed by a vote of 51-48 with six Democrats voting in favor of it, when he offered it as an amendment during consideration of the FY 2014 Senate budget resolution. Portman acknowledged that there is a legitimate debate over which models and assumptions should be used, but he encouraged detractors to support presenting dynamic estimates as supplemental information, as his amendment would, not for official purposes. It would be apparent if dynamic estimates are significantly different than current methods, and policymakers would be able to look back to see which estimates were more accurate.

Congressman Van Hollen gave the opposing viewpoint. Van Hollen argued that dynamic scoring inherently demands that CBO or JCT adopt a specific ideology when estimating a bill. He mentioned estimates from the Heritage Foundation predicting revenue increases from the 2001/2003 tax cuts and claims that the 1993 tax increases would harm the economy. He also said that many models for dynamic analysis make assumptions about future actions to offset the cost of tax cuts, effectively giving legislation credit for policies not in the bill. He drew a distinction between the CBO estimate of immigration reform legislation, which took into account the direct impact of additional workers in the labor force, and dynamic estimates which incorporate the estimated indirect economic effects of legislation. Van Hollen reminded audience members that CBO and JCT already use microdynamic analysis in scoring bills—they weigh behavioral responses from individuals and businesses and the factors of supply and demand. He also acknowledged that dynamic analysis is useful as supplemental information, as long as policymakers understand the underlying assumptions.

The panel that followed included John Buckley, a visiting professor at Georgetown who formerly served as Chief of Staff of JCT and chief tax counsel for the House Ways and Means Committee; Scott Hodge, President of the Tax Foundation; Bob Dennis, the former Assistant Director of the Macroeconomics Division at CBO; and Bob Carroll, National Director of Ernst and Young's Quantitative Economics and Statistics (QUEST) practice and former Deputy Assistant Secretary of the Treasury for Tax Analysis.

John Buckley argued that dynamic analysis was appropriate when designing tax reform, but using a dynamic score on legislation would be a poor choice. If used in legislation at all, dynamic scoring should include both revenue and spending effects, otherwise there would be a strong incentive to move spending programs to the tax code to get credit for the dynamic effects. He argued that it would be inappropriate for the non-partisan Congressional staff at JCT and CBO to make assumptions about future actions that Congress would take as most dynamic models require. He also criticized dynamic models for making assumptions that do not reflect the realities of the economy.  Finally, he pointed out that it would be difficult for scorers to have significant gravitas with their estimates as there is little to no consensus over the appropriate models, assumptions, and views on the economy.

Scott Hodge of the Tax Foundation had a much more favorable view about the ability and benefits of CBO and JCT providing dynamic estimates. The Tax Foundation has written extensively on the subject and routinely uses their own models to score major bills including Chairman Camp's tax reform plan. Hodge considers much of the debate over whether or not CBO should use dynamic scoring as moot because they already provide some dynamic analysis. The key part of the debate for him is whether CBO or JCT should use dynamic analysis for the primary score or to provide it as additional information. He called it “economic malpractice” to have Congress vote on a bill and not know its impact on the economy. He concluded by saying that while current dynamic estimates may not be perfect, they are better than the methodology scorers currently use.

Bob Dennis provided a technical background based on his experience in developing dynamic estimates during his tenure at CBO. He emphasized that the concept of scoring a bill revolves around providing one single number to allow for decisions and evaluation of procedural requirements like spending caps and budget rules. Dennis explained that there are time and resource limitations on scorers, and providing dynamic analysis adds several days to the turnaround time for CBO. Additional dynamic analysis requires CBO to make assumptions and estimates about what a future Congress will do regarding changes in deficits, which greatly impact the model's outcomes. He also noted that producing a single dynamic estimate of legislation would require estimators to make decisions in which they would have to choose sides on issues where there are deep philosophical differences, which would create challenges for the non-partisan nature of CBO. He indicated that it is easier for CBO to provide dynamic analysis for comprehensive policy proposals such as the President's budget because they provide full paths for spending, revenues, and deficits rather than individual bills in which assumptions need to be made about other legislative actions.

Bob Carroll brought up that a key objective in the policy space is to provide a range of economic analysis of policy to policymakers to help inform the decision making process. Economic analysis that accounts for a range of economic impacts of bills would be a useful addition, instead of the more narrow focus on just the revenue score. Carroll reminded the audience that there exists a general consensus on tax reform that it should create a simpler tax code and be pro-growth. However, agreement evaporates once details are introduced. He explained that dynamic analysis is similar: generally, people agree that dynamic analysis can be a helpful to inform policy, but disagreement appears when you get past that 10,000 foot view. Carroll spoke to the fact that even conventional scoring involves estimates that are uncertain as well. If conventional scoring is already uncertain, then dynamic scoring adds even more uncertainty.

Hopefully, the panel discussion is the start of a bipartisan conversation on dynamic scoring and dynamic analysis for the next Congress. We thank Senator Portman and Congressman Van Hollen for their participation and all of our panelists for their thoughtful comments.

For additional reading on dynamic scoring see: