Dynamic scoring has once again entered the news headlines. In this report, we dissect the Joint Committee on Taxation's fiscal and economic analysis and discuss "crowding out," a key factor in dynamic scores.
Dynamic Scores Crowding Out Static Ones
Last week, the Joint Committee on Taxation (JCT), Congress's official revenue-scorekeeper, released its "dynamic" score of the Senate GOP tax plan. In short, a dynamic score differs from a "static" score in that it attempts to account for changes to aggregate economic variables such as GDP and interest rates when estimating the revenue impact of a tax bill. The score from the JCT showed the Senate plan increasing the deficit by roughly $1 trillion over the next 10 years, down from the $1.4 trillion "static" score (top chart).
Parsing out an annual GDP growth rate from JCT's score is a tricky exercise. The JCT describes the economic impact as follows: "We estimate that this proposal would increase the level of output by about 0.8 percent on average over the 10-year budget window." Since JCT is referring to the level of GDP, and since growth compounds over time, the exercise is more involved than simply adding 0.8 percentage points to the baseline growth rate each year.
There are countless scenarios under which this could occur. For instance, the level of GDP could jump 0.8 percent in the first year, before returning to the baseline growth rates, creating a level of GDP that is, on average, 0.8 percent higher with an average growth rate that is 0.08 percentage points faster per year. Alternatively, GDP could grow 0.14 percentage points faster per year, yielding a level of GDP that is 0.8 percent higher than the baseline, on average, over the next decade. Perhaps more likely than either of these cases is a hybrid where growth jumps in the first few years by 0.3 percentage points or so, but then begins to slow as the temporary provisions, such as full expensing and the individual tax cuts, get closer to expiring.
A slew of factors can influence the output of a dynamic score, but one of the most important is a phenomenon known as "crowding out." Crowding out occurs when an increase in the federal budget deficit leads more national savings to be used to buy Treasury securities rather than to fund private investment. The Congressional Budget Office, a cousin to the JCT, estimates that when the deficit goes up by one dollar, private savings rise by 43 cents and foreign capital inflows rise by 24 cents (as higher interest rates increase the incentive to save and attract foreign capital), leaving a net decline of 33 cents in savings available for private investment.
Some think tanks have also produced dynamic scores. The Tax Policy Center and the Penn Wharton Budget Model produced estimates suggesting that the House-passed bill would lift growth by less than 0.1 percentage point per year. The Tax Foundation, a leading conservative think tank whose model assumes a much smaller crowding out effect, estimated the House bill would lift growth by about 0.3 percentage points per year over the long-run. As we wrote in a piece earlier this year, with real potential GDP growth less than 2 percent at present, the hurdles to sustained economic growth of 3 percent or more are high (bottom chart).