Ending Stimulus and the Shape of the Recovery

by Tom Bozzo

Brad DeLong observes that the FY2011 budget features “big, very big” tightening on the revenue and spending sides (2.5% of GDP “from 2010 to 2011”) for the prevailing labor market conditions. DeLong wants his “morning in America” (don’t we all?), and is understandably alarmed at the pessimistic forecast of the rate of labor market improvement. Paul Krugman echoes the sentiment on “near-term” fiscal tightening.

As is always the case, the tightening question has to be “relative to what?” [1] Receipts as a fraction of GDP are expected to increase fairly substantially, for example, but that’s largely a consequence of expected economic growth.

Compared to the current-policy baseline, the FY 2011 (10/2010-9/2011) budget increases the FY 2011 deficit by around 0.8 percent of GDP. In FY 2012, the budget would subtract around 0.7 percent of GDP from the current-policy deficit. Krugman is correct to attribute this to the winding-down of ARRA stimulus and of our “overseas contingency operations” better known as the wars in Iraq and Afghanistan. Go see Table S-2 here [PDF]. Additionally, current policy has some stimulus on top of current law. Allowing most of the Bush tax cuts to become permanent reduces FY 2011 receipts by around 0.9 percent of GDP. (See Table 14-2, here.)

As for the timing, the budget assumes (see Table 2-1) that real GDP in quarter 4 of calendar year 2010 will be 3 percent higher year-over-year; in Q4 of 2011 (a/k/a Q1 of FY 2012), real GDP is expected to increase another 4.3 percent. Even with the tightening, Q4 2012 real GDP would increase another 4.3 percent y/y. So the FY 2012 tightening only arrives after two years of modest growth.

If measures labeled as such are actually to be temporary economic stimulus measures, they obviously must end sometime. Ending them after the expansion ends is stupid — the tightening would reinforce the subsequent downturn — so it’s going to take some steam out of the expansion one way or another. The most pressing timing concern would be not to take away the stimulus before it’s clear that the recent GDP growth is sustainable; I’d argue that after two years of growth, should we get there, the case that measured GDP growth is a matter of one-time shots and/or statistical glitches will be fairly weak.

The slow assumed labor market recovery Brad DeLong notes might be seen as a mirror-image of the GDP recovery assumed in the budget baseline:

The budget’s baseline economy (with the triangle marks) isn’t as pessimistic as OMB is willing to imagine in public (and if you’re Ken Houghton, you might see all of these as irrationally exuberant), but the ‘output gap’ opened by the recession is assumed to close very slowly. While higher-frequency data are not equally optimistic, there’s building evidence (so far, outside of employment) for a reasonably strong recovery. And as Maynard explains at Creative Destruction, it’s arguably in the administration’s interest to err on the pessimistic side since people (again, even including some economists) don’t understand counterfactuals and thus tend to inappropriately place blame (or credit) for surprises.

[1] Every economics professor who disparages the “jobs created or saved” concept should be immediately stripped of tenure and exposed to the current labor market for forgetting that all economic analysis is counterfactual.