Re-Examining that "Massive Stimulus"

I keep on seeing references to "massive stimulus", so much I have this feeling of innumeracy everwhere. Here's one example, provided by a commenter on Free Exchange the day before yesterday, responding to the linked article of Brad Delong:
Why doe someone have to keep reminding Krugman and Delong of recent history? It’s absurd that PhD economists ignore what actually happened in the history of the recent crisis.

The US tried massive fiscal stimuli in response to the crisis and the Fed did not raise interest rates at all to stop price inflation. Nevertheless, the multiplier was close to zero. All the likes of Krugman and Delong could say was “it would have been worse.”

It constantly amazes me that people throw around words like massive without really thinking about the fact that the relevant things to normalize by are also massive. So, I thought I would plot total nondefense spending in the US (ex-interest), and normalize by potential GDP, to obtain the following graph.

Figure 1: Ratio to potential GDP of Federal nondefense outlays ex.-interest payments (red) and of state and local outlays ex.-interest payments (blue). NBER defined recession dates shaded gray. Source: BEA, 2011Q4 second release, CBO Budget and Economic Outlook (Jan. 2012) supplemental information, NBER, and author's calculations.

Total nondefense spending at all levels rose from 27.2% of potential GDP in the last quarter of the Bush Administration (2008Q4) to a shocking 29.9% by 2010Q4, before declining to 28.4% in the last quarter for which data are available for.

That's not to say that the Federal government did not increase spending; merely that to a large extent it was offsetting the contraction at the state and local levels of government, as I pointed out in this March 2010 post. The fact that real interest rates have zoomed toward negative merely confirms the fact that government spending could not compensate for the massive (here the adjective is correctly used) collapse in private consumption and investment spending.

Figure 2: Ten year constant maturity yields, nominal (blue) and real, from TIPS (red). Source: St. Louis FRED.

Even if the state and local governments had not contracted expenditures, the fiscal stimulus arising from the ARRA (spending and tax provisions) would not have been particularly large, when expressed as a ratio of GDP. As I noted on February 16, 2009, as the bill was close to being signed:

I want to stress the adjectives "massive stimulus" conjoined to the noun "bill" is a matter of context. Dividing by baseline GDP shows that in a proportional (rather than dollar) sense the bill is rather modest. The fiscal impulse to GDP ratio never exceeds 2.5 ppts in any given fiscal year.

But then, it was clear that many of the critics never wanted to do the hard work of long division. (Oh, and I still want to talk to all the critics of the bill who said the spending would kick in long after the economy had recovered...).

A parting observation. The impact on the budget (tax reductions and outlays) can be placed in context by comparing to other major undertakings.

Figure 3: Impact on budget balance, in billions of FY2010$, for EGTRRA; for JGTRRA; cumulative budget authorization for operations in Iraq (Operation Iraqi Freedom, not including incremental debt servicing costs) through FY2012; for Patient Protection and Affordable Care Act; for American Recovery and Reinvestment Act, all in billions of FY2010$, deflated using CPI. Source: CBO, Budget and Economic Outlook: An Update (August. 2001), Table 1-4; CBO, Budget and Economic Outlook: An Update (August 2003), Table 1-8 (revenue implications only); and CBO, "H.R. 4872, Reconciliation Act of 2010: Estimate of direct spending and revenue effects for the amendment in the nature of a substitute released on March 18, 2010," (March 18, 2010), Nominal figures from Amy Belasco, "The Cost of Iraq, Afghanistan, and Other Global War on Terror Operations Since 9/11," RL33110, Congressional Research Service, March 29, 2011, Table 3. Data for FY2011 Iraq operations is for continuing resolution, for 2012 is Administration FY2012 request (see notes to Figure 2 of this post for calculations); for ARRA, CBO Budget and Economic Outlook (January 2012), Box 1-1, page 9; for CPI, historical from FREDII, and forecasts/projections from CBO (January 2012), Table 2-1 (using calendar year forecasted inflation for FY inflation).

While the expenditures in the Iraqi theater of operations (including searching for WMDs) surely had some stimulative impact on the economy, it is debatable whether those expenditures were the best way to undertake military Keynesianism, given the peak of Iraq-related expenditures was in FY2007-08, when the economy was at or near full-employment, and before the collapse in GDP beginning in 2008Q4 (that is, "real" believers in counter-cyclical stabilization policy believe that you undertake stimulative policy when you are below the output gap...).