You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

How Much Of The Stimulus Has Actually Been Implemented?

It's pretty common these days to point out that only about 10 percent of the stimulus has been implemented so far (I've done it myself), making a second round politically difficult to justify even if we know the first will almost certainly be insufficient. But while that 10 percent figure is technically accurate (give or take), the economic research team at Goldman Sachs argues that, as a practical matter, we've already felt about 40 percent of the benefits. As they put it:

This [the 10 percent figure] is approximately correct if we focus only on the spending measures and exclude the personal and business tax provisions; if the latter are included, the share rises to just over 15%. But even this 15% share is misleading if we care about the ability of the stimulus package to push the economy back to a more satisfactory growth rate. This essentially depends on four factors: 1) the change in the monthly or quarterly stimulus “run rate”, 2) the marginal impact of different types of stimulus on spending/output, 3) the lags between disbursement of the stimulus and spending/output, and 4) the multiplier effects via induced effects of stimulus-induced spending/output on income. Taking all of these factors into account, we find that we should have seen about 40% of the cumulative positive impact on quarterly real GDP growth as of the second quarter. 

If that's the case, then it's obviously harder to accuse the administration of moving too slowly, but also a lot harder to take a wait-and-see posture...

P.S. Reasons Factors 1.) and 3.) are a little opaque to me. I've contacted the Goldman economist who wrote this to see if I can get some elaboration. I'll update if I hear back.

Update: I briefly spoke just now to Jan Hatzius, the author of the aforementioned report. If I'm understanding him correctly, point 1.) is just that, if you start with no stimulus and add stimulus in increments, that first piece of stimulus gives you a big bump, GDP-growth-wise, while the next increment just kind of maintains it. To take a simple example, suppose GDP is $100 without any stimulus and you have two one-year increments of stimulus of $5 each (and assume away multiplier effects, etc.). The first year, the stimulus alone contributes 5 percent to GDP growth. But the second year you may need that $5 just to keep GDP at $105. Even if not, you're not going to get another 5 percent boost. So the impact of that first $5 on GDP growth is far greater than 50 percent, even though it's only technically half the stimulus.

As for point 3.) the confusion here is why a lag between disbursement and spending would cause the 10 percent figure to understate the benefits of the stimulus, when it seems like it would overstate it. (That is, 10 percent would have been disbursed, but less than 10 percent spent.) The mistake here is my own misreading of the report. The four factors aren't four reasons why the 10 percent figure understates the impact of the stimulus so far; they're just four factors you have to take into account when determining its impact. This particular factor actually does make that initial 10 percent figure an overstatement. But, together, the four factors suggest the 10 percent figure is pretty understated.

--Noam Scheiber