Home > Main > Revision to GDP makes it clear Fiscal Multiplier was Gigantic

Revision to GDP makes it clear Fiscal Multiplier was Gigantic

August 12, 2011

Lots of conservative economists claim the stimulus didn’t amount to much.  Turns out – quelle surprise! (heh.) – that they were totally wrong.

Their claims were based on the initial measurement of the depth of recession – and these measurements turned out to be wildly optimistic.  The recession was far, far deeper than previously admitted.

The stimulus worked great.

It’s pretty clear if you use the exact same methodology used by prominent conservatives to show the stimulus didn’t work with the new, massive numbers on the depth of the recession, you’d get a multiplier for the fiscal stimulus that was totally huge.


I meant to write a TC post about it, but the Center for American progress did the work, and beat me to the post.

CAP’s Michael Linden via Kevin Drum:

“Using the most updated data, we can see that in 2009 there is actually about a $544 billion difference between what GDP would have been had it continued to contract as rapidly as it did during the fourth quarter of 2008 and what it actually was. As Holtz-Eakin points out, the total amount of fiscal stimulus during that year was $260 billion. This suggests the Recovery Act produced about $2.10 in economy activity for every $1.00 in spending or tax cuts. That’s a pretty good multiplier.

And if we apply the same methodology to the entire lifespan of the Recovery Act, not just to 2009, the multiplier becomes even more impressive. The total cost of the stimulus bill was about $800 billion, delivered over the course of two years. The difference between actual GDP through the first quarter of 2011 and what GDP would have been had it continued “falling off a cliff” is around $3.3 trillion—implying a multiplier of more than 4.”

A multiplier of more than 4.  This isn’t the Romer method, or the barro method.

But it is one method. and no matter what method you use, you need to use the new GDP data. Using that data will show that the stimulus was extremely effective, and was a 2 or 3 to 1 return on investment in a few short months.

This is more evidence that fiscal policy works great when monetary policy can’t get in the way.  A reminder: Monetary Policy sucks.

Categories: Main Tags: ,
  1. RonT
    August 16, 2011 at 8:05 am

    I don’t get this talk about multipliers. Money is free for the govt that is (er, should be) in charge of maintaining full employment. Since the stock of debt can never become a problem, we should print as much as it takes to achieve full employment, the multiplier be whatever it wants to be, even 0.0001.

    Worrying about multiplier implicitly assumes money is scrce for the govt that should thus watch its “return on investment”, nonsense.

    If at any point in the future the stock of debt proves to big, just tax the money back, what is the problem?

  2. Peter D
    August 25, 2011 at 11:17 pm

    It is so frustrating having to argue that stimulus worked with people whose level is to say “there is still unemployment, see? it did not work”.
    Here, by the way, is Ezra Klein’s intern surveying 9 studies of stimulus:
    6 say it worked, 3 say it had little or even negative impact. Out of those 3, two are obviously garbage (one uses Riccardian Equivalence an the other produces statistically insignificant results, hahaha.) The last one is from John Taylor and basically says that it actually was too little to work well (another hahaha) Or, as Taylor himself summarized it:

    To sum up: the federal government borrowed funds that it mainly sent to households and to state and local governments. Only an immaterial amount was used for federal purchases of goods and services. The borrowed funds were mainly used by households and state and local governments to reduce their own borrowing. In effect, the increased net borrowing at the federal level was matched by reduced net borrowing by households and state and local governments.
    So there was little if any net stimulus.

    Not only does he say the stimulus was too small, he also basically admits that the issue was not malivestment and the supposedly rampant govt waste (“only an immaterial amount was used for federal purchases of goods and services”) but the lack of direct govt spending on output! Geez, how do thinking conservatives avoid cognitive dissonance uis beyond me…

    • TC
      August 26, 2011 at 8:38 am

      If this was my full time job, I’d go through these studies and plug in the new GDP numbers into the studies.

      I really think the multiples would be shockingly high for any real study, and probably flip 2 of the 3 negative studies over to positive. I’d have to look at the exact methodologies to be clear.

      But the new lower estimates for GDP should take most of these models and push the expected future GDP down to much lower levels if I read them correctly. Therefore, the realized level of GDP after the stimulus ends up being much, much higher than expected.

      I can’t believe DeLong or Krugman hasn’t picked up on this yet.

  3. RonT
    August 30, 2011 at 9:02 am


    you ingored my comment completely:)

    Why care about multpiers at all?

    Is money scarce for the govt so that it needs to watch its return on investment? No. So why care? If multiplier is 0.1 and the output gap is 1T, spend 10T: 9T will be stashed away under the mattress and 1T spent. What is the problem? If at any point people start spending the 9T too fast (so we have full capacity and inflation), you raise taxes without limiting anybody’s real consumption (since we are at full capacity, taxes don’t limit consumption, only prices, because people couldn’t consume more if not for the taxes – they can consume only as much as the economy at full capacity provides) – so it comes at no economic cost.

    • August 30, 2011 at 9:35 am

      Multipliers are useful in terms of where you direct your firepower. It’s still best to direct the firepower where it can have the most real impact if you can.

      Politically the best approach is to pick the deficit increase pattern that generates the most real impact and has the most political support. So although an ELR scheme is certain to have a massive multiplier it might not have sufficient political support in the short term.

      Therefore you go with the weaker tax cut approach for political expediency.

      The other problem with fiscal waste is that it tends to concentrate political power. People with command over a lot of financial tokens have a lot of political clout.

      • RonT
        August 31, 2011 at 7:53 am

        Thanks Neil,
        I would agree that it may make sense to choose the stimulus with the largest multiplier, but do we care if it is above one? It seems to me we shouldn’t.

  1. No trackbacks yet.
Comments are closed.
%d bloggers like this: