Sunday, December 2, 2012

Paul Krugman Will Not Like These Figures

[See update below]
See if you can figure out why:

This first figure shows that aggregate demand growth has not been affected by a tightening of fiscal policy since 2010.  Specifically, it shows that nominal GDP (NGDP) growth has been remarkably stable since about mid-2010 despite a contraction in federal government expenditures. The same story emerges if we look at the budget deficit relative to NGDP growth:

Both figures seriously undermine the argument for coutercyclical fiscal policy and suggest a very a low fiscal multiplier.  They also indicate that the Fed has been doing a remarkable job keeping NGDP growth stable around 4.5%. Monetary policy, in other words, appears to be dominating fiscal policy in terms of stabilizing aggregate demand growth.  This in turn implies that the Fiscal Cliff should not be a big deal if the Fed continues to stabilize NGDP growth around 4.5%. Yes, there would still be distributional and incentive changes if fiscal consolidation occurs, but the fiscal tightening itself should have no bearing on aggregate demand if the Fed continues to do its job. As I noted before, the Fed is the other solution to the Fiscal Cliff.  Lars Christensen agrees.

P.S. Though the Fed has been doing a remarkable job keeping NGDP growth stable since mid-2010, it has yet to allow a period of catch-up nominal spending growth that would return NGDP to its pre-crisis trend. So the Fed's work is not complete.

Update: Noah Smith and Daniel Kuehn provide some pushback to this post. Noah invokes Nick Rowe's thermostat reasoning to conclude the graphs could be consistent with a large fiscal multiplier.  That is a possibility, but given what we know about fiscal policy and the last two years his alternative story does not fit.  The economy has been hit by multiple negative shocks--Eurozone crisis, 2011 debt cliff talks, concerns about China slowdown--over the past few years that have kept demand for safe assets elevated and thus the economy below full employment.  For Noah's interpretation to work fiscal policy must be nimble enough to counter these shocks as they happen. That hasn't happened; fiscal policy is clumsy and has been winding down despite these shocks and despite the consternation of Paul Krugman who has endlessly complained that it hasn't been enough.

Daniel, on the other hand, thinks if we put the variables on the same scale all is well.  No, here are total federal expenditures and NGDP in the same growth rate form and in dollar levels. In both cases, total federal expenditures have been trending down since 2010. 

24 comments:

  1. David
    And that´s what we observe everywhere. It´s not fiscal stimulus that "does it" and it´s not fiscal consolidation that "undoes it".
    http://thefaintofheart.wordpress.com/2012/02/08/nominal-spending-determines-outcomes-not-fiscal-stimulus/

    ReplyDelete
    Replies
    1. Yes Marcus, and as you note we call that the Sumner Critique.

      Delete
  2. So, contrary to the conservative and Republican and Romney attacks over the past three years, "the economy is doing fine"? Isn't that a statement that Obama made that resulted in attacks of ridicule by Republicans?

    I assume the construction sector is robust as well given the reduced public and private spending, and the ongoing deterioration of roads and bridges and rail lines in the aggregate does not represent a problem - our transportation infrastructure has continued to be excessive with too little congestion and delay for optimal economic efficiency?

    ReplyDelete
  3. This comment has been removed by the author.

    ReplyDelete
  4. "This first figure shows that aggregate demand growth has not been affected by a tightening of fiscal policy since 2010."

    I think everyone in the room knows better than to make causal claims based on the short-term trend lines in these graphs.
    But if one were to do so, shouldn't it come up that the slopes of these two lines are possitively correlated?

    ReplyDelete
  5. Couldn't Krugman respond by saying that the NGDP growth rate was rising when fiscal stimulus was high and then peppered out when fiscal stimulus started falling?

    ReplyDelete
    Replies
    1. Krugman has been saying over the same period that fiscal policy is not enough. See my updated.

      Delete
  6. David, i am a fan of your basic campaign, but surely these graphs don't prove anything as glibly as you imply. The RHS chart is g/ngdp isn't it? So if NGDP grows at 4 or 5 pc and federal spending by 2 or 3, the ratio falls. Ditto the deficit. That scarcely proves that NGDP defies austerity, just that nominal GDP growth can outpace nominal federal spending growth.

    G/Y is a funny money currency. Put both axes in honest dollar amounts or growth rate of dollar amounts and i doubt Krugman has anything uncomfortable to explain. I think your case is stronger than this ...

    ReplyDelete
    Replies
    1. Giles, sorry to disappoint. See my update to the post where I provide links to government expenditures by themselves. Still a downward trend.

      Delete
    2. I find that much more useful so thank you!

      Final quibble: how big a part of total US spending is the Federal as opposed to other government?

      Delete
    3. And sorry to appear sceptical when I basically agree with so much else you write here! E problem is that over here in the UK we read so much about the less austere fiscal path that the US has taken. And I read on the White House site that Federal US spending has risen from 2.9trn to 3.7 trn over 2008-2012 .... Is that not accurate, and is the Right therefore incorrect when it claims that the Federal government has grown fast under Obama?

      Delete
  7. NGDP has been amazingly stable since 7/2009 - more stable even than before the crisis!

    ReplyDelete
  8. Nice post.

    Yes, by themselves, the charts of this post are not totally compelling.

    But they sure are suggestive.

    My bone with this post is that the Fed is somehow being given credit for keeping NGDP growth at 4.5 percent.

    In truth, the Fed should be criticized for holding growth so low, and maintaining a passively tight money supply. We should be targeting 8 percent growth NGDP.

    The proof that money is too tight is seen in low interest rates and dead inflation. The Fed is undershooting even its dubious target of 2 percent inflation.

    Yes,, the fiscal cliff is not that important, if the Fed gets aggressive.

    ReplyDelete
    Replies
    1. The Fed cannnot push a chain. Are you being sarcastic when you say" The proof that money is too tight is seen in low interest rates and dead inflation" or do you actually belive that up is down? Fiscal policy refers to government spending. Monetary policy is controlled by the Fed. What else would you have the Fed do?

      Delete
  9. To compound my confusion (now I am on a proper computer):

    http://www.taxpolicycenter.org/taxfacts/displayafact.cfm?Docid=200 tells me that US government outlays from 2007 to 2012 look like they rise from $2.7trn to $3.7trn. About 4-5% per year for the last two years, 40% over the whole period.

    And government dollar spending doesn't just hit like a surgical strike: it leaks, splurges etc. In the UK, projects are still draining away from the Labour govt capital splurge (but on a simple graph it would look like a 2009 event).

    I think your basic crusade is arguably the most important anywhere, and extremely right. But I honestly can't see how you or Tyler showing these charts on their own disprove the Fiscalist case that higher government spending in this situation has, other things being equal,increased velocity and hence NGDP. I doubt Krugman is suddenly feeling uncomfortable.

    Put it this way: since 2008, UK total expenditure has increased from £618bn to £712bn. That is 16% up in cash terms. US total outlays have increased by 27%. The anti-Austerians will continue to use that kind of comparison to prove stuff ...


    ReplyDelete
  10. Giles, the data you cite is in annual form and has estimates for 2012. And I am not sure it is total expenditures. If you look at the quarterly data for for federal total expenditures there is a drop in spending since about mid-2010. It is not sharp, but it is there. Maybe this longer-term graph will make it more apparent: http://research.stlouisfed.org/fred2/graph/?g=drK
    The points is that these expenditure declines since mid-2010 have happened in the midst of a weak economy, something that Keynesian economics says is a no-no.

    Total expenditures, however, miss the full effect of fiscal policy that is why I included the budget deficit. It reflects both tax cuts and increases in spending that can be used to stimulate the economy. And on this count there is no mistaking a very sharp contraction. See this figure: http://research.stlouisfed.org/fred2/graph/?g=drL

    ReplyDelete
  11. David, thank you very much again. The near stopping of Federal expenditures does seem clearer now, particularly compared to the earlier splurge.

    I am less convinced that throwing in revenues is necessarily useful, because particularly at this stage of the cycle they are very highly geared to NGDP, and untangling a structural component is very difficult; in the UK, a fall in forward NGDP has seen a 57% fall in forward tax revenues, even though tax/NGDP is 36%.

    One point that you and the others (Lars, Scott, etc) make very clear is that even measuring what the stance of fiscal policy IS can be frustrating - is it relative to expected pathways, for example - and this conversation really nails that point.

    I have clearly taken more of your time than I deserve; many thanks for it.

    ReplyDelete
  12. Could we get some kind of reference from 2001 to 2008?

    ReplyDelete
    Replies
    1. Just use these links to play around with the data. I haven't figured out how to send graphs like those above, but the site is pretty intuitive.
      http://research.stlouisfed.org/fred2/series/GDP
      http://research.stlouisfed.org/fred2/series/W019RCQ027SBEA

      Delete
  13. has yet to allow a period of catch-up nominal spending growth that would return NGDP to its pre-crisis trend.

    The 1995-2006 housing bubble expansion was unsustainable. Returning to an unsustainable trend is a lousy goal.

    With baby boomers now retiring, for the first time since WW II (naturally!), all pre-BB retirement trends will be changed to have higher dependency and less production; thus less growth. Demographics is (a big influence on) destiny.

    ReplyDelete
  14. Which way does the causation run? Does it go from fiscal variables to GDP or vice versa? Surely one needs to use a cyclically adjusted measure of fiscal policy to make causal inferences?

    ReplyDelete
  15. This tiny sample of short-term data can't disprove the counterargument:

    Growth would have been faster with more spending/less austerity.

    That might be wrong, but long-term (and especially multi-country) studies strongly suggest it's right. This short-term splash of some recent numbers tells us nothing. And I have to presume that given his training, Professor Beckworth knows that on some level.

    Really, this "proof" is worthy of an internet econocrank like me, not a professional economist supposedly well versed in the methods, practices, and standards of his discipline.

    Heck, even yahoos like me understand the importance of feeding their judgment by looking at multiple (long) periods, of various lengths, and various time lags over which effects might appear.

    Viz:

    http://www.asymptosis.com/?p=5376

    These are the kind of studies Professor Krugman cites. Who you gonna believe?

    ReplyDelete
  16. does this graph properly account for time delay in cap projects; eg if the stimulus/porkculus bill had a lot of 5 year cap projects, like roads, would that be enough to skew the data ?

    the change in NGDP is about 7%; the change govt spending is about 3.5%
    That implies a large effect elsewhere in the economy.
    How do we know that this isn't a business cycle growth, and that the stimulus of 2008/09 helped that growth ?

    ReplyDelete
  17. Great article. I think, everyone in the space knows better than to create causal statements depending on the short-term pattern collections in these charts. Both graphs seriously weaken the discussion for coutercyclical financial plan and recommend a very a low financial multiplier. Country Reports

    ReplyDelete