Glasner demolishes Cole and Ohanian

Cole and Ohanian are right that the NIRA and other similar policies greatly retarded the recovery from the Depression.  So why do they have to push things too far, and try to suggest that demand stimulus didn’t play an important role?  David Glasner demolishes their argument in this post.  I’m really surprised they are still making these misleading claims.  I sent Ohanian papers with all the information in Glasner’s post, so I can’t see how they would be unaware that their data is extremely misleading.  At monthly frequencies wholesale prices and output were highly correlated throughout the Great Depression.  Period.  End of story.

I also keep pointing out that Paul Krugman continually misrepresents Lucas’s views of macro.  But he keeps doing it.   Lucas has indicated that the big drop in nominal spending in late 2008 and early 2009 depressed real output.  He does accept Friedman and Schwartz’s explanation of the Great Contraction.  But Krugman keeps implying Lucas believes demand shocks don’t matter.  Maybe Lucas doesn’t use Keynesian terminology, but he certainly buys the standard view that nominal shocks can matter in the short run, but not the long run.   BTW, I don’t agree with everything Lucas says about the Great Recession, but his views ought to be characterized accurately.

PS.  My email box keeps getting busier and busier.  Don’t be surprised if I’m not able to answer requests, or just give a one sentence explanation.  I encourage people to rely more on my previous posts, now that I am somewhat better organized.  The link at right called “Links to key blog posts and papers” is the place to go.  You need to scroll down on the right side of the blog to find that link, I’ll try to rearrange it later.  Eric Morey also recommended I put a link for RSS feeds for all comments, and I’ve done so in the same section.  I expect further improvements soon, and will let you know.

I’d also like to better organize posts by topic.  I spent 2 hours yesterday creating a new category; “China.”  There must be a better way.  I will gradually try to add more.



13 Responses to “Glasner demolishes Cole and Ohanian”

  1. Gravatar of Josh Josh
    26. September 2011 at 16:24

    Their presentation does seem rather dishonest, but it is consistent with their other work on monetary factors and the Depression/s:

    Note that I probably take the Lucas view on this, which is a synthesis of Friedman/Schwartz and their work, and not drastically different from what your view seems to be.

    And yes, Krugman is consistently not worth listening to on Lucas, Prescott et all… but especially Lucas. He takes what is basically a methodological difference and tries to morph it into an ideological difference, because there’s basically no chance that his readership will understand the former.

    So we end up with:

    “The Fed’s response was definitely needed and the fiscal stimulus was probably fine too and also I voted for Obama”


  2. Gravatar of StatsGuy StatsGuy
    26. September 2011 at 18:07

    Hire an undergrad. Or an unemployed grad student somewhere – there are plenty, and they’re cheap. 🙁

  3. Gravatar of Benjamin Cole Benjamin Cole
    26. September 2011 at 18:48

    Kudos to Glasner and Scott Sumner. Dang–can;t we get someone, or group of people, to make a presentation to Ben Bernanke?

  4. Gravatar of JimP JimP
    26. September 2011 at 19:17

    GDI vs.GDP

  5. Gravatar of Jon Jon
    26. September 2011 at 20:13

    As soon as I read that essay in the WSJ, I was on my iphone checking for your reply. It’s too bad the WSJ doesn’t have a comment-reply journal style going.

  6. Gravatar of Kevin Donoghue Kevin Donoghue
    27. September 2011 at 01:01

    “But Krugman keeps implying Lucas believes demand shocks don’t matter.”


    In the 1970s, Lucas and disciples take it up a notch, arguing that we should assume rational expectations: people make the best predictions possible given the available information. But in that case, how can we explain the observed stickiness of wages and prices? Lucas argued for a “signal processing” approach, in which individuals can’t immediately distinguish between changes in their wage or price relative to others “” changes to which they should respond by altering supply “” and overall changes in the price level.

    He’s not implying Lucas believes demand shocks don’t matter. What he says is that the Lucas project relies on “rational confusion” to model demand shocks, which might be plausible if recessions were always short; but it’s a stretch to say that rational confusion explains the recessions we actually observe. That’s fair enough, surely?

    If Krugman continually misrepresents Lucas that’s important. So please point me to an instance where he actually does so.

    Also, kudos to David Glasner.

  7. Gravatar of Jason Odegaard Jason Odegaard
    27. September 2011 at 05:07

    Maybe I’m being a bit naive here, but perhaps yourself or David Glasner should write a letter to the editor of the WSJ expressing the inaccuracies presented in Cole and Ohanian’s article? Who knows, could help.

  8. Gravatar of Scott Sumner Scott Sumner
    27. September 2011 at 18:03

    Josh, Very well put.

    Statsguy, For some reason I’ve always felt I had to do things myself to get it right. It probably hurt my career, as I spent so much time getting the Depression data right that it limited my publications. Now I notice that many other academics are very sloppy, often using in correct data in their published papers. But from a cost benefit perspective, that may be optimal.

    Thanks Ben.

    Thanks JimP, I saw that and am already planning a post. From now on it’s NGDI targeting.

    Jon. Yes, that would be a good idea.

    Kevin, It’s not the precise words (Krugman is very artful with language) it’s the way he leads readers to believe Lucas is a clueless RBC-type who doesn’t think demand shocks matter.

    It’s also true that Krugman almost never said monetary policy is ineffective at the zero bound, but 99.9% of his readers thought he said that, and he must have known that.

    Jason, I used to view the world that way, but I’ve become very cyclical in recent years.

  9. Gravatar of “Bashing Cole & Ohanian” from another angle | Historinhas “Bashing Cole & Ohanian” from another angle | Historinhas
    27. September 2011 at 20:50

    […] Monetarists” have had a go at it. David Glasner did it in elegant and convincing fashion. Scott Sumner was glad Glasner “demolished” it, Bill Woolsey was “circumspect” and Krugman (a […]

  10. Gravatar of Ritwik Ritwik
    28. September 2011 at 21:21


    Questions of good faith and bad faith aside, don’t you think you/Glasner/Krugman need to define better what ‘aggregate demand’ is? (I understand the AS/AD model, btw. Or atleast I think I do. )

    It seems to me that what Glasner and you have proved is that the price level was procyclical during the great depression. This is prima facie some evidence of the AD-driven business cycle hypothesis. Which is fine.

    But in their argument (at least the version presented in the defence to these criticisms), Cole & Ohanian talk of the aggregate demand as operating through the aggregate labour ‘channel’ which in turn is quantified by the number of hours worked. Is there something fundamentally wrong with this characterization? If not, then Cole and Ohanian have their own correlations and graphs to match up with yours.

  11. Gravatar of Scott Sumner Scott Sumner
    29. September 2011 at 05:06

    Ritwik, I agree that AD works through hours worked, but I do not think that C&H have any data that contradicts my view of the Great Depression. I believe the NIRA program (an adverse supply shock) aborted the recovery from the Depression after July 1933. So do C&O. I gather Krugman disagrees.

    I believe a massive AD shock pushed up industrial output by 57% between March and July 1933. Do C&O have any data to contradict that view? I believe an AD shock pushed up output and hours in 1936 and early 1937. Can they contradict that? I believe an AD shock pushed up output and hours from mid-1940 to late-1941. Can they contradict that?

    Also, Does C&O data include hours worked in government jobs programs? As you may know, the official unemployment data from that period treats the government jobs program workers as “unemployed.”

  12. Gravatar of Barry Barry
    29. September 2011 at 05:06

    “Cole and Ohanian are right that the NIRA and other similar policies greatly retarded the recovery from the Depression.”

    I’m sorry, but where did they support the point of ‘greatly retarded’? Several problematic months at that point was not ‘greatly retarded’.

    I’m beginning to think that the right-wing revisionists are walking slowly backwards; they can’t show that FDR caused the Great Depression, nor that FDR lengthened it or made it worse, so they are finding minor problems, and trying to present them as doing this.

  13. Gravatar of Scott Sumner Scott Sumner
    29. September 2011 at 05:18

    Barry, You misread my comment, I said they were right, not that they had good arguments. I haven’t even read the paper yet. I spent much of my life studying the effect of the new Deal, and the NIRA basically put the recovery on hold for two years. Here are the facts:

    1. FDR’s program of leaving the gold standard and devaluing the dollar pushed industrial production up by 57% in the 4 months from March to July 1933. We recovered 1/2 of the ground lost in the previous 3 and 1/ years.

    2. In July 1933 FDR raised nominal and real wages by 20% (NIRA program) for most workers. Monthly industrial production fell immediately, and didn’t regain July levels until a few months after the NIRA was declared unconstitutional in May 1935. There were a total of 5 wage shocks during the New Deal, in each case the growth of industrial production plunged right after the wage shock.

    4. By the spring of 1940 the US was still deep in the Great Depression, with very high unemployment. (The data is unclear, but all the estimates are quite high.) That’s 7 years after FDR took power. Under Harding the recovery from the very deep 1921 downturn took a bit over one year. If we had recovered at that pace the GD would have been over around 1936.

    5. FDR’s decision to devalue the dollar was a great move, and largely explains the recovery up to 1940. But the recovery was still disappointing, even if light years better than the Hoover administration. The German invasion of France in the spring of 1940 caused US military production to soar, and the Depression ended about at the end of 1941.

Leave a Reply