Talk:Recession of 1937–1938

Causes link
I think that the link to causes should be removed and the causes be placed directly on the page. - May 29th, 2007

Needs revision, clarification and expansion
This sentence from the first paragraph doesn't make a lot of sense:

Economic historians have not agreed on the causes [1], but many of the causes show that because the New Deal involved spending money from the Federal budget, President Roosevelt had to end New Deal spending, and thus programs, as a result.

Similarly from the last paragraph:

It began to get better in mid-1938, and every month it was better.

And:

The farm population had fallen 5%, but farm output was up 19%, so the remaining farmers were better off than the average farmer in 1939.

teneriff (talk) 01:14, 15 November 2008 (UTC)

Wait, Hemp Production? I mean if you want to get detailed go right ahead, but you're going to have to talk about something more than a niche manufacturing industry which wasn't even outlawed 01:57, 27 January 2012 (UTC) — Preceding unsigned comment added by 98.24.15.126 (talk)

Reports of New Deal death are greatly exaggerated
The following is not supported by the historical record and the New Deal Wikipedia article:

"Economic historians have not agreed on the causes [1], but many of the causes show that because the New Deal involved spending money from the Federal budget, President Roosevelt had to end New Deal spending, and thus programs, as a result."

I'm going to update it to read:

"Economic historians have not agreed on the relative importance of the causes the Recession of 1937. The following are generally cited as contributing factors:  The reduction of net government contributions due to balancing the budget, excessive business inventory build up, government actions that increased the cost of credit in response to rapidly rising commodity and labor costs, insufficient business investment and labor costs that continued to climb even after commodity prices collapsed."

I'll add the appropriate references when I post.

Does anyone who is a better writer than me want to do this update or are there any factual problems with the above? Oneeyedguide (talk) 16:48, 11 February 2009 (UTC)


 * related is the unsourced claim that employment didn't return to 1937 levels until the war started while the supporting sourced employment graphic shows employment higher than in 1937 by 1939. Mulp (talk) 22:12, 10 March 2009 (UTC)

What changed in 1937?
Some have said that Wartime spending is what ended the recession (and the aftershocks of the Great Depression) for good. However if this was the case, then the Depression should have started back up at the end of WWII. In fact the opposite happened. A new boom time started. You could suggest that the pump was primed by the spending, but most of the capital expenditures in a war time economy don't really apply to a peace time economy, so you have the same primed pump and production surplus situation in 1944 as in 1928. The only real difference was > 1944 was a tax and spend regime with 90% top tax bracket, strong unions, and a growing middle class. 1928 had a locked in Gold standard, lots of financial leverage, a disappearing middle class, and an extremely low top tier tax rate, all of which had depressing results. Since all of these changes started during the recession of 1937, they are relevant to the discussion here. — Preceding unsigned comment added by 70.113.69.97 (talk) 05:33, 11 July 2011 (UTC)
 * There was a corrective recession in 1945. The likely reason for not having a larger recession for a few years is that the US was the only country with any productive capacity left. Europe HAD to buy certain goods, so they bought from the US, which helped the US economy.--Metallurgist (talk) 20:24, 11 August 2011 (UTC)

I'm confused about this page in general, and this section in particular. Christina Romer's work, especially "What Ended the Depression?" (1992) shows gold sterilization by Treasury is the primary cause of the 1937-38 recession. This is supported and expanded by Douglas Irwin's "Gold Sterilization and the Recession of 1937-38". Sterilization, in response to the enormous inflows of gold since the U.S. reset the price of gold in 1933, was implemented to offset the risk of inflation. It was accompanied, in an uncoordinated effort, by the Federal Reserve increase reserve requirements for banks, effectively reducing credit. The Fed's move had a much less significant impact because banks were relatively cautious themselves. Treasury's action was catastrophic to the recovery. Friedman and Schwartz (1963) give somewhat more credit to the Fed, explaining "the sterilization program sharply reinforced the rise in reserve requirements in producing monetary tightness: the rise in reserve requirements increased the demand for high-powered money; simultaneously, the Treasury's action virtually brought to a halt an increase in the stock of high-powered money which had been proceeding with only minor interruptions since 1933." (pg. 510 of 1993 Princeton edition)  Carolinechurchill (talk) 22:46, 28 November 2020 (UTC)CarolineChurchill

Keynesian and Monetarist views are blatantly false
The 1937 deficit was $2.2 billion while the 1936 deficit was $4.3 billion (-$2.1 billion). This is true. However, total government spending only decreased by $0.6 billion. This is due to increased tax revenues. Total government spending in 1935 was $6.4 billion, while in 1937, it was $7.6 billion. The Keynesian view that the deficit spending reduction caused the recession is either a lie or mistakenly false. Further, we can go into 1938 and the deficit was only $0.089 billion, despite economic recovery. Again, the Keynesian explanation is completely false. There was less spending in the middle of the recovery.

Decreased lending and borrowing contracted the money supply. This was due to the recession, not what caused it. The contraction occurred after the recession started.

So both theories are absolutely wrong and should be removed. I would have removed it myself, but I wanted to make a clear case and explanation for why it should be removed.

Source --Metallurgist (talk) 20:36, 11 August 2011 (UTC)


 * Just because it's false doesn't mean the interpretation shouldn't be here. We should have an "interpretations" section. This page might help out:

http://mises.org/daily/4039/Dangerous-Lessons-of-1937 byelf2007 (talk) 11 August 2011
 * Yeah I put that. I can understand putting the monetarist view. Thats a bit more complex, but the Keynesian view is completely false. This is easily verifiable. It would be like having creationism prominently displayed on the Evolution page.--Metallurgist (talk) 07:16, 12 August 2011 (UTC)
 * We're not talking about having the Keynesian interpretation on the "Austrian School" page, we're talking about it being on a recession page. Whether we like it or not, Keynesianism dominates the economics field, so we're going to have to have their interpretation in this article, along with the monetarist view (which is BS as well), and the Austrian view (a 45% expansion of the money supply from 1933 to 1937 caused it). Will you please make an (unbiased) "interpretations" section? You seem to have the most interest in it. byelf2007 (talk) 12 August 2011


 * What the heck are you talking about? Keynesian economics say the economy shrank because of a decrease in government spending and an increase in taxes.  Both of those things are demonstrably true.  You seem to only be concerned with the government spending side of the equation but even you realize government spending shrank in 1937.  I'm not sure what you're arguing.  The Keynesian theory is clearly wrong because government spending only shrank 8%, ignoring the tax increases? Roofle (talk) 18:21, 14 August 2011 (UTC)
 * He's pointing out that the idea that "decrease in deficit spending = bad for economy" is demonstrated to be false. The budget deficit continued to decline from 1937 to 1938 while the economy began to recover from the recession. It fell to $89 million. This means the budget deficit fell by $2.2 billion between 1936 and 1937 (before the recession) and by another $2.2 billion between 1937 and 1938 (during the recession), yet the economy improved between 1938 and 1939. byelf2007 (talk) 12 August 2011
 * No one says that. Decrease in aggregate demand is bad for the economy.  A decrease in aggregate demand is caused by decreasing government spending or by raising taxes, both of which the American government did in 1937.  The budget deficit is irrelevant to this discussion. Maybe you should read up on Keynesian theory before declaring the entire thing to be false? Roofle (talk) 03:35, 15 August 2011 (UTC)
 * We're not criticizing "Keynesian theory," but certain depictions of 1937 usually made by Keynesians. I wouldn't say "no one" ever said cutting the deficit (as such) was a bad move in 1937 (Krugman has said this many many times). byelf2007 (talk) 14 August 2011
 * Oh cool you're not criticizing Keynesian views. That's why we're posting under the heading "Keynesian views are blatantly false".  Also, cutting the deficit WAS a bad move because that policy resulted in less government spending and higher taxes in 1937.  The deficit itself is irrelevant to aggregate demand but the government choices when constrained under a "lower the deficit" policy certainly had an effect on aggregate demand.  I don't think you're arguing honestly with me because you first claimed "decrease in deficit spending = bad for economy" was a truth held by Keynesian views which is blatantly false and have now changed to say "the decrease in deficit spending in 1937 was a mistake" which is absolutely true. Roofle (talk) 17:08, 15 August 2011 (UTC)
 * A "Keynesian view" can include the "views of a Keynesian." Get it? byelf2007 (talk) 15 August 2011


 * So please cite which Keynesian says that a decline in deficit spending is always bad, and which Keynesian said that only government spending matters and not taxation levels.

No mention anywhere of an actual empirical savings rate. This entire article is bullshit. — Preceding unsigned comment added by 2605:A601:A78C:9900:F0E9:6031:82DB:8D4 (talk) 11:12, 24 May 2020 (UTC)

Unemployment Rate Discussion is Very Inaccurate
The numbers are based on Lebergott's estimates. These have two major problems and then a third interpretive problem when compared to modern unemployment numbers.

1) They exclude people directly employed in WPA and other New Deal programs. This is an issue Michael Darby covered over four decades ago an in article entitled ""Three-and-a-Half Million U.S. Employees Have Been Mislaid: Or, an Explanation of Unemployment, 1934-1941".

Now for certain purposes it may make sense to include WPA and other direct employment along with unemployed workers (notably, in estimating a reserve army of labor that is avalaible to be hired by private employers) but it is not the current methodological practice and it misleads lay readers in the understanding of the 1930s economic experience. At the very least, both numbers should be published side by side so that readers can get a more accurate sense of what was going on, as Professor Mathy (2018) does in this recent paper on long term unemployed during the 1930s.

2) Even the Lebergott estimates as adjusted by Darby are misleading because they are not monthly series like we have today and are based on a number of estimation strategies that leads to "spurious volatility" and incorrect estimates, or at least, estimates that are at variance with modern methodologies. Professor Christina Romer (1986)showed this in a lot of detail. To quote her at length.

"If the labor force were also procyclical in the 1900-1930 period, then it is clear that Lebergott's labor force numbers are too high in recessions and too low in booms. This implies that the unemployment rate calculated as a residual is artificially high in recessions and artificially low in booms. Thus the historical unemployment rate is, by construction, more volatile than the truth [...]

To estimate annual employment, Lebergott uses more complicated procedures. He estimates it as the sum of several component series on employment in various sectors and among various classes of workers. To form these component series he begins with basic data on employ- ment in each sector in whatever base years are available. He then interpolates each employment series using some annual variable he believes to be related to employment in that sector. The most com- mon interpolating variables are measures of output, fragments of employment data, and indexes of labor demand"

She goes onto explain the problems with this procedure. Meanwhile, we do have a number of estimates, the best of which is the National Industrial Conference Board monthly data as Mathy (2018) points out. These estimates continued after the war and thus we can compare them to official statistics and see how they are reasonably close. These numbers broadly conform to Lebergott's estimation strategy but, as Romer predicts, the unemployment rate is much lower than his estimates. In addition, we get to see month to month patterns we don't in Lebergott's data that qualitatively change our narrative from the era.

3) The final issue is that this number, as low as it is, is still an overestimate of unemployment when we compare it to modern headline unemployment. Today's measure, U3, doesn't include "discouraged workers" i.e. workers who desire employment but have given up searching. Meanwhile, the methodology used for the 1930s number, previous to the the search concept of unemployment developed by the WPA in the late 1930s which is used today, does capture what we now consider discouraged workers on the idea that all able bodied males are supposed to be working. Thus the appropriate comparison for the unemployment rate is U5, not U3. When compared to U5, a different narrative emerges. For example, in June 2016 U5 was 6 percent while in June 1937, U5 was 5.4 percent. This suggests the economy, while not at full employment, was recovered from the great depression in 1937 before the recession of 1937-1938. The article should be updated to reflect the best avalaible data on the 1930s economic performance and the true meaning of the "Roosevelt Recession". See e.g. Card 2011.

''References '' Card, David. "Origins of the unemployment rate: the lasting legacy of measurement without theory." American Economic Review 101.3 (2011): 552-57.

Darby, Michael R. "Three-and-a-half million US employees have been mislaid: or, an explanation of unemployment, 1934-1941." Journal of Political Economy 84.1 (1976): 1-16.

Romer, Christina. "Spurious volatility in historical unemployment data." Journal of Political Economy 94.1 (1986): 1-37.

Mathy, Gabriel P. "Hysteresis and persistent long-term unemployment: the American Beveridge Curve of the Great Depression and World War II." Cliometrica 12.1 (2018): 127-152. — Preceding unsigned comment added by Nathan Tankus (talk • contribs) 20:39, 13 June 2019 (UTC)

Is this language worthy of an encyclopedia?
Article says: "But until 1937 Roosevelt had claimed responsibility for the excellent economic performance. That backfired in the recession and the heated political atmosphere of 1937." What is the antecedent of "That"? Or did the writer mean to use "that" as a relative pronoun and cut off a subordinate clause from the noun it modifies? How could an excellent performance backfire? The statement implying an excellent economic performance during the depression seems fantastic. (AltheaCase (talk) 03:09, 12 August 2022 (UTC))