Did Keynes understand his own theory? Apparently not.

At least if we are to believe modern Keynesians like Paul Krugman, who insist that the deep 1937-38 recession was produced by a tightening of fiscal policy.

In previous posts I’ve pointed out that there are all sorts of problems with the modern Keynesian explanation for 1937.  The tightening was mostly in taxes and transfers, not government output.  In contrast, modern Keynesians suggest that stimulus involving government output is far more effective than tax cuts.

The timing is also wrong.  Investors were fully aware of fiscal tightening in early 1937, yet stocks were at very high levels.  In contrast, my explanation of the deep recession (mostly a turnaround in the world gold market which led to deflation) correlates very closely with movements in the stock market.  (I also think FDR’s high wage policy played a role in the recession.)

But who cares what I think.  I’m more interested in Keynes’s views.  Would he agree with me or Krugman?

Keynesians like to make a big deal of Keynes’s skill in investing, which is mostly (albeit not entirely) a myth.  He lost lots of money in the 1920s and had to be bailed out by his father so that he could start over again.  In the spring of 1937 he thought it was a great time to own US equities, indeed so much so that he purchased many on margin.  How’d that work out?  Here John Hussman:.

As biographer Robert Skidelski observes, “In the year of the ‘terrific decline’ which had started in the spring of 1937, he lost nearly two-thirds of his money.”

Keynes was fully aware of the fiscal tightening by this time.  And he was fully aware that a steep recession would cause a sharp sell-off on Wall Street.

So which is it?  Is the Keynesian theory wrong?  Or did Keynes not believe his own theory?

I say the Keynesian theory is wrong.  There was no reason for Keynes not to own stocks in 1937.  The events that caused the severe recession also caused the stock market crash–and those occurred in the last half of 1937.  EMH + market monetarism >>>>>> Keynesian theory.



21 Responses to “Did Keynes understand his own theory? Apparently not.”

  1. Gravatar of marcus nunes marcus nunes
    27. November 2011 at 06:34

    Flash forward, not even the New York Fed believes monetary policy had anything to do with the “Great Recession”. The conventional “causal chain” is from house price to financial crisis to the steep fall in RGDP! I did a post:

  2. Gravatar of david david
    27. November 2011 at 07:50

    Scott, you -are- aware that the Hussman article you link to is screaming about inflation, etc., right?

  3. Gravatar of Blue Aurora Blue Aurora
    27. November 2011 at 08:02

    While I disagree with your position that Keynes did not understand his own theory, I have a question for you, Scott. Have you seen Michael Emmett Brady’s review of Allan Meltzer’s “Keynes’s Monetary Theory: A Different Interpretation”?


    I have a feeling that you might like that review, even if you have yet to read the book. Expectations are important to the economics of Keynes and Keynesian economics, after all.

    P.S. Would it be possible for me to e-mail you at some point? I’d like to engage in some correspondence.

  4. Gravatar of Greg Ransom Greg Ransom
    27. November 2011 at 08:43

    What is “government output”?

  5. Gravatar of Greg Ransom Greg Ransom
    27. November 2011 at 08:50

    The high wage policy was THE fashionable strategy for increasing business demand in U.S. circles — a policy position wildly popular in America during the 1920s and 1930s .. a conception, famously, which American economists could not find to contain any flaw.

  6. Gravatar of ssumner ssumner
    27. November 2011 at 09:49

    Marcus, That’s a very good post.

    david, Inflation? What’s that?

    Blue Aurora, Yes, I have read the Meltzer book.

    You can email me but I can’t guarantee a response, as I get more emails than I can deal with as it is. But I’ll try to give you a response if your email is short and to the point.

    The post title was sarcasm. Obviously I know that Keynes understood his own theory in its own terms, but it’s also 100% obvious that he understood it much differently than modern Keynesians. He did not believe that the US fiscal tightening was likely to lead to a deep recession, we know that for sure. Modern Keynesians say Keynesian theory predicts that tightening would cause a recession.

    Greg, It is output produced by the government.

    Yes, the high wage policy was a disaster.

  7. Gravatar of marcus nunes marcus nunes
    27. November 2011 at 10:19

    And monetary policy continues to be discriminated!

  8. Gravatar of Blue Aurora Blue Aurora
    27. November 2011 at 10:28

    Scott Sumner: I see. What did you make of Michael Emmett Brady’s review, though? I had a feeling that you would be bomnbarded by e-mails. What is your e-mail address?

  9. Gravatar of Carl Lumma Carl Lumma
    27. November 2011 at 12:44

    I humbly suggest that instead of “EMH”, you use “rational expectations”. I reckon the latter more specifically applies to the market monetarist approach. Plus, it just won a Nobel and doesn’t have the PR problems that (unfairly or not) plague “EMH”.

  10. Gravatar of UnlearningEcon UnlearningEcon
    27. November 2011 at 13:51

    Keynes understood that both the 1929 recession and 1937 were caused by catastrophically high long term rates.

  11. Gravatar of Rajat Rajat
    27. November 2011 at 14:50

    To pick up on David’s point, Hussman seems to disagree with you on many things. He has been a big critic of the Fed’s unconventional measures over the past few years and has consistently pushed instead for the restructuring of residential mortgage obligations in exchange for principal appreciation rights. In particular, he thinks that the increases we have seen in the money base *to date* will almost inevitably lead to inflation – or very high rates of NGDP if you prefer – over the second half of this decade. Yet he doesn’t think that would be a good thing, hence his criticism of the Fed. It seems to me that you would disagree with him on both the positive and the normative, so is there any reason you quote him other than as the source of a Skidelski quote?

  12. Gravatar of Tommy Dorsett Tommy Dorsett
    27. November 2011 at 19:36

    Hussman has been calling for a stock market crash/recession every year since1995. In other words, even with the last decade of two ‘bad bear markets’, he has been wrong, on net, for hundreds of S&P points. Having one view that never changes — and attacking others in a dishonest way – is par for his course. The permabears love him, but the emperor has no clothes and it sure looks like the pool was cold.

  13. Gravatar of Rajat Rajat
    28. November 2011 at 06:52

    Well, it’s not for me to defend Hussman, but I don’t recall him predicting a recession in 2004, 2005 or 2006. I also note that there has been no nominal increase in the S&P500 for the last 13 years, which suggests that there may be something to his views on equity valuation. And I suspect that Treasuries have outperformed equities since 1995.

  14. Gravatar of TheMoneyIllusion » The Eurozone must not “move forward” TheMoneyIllusion » The Eurozone must not “move forward”
    28. November 2011 at 08:00

    [...] Karl Smith scolded me yesterday for my post on Keynes and stocks during 1937.  He argued that this is a very serious topic that affects the lives of millions.  I [...]

  15. Gravatar of ssumner ssumner
    28. November 2011 at 15:54

    Marcus, Yes, the entire profession seems to have amnesia about monetary policy.

    Blue Aurora, I don’t actually recall much of Meltzer’s book, as I read it long ago. I do recall one interesting footnote–he asked why if Keynes was worried about liquidity traps, he didn’t simply recommend a higher inflation target. Still a good question.

    I don’t have strong views on that review, as Keynes can be read in many different ways, and all are arguably correct. There is no such thing as “what Keynes really meant.”

    email: ssumner@bentley.edu

    Carl, Both have PR problems, but I’ve never cared much about conventional wisdom. It is usually wrong.

    UnlearningEcon, Perhaps, but I don’t recall long rates being that high in 1937—what were they?

    Rajat, Yes, it was simply the first quote I googled. I knew about the 1937 stock losses already, but couldn’t find my Lords of Finance book for a quote. I don’t agree with his views on monetary policy.

  16. Gravatar of UnlearningEcon UnlearningEcon
    28. November 2011 at 16:11

    Hey Scott,

    I have a poorly taken photo:


    They spiked up to about 6%, despite Keynes’ advice in his letter to FDR to keep them down at 2.5%.

  17. Gravatar of Blue Aurora Blue Aurora
    29. November 2011 at 07:06

    I agree with you that the whole “what Keynes really meant” issue is a pointless wild goose chase. Also, I’ve sent you an e-mail.

  18. Gravatar of Scott Sumner Scott Sumner
    29. November 2011 at 07:22

    UnlearningEcon, Those are corporate yields, what were government yields?

    Blue aurora. Under what name? I don’t think I got it.

  19. Gravatar of Blue Aurora Blue Aurora
    29. November 2011 at 07:30

    I prefer not to give out too many details, but it addresses the econophysicists, a group I have told you about before.

  20. Gravatar of Greg Ransom Greg Ransom
    29. November 2011 at 16:01

    Government output “is output produced by the government.”

    What market establishes the value of this “output”, i.e. with out pricing in the market via exchange, “government output” is best classified as direct consumption priced at multiple times its value to consumers.

    Examples: government pork air ports without significant customers, government museums with significant customers, most humanities courses paid for by government –a list which could be extended endlessly.

    The government could equally produce “output” by burning large stadiums full of $100 bills, or by paying people millions to set forests and neighborhoods and oil wells on fire.

    All of it would be a equivalent sort of “government output” .. equally lacking an valuation using exchange ratios in the market.

  21. Gravatar of Scott Sumner Scott Sumner
    1. December 2011 at 05:24

    Greg, Yes, government output is often valued at much less than cost. But how does this relate to the post?

Leave a Reply