Tuesday, May 17, 2011

1937, the Fed, the stock market, and the Treasury

Sitting on cash when people want cash drives up interest rates. High interest rates discourage investment. They discourage investment even more when investors are already predisposed against expanding operations to provide for future demand that they're uncertain about. They discourage investment even more if there are arithmetic barriers to relief on interest rates (like a nominal price floor and weak inflation).

As far as the economy is concerned, it doesn't matter who does it: households increasing savings, businesses demanding and retaining higher profits, central bankers mandating higher reserves, or fiscal authorities sitting on money. As far as history is concerned, of course, sometimes one entity is more to blame than another.

Paul Krugman recently wrote a post on the "recession within a depression" of 1937, which he says eerily parallels the discourse we see today. He says (citing Friedman and Schwartz) that in 1936-1937 the Federal Reserve started getting concerned about the inflationary potential of the large growth in excess reserves. These concerns are not unlike those we are hearing today. They responded by raising reserve requirements, a decision which Friedman and Schwartz consider to be the source of the 1937 downturn. Krugman raises doubts about how important this policy really was, citing the Romers. Recent research looking at Federal Reserve member banks by Calamoris, Mason, and Wheelock (2011) confirms Krugman and the Romers' suspicion that Friedman and Schwartz were too quick to blame the Fed. They demonstrate that the reserve requirement rules were non-binding. This is all in the spirit of Krugman's own mentor, James Tobin, who said of the 1936-37 Fed policy that "raising reserve requirements may have been a mistake but it was probably a relatively harmless one."

Jonathan Catalan presents a response that ties Krugman to Friedman and Schwartz (which seems a little odd to me). Aside from the question of what Krugman thinks fo the Friedman and Schwartz argument, Jonathan does identify the common thread between the three economists, that "the point Krugman is trying to make is that tight monetary policy will definitely set a recovery back". Jonathan, of course, disagrees - a position which he has staked out earlier in his Mises Daily article on the 1937 recession. He blames the stock market crash for the contraction of the money supply. Falling stock prices increased precautionary demand for money, leading to a contraction in the money supply.

This is a little confusing to me. As I understand it, the stock market crashed in August of 1937, the monetary base started falling at the beginning of 1937, and output started falling earlier in the spring. How does Jonathan get the causality going from the stock market to the monetary base? I don't entirely understand the argument.

So barring an explanation from Jonathan, he seems to be wrong in pointing to the stock market (and thus businesses and households) as the source of the reduction in the money supply. Friedman and Schwartz are wrong that the Fed is to blame (according to Jonathan, Krugman, Calamoris, Mason, Wheelock, Tobin, Romer, and Romer). So what happened in 1937?

Well once again we can look to Calamoris, Mason, and Wheelock (2011), who suggest that the real culprit may have been the U.S. Treasury. In 1936 the Secretary Morgenthau decided to play central banker and sterilize billions of dollars of incoming gold by stock-piling it and paying for it by selling government securities (rather than by depositing it at the Federal Reserve). This is just as contractionary as it would be if the Federal Reserve had engaged in open market sales.

Of course taxes were raised and the deficit was cut too - that doesn't help. But the gold sterilization point is interesting because you don't hear people talking about it as much.

Does this history make sense? I'm no expert on the depression, but the literature seems to point to this explanation. Perhaps Friedman and Schwartz defenders can reinvigorate the case for the impact of reserve requirements or Jonathan could explain how the stock market is causal here.

23 comments:

  1. "They demonstrate that the reserve requirement rules were non-binding."

    In law or in reality?

    One thing you realize rather quickly about FDR, et. al. during the 1930s is that the rule of law meant nothing to these people - which is why they could imprison thousands of American citizens for their skin color or ethnic origin and call it legal.

    ReplyDelete
  2. "Non-binding" here refers to economically non-binding. The reserve requirement was found not to exceed desired reserves in the absence of a requirement. It's like a minimum wage law of $0.50. We'd call that non-binding.

    FDR had absolutely nothing to do with the reserve requirement policy.

    ReplyDelete
  3. Sitting on cash when people want cash drives up interest rates.

    Prove it. Sitting on cash only increases the value of money. "Cash" has nothing to do with interest rates.

    ReplyDelete
  4. http://scholar.google.com/scholar?hl=en&q=liquidity+preference+theory+of+the+interest+rate+empirical+evidence&as_sdt=0%2C9&as_ylo=&as_vis=0

    ReplyDelete
  5. You should check out what Scott Sumner has to say in some of his past entries. He's spent a great deal researching it:

    http://www.themoneyillusion.com/?p=6439
    http://www.themoneyillusion.com/?p=3192

    ReplyDelete
  6. Gold sterilization may have played an important role. However, the main culprit was a sudden fall in business activity largely because of a fall in profit margins (high wages are frequently cited).

    ReplyDelete
  7. Let me rephrase:

    Why don't you explain to me why cash should have any effect on interest rates.

    ReplyDelete
  8. People make trade-offs on all relevant margins, not just one. One margin they may make trade-offs on is savings versus consumption, and that trade-off is modeled by the classical picture of the loanable funds market. But another margin is that of holding more liquid versus less liquid savings. As people's preferences shift towards liquidity, borrowers will have to pay more to borrow long term.

    ReplyDelete
  9. High wages are tricky to diagnose in the Depression because we don't have microdata on wages. Compositional effects bias wage estimates upwards in downturns. We know this from essentially every study that has ever been done on post-war recessions comparing micro to macro wage data. Google "real wage cyclicality" and you'll get a lot of it.

    What happens is low marginal productivity industries and low marginal productivity workers get hit the hardest during downturns, they fail, and observed wages are of high marginal productivity workers in high marginal productivity industries. That means wages that are higher in aggregate. If you look at them disaggregated, you never see higher wages in the post-war period.

    The problem is, we don't have a lot of disaggregated wage data during the Depression. But it seems reasonable to me to conclude that the same phenomenon holds. So I'm skeptical of pointing to aggregate wage statistics and assuming high wages are the problem. Maybe, but a mountain of evidence on the postwar period suggests it's unlikely.

    Brad DeLong absolutely eviscerated Vedder and Galloway on this point back in the 90s when their book came out. I have a comment in to the QJAE on the same issue.

    ReplyDelete
  10. Mattheus - I don't know how many times we've gone over this. You just need to accept that we don't agree.

    What's amazing is that liquidity preference theory is not controversial among financial economists. I think you're just balking because it has macroeconomic implications that bother you.

    ReplyDelete
  11. Daniel,

    "FDR had absolutely nothing to do with the reserve requirement policy."

    Except when he was threatening firms to get in line with the non-binding reserve policy.

    ReplyDelete
  12. Daniel,

    While liquidity preference theory is not controversial amongst a subset of financial economists, how often it happens is. This was Milton Friedman's point; yeah, it is possible to get into a liquidity trap (though it is not very likely to ever occur), but to build your entire notion of economics around it (as Keynesians of all stripes do) is silly.

    ReplyDelete
  13. Gary -
    re: "Except when he was threatening firms to get in line with the non-binding reserve policy."

    WTF are you talking about? When did FDR ever threaten firms to get in line? If firms find reserve requirements too costly they just leave the Federal Reserve System, which has always been a major policy concern around reserve requirements - if they're set too high and there's no compensation (interest paid on reserves - which they are now doing), then less banking is done in the Federal Reserve System, weakening the traction they can get with monetary policy. Who is being threatened here? And if anyone is being threatened what does FDR have to do with it???

    ReplyDelete
  14. re: "This was Milton Friedman's point; yeah, it is possible to get into a liquidity trap (though it is not very likely to ever occur), but to build your entire notion of economics around it (as Keynesians of all stripes do) is silly."

    Not to be a broken record, but WTF are you talking about? A liquidity preference theory of the interest rate and a liquidity trap are two entirely different issues, Gary. A liquidity trap occurs when the money demand curve is horizontal. It's very rare and the only two times in American history even the most die-hard Keynesians think this has happened, what they're actually refering to is a zero interest rate floor, rather than a horizontal Md curve specifically. A liquidity preference theory of the interest rate, on the other hand, holds true and is accepted by most financial economists as a good explanation for the shape of the yield curve at all times - whether we're in a liquidity trap or not.

    ReplyDelete
  15. Daniel,

    "When did FDR ever threaten firms to get in line?"

    All the time. This is very well documented.

    There has never, ever been a liquidity trap in U.S. history (or world history for that matter); as Friedman stated, it is ludicrous to even think that such a thing would happen in the real world.

    Liquidity preference (demand for money*) and the trap are intimately related according to Keynesians (whatever Keynes thought about the matter is another story entirely) and there really is no reason to talk about the latter without talking about the former (or vice versa).

    *The notion of liquidity preference as Keynes stated it is fatally flawed because of a rather bad, unsubstantiated assumption Keynes makes in arguing for the concept - but there is no reason to get into that now.

    ReplyDelete
  16. re: "All the time. This is very well documented."

    Show me where it is documented that FDR threatened a firm over the question of reserve requirements.

    Stop saying it - show me a shred of evidence. If it's well documented it shouldn't be hard for you to back up these claims. Show me.

    ReplyDelete
  17. re: "Liquidity preference (demand for money*) and the trap are intimately related according to Keynesians (whatever Keynes thought about the matter is another story entirely) and there really is no reason to talk about the latter without talking about the former (or vice versa)."

    Gary, NO ONE WAS TALKING ABOUT THE LATTER until you came along. You can't talk about liquidity traps without talking about demand for money. You can talk about liquidity preference without talking about liqidity traps.

    ReplyDelete
  18. Liquidity traps are always seem to be underneath the surface of a discussion of liquidity preference. I just brought it out into the open.

    ReplyDelete
  19. Daniel

    Mattheus - I don't know how many times we've gone over this. You just need to accept that we don't agree.

    Clearly we don't agree. I'm not in any confusion over whether we agree or disagree on the idea of liquidity preference.

    What's amazing is that liquidity preference theory is not controversial among financial economists. I think you're just balking because it has macroeconomic implications that bother you.

    You're suggesting that I "balk" because I don't like the implications. How sophomoric. I would hope you consider me intellectually honest enough to not place my own opinions on how I should like things to work as a determinant of the theories I hold (just as I've never argued that you "balk" at natural rights theories because you "like" governments).

    The truth is, my intellectual process follows from premises to conclusions, not the other way around. Did you think I'd always been a libertarian?

    Honestly Daniel - I know we've been over it many times, but I've never been convinced. I've never been shown any deductive reason why cash balances ought to affect the interest rate. The most you've ever done is give me a GOOGLE SEARCH link towards a lot of correlative evidence that, as we know, can be interpreted in many ways.

    If an economic proposition is true, it's truth can be determined by deductive reasoning. I'm sure you'll agree. That's what I've been asking for, and that's what I've never gotten. On the other hand, I have gotten a pretty good deductive argument against liquidity preference theory - so you'll have to excuse me when I don't join you all because the majority of financial analysts happen to agree with you.

    As people's preferences shift towards liquidity, borrowers will have to pay more to borrow long term.

    Why, Gene? One person borrowing money implies that another has saved it - not that another is currently holding it in their cash balances. Please substantiate your claim further.

    ReplyDelete
  20. re: "Clearly we don't agree. I'm not in any confusion over whether we agree or disagree on the idea of liquidity preference."

    I meant "accept" in the sense of "be content with", not "resolve your confusion over". I know you're not confused about it. Be content with it. I'll note from time to time with you guys how a point changes when you accept liquidity preference theory, but I don't go around asking you to justify your position every time you state it. That seems futile to me.

    re: "You're suggesting that I "balk" because I don't like the implications. How sophomoric."

    I think you're misusing the word "sophomoric" here.

    re: "I would hope you consider me intellectually honest enough to not place my own opinions on how I should like things to work as a determinant of the theories I hold"

    Well it's not a matter of your opinion. You have worked out a theory of the macroeconomy that you find convincing from an entirely objective standpoint. It makes sense that you're loathe to integrate points that would make that theory work differently. You've made a reasoned judgement about parsimony and explanatory power with your macro theory. It's not a matter of your opinion.

    re: "(just as I've never argued that you "balk" at natural rights theories because you "like" governments)."

    Well I do like governments, actually. I like that we've evolved institutions to collectively deal with certain problems. I like that. Government is a good thing. But I'm not sure how any of that would change with natural rights. Anyway - I just wanted to clarify that. Just because government can be or do bad doesn't mean I don't like it.

    re: "I've never been shown any deductive reason why cash balances ought to affect the interest rate. The most you've ever done is give me a GOOGLE SEARCH link towards a lot of correlative evidence that, as we know, can be interpreted in many ways."

    Now you are being dishonest. My arguments aren't always deductive, of course, but I've provided far more than a google search link. That was a way of dismissing a question that we've gone through many, many, many times. You haven't convinced me either. That doesn't give me license of accusing you of not providing an argument. You made and cited lots of attempts to convince me. I recognize that.

    re: "If an economic proposition is true, it's truth can be determined by deductive reasoning."

    Prove that claim deductively (and therein lies the downfall of foundationalism).

    re: "I'm sure you'll agree."

    Really??? Of all statements you expect me to agree with that one???

    re: "On the other hand, I have gotten a pretty good deductive argument against liquidity preference theory - so you'll have to excuse me when I don't join you all because the majority of financial analysts happen to agree with you"

    The Hoppe argument? The Hazlitt argument? Which one? I've gone over each of these with you and pointed out that it's not a good argument. Don't act like I have dodged this discussion, Mattheus. I have run out of arguments to convince you and you have run out of arguments to convince me. We appear to be at an impasse. Don't act like you've scored a resounding win and I have stayed silent.

    ReplyDelete
  21. Mattheus, I'd like you to explain to me why I get a higher interest rate on my twelve month CD than I do on money I hold in my savings account for the same twelve months.

    ReplyDelete
  22. I said I'm sure you'll agree because you agreed with me on the Kant post awhile back about a priori and a posteriori truths.

    If we've run out of arguments, then we've run out of arguments. It just doesn't seem like you've offered much in any attempt to convince me. But I'll shut up about it from now on, if thats what you want.

    I can't explain well why that's the case. It most likely has to do with the fact that savings account are becoming more and more like checking accounts, and the line between purely time deposits and purely demand deposits is blurred. But I'm not an expert in finance. Go ask Bob Murphy, he handles the FRB at the Mises Institute haha

    ReplyDelete
  23. Prove that claim deductively (and therein lies the downfall of foundationalism).

    Because the contrary is contradictory. There exist intersubjective concepts that positivism denies, and in so doing, denies its own ability to determine truth. The preconditions for experience cannot be determined by experience, but by logical inference - a la Kant. I don't have to "prove" my claim deductively (after which you would ask for another deductive proof for THAT argument, and so on ad infinitum), I just have to illustrate that deduction is at the heart of all knowledge, and empirical observation relies on intersubjectively established logical categories.

    ReplyDelete

All anonymous comments will be deleted. Consistent pseudonyms are fine.