Ben Bernanke: Falling NGDP, not banking crises, caused the Great Depression

At least that’s how I read the passage sent me by Declan Trott, from Bernanke’s seminal 1983 paper on the impact of the banking crisis:

Did the financial crisis of the 1930s turn the United States into a “temporarily underdeveloped economy” (to use Bob Hall’s felicitous phrase)?  Although this possibility is intriguing, the answer to the question is probably no.  While many businesses did suffer drains of working capital and investment funds, most larger corporations entered the decade with sufficient cash and liquid reserves to finance operations and any desired expansion (see, for example, Friedrich Lutz, 1945).  Unless it is believed that the outputs of large and of small businesses are not potentially substitutes, the aggregate supply effect must be regarded as not of great quantitative importance.

The reluctance of even cash rich corporations to expand production during the depression suggests that consideration of the aggregate demand channel for credit market effects on output may be more fruitful.”

Many people read this paper as suggesting that the banking crisis was an important problem, even apart from the decline in AD.  And that’s probably how Bernanke intended it.  (You don’t get published in the AER by agreeing with previous studies.)  But in the end Bernanke has to acknowledge that the major contractionary effect worked through falling NGDP.  If the Fed doesn’t allow NGDP to fall in half, there is no Great Depression.

And if the Fed doesn’t allow NGDP to crash in 2008-09, and stay low in 2010-11?

PS.  It’s quite possible that in1929-33 the Fed wouldn’t have been able to prevent the crash in NGDP without leaving the gold standard—which it had no authority to do.  I have mixed feelings on that debate, but of course it is of no relevance to the current situation.



24 Responses to “Ben Bernanke: Falling NGDP, not banking crises, caused the Great Depression”

  1. Gravatar of marcus nunes marcus nunes
    15. July 2011 at 14:13

    Yes, over the years, as a researcher Bernanke has made “all the right sounds”. He still does, but they are almost inaudible and his actions contradict them!
    He´ll go down in history as the “man who knew it all but just couldn´t make it happen”.

  2. Gravatar of John John
    15. July 2011 at 15:54

    I think Bernanke is simply realizing that believing you can create prosperity with a printing press is the ultimate money illusion. It’s similar to how Roosevelt’s cabinet (Tugwell and Morgenthau) had realized the futility of government spending by the end of the 1930s. By the way, the idea that WWII pulled the US out of the Depression is the ultimate economic myth.

  3. Gravatar of Marcelo Marcelo
    15. July 2011 at 16:23

    I still have trouble understanding how a run on banks could not lead to a recession/depression. While there was no official nationwide bank run due to the government stepping in and shoring the banks, I still wonder if that couldn’t have played a role in the Great Depression. WHen banks were closing and calling in loans on businesses, I can see the relevance of a liquidity trap, where no amount of money is really going to affect the amount of goods created because no one has confidence in the market. Perhaps, after the bank run it would be possible for the Fed to shore up confidence with NGDP raises, but I suppose I just haven’t seen a case where any central bank has used NGDP targets to affect such a depression. I suppose that I’m wondering if the liquidity trap can really hold if there is no faith in future market performance/ consumer confidence…

  4. Gravatar of John Thacker John Thacker
    15. July 2011 at 16:48

    While there was no official nationwide bank run due to the government stepping in and shoring the banks, I still wonder if that couldn’t have played a role in the Great Depression.

    Well, isn’t it useful to use Canada as a comparison? Canada had not a single bank failure during the Great Depression (possibly thanks to widespread branch banking, and possibly other weird factors like free banking and so forth, not trying to explain why), but its slump was in percentage terms worse than that of the US.

  5. Gravatar of Scott Sumner Scott Sumner
    15. July 2011 at 17:51

    Marcus, That’s right.

    John, It’s much worse, he thinks we are prosperous, at least in nominal terms. He doesn’t think the economy needs any more stimulus.

    Marcelo, Between March and July 1933 much of the US banking system was shutdown, due to a severe banking crisis, and yet industrial production rose 57%, the fastest rate in history. How did it happen? Expansionary monetary policy.

    My second response is that if the Fed hadn’t let NGDP collapse in 1929-33, there would have been no banking crisis anyway.

    John Thacker, Good point, and Bernanke discusses that in the paper.

  6. Gravatar of Jim Glass Jim Glass
    15. July 2011 at 20:59

    Beckworth: “It’s amazing to see Bernanke list all the things the Fed could do to help the economy, but choose not to act.”

  7. Gravatar of Bogdan Bogdan
    16. July 2011 at 03:25

    But with regard to the persistent high unemployment throughtout the 1930s, the price controls, the forced unionisation and cartelization, the anti-business rethoric, legal and administrative decisions etc must have played a very significant role also.

  8. Gravatar of Bill Woolsey Bill Woolsey
    16. July 2011 at 05:26


    If the economy starts in monetary equilibrium, then “printing money” might create what might be characterized as “prosperity” for a short time, but nothing but higher money prices and incomes in the long run.

    However, if the economy starts off with a quantity of money too low, or, in other words, prices and nominal incomes too high for long run equilibrium, then “printing money” just returns the economy to equilibrium without the need to deflate all prices and wages.

    Unless there is some reason why deflating prices and wages is desirable (maintaining a traditional gold parity, transering real wealth to creditors, or something,) printing money is equilibrating.

    Further, because moving to an disequilibrium with a shortage of money (prices and wages have yet to fall enough) to one with sufficient money allows a recovery of output, employment, and real income. It creates prosperity.

    The only way to avoid this logic is to assume that prices and wages are always perfectly flexible. The current level, then, is always the level where the real quantity of money is equal to the demand to hold money.

  9. Gravatar of William J McKibbin William J McKibbin
    16. July 2011 at 05:43

    Dr Bernanke leads the Federal Reserve, which is filled with bankers who are devoted to near zero inflation irregardless of the costs in terms of growth and employment — said another way, the Federal Reserve is sitting quite happy right now with inflation held at the lowest levels in decades — no one at the Federal Reserve is seriously committed to creating growth or employment — that’s how bankers think — for the Federal Reserve, it’s all about holding inflation to near zero levels — anything else the Federal Reserve spouts is propaganda…

  10. Gravatar of William J McKibbin William J McKibbin
    16. July 2011 at 05:47

    Let me add that QEII was not really “printing money” — the reality is that money was used to remove treasury paper from the street — however, the total amount of US paper out there still balances out to the same number — yes, there’s more money out there, but yes, there’s less treasury paper out there to offset that trade — the Federal Reserve has yet to truly start “printing money” (as in helicopter distributions of cash to spenders on Main Street) — monetary fiscal expansion is not the plan of the Federal Reserve (at least so far)…

  11. Gravatar of William J McKibbin William J McKibbin
    16. July 2011 at 05:52

    One last comment, anyone interested in knowing what the Federal Reserve is contemplating for small businesses in America will enjoy this video:

    Federalism is more that happy to wipe out small business in America — let’s face it, Walmart and MacDonald’s are higher priorities for the Federalists sitting the Federal Reserve Bank and US Capitol at this point…

  12. Gravatar of Scott Sumner Scott Sumner
    16. July 2011 at 08:45

    Jim Glass, Yes, he listed those before.

    Bogdan, That’s right, the statist policies back then were far worse than today.

    Bill, Good explanation. William, Yes, bankers have too much influence. But their policy doesn’t even help the banks, it hurts them.

  13. Gravatar of Morgan Warstler Morgan Warstler
    16. July 2011 at 14:53


    “William, Yes, bankers have too much influence. But their policy doesn’t even help the banks, it hurts them.”

    Makes you sound like a fool.

    There is no long term for bankers, for the right things to happen, they have to be INSOLVENT.

    So to the bankers, anything is better than insolvency.

  14. Gravatar of flow5 flow5
    17. July 2011 at 10:08

    It is indisputable that the banking crisis was principle cause. The loss of faith in the private commercial banks had become so pervasive by the end of 1932, banks were being forced to liquidate by the thousands. People everywhere were attempting to convert their demand and time deposits into currency. Thousands of towns and cities throughout the country were attempting to finance their daily commerce without a single operating bank. And by March, 1933, just before Roosevelt’s “banking holiday” there were even entire states without a single operating bank.

  15. Gravatar of david glasner david glasner
    17. July 2011 at 13:18

    Until early 1933 the US dollar was convertible into gold at the fixed parity of $20.86 per ounce. The US price level was determined not by the quantity of money, but by the value of gold and the fixed parity. The value of gold was determined in an international market, so it is simply fatuous to believe that banking conditions in the United States had any significant effect on the US price level except insofar as they affected the overall global demand for gold. It was the overall global demand for gold in relation to stock of gold that was determining the rate of deflation for all gold standard countries including the US. So Bernanke’s entire discussion of the of the banking crisis on overall output relates to nothing more than a minor perturbation in the decline in output determined by the internationally determined rate of deflation. The same observation applies to explanation of the Great Depression advanced by Friedman and Schwartz, as is clearly implied by the case of Canada as John Thacker correctly recognized.

  16. Gravatar of Scott Sumner Scott Sumner
    17. July 2011 at 16:20

    Morgan, You seem to relish the thought of lots and lots of people suffering. Shouldn’t we be making life better for everyone?

    flow5, You said;

    “It is indisputable that the banking crisis was principle cause. The loss of faith in the private commercial banks had become so pervasive by the end of 1932, banks were being forced to liquidate by the thousands. People everywhere were attempting to convert their demand and time deposits into currency. Thousands of towns and cities throughout the country were attempting to finance their daily commerce without a single operating bank. And by March, 1933, just before Roosevelt’s “banking holiday” there were even entire states without a single operating bank.”

    You’ve got it exactly backward. The mega-crisis you refer to occurred in February-March 1933, AFTER THE GREAT CONTRACTION WAS OVER. Over the next 4 months industrial production rose at the fastest rate in American history, despite much of the banking system being shutdown.

    David, I mostly agree, except the banking crises did increase the demand for US currency, which did create a derived demand for gold to back that currency. But you are certainly right that gold hoarding was the main issue, and bank difficulties were a side issue, and mostly a symptom of deflation.

  17. Gravatar of Morgan Warstler Morgan Warstler
    18. July 2011 at 04:19

    No Scott, we shouldn’t. Monetary policy makes things best for the most people when it allows the game to be played fairly.

    And the game is not “democracy,” the game is “free markets.”

    That’s why using monetary to undermine DeKrugman democracy “vote to get free shit” – like DeKrugman wants with mortgage relief, is GREAT.

    It is also why, using QE to prop up the banks rather than force them to lose all the hard assets and be insolvent, like real Austrians want, is HORRIBLE.


    As a utilitarian, you have to admit there is a date – a length of time at which, it WOULD HAVE BEEN BETTER to do it my way.

    How long do we have to screw through all of this before you eventually just wish policy would have been to liquidate the housing market?

  18. Gravatar of flow5 flow5
    18. July 2011 at 08:12

    “You’ve got it exactly backward” Right, the legislation that followed had nothing to do with correcting bank failures.

  19. Gravatar of flow5 flow5
    18. July 2011 at 08:34

    Before the New Deal, widespread bank failures were epidemic even in so called prosperous times In the 1921-29 period 5,411 banks failed; 8,812 in the 1930-33 period; but only 336 insured banks failed in the 28 year span from 1934-1962.

    In the first 20 years under the Federal Reserve Act of 1913, there were over 20,000 bank failures. The intention of the framers of the Act was to establish a unified banking system under 12 central banks. There were many flaws in the original Act, one being the establishment of 12 rather than one central bank

    The fatal flaw was not making membership in the System compulsory for all money creating institutions. And had not Franklin Roosevelt declared a “banking holiday” in March 1933, the lack of confidence in the banking system would have resulted in the failure of virtually every bank in the United States.

    We now actually have a central bank. It is called the Federal Reserve Bank of New York. An amendment to the Federal Reserve Act in 1933 established The Federal Open-Market Committee and gave it the power to control Total Reserve Bank Credit. The Fed can now buy an unlimited volume of earning assets. (With the federal debt at over 12 trillion, and expanding, and billions of dollars of “eligible paper” available, the term “unlimited” is not an exaggeration in terms of any potential needs of the Fed.) In the process of buying Treasury Bills etc., new Inter-Bank Demand Deposits (IBDDs) are created. These deposits can be cashed by the banks into Federal Reserve Notes, without limit, on a dollar-to-dollar basis.

    Today, the public, seeking to cash their deposits, would soon have a surfeit of paper money. A general run on the banks is impossibility. Where the Federal Deposit Insurance Corporation cannot handle the situation (Continental Illinois, for example), the Fed will guarantee the liquidity of the bank’s deposits.

    In other words, a liquidity crisis leading to the wholesale failure of commercial banks is impossible. Where banks are allowed to fail, or are absorbed into solvent banks, customers never suffer losses if their deposit does not exceed $250,000. The fed intervened in the Continental case because many corporations, foreign and domestic, had deposits far in excess of $250,000. These institutional changes plus the numerous “safety nets” now provided business and consumers preclude a recurrence of a “Great Depression”.

    In the period from 1929 – 1932 stocks were spiraling down, unemployment was becoming endemic, businesses were failing in increasing numbers, BANK FAILURES WERE ACCELERATING, and millions of people were suffering severe malnutrition, there was not a single piece of legislation passed by Congress or action taken by the administration which had any significant effect in stemming the tide of economic disaster.

    In retrospect, the answers to the depression seem simple. We needed a central bank that could and would pump IBDDs into the commercial banks in a volume SUFFICIENT TO SATISFY THE PUBLIC’S DEMAND FOR CURRENCY, specifically paper money (currency is an asset the Fed does not, should not, and cannot control). In the control of the monetary aggregates, the monetary authorities are completely dependent on their power to control the volume of bank credit. They have no power over the volume of the Treasury’s General Fund Account or the currency holdings of the public.

    It was not until 1933 that we began to unshackle our paper money from the numerous and unnecessary restrictions pertaining to its issuance. With the numerous types of paper money in circulation at the time, this would seem to have been a non-problem. Here is the list: gold certificates, silver certificates, national bank notes, United States notes, Treasury notes of 1890, Federal Reserve Bank notes, and Federal Reserve notes. With that array of paper money there should have been plenty to meet the liquidity demands placed on the banks by the public. But the volume of each type that could be issued was so circumscribed by restrictions that even the aggregate group could not begin to meet the panic demands of the public.

    Today we have only the Federal Reserve Note, and there is only one restriction placed upon its issuance. No Federal Reserve Note can be put into circulation unless there is a prior transaction involving the relinquishing by the public of an equal volume of bank deposits, and an equal diminution of the holdings of IBDDs on deposit with the Federal Reserve Banks; In other words, the issuance of our paper money contains no inflationary bias. Its issuance does not increase the volume of money. It merely substitutes one form of money for another form

    In 1933 the Federal Reserve Note had to be collateralized by at least 40 percent in gold bullion or coin, and the remaining collateral had to consist of eligible comer paper, principally Trade and Banker’s Acceptances. The problem was the banks had practically no eligible collateral.

    The first tentative step was to reduce the gold requirement to 25 percent and allow U.S. government obligations to provide the remaining collateral. The framers of the Federal Reserve Act did not believe that the credit of the U.S. government was inferior to that of the Federal Reserve Banks and the short term commercial paper of business; they merely believed that the volume of paper money should rise and fall with the level of business activity. They also had the naïve belief that this country was so big, so diverse in its commercial needs, that it needed twelve central banks.

    Had the present Federal Reserve System been in place at the beginning of the Great Depression, there would have been no Great Depression. We were not reduced to practically a barter economy because the banks were insolvent; we needed that condition because perfectly sound banks could not meet the liquidity tests imposed upon them by a panic-stricken public.

    One of the preconditions the U.S. needed in 1929 was a much larger national debt, and a willingness on the part of the Congress, the Administration, and the business community to tolerate an adequate expansion of the national debt. In 1929 the national debt was less than $17 billion, and the banks held only a small proportion of that amount. We needed a larger debt and a much more rapidly expanding debt in the 1930’s, not only to “prime-the pump”, but to meet the monetary management needs of the Fed. Note: Both Roosevelt and Hoover in 1932 ran on platforms calling for balanced budgets.

    The open market operations of the Fed require a depth of market that will enable the Fed to buy or sell billions of dollars worth of treasury bills on any given day without deeply disturbing the bill rates. Another of the many lessons from the Great Depression was the realization that if a financial panic is allowed to reach crisis proportions, monetary policy becomes useless, totally ineffective.

    For all of the Great Depression legal reserve management was impossible even though the Banking Act of 1933 provided for the coordination of all open market operations through the New York Reserve bank. (that is to say, before 1933 one FRB could be conducting operations of the buying type — expanding credit, creating bank free-gratis legal reserves and laying the foundation for a multiple expansion of money, while another FRB was doing the opposite, — conducting open market operations of the selling type) Before April 1933 any excess free-gratis legal reserves in the system were quickly wiped out by the massive “runs” on the banks. Between mid-September and the end of December 1931, currency held by the public rose $544 million (11 percent).

    But even after bank failures were brought under control business confidence remained so traumatized the expansion of free legal reserves remained to a large extent free excess reserves. There were not enough credit worthy borrowers in the private sector (according to the bankers), and in the public sector there was an insufficient volume of government debt to absorb excess bank lending capacity. From 1933-1942 the centralization of the open market power was of little consequence. It was not until about 1942 that the member banks operated with no excessive amount of free excess legal reserves. The FOMC couldn’t control the money supply.

    After 1933, after we had a central bank and a coordinated Fed credit policy, the Fed pumped billions of dollars of costless legal reserves into the banks; and nothing happened. There were years during this period when the free excess legal reserves held by the member banks were larger than the volume of required reserves. The exercise of Fed policy was likened “to pushing on a string”.

  20. Gravatar of flow5 flow5
    18. July 2011 at 08:35

    Estimates by the Department of Commerce put the net debt figure as of the end of 1939 at $183.2 billion compared with a figure of $190.9 billion as of the end of 1929. I.e., for the period encompassing the Great Depression there was no over all debt expansion.

  21. Gravatar of Scott Sumner Scott Sumner
    18. July 2011 at 08:39

    flow5, I agree with much of your criticism of Hoover’s polices, and indeed have written an entire book (unpublished) on the subject. But the only good thing FDR did was devaluing the dollar. The fastest growth in American history occurred when much of the banking system was shutdown The 1920s were another prosperous period–full of bank failures as you correctly noted.

  22. Gravatar of Scott Sumner Scott Sumner
    18. July 2011 at 08:40

    Morgan, Economic policies that go after certain groups don’t usually end well.

  23. Gravatar of david glasner david glasner
    18. July 2011 at 19:20

    Scott, And I, of course, agree that banking failures insofar as they led to an increased demand for gold intensified deflationary pressures, but deflationary pressure was distributed more or less uniformly across all gold standard countries, not confined to the US, as the Friedman-Schwartz account would lead one to believe. I just wanted to spell out the details; I knew that you understand the analysis at least as well as I do.

  24. Gravatar of The new monetarism: Fed failures | PARTISANS The new monetarism: Fed failures | PARTISANS
    30. January 2012 at 16:02

    […] Scott Sumner: Falling NGDP caused the Great Depression […]

Leave a Reply