This archived content originally appeared at, the predecessor site to, and does not carry the sponsorship of the Cato Institute.

Keynes to FDR: Forget Quantitative Easing

From Keynes' "Open Letter to President Roosevelt," published December 16, 1933:

"Rising output and rising incomes will suffer a set-back sooner or later if the quantity of money is rigidly fixed. Some people seem to infer from this that output and income can be raised by increasing the quantity of money. But this is like trying to get fat by buying a larger belt. In the United States to-day your belt is plenty big enough for your belly. It is a most misleading thing to stress the quantity of money, which is only a limiting factor, rather than the volume of expenditure, which is the operative factor."

This near the very bottom of the Great Depression. Perhaps Keynes was wrong then. But is there not a strong case to be made, nevertheless, that the recent rounds of QE were, what with all that heaping-up of excess reserves, just so much unhelpful belt-loosening?

What say ye, my Market Monetarist friends?


  1. J.M, Keynes does not mention the actual causes of he mess the United States was in.

    He does not mention the expansion of credit-money (led by Benjamin Strong of the New York Federal Reserve) in the late 1920s – that led to the boom-bust.

    Nor does J.M. Keynes mention the terrible reaction to the (inevitable) crash of 1929 by the Hoover Administration – where (the reverse of the policies of the Harding Administration in 1921 – the response to the crash of the World War One credit-money bubble) they increased (rather than cut) government spending and taxation, and (worst of all) actively intervened to PREVENT prices and wages (a wage is a price – the price of employing someone) adjusting to the monetary bust.

    In 1933 it was vital that wages be allowed to adjust to the credit-money bust – but we do not get anything on that, instead we get waffle about "expenditure".

    More generally the source of all great bubble economies (whether government or commercial banking generated) is the fallacy that by clever manipulations one can (without terrible consequences) lend out more money than was REALLY SAVED (make lending bigger than REAL SAVINGS) – any effort to do this leads to terrible consequences.

    1. Typical Austrian thinking.

      Suppose the economy is working at well below capacity. And suppose private banks create new money from thin air and lend it out. Or suppose the state prints more base money and dishes it out or spends it into the economy. In neither case has any corresponding “real savings” taken place (shock horror). And suppose that brings about full employment, without excess inflation. Where exactly are the “terrible consequences”?

      Of course I’m not suggesting it’s easy to get the above sort of solution to recessions exactly right: e.g. the money printing may inadvertently go too far. But IN PRINCIPLE, creating new money which is unbacked by any saving can be beneficial.

        1. No it is not Phillippe – however the fact that you think that a basic principle of reason (that one should not try and lend out more "money" than was really saved – i.e. that lending should not be greater than the actual sacrifice of consumption)is "meaningless", shows a lot of what is wrong with modern "economics".

        2. Would realizing that bank-lending would have zero effect on anyone's savings make a difference there, same for Guv chopper-drops of $20 Bills.
          Or, was there something about that capped REALLY SAVED?

          1. Yes there was Joe – if lending is financed from monetary expansion (rather than real savings) then there terrible long term economic effects – the basic capital structure is distorted.

          2. Gawd,
            Do I really need to reply to another Paul Marks inanity?
            We want ALL borrowing to be done with real money savings, you &^%$**&#$ .
            The government spending money into existence on its budget is in a precursor loop (creation-issuance-seigniorage) to its savings designation and entering the credit loop.
            Only Lostrians think we can have money from something other than the law that establishes its use.

  2. Intriguingly, I came across a new paper today suggesting Keynes was quite market monetarist ( taking pretty much the opposite line implied in the letter there.

    1. There is a good amount of evidence that recent QE is not pushing on a string. Look at the Yen or Nikkei when BOJ's Kuroda announced latest easing. Similarly, there are lots of recent papers suggesting that QE is effective even now at the ZLB. (e.g.,,,,,,,,

    2. We should expect banks to hold portfolios based on their preferences ( Thus, if banks hold lots of liquid low-risk assets (i.e. money) we should expect this is because of their preferences, expectations, judgements and so on. These depend on real factors. If QE changes real factors they may change (e.g. if QE works then they will want to hold less in excess reserves). But simply printing more money would not be expected to change the portfolio they want to hold. It would be very strange if we didn't see assets going where incentives directed them, but we in fact do (e.g. see recent papers & David Beckworth has made this case persuasively before (e.g.

    I apologise for the link-heavy post but it seems relevant given the empirical nature of the question 'does QE work'.

    1. Thanks for the links, Ben. I'd be very curious indeed to know what Prof. Sutch makes of the very plain language of Keynes' open letter to FDR, or how he would have Keynes concoct an argument to the effect that QE, though not warranted in December '33, was warranted, not only when it began in 2009, but years (and several trillion dollars) later!

      Your second point seems to me to beg the question: if banks' "preferences" were such, as they evidently were, as caused just about every $ of Q-Easing to pile up in their reserve accounts, then was that not a reason for concluding that the fundamental problem was not a lack of reserve money, but something else–like IOR, say; or the huge sword that Obama and Co. first held by a horse's hair over the banking system's heart–and later loosed on it–known as the Dodd-Frank Act?

      While bank lending, the real key to recovery, was thus kept prostrate, many MM's were busy cheering and shaking pom-poms as the QE parade marched by, and might have kept it marching still longer if they could. And to what purpose? What grounds is there for attributing such meager recovery as there has been, and so long delayed, to all that easing, when mere adjustment of price-expectations might have been expected to accomplish nothing less, and in fact has, in the wake of past downturns, accomplished a lot more? If you ask me, it achieved no good purpose at all. It did, on the other hand, supply welcome intellectual cover, albeit perhaps unwittingly, for what was really a vast bailout operation that merely masqueraded as monetary policy.

      1. I disagree on all counts. But let me focus on one: certainly I agree that QE has not 'gone into' bank loans, but this is part of the market monetarist platform, because 'we' don't think there is or has ever been a lending channel to QE, and indeed 'we' think it would be a bug, not a feature, of QE, if there was a lending channel. We shouldn't have any preference over what kind of finance firms use. We know that QE raises equities and indeed all assets. We have quite a lot of evidence that stock markets are not 'side shows' (see e.g. and that they provide funds for investment.

        But even here that's not the point. All we need for recovery from a recession caused by falling below the NGDP trend is time or inflation that brings us back to trend. We know QE can cause inflation. It's deeply implausible to suggest it can't if permanent & tied to a credible target. Credit is irrelevant.

        1. Well, if the lack of any substantial, discernible response of NGDP (or inflation–which you seem here to confuse with same) to trillions of dollars-worth of QE, however "implausible" that lack may seem, doesn't suffice to convince you that, in this particular case, things haven't quite worked in accordance with theory, I wonder whether anything could.

          I quite agree with you, on the other hand, that "time" itself ought to be capable of bringing an economy back to trend even in the absence of QE. To me, however, in light of the very lackluster nature of a recovery that's now approaching its 5th year, that's all the more reason for holding that QE itself didn't accomplish what MMs hoped it would.

          1. You might disagree with the empirical work but there is a good amount of evidence that monetary shocks like QE boost NGDP. I use NGDP and inflation interchangeably in some contexts because I think it's OK for those purposes.

          2. Ben, I don't dispute at all the possibility that QE is capable in many instances of boosting NGDP. The question is whether it did so to any great extent in the US after 2009. Arguments concerning the theoretical potential, or empirical evidence from Japan and other countries (or times), simply don't address the issue. I have seen some careful econometric studies affirming positive QE results, but these refer only to QE1, and not even to the full period for that. If there's a study addressing the full QE episode, or most of it, and properly controlling for other possible causes of such recovery as occurred, I'd be very happy to have it pointed out to me. In the meantime, I plan to remain a skeptic.

    2. Ben,

      I don't have my comments on Fama's paper with me, but from what I remember it didn't strike me as exceptional, and rather out of touch with reality (I'll take a look again at my notes tonight though as my memory is perhaps bad).

      In today's world in particular, banks' preferences are massively influenced by regulation, which dictate both capital, liquidity and funding requirements. Banks' own preferences are as a result quite meaningless in such a world (not entirely, as they still have a limited range of options available). By directing loanable funds to asset classes favoured by regulation, prices/yields of a whole bunch of securities end up distorted (as you know I have argued for a while on my own blog).

      George is right to blame Dodd-Frank (though I'd rather blame Basel), and MM should pay a lot more attention to what's happening in the banking system.

      I also don't believe that booming stock markets are necessarily a sign of investor confidence in economic recovery. It can simply be a sign of investors expecting further trading gains and hoping to go out of their positions in time. This has happened countless times in history. And it's the job of hedge funds.

      1. Couldn't agree more, JN: and you are right to point a finger at Basel, as I neglected to do.

      2. The Fama paper is a peerless framework for looking at what banks are and how they work. They are not special, they are simply firms that provide payments services and there happen to be synergies between those and holding certain types of portfolios. Your points about regulations fit neatly into the framework.

        1. I don't disagree.
          Just saying that those 'preferences' have been far from free market ones for a few decades (at least). Hence banks' portfolio allocation may not be an accurate measure of what's going on (and real factors may not have that much of an influence).

          (still have to check my notes though!)

    1. Was there a liquidity trap in Dec. 1933? I doubt it (Chris Hanes has a good paper on that). Was there one after 2008? Yep–though not necessarily for the reasons Keynes had in mind.

  3. Isn't "it is most misleading to stress the quantity of money, which is only a limiting factor, rather than the volume of expenditure, which is the operative factor" entirely consistent with market monetarism? (Assuming Keynes meant nominal expenditure).

    As for MM support of QE, I thought they favored it not because it created a lot of money, but because it did have some effect on NGDP. Scott Sumner has often argued that the Fed could have provided more stimulus -with a smaller balance sheet – by targeting NGDP or the price level – I don't think QE as it was done by the Fed was his preferred policy.

    As to whether QE was a back door bailout. Does it matter that they weren't just given reserves, they traded bonds for the reserves? I don't know – it does seem like a bailout to me because there were other alternatives for addressing a demand shortfall. But why are the band better off with all this cash than they would be if they just had bonds.

    1. Consistent with MM? Consistent in thinking spending matters, but not in thinking it can be achieved my means of monetary expansion. So far as I'm aware, MMs see monetary expansion as the best if not only means for combating a spending shortfall.

      MM support for QE: The question is precisely whether or not QE really succeeded in increasing NGDP. I don't think there's much evidence that it did. NGDP growth recovered, but I don't think QE had much to do with it, because of the previously-mentioned moribund state of bank lending.

      Bailout: the sellers of securities purchased during the various rounds of QE (not necessarily banks, these) benefited because they got more cash for those securities than they could ever have hoped to get otherwise. The banks that ultimately sat on the new reserve money were themselves not the prime bailout targets, though having lots of fresh reserves come your way, on which you can earn risk-free interest without any lending muss-or-fuss, ain't a bad deal either.

      1. MMs believe that the stance of monetary policy is a function of expected future policy. Insofar as QE signaled that the Fed would keep the monetary base higher/rates lower in the future, it was expansionary.

        Given equity/bond/currency/commodity market reactions to QE announcements, financial markets certainly seem to view QE as beneficial (Japan being the latest and one of the more dramatic examples). I would note that they haven't been proven wrong and retreated, except temporarily in times when Fed policy seemed to be moving towards tightness again.

        1. Of course certain "financial markets" benefited from QE, as I myself acknowledge above. But boosting various asset prices isn't the same as forwarding general recovery. One (admittedly simplistic) way to think of this is to consider that stock prices, for example, can rise either because of a decline in anticipated interest rates, and hence in the discount rates applied to arrive at present values, or in the un-discounted expected value of net nominal revenues. QE might do nothing to enhance expected nominal returns, yet still boost securities prices through its effect on expected interest rates.

          1. Wouldn't this be a one-time effect, though? Prices would adjust to higher level as the necessary discount on the NPV of future revenue is worked downwards and then they would return to whatever path they would follow based on their 'fundamentals'? That wouldn't, in itself, be able to drive a sustained rising path for asset prices, right?

            I think the mechanism for the Market Monetarists is that expansionary monetary policy acts as a salve for the worries of the financial sector: The central bank loudly and publicly commits to a stable growth path for some important aggregate (NGDP in this case) and this reassures lenders that they aren't going to be hit by suddenly adverse financial conditions (sudden hikes in the riskiness of given asset portfolios). This causes them to be more willing to lend and it is that which then drives economic recovery. It's perfectly consistent with the Keynes quote in your original post — although not necessarily consistent with everything Keynes has ever said, of course. It's not the volume of money that matters, it's how much money is being spent. The volume of money just has a loose connection with the level of expenditure through the expectations about future financial conditions on the part of the financial sector

            Banks prop up spending with their loan activity and the central bank props up bank lending with its set of policy tools and a firm, public commitment to use those to hit an NGDP target. Current implementations of QE have failed to stimulate the economy to satisfaction simply because the central bank has failed to adequately communicate a commitment to the necessary financial stability.

            This is my understanding, anyway. I'm not entirely on board with the whole thing. Maybe there are serious real factors that are outside the control of just one central bank. Who really knows?

  4. "Consistent with MM? Consistent in thinking spending matters, but not in thinking it can be achieved my means of monetary expansion. So far as I'm aware, MMs see monetary expansion as the best if not only means for combating a spending shortfall."

    I am probably out of my depth here (as well as guilty of looking through MM-colered glasses), but what I thought of when reading Keynes quote was Nick Rowe's "people of the concrete steppes". To me, at least, one of the key implications of MM is that there is no simple formula whereby printing $X in dollars will lead to $Y in nominal spending. As the effect of expanding the base by any particular amount on nominal spending is not really predictable, the correct monetary policy target should be spending rather than the quantity of money… Maybe mine is a naive and biased view of what Keynes was trying to say here, I don't claim any particular expertise.

    Is monetary expansion the best if not only means for combating a spending shortfall? I would say that expansionary monetary policy may be the best if not only means for doing so, at least under our current monetary regime. But I would not call monetary expansion (such as that which we have seen since 2008) the best if not only means for combating such a shortfall. The Fed could have gotten a lot more bang (in spending) for its printed buck if it had pursued a more expansionary policy.

    I can certainly buy the idea that the Fed wanted to expand the base more than it wanted to combat a spending shortfall. I'm not 100% convinced but as a theory it is more than plausible.

    1. The problem with the argument that, so long as spending appears deficient, more monetary expansion is called for, is precisely that it overlooks the possibility being emphasized here, namely, that conditions are such that banks simply aren't inclined to lend more, for whatever reason. In that case, some deeper problem must first be addressed before monetary expansion can serve any useful purpose; and then, if it is addressed, the expansion may prove unnecessary because spending revives without it.

      If I were a doctor and a patient of mine became undernourished owing to a tapeworm of which he is unaware, would I be judged competent were I to send him home with instructions to try eating more?

      1. No, but this begs the question of the viability of a debt-based money system, and it being capable of providing our money system needs in a balance sheet recession.

        Banks will not lend because the economy doesn't have enough money to pay back loans….. prudent they be in a debt-saturated, but money-starved national economy.

        If you read Martin Wolf's book, you;ll see this is why he has called for the end of relying on banks to provide the money, just as did Lord Adair Turner over a year ago.
        We need money, without debt.
        That ain't the bankers' business.
        It's the government's.

        1. No we do not need more money – the idea that more money means more wealth is a fallacy (one that economists refuted centuries ago).

          1. Well,
            Perhaps Paul Marks, and a cadre of money-hoarders, need no more money.
            But the national economy does.
            The ravenous unemployed, the low-wage earners and average debt-laden consumers all need more money, so that more goods and services might be produced and consumed.
            Given that elsewhere you raise Fisher's money and exchange equations were originally penned in the 1930s mostly, how could they possibly have been refuted centuries ago, except by Marks-ist logic of course?

        2. No we do not need additional currency – the idea that additional currency equals additional wealth is a fallacy (one that was refuted by thinkers such as J.B. Say centuries ago). As for the Mr Wolfe – he works for the "F.T." an infamous establishment organ.

          1. The priority use of additional currency (money in circulation) would usually mark additional incomes and subsequent increases in economic throughput, and that throughput usually, takes 'wealth' forms, for the most part.
            I would SAY that those who are waiting for the 'markets' to lead to more production to lead more consumption seem to out in the cold.
            We are stuck in a balance sheet recession, where only first becoming un-leveraged can lead to any increase in personal consumption, and therefrom, more production.
            Got anything that's not a few centuries old?

  5. Keynes contradicts himself. He says “Rising output and rising incomes will suffer a set-back sooner or later if the quantity of money is rigidly fixed.” Then he says “Some people seem to infer from this that output and income can be raised by increasing the quantity of money.”

    Well if you’re in a situation where the economy has been “set back” by an inadequate quantity of money, then it follows that “output and income can be raised by increasing the quantity of money”. Or have I missed something?

    I’ve looked at the surrounding text in Keynes’s letter, and he does not clear up the above self-contradiction, so I’ll do it for him.

    It’s pretty obvious that output cannot be increased without limit by simply printing money. On the other hand if the economy is at well below capacity, then unused resources can be brought into use by printing.

    I’m sure Keynes, looking down from Heaven, will approve of my clarification of his words.

    1. Hi Ralph. I wouldn't be so sure about Keynes approving. His liquidity trap argument, first of all, suggests that there are circumstances in which AD may be insufficient, yet monetary expansion won't help. And though Keynes' own theory regarding the necessary or sufficient conditions for such a trap may not be valid, it doesn't follow that liquidity traps can never arise for other reasons.

    2. Ralph,
      I believe he is simply expressing the fact that "you can pull on a string, but you can't push on it." In an expansion/boom, tight money (which implies rising interest rates) can put the brakes on expenditures; in a recession, with already low interest rates, more money injected by the central bank won't induce increased spending by firms and households.

  6. The votes in many States yesterday to have higher "Minimum Wage Laws" – is classic Hoover (as well as classic Roosevelt) thinking. The fallacy of "demand" of thinking that more "spending" means more jobs.

    In reality (of course) such government interventions mean that unemployment will be higher (not lower) than would otherwise be the case.

  7. Those who claim that the spending of World War II got rid of unemployment are mistaken – it was actually the fall in REAL wages (actual prices being "Black Market" prices for goods and services – not government prices).

    1. Not so, Paul: see Bob Higg's careful analysis, in his paper "Wartime Prosperity?" which argues that employment–that is, employment not counting combatants and those employed by the defense industry, didn't recover until after WWII, when price controls were lifted.

      1. Correct George. However my point was that even the "full employment" of the latter part of World War II (in military service and so on) could not have been achieved without cuts in real wages (disguised cuts in real wages – but cuts in real wages just the same).

        What the late 1940s witnessed was rising real (not notional) living standards – due to an end to many government regulations and a massive cut in taxes and government spending.

        Of course a similar policy was followed by Warren Harding in 1921 (although there had to be a period when wages fell – when the markets were allowed to clear, living standards could rise) it is the exact opposite (in almost every respect) to the policy followed by Herbert "The Forgotten Progressive" Hoover in 1929.

        Cut taxes (not increase them), cut government spending (do not increase it) and (above all) deregulate and allow markets to clear – even though this may mean a reduction in wages for a period (as it did in 1921).

  8. I think Keynes was a nut, so I don't care what he wrote.

    People who think Keynes was a great thinker should respect his own sense of consistency. He said that he changes his mind when the facts change, and the fact are always changing. Whatever he thought in 1933, he might not think that now.

    Fuzzy inputs, fuzzy thought, fuzzy outputs, confusion. That is JMKeynes.

    1. Agreed Andrew.

      And as for the idea that economic law is relative (depending on empirical "data" or whatever) this was refuted by Carl Menger in his debate with the German "Historical School" of the late 19th century.

      The final shot being fired in the shape of "The Errors of Historicism" (establishing the universal nature of economic law) published in 1883 – ironically the year J.M. Keynes was born.

  9. If a "monetarist" is anything it is someone who believes that the money supply should be increased in line with economic growth so that the "price level" does not go down (or up) – it is the Irving Fisher 1920s fallacy.

    "Market monetarists" do not seem to fit with this description at all – not that they are better than monetarists (far from it), they are actually worse (much worse).

  10. Off-topic, but I can't help but ask: will the Cato Monetary Conference presentations be available online? I'd very much like to see your talk on the monetary role of gold.

  11. Joe B. – your own principles are also "centuries old" and your principle (that more money means more real wealth) is not only a centuries old idea, it is also totally wrong.

    The "Monetary Cranks" (the people who claimed that government producing more money would improve long term economic prosperity) were a favourite target for refutation by real economists.

    1. LOL.
      Only my 'money is law' principle is centuries old.
      Everything else I know from Soddy to Fisher and Friedman, so, in a lot of peoples lifetimes or a generation hence.
      Most important is that age has failed the Goldies, while modern pioneers like Turner (Lord Adair), Kumhof, Huber, Martin Wolf and Evans-Pritchard are all witness to the need for money without debt, and debt only with money.
      But I know that's way over your head, Paul.
      Only 10 percent of lawmakers know how money is created. We're hoping that changes next week.

      Then we can have a real discussion of money-issuance options.

      1. Joe – the idea that the government can (without bad effects) print money and call it savings, is not "over me head", it is just wrong.

        As for "money without debt" – what sort of money are your offering me? Gold? Silver? Rubies? What amount and what quality (what level of purity)?

        But do not go around making up "money" from nothing (like a credit bubble type banker) or just printed – like a government (or private scam artist).

        Money should be a store-of-value not just a medium-of-exchange, economic value being subjective does not change this. And to be a true store-of-value (even with economic value being subjective) money must have nonmonetary (NON monetary) value – people must value it apart-from its uses as money.

        That is why I can you several real alternatives to consider – all "money without debt". Physical gold, silver……..

  12. Joe B – real saving means the sacrifice of consumption from real income. I can not just print money in my basement and, legitimately, call it "savings" – and neither can the government.

Comments are closed.