I find it amazing how many people still believe that QE is adding all sorts of liquidity to the markets.  There is still a widespread belief that QE is inflationary and an addition of net new money (despite the rather simple accounting behind a QE transaction).  Well, it looks like a few notable people are finally starting to get it.  Some recent commentary by notable central bankers makes it clear that some are beginning to notice that QE has no real relationship with higher inflation (via Warren Mosler)::

Don Kohn (Former FRB Vice Chair): “I know of no model that shows a transmission from bank reserves to inflation”.

Vitor Constancio (ECB Vice President): “The level of bank reserves hardly figures in banks lending decisions; the supply of credit outstanding is determined by banks’ perceptions of risk/reward trade-offs and demand for credit”.

Charlie Bean (Deputy Governor BOE) in response to a question about the famous Milton Friedman quote “Inflation is always and everywhere a monetary phenomenon”:

“Inflation is not always and everywhere a monetary base phenomenon”.

In this week’s letter John Hussman dispels the myth that the increase in the monetary base is adding new liquidity:

“From the standpoint of prospective investment returns, it is important to recognize that the main effect of quantitative easing has been to suppress the expected return on virtually all classes of investment to unusually weak levels. It’s widely believed that somehow, QE2 has created all sorts of liquidity that is “sloshing” around the economy and “trying to find a home” in stocks, commodities, and other investments. But this is not how equilibrium works.

Here’s how equilibrium does work. Every security that is issued has to be held by someone, in precisely the form in which it was created, until that security is retired. Period. That means that if the Fed creates $2.4 trillion in currency and bank reserves, somebody has to hold that money, in that form, until those liabilities are retired. The money ultimately can’t go anywhere. If someone tries to get rid of their cash in order to buy stock, somebody else has to give up the stock and hold the cash. In the end, every share of stock that has been issued has to be held by somebody. Every money market security that has been issued has to be held by somebody. Every dollar bill that has been created has to be held by somebody. None of these instruments somehow “find a home” by going somewhere else or becoming something else. They are home.

So what is the effect of creating an extra $600 billion dollars of monetary base by having the Fed purchase $600 billion dollars of Treasury debt? The same thing that happens anytime any security is issued. Somebody has to hold it, and the returns on all other assets have to shift by just enough to make everyone in the economy happy, at the margin, to hold the outstanding quantity of all of the securities that have been issued. In practice, the only way you can get people to willingly hold $2.4 trillion in non-interest bearing cash is to depress the return on all close substitutes to next to zero. So short-term Treasury bill yields have been pressed to nearly nothing.

I’ve been hammering on this point for 6 months now, but an increase in reserves has no impact on net financial assets.  As Mr. Hussman describes it merely alters the mix of assets.  This is not inflationary in the sense that some Friedmanites might have you believe.  And it certainly does not increase the odds of some sort of hyperinflationary collapse.  There really isn’t new money in the system.  And the notion that QE is helping to fund the deficit is beyond nonsensical and displays a terrible lack of understanding when it comes to how the US monetary system functions.  The only thing QE is doing is generating a huge amount of confusion, making investors believe the Fed is printing money and altering investor perception so as to to induce a speculative boom in commodities that is now helping contribute to global turmoil….

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services.

More Posts - Website

Follow Me:


  1. Scott, are you saying that no new liquidity is created if the dealer repos out a treasury it owns in order to buy a treasury from say a hedge fund that it sells to the Fed?

    • Hi William,

      First, I’m actually saying the opposite.

      Second, I think “liquidity” is one of the most misused terms there is, like saying “money.” “Liquidity” isn’t spending power in the aggregate. In the aggregate, “liquidity” can always be created by the non-govt sector, except in a crisis.

      • The point is, there is no difference, except that the Tsy can be repo’d over and over again, and thus create more “liquidity” than the deposit can.

  2. Rates bottomed on 8/31/2010. The monetary transmission mechanism is not QE2, it is via interest rates. Thus the degree of ease or restraint is related to the changing portion of the yield curve covered under the remuneration rate’s umbrella.

    The higher rates become, the looser policy becomes (as the portion of the yield curve which is lower than the remuneration rate @.25% continues to shrink). I guess you could call that fractional yield-curve banking.

    I.e., QE2 is not a neutral monetary policy. It is an increasingly easier monetary policy. The banks are not reserve constrained. As long as it is profitable for borrowers to borrow, & lenders to lend, money creation is not self-regulatory. That’s why the Great-Recession is called a balance-sheet recession.

  3. Could the speculative inflation we are seeing simply be due to the enormous amount of deposits built up during the credit boom.
    These were not defaulted on during the bust of 2008 (these bank liabilties would have declined if the assets behind them were written down to reflect their true cost) and so therefore are seeking a yield in a world without real capital investment.
    In a world without capital investment it is wise to buy commodities as they decline over time in this environment.
    Whats the point in holding on to deposits at 1% when resourses are going south at a faster rate then this yield ?
    I think its called opportunity cost if I am not mistaken.

    Long term investment cannot survive the onslaught of the “markets” – there is no place for multi decade utility investments as they quite rightly are seen as having a low yield potential.
    However without these utilities companies cannot survive for long in a degraded economic ecosystem and so therefore its best to speculate on the 4 horsemen rather then some shiny knight.

    Utilities will have to become low yielding or zero yielding institutions before wealth / capital can be restored – then bean selling companies will be in a postion to make a real profit but not before.

  4. IORs are the FUNCTIONAL EQUIVALENT of required reserves, not short term T-bills. T-bills are settled upon maturity, or are liquidated at a discount. IORs are bank earning assets (investments).

    IORs sport a “floating” overnight remuneration rate (currently consonant with the 1 year “Daily Treasury Yield Curve Rate”).

    I.e., the policy rate “floats” (like an adjustable rate mortgage), via a series of cascading or stair-stepping interest rate pegs. I.e., as with ARMs, a “note is periodically adjusted based on a variety of indices”.

    “Among the most common indices (for ARMs), are the rates on 1-year constant-maturity Treasury (CMT) securities, the Cost of Funds Index (COFI), and the London Interbank Offered Rate (LIBOR).” – Wikipedia

    I.e., the remuneration rate is the Central Bank’s target rate’s “floor” (see the “Friedman rule”).

    “When the effects of financial intermediaries and credit spreads are taken into account, the welfare optimality implied by the Friedman Rule can instead be achieved by eliminating the interest rate differential between the policy nominal interest rate and the interest rate paid on reserves by assuring that the rates are identical at all times.” – Friedman Rule

    “The Central-Bank Balance Sheet as an Instrument of
    Monetary Policy” — April 11, 2010

    As Dr. Richard Anderson (V.P. St Louis FED), states: “Remember that “excess reserves” is an accounting concept, not a physical item. The physical item (asset) is deposits at Fed Res Banks. These deposits may be used to satisfy statutory reserve requirements; any “excess” deposits are labeled as “excess reserves.” This terminology dates from the 1920s, and I find it obsolete.”

    Those who point to the “monetary base” which is not now, nor has ever been, a base for the expansion of new money & credit, are the: “could have, would have, should have” dinosaurs.

  5. The “Friedmanites” (quantity theorists), may be iconic, but do not espouse monetarism. The St. Louis FED reports that the latest M2 “money velocity” figure turned over only 1.7 times in the 4th quarter of 2010, & that M1 “money velocity” turned over only 8.2 times (all “FRED’s” velocity figures are declining).

    It is obvious that while INCOME velocity (FRIEDMANITE’s contrived figure), is declining, the money actually exchanging hands (IRVING FISCHER’s transactions velocity of money) is patently rising. Thus, according to the monetarists, QE2 is not neutral.

  6. I think yee guys get too entangled in monetory mechanics to figure out what this money stuff is doing even if it goes beyond the FED.
    I don’t really see real capital formation since 71 – the entire matrix is net energy negative.

    • Thank you for your two comments: (too entangled in mechanics and speculative inflation) I think you are spot on.

  7. Yes & no. It’s a convoluted topic. From the standpoint of the banking system your analysis is indeed correct. I read you because of your focus & understanding.

    But consider that Investors discount (company earnings), i.e., nominal gDp (up to 9 months in advance), by purchasing equities. It is equally probably that the buyers of commodity ETFs do exactly the same thing (which does increase aggregate demand). Whether prices are “sustainable” (& not supply-side shocks), depends upon whether the “administered” price increases are evenutally (in succeeding years), “validated” by the FED’s subsequent policies.

    But is that foresight driven by QE2 expectations, normal reflation, or the anticipation of even an anemic recovery?

  8. Cullen, to be sure I understand your position, with the Fed buying $600 billion worth of Treasurys under QE2, other things equal, you are saying that has no effect on the total of bank deposits. Is that correct?

    • I am not sure if my comments are exactly correct, but I’ll give it a go.

      Total bank deposits MAY change, depending on who is the eventual net seller of the Treasuries. Though the QE transaction is between the Fed and banks, banks may have bought the Treasuries from the household/corporate sector beforehand, meaning that banks are just the middlemen.

      If this is the case, and households are the net sellers, then yes, bank deposits will increase. The swap (for the household) will simply be Treasuries for cash. The bank will then have an increased deposits liability, and an increased reserves asset.

      This will increase the broader ‘money supply’, however this does NOT mean the transaction is inflationary. This is because:

      - net financial assets have not changed (the household’s equity remains unchanged)
      - grandma might now have cash instead of bonds, but that does not mean she going to buy a new TV. If grandma wants to save, that is what she will do. Not only that, but if she wanted to spend whilst she had the bonds, a simple trade in a very liquid market would be the only thing stopping her.
      - this is quite comparable to the issue of debt after government spending (in the reverse way though). Officially, the issue of debt reduces the ‘money supply’ back to where it was pre-spending, however net financial assets have still increased – which is potentially inflationary. This means that government spending is no more inflationary if it is not followed by bond issues, than if it were. Hence, at least in this issue, it is the net financial asset number that is important, not the number of bank deposits, or composition of assets that alters the ‘money supply’.

      • Oh, and if it were the banks that were selling the bonds (if it had nothing to do with the household/corporate sector), then bank deposits would remain unchanged. So it depends who is the eventual net seller of the bonds.

      • Right Alex. We can actually go into the Fed website and see that demand deposits have increased. There is no question that QE has altered deposits to some degree. But that didn’t make these savers dissavers.

  9. The author writes as if the perception is unimportant. If a policy is not clear and transparent, as obviously quantitative easing in its current form is not clear and transparent, then it is a flawed policy on its face. It will continue to produce unintended consequences and should be ended immediately if not sooner. Duh!

    • ANOTHER comment from someone who has no clear idea what I believe. I have written EXTENSIVELY about the unintended consequences of QE….

  10. The FED finally got religion. EVERYBODY has missed their hint.


    “For this dynamic to work correctly, the Federal Reserve needs to set an IOER rate consistent with the amount of R-E-Q-I-R-E-D reserves, money supply and credit outstanding”

    The Federal Reserve has never gauged the volume and timing of its open market operations in terms of the amount and desired rate of increase of member commercial banks COSTLESS legal reserves, but always in terms of the levels of the federal funds rates (the interest rates banks charge other banks on excess balances with the Federal Reserve).

    By using the wrong criteria (interest rates, rather than member bank reserves) in formulating and executing monetary policy, the Federal Reserve was always the bubble’s engine.

    While IORs may initially continue to fool the masses, the technicians at the FED will gradually change the course of monetary history that has existed since 1965. That actually might mean MMT’s toast.

  11. 1. Fed’s QE buys bonds from Government due to deficit spending – this is money printing due to Government spending requests and is inflationary. This inflationary process is more than offset by the deleveraging that is occuring in the economy.

    2. Fed’s QE buys bonds from banks to increase their reserves – this has no effect on the inflationary aspect of the economy because banks are not able to make loans due to lack of demand for those loans – a balance sheet recession.

    3. Senitment towards QE is what is driving trends. But the underlying fundamentals do not agree with the perceived inflationary trends.

    4. Velcoity – The rate at which money exchanges hands is very important, and since 1995 velocity of money has been collapsing which is deflationary. It continues to fall.

    5. M1 minus M3 – a large part of this difference is the reserves held at the banks. As long as banks cannot lend out their reserves, it is deflationary.

    6. The US Dollar will rally to all time record highs on any deleveraging crash as a rush to sell US Dollar based asset ensues and puts the largest credit crunch bid on this currency.

  12. Take out the Fed for a moment and assume a fixed money supply where bank lending and bank debt cancellation are in equilibrium: The flow of funds in the primary Treasury market is that REPO cash funds flow from holders of cash deposits through primary dealers into Treasury’s general account joining funds collected as tax receipts and then emitted back into the economy via public sector spending. REPO cash is then replenished from that portion of savings which flows into public debt. This flow of funds increases when government runs deficits and debt accumulates. More savings divert into public debt and out of private assets….see Japan. The result: Private asset price deflation, public asset price inflation, private sector investment down, public sector investment up.

    Now add the Fed into the mix. The Fed effectively displaces a portion, all the way from a tiny to a progressivly signficant one during quantitative easing, of the total amount of private domestic and foreign offical savings otherwise destined for Treasuries. The act of purchasing Treasuries from primary dealers replenishes dealer cash ultimately used to repay a REPO loan. To the extent that this cash comes from the Fed, it displaces cash that would otherwise come from private domestic and foreign offical sources. That cash ends up elswhere inflating other asset class prices.

    Now to inflation: The main question to ask is if there is a link between consumer price inflation and asset price inflation. Commodity price inflation leaks pervasively into consumer prices to the extent that products and services are affected by commodity use and producers pay higher prices which they are able to pass on to customers.

    So is there a link between QE and inflation? Absolutely!
    Is the Fed monetizing government debt with QE? Indirectly yes.

    The confusion arises due to the indirect link between base money reserves and broad money deposits which traditionally are linked via bank lending. If net bank credit is stagnant or even contracting, increasing reserves would not inflate the money supply absent other channels through which incomes and credit can be extended. However, government deficit spending is the Keynesian compensation channel for a private sector-bank credit contraction.

    Even though the net effect of a public sector-monetary intervention may seem non-cumulative when balanced against a private sector-bank credit contraction this is not the case. Public sector spending is not effectively sanitized by productivity growth to the extent that private investments are. The net effect is that the economy while quantitatively running on the same money supply is qualitatively less efficient….and that leads to inflation and even stagflation.

  13. So clearly the difference between Japan’s deflationary decades and us despite identical circumstances otherwise (RE price bubble followed by collapse and the resulting monopolized zombie bank system sitting on the side lines while the private sector delverages its balance sheet) is that the Fed is quantitatively doing what the BofJ did not: (Indirectly) monetizing government deficits created by the spender and investor of last resort (as part of classic Keynesian intervention) and thus reflating private sector asset and consumer economic transaction prices by QUANTITATIVELY displacing private capital from the public sector capital market (as dictated by neo-Keynesian doctrine) thus undercutting the natural rate of interest on private capital by increasing its supply.

    Austrian economics points out that fiat monetary systems by nature undercut the natural cost of pooled savings turn capital which would prevail otherwise based on the supply of savings. That unhinges the relationship between consumer prices and the proclivity to save on the one hand and the cost of capital and the proclivity to produce on the other. The Fed did this by first controlling the supply then the cost of capital available to banks in the form of reserves yet both paradigms failed to prevent credit floods and inflation of either consumer or asset prices. Since 2008, interest-on-reserves has empowered the Fed to control reserve flow independent of reserve supply making it possible to buy large quantities of assets from private and public markets without affecting bank credit formation.

    Only those who still look at the new Fed in terms of the old Fed will interprete QE within the framework of the banking sector, ie as harmless with respect to inflation. This erroneous focus is by design to divert attention away from what QE is really all about.

  14. my question is this- If the government spends money into existence- by what mechanism is it forced to issue government bonds? You state that government bonds merely soak up excess liquidity created through government spending as a means of controlling inflation- and you may be correct, but this is clearly not the thinking of most treasury officials. Clearly treasury officials feel that their spending is constrained- even if as you point out it is an artificial constraint. Given that they feel the need to issue debt, then surely QE is necessary as it lowers return on the debt to a level where the interest payments do not take up too large a part of the tax take?

  15. Cullen.


    To the extend that banks are swapping longer term negotiable government securities M2 in exchange for the shortest notes “cash” M1

    At this time banks still do not seem to be lending out against their reserves but could we say that if we where in a more prosperous period banks would be able to lend out more money from a QE operation “swaps” ? Banks can not use M2 money to create M1 but they can use Cash (time 9)and increase M1.

    I know that this is not the goal of the FED at this time since the reserve certainly do not need to be increased and the FED most likely wants money to be moved from Sleepy Town to more aggressive investments but my question as more to do with the consequence replacing longer term bonds with cash (other than changing maturities and influencing rates)

  16. ok i understand they are self imposed constraints… but given that you need to issue debt to reduce the money supply to prevent inflation, then surely when you have an explosion in government spending and a corresponding increase in debt issuance to control inflation… you do need QE to keep the interest payment cost down. Ok theoretically you could spend more money into existence to pay the interest payments but then you would need to issue more and more debt. So in effect despite the fact that the US is not spending constrained by the amount of debt it issues, QE does in fact enable government spending. I am trying to get to a point where I accept your theory of the way the monetary system works, and at the same time say that QE is facilitating deficit spending…

    • “but given that you need to issue debt to reduce the money supply to prevent inflation”
      Loans create deposits. Reserves do not. This means you don’t control the money supply (and don’t prevent inflation) by issueing “debt” = USTs (which soak up reserves), you only control the overnight rate at which banks lend reserves to each other. But when you pay IOR = interest on reserves you don’t need to issue USTs at all.

      “you do need QE to keep the interest payment cost down”
      US Bonds are no fiscal financing tool.
      The US can ALWAYS pay the interest on bonds, no matter how high the yields are. You don’t need QE to hold yields down to be able to “finance” the US.
      The US can and will always be able to soak up the reserves spent into existence by deficit spending through UST issuance. This is because for banks USTs are always “better” than reserves (higher interest, no “liquidity” disadvantages) and reserves can’t be redeemed for Gold or something else (like under the Gold Standard).

      So this:
      “QE is facilitating deficit spending”
      is wrong.

  17. first and DIM

    The act of quantitatively creating new reserves displaces deposits away from the public money cycle (income–>savings—>buying public debt—->government spending—->public sector spending recipient income) and into the private money cycle (income—>savings—>private investments—-business spending—->private sector income). This effect is independent of bank lending. In fact this is what neo-Keynesian doctrine is based upon. It is the countercyclical reflation channel used by the public sector and the Fed when the private sector contracts threatening to shut down the fiat system of credit creation via the banks.

    To the extent that these displaced deposits end up in commodities they are inflationary to prices. This is exactly the intent of QE2, not just an apparent coincidental occurance.

  18. You are dead wrong and so is John Hussman who I really respect.
    If you knew what Qunatitative Easing is, you wouldn’t say that.