The Futility of Stimulus

federal reserve, sisyphus, monetary stimulus

federal reserve, sisyphus, monetary stimulusGeorge Selgin has recently focused on the failure of Federal Reserve policy to finance a normal recovery.  The Fed has greatly expanded its balance sheet and created a large quantity of excess reserves, which, for a variety of reasons, commercial banks have not mobilized into credit creation.  Instead, banks seem content to earn the 25 basis points of interest the Fed now pays on reserves.

This anomalous behavior shows up in the M1 money multiplier, which is at record lows – less than half its value before the financial crisis.  The Fed is creating reserves, but commercial banks are not creating as much bank money as has been historically true.  Compounding this is the fact that the velocity of M1 – the rapidity with which each dollar is spent annually – has hit a 40-year low.  Consequently, the Fed’s efforts to produce monetary stimulus have failed.

(A similar story can be told for other money supply measures.  Data and charts can be found at FRED, the online data center at the Federal Reserve Bank of St. Louis.)

I do not think economists fully understand all of the factors contributing to this policy failure.  But Selgin has surely identified one relevant factor, the payment of interest on reserves.  On the margin, it creates a disincentive for commercial banks to create money and credit in a normal fashion.  There are also fiscal reasons for ending the payments, as they reduce the payments the Fed makes to the Treasury.  As it is, the payment of interest on reserves constitutes a fiscal transfer from taxpayers to commercial banks.  In a normal world, I would endorse his call to end the interest paid on reserves.

We do not live in a normal world.  The Fed has replaced liquid, short-term assets on its balance sheet with illiquid, long-term assets.  Normally, to raise the Fed Funds rate, the Fed would sell Treasury bills.  It has none to sell.  Analysts and pundits speculate on when the Fed will raise interest rates.  They should be asking how the Fed will raise interest rates.

Stanford’s John Taylor thinks the Fed will need to increase the interest rate paid on reserves to accomplish that goal.  Markets through arbitrage would then increase the interest rates banks pay each other to borrow reserves.  I suspect he is correct, with two caveats.  First, there is no longer much of a market for federal funds.  Banks aren’t lending each other reserves.  Second, there are other possible mechanisms for raising short-term interest rates like the tri-party, reverse repo facility at the New York Fed.  This, and other facilities, are untested as a means to implement a policy change.  Their use would put monetary policy in unchartered waters.

To sum up, monetary policy has failed to simulate economic activity.  It has failed even to finance a normal economic recovery.  In pursuing a failed stimulus policy, the Fed has tied its policy hands going forward.  At some point, interest rates will need to rise.  The Fed will need to rely on novel means to accomplish a turn in policy.  Paying higher interest rates on bank reserves may be one method.  It is an unpleasant reality.  It is only one consequence of the Fed’s experiment with extraordinary monetary policy.


  1. I don't know, Gerry. I think the Fed could normalize without much consequence if they had the will to do it. In fact, I think financial markets would work better. There's a high demand for Treasuries and a collateral shortage, and I think intermediation would pick up with a steeper yield curve if they un-"Twisted."
    I believe they could stop paying IOR now, normalize the composition through rolling off maturities and some OMOs over the next year or so, and then downsize the balance sheet gradually as needed. I mean, from our perspective, would it be so bad if lending and velocity picked up therefore they would need to halve the balance sheet? I'd call that a win.

  2. I second Jerry O'Driscoll's analysis above. — Walker Todd [Justin Merrill's ideas also have merit. Easiest way to get back to "normal" monetary policy is to stop reinvesting in mortgage-backed securities as they mature, and then the same for Treasuries maturing in over 5 years. Eventually the balance sheet would resemble a 6 to 10 pct. share of GDP, a "normal" pct. for central bank balance sheet vs. GDP. Above 25 pct. (about where we are now), it's corporatism, and financial corporatism at that.)

  3. Assuming the fear is that economic recovery will unleash a tsunami of excess reserves into the money multiplying economy thereby ushering in inflation, the increased interest rates that the Fed would need to pay to compete may be high indeed. The reason is that such a recovery would itself tend to push rates up.

    Therefore, attempting a controlled release of excess reserves in this way would come at the expense of further expanding the monetary base.

    Could this possibly work to restrain inflation?

  4. "the Fed’s efforts to produce monetary stimulus have failed."

    To What are you comparing success and failure? I really don't think that it is fair to judge the Federal Reserve relative to previous monetary expansions. I think it is more realistic (albeit imperfect) to contrast the US economy against the EZ and their 2011 tightening. On this basis, QE worked! Surely the Fed could have pursued better policies (unlimited QE from the get go, no/negative interest on reserves, etc.), but to say that their efforts to produce monetary stimulus have failed is incomplete without considering the state of the economy absent the Fed's monetary stimulus.

    I am not concerned about the Fed's ability to drain money from the system when the time comes at all. The monetary policy that the Fed has pursued is unusual only in its size. It added money to the system. It can take that money back.

    1. "I am not concerned about the Fed's ability to drain money from the system when the time comes at all."

      If the Fed now lacks Treasuries or other sort term or liquid assets, then how?

      1. Reverse repo, selling their portfolio, creating short-term Federal Reserve bills backed by their long-term portfolio and selling the hose securities. There are as many ways to take money out of the system as there are to put money into it.

        You have to distinguish what the Fed is doing (adding/removing money from the system) from the mechanism by which they are doing it (buying/selling short-term securities). The what matters, the how is less important as long as they adequately achieve the what.

          1. The only thing extraordinary about recent monetary policy is the size. On a macro level, does it really make a difference if the Fed buys T-bills or if they buy stocks?

          2. Probably, given the quantities that they deal in, and the politics inherent in any central planner's decision making. But whether or not it is an experiment or extraordinary has to do with whether or not they've done it before. If they haven't done it before, then history tells us to expect unintended or unexpected (and undesirable) consequences.

  5. Not sure why the Conclusion to be reached is not that the federal reserve has an economic mandate (inflation-free, quasi-full employment economic growth) and exactly zero policy tools with which to achieve that mandate. (They neither create nor issue any money, only reserves that are not manifesting borrowing and lending.)
    "" The Fed is creating reserves, but commercial banks are not creating as much bank money as has been historically true.""

    The article correctly points out that important monetary variable ….. related to the velocity ( herein the lack thereof) and if we just throw in the other important monetary-economic variable …… that of the increasing amount of debt necessary to produce a unit of new GDP …… then what we have is a monetary-economic system itself that is broke, broken and insolvent.

    A broken monetary system portends financial chaos, and therefrom, economic stagnation.

    But, keep on collecting the data.
    And, whatever you do about the monetary system, don't do anything about it ……. off-limits.

  6. "[M]onetary policy has failed to simulate economic activity." Compared with what? I think you are discussing the policies followed by the Fed since late 2008. (Aside: these policies have varied over time; the different policies followed serially by the Fed really should be treated separately.) It seems obvious that these policies have stimulated economic activity *more* than some alternative policies would have done, but *less* than other alternative policies. Do you have in mind some particular set of alternative policies as your baseline?

    Perhaps you mean that the Fed's actual policies have been less stimulative than they should have been–less stimulative than *the ideal monetary policies* would have been. Given the actual economic weakness, together with the Fed's repeated assurances that it can do (could have done) more, that is obvious. Or does your reference to "extraordinary monetary policy" suggest that your baseline is
    *ordinary monetary policy*? If so, what would that have looked like over the last seven years?

Comments are closed.