Recent Posts

Sunday, October 23, 2016

Money-Financed Fiscal Stimulus: Nothing To Do With "Money"

As I have noted repeatedly, the problem with writing about "Helicopter Money" is that the people advocating it are extremely vague about what it actually is. I tried going through the Jordi Galí paper - "The Effects of Money-Financed Fiscal Stimulus" - as that paper allegedly provided a formal explanation of how "overt monetary financing" is supposed to work. Unfortunately, "money" has nothing to do with the results in that paper.

Overt Monetary Financing


I will immediately note that some authors use the term "Overt Monetary Financing" for what I consider to be a completely different policy. Modern Monetary Theory (MMT) authors, particularly Bill Mitchell, and Ralph Musgrave (who previously responded to my writing here on this lines) use "Overt Monetary Financing" to refer to what I would call "permanent ZIRP" (zero-interest rate policy).

In permanent ZIRP, the government makes a credible promise to set the policy rate at zero forever, and then the government can finance itself using money or Treasury bills yielding 0% forever. I am not going to debate the wisdom of such a policy here, but I will just note that such a policy involves reducing the power of the central bank, not increasing it. (The mainstream versions of "overt monetary financing" involve giving the central bank more power.)

Why is Fiscal Policy Allegedly Ineffective?

The pre-Financial Crisis consensus was that fiscal policy is ineffective with an independent inflation-targeting central bank: any attempt to stimulate growth by fiscal policy would be cancelled out by the central bank. (There is also the debatable question of Ricardian Equivalence, but that is a second order effect,)

(The mainstream consensus is shifting; but this basic principle is still embedded in most DSGE theory, including the Galí paper.)

Therefore, it is not correct to say that the (original) mainstream position was that fiscal policy was ineffective; it's just that the central bank just cancels it out as part of its economic stabilisation mandate. (Please note that I do not agree with that assessment, but that is neither here nor there.)

The "permanent ZIRP" policy would make fiscal policy effective again -- because we have shut down the central bank cancellation. (That said, mainstream economists would scream about the price level jumping to infinity because of expectations.)

Otherwise, with such an analytical framework, "monetary financing" does not look like it would work: the central bank's inflation-targeting mandate would wipe out the effects of "money-financed" fiscal stimulus. (The zero lower bound cannot be used as a cop out; at a zero interest rate, Treasury bills are indistinguishable from money.)

Returning to the Galí Paper

The Galí paper has a lot of mathematics in it, which buries what is happening. In my view, the mathematics provides a good distraction from what the paper says.

The results can be summarised: if the central bank no longer cancels out fiscal policy by following a Taylor Rule, fiscal policy is once again effective. If you accept the assumptions of mainstream economics, that is an obvious result. Furthermore, money financing has nothing to do with the conclusions.

The only reason that money appears within the discussion at all is that the paper assumes that there is a stable relationship between "money" and interest rates via a money demand function. Instead of following a rule setting the policy rate -- a Taylor Rule -- the bank sets the monetary base.

The fact that a mainstream economist is publishing a paper in 2016 which is based upon the concept that "money" is exogenous -- an idea which was thoroughly discredited in the 1980s -- is rather telling. In the real world, no such stable "money demand function" exists, and the policy would only consist of setting interest rates "too low" in order to generate inflation. "Money financing" only appears in the paper because of the use of this discredited theoretical concept.

DSGE Models Have Finance -- Yeah, Right

Apologists for Dynamic Stochastic General Equilibrium (DSGE) models say that the models have improved after the crisis, and include a financial sector. That argument is disproved by the Galí paper: a banking system is nowhere in sight. In reality, the DSGE modelling framework cannot produce realistic results, and all that can be done is that a particular model can attempt to make one particular improvement, but the improvements are not cumulative -- other things have to be thrown out to make the model tractable. That is, if a financial sector is added to a model, it has to throw out other things.

"Money financing" enthusiasts like Lord Adair Turner write about the monetary base in the real world, and want to consider that to be the same thing as money in a DSGE model. This is an elementary error, but they would need a DSGE model with a banking system to model the distinction. All of the stochastic calculus that the DSGE paper indulge in are a distraction from the fact that they provide exactly zero insight into real world policy decisions.

(c) Brian Romanchuk 2016

10 comments:

  1. Brian,

    It seems to me that a new framework is needed. One potential new framework compares "money" to a gift certificate. I made an opening stab at this with the post "Money is Like a National Gift Certificate".

    Now if we applied this model to either SFC or DSGE models, gift certificates could easily be traded value-for-value where supply always equals demand. A discontinuity comes from the creation of the gift certificate, which can originate from two sources.

    ReplyDelete
    Replies
    1. A gift certificate is a liability to the issuer, like money. (One difference is that gift certificates have an expiry date, whereas most government money does not. "Stamped money" sort of does.)

      To paraphrase Minsky, anyone can issue a liability and hope it is treated as "money". Whether or not it is depend on economic institutions in place.

      I will be writing an introduction on my SFC models package later this week (probably Wednesday). I have deliberately left "money" out of the package; all entities have a financial asset holding ("F") which may be held in whatever instruments are in the model. I am leaving the financial side of things on the back burner for now, but I just wanted to point out we can build models without worrying about the nature of money.

      Delete
    2. This comment has been removed by the author.

      Delete
    3. "but I just wanted to point out we can build models without worrying about the nature of money."

      YES! Way to not get sucked into essentialism.

      I know you don't like to put money in formal models. I don't have a dog in that fight, in part because I am not well enough informed. You surpassed me a while ago, Brian.

      But what a waste of time enquiring into the "nature" of money would be. Money is a flower. No, money is a manifestation of love! No, money is an inside asset, which REALLY helps in the understanding.

      By the way, I share your take on what people call helicopter money. It seems like macroeconomists can be even more fraudulent than banksters or politicians. Some of their claims on this issue were pretty darn dishonest, rather than just wrong.

      Back to money, as I see it, the government has -- for whatever reason -- a unique ability to create zero-interest-bearing liabilities. You could call that an asset if you wanted. One of the functions of the central bank is to realize, dare I say, monetize this advantage. You can call the stuff they issue as part of that process liabilities if you want. But who cares about the words?

      Sorry for double posting. There was a typo. And then it got longer! Maybe I'll see you around after Wednesday. Hope not, but maybe.

      Delete
    4. They can set the rate of interest on all of their liabilities, as long as the currency is non-convertible. The issue is that they want to set the policy rate to control inflation, and they let bond yields float to create a yield curve. (And when places like Australia threatened to retire their bonds, the various free-market loving, fiscal disciplinarians in the financial sector screamed bloody murder to keep their yield curve.) In other words, the government *wants* to pay interest for policy reasons. So the rhetorical question is: why is 0% financing an advantage?

      If you get deported after Wednesday, I might be seeing you fairly soon!

      Delete
  2. I agree with all of this. (Including the snarky asides.)

    ReplyDelete
  3. Most of the technical arguments made as to why money financing might be more effective than debt financing, even at the ZLB, rely on the idea of a permanent difference. If it assumed that the finance mix reverts to the same blend after a certain period of time, then the supposed benefits of money financing disappear.

    But, it's hard to see how you can take a permanent measure in the real world, as opposed to in models. The monetary authority today cannot bind the monetary authority tomorrow. Long run policy must be viewed as endogenous. You have to ask whether the funding mix today will somehow change what the authorities decide to do in the future.

    ReplyDelete
    Replies
    1. The idea of it being "permanent" pretty much relies on money being exogenous, which isn't a feature of many models currently in use. Even if one believes that money demand is stable, a "permanent" increase in "money" is just "permanently lower interest rates", and that is the thing that is driving model outcomes.

      Since it is mainstream economists who usually talk about things like "permanent increases in the money supply," and then they ignore post-Keynesianism because it is "too verbal" and "there are not models", I just shake my head.

      Delete
  4. Overt monetary financing is surely not necessarily ZIRP. The policy interest rate could be at any level. It is just that the policy interest rate becomes and remains the floor rate the central bank pays on exchange settlement balances. It is not a requirement of MMT that the official interest rate is set at zero, even though this is often advocated by MMT authors.

    ReplyDelete
    Replies
    1. My wording was vague, but yes, not all MMT authors advocate the permanent ZIRP policy. However, Bill Mitchell does use "overt monetary financing" to refer to the policy, while more mainstream economists allow for non-zero interest rates. Although both groups use the same term, I see the policies as being very distinct.

      Delete

Note: Posts may be moderated, and there may be a considerable delay before they appear.

Although I welcome people who disagree with me, please be civil.

Please note that my spam comment filter appears to dislike long "anonymous" posts. I get no warning about this, and only go through my "spambox" infrequently. The best bet it to keep comments short, and if you think the spam filter struck, let me know with a short comment.