Recent Posts

Wednesday, June 12, 2019

MMT And "Printing Money"

Yet again, the question of "printing money" and Modern Monetary Theory (MMT) has come up on Twitter. In my view, these debates are confusing because critics of MMT tend to mash multiple concepts into a pile of textual sludge. Unfortunately, MMTers are forced to respond to those attacks, and the end result is that everyone is confused.

One of the advantages of mathematical training is that you are forced to take definitions seriously, and step cautiously from premise to premise. This enforces clarity in logic.

The whole "MMT is pr1Nting M0N3y$!!!" online debate is a classic example of the logical mish-mash we run into. By my count, there are at least three concepts being pushed together.
  1. The MMT description of monetary operations in the real world, which involve the government "writing cheques" (or using electronic transfers) to pay for expenditures. I.e., "everything is paid for by printing money."
  2. MMT proposals about abolishing the government bond market.
  3. MMT criticisms of how "mainstream" economists employ the "governmental financial constraint" logic.*
I will focus only on the second argument here, for very good reasons. 
  • The first is a description of the legal reality of government financing, and the scholarly MMT literature appears to have that subject nailed. (One might argue about presentations in internet primers, a subject that I have exactly no interest to discuss.)  If one wants to debate it, one needs to get into legal technicalities that are jurisdiction-specific.
  • The third point is an example of how many "conventional" economists have a very hard time understanding their own neoclassical models. In the question of government finance, the MMT description of current operations is "more neoclassical than the neoclassicals": bond yields are determined largely by expectations for the policy rate. They are not determined by "fiscal risk premia," which is the siren call that claimed many conventional analysts in the Great Widowmaker Trade (Japanese edition).
Only the second topic is tied to an actual MMT policy proposal, which is usually how the subject is presented in criticisms. I will only outline the topic, since I already covered this in Section 6.7 in Understanding Government Finance (link).

The proposal is easily understood: instead of the central government issuing bonds, government liabilities are only in the form of base money: deposits at the central bank, and notes and coins (which tend to be stable with respect to nominal incomes). If we use American parlance, the liabilities will mainly end up as "excess reserves," but that term only makes sense in countries with bank reserve requirements. Depending on the jurisdiction, it might take some changes to allow such a form of financing to work, but these are only technical issues (laws concerning government financing have changed in the past).

Will this have a radical effect on anything? Not really. Such a policy is essentially the same thing as Quantitative Easing (QE), which has had no measurable effect on any macro variables in places like Japan. (There is certainly a feverish mythology about QE in some corners of the markets, but financial market participants also used to worry about things "eyeball counts.")

What are the differences between such a regime and now?
  • Bond vigilantes out of business. Since there are no bond holders, there are no vigilantes to "discipline" fiscal policy. This is obviously very distressing for those of a neoliberal bent - since the definition of neoliberalism is that the government should submit to market forces.
  • Interest rate policy is not an option - unless interest is paid on deposits at the central bank ("reserves"). Paying interest on "reserves" somewhat dilutes the previous point, but it could be taken as a half-way house. However, since the duration of government liabilities would be (near) zero**, there would be no term risk-free instruments to price a yield curve off of.
The second point is rather critical for people who believe in neoclassical economists. It is entirely possible that they would argue that fixing interest rates ("an interest rate peg") is unsustainable (as was "proved" in various DSGE macro textbooks***). I am agnostic on that debate, but we have to accept that this is a point of dispute between MMTers and neoclassicals: we cannot assert that one side or another is wrong without digging into the gory details of neoclassical theory. (In fact, that digging is exactly what is being pushed into my second volume on my recession text.) In other words, this morphs into yet another concept: are neoclassicals incorrect about interest rate policy, as MMTers argue?

One could easily argue in good faith whether this is a good policy decision; I noted reservations myself in my book. However, it should be noted that this ends up being a really technical debate about the role of monetary policy and financial stability (for reasons I discussed therein). Importantly, it is not an argument about the size of deficits that are desirable -- which is yet another concept ("Functional Finance") that critics drag into the discussion. That is, one could favour a "no bonds" policy and be a "fiscal conservative" - although that would be an unusual combination of views.

Concluding Remarks

If one wants to criticise MMT in good faith about "money printing," you need to do a fair amount of effort pinning down exactly what you are referring to. In particular, you can't rely on summaries provided by non-MMTers, since they mangle together too many concepts at once.


Footnotes:

* For those of you who have not run into this argument, it is in response to common assertions about a governmental accounting identity: "government spending is financed either by taxes, borrowing (bond issuance), or by printing money." The person making the argument then goes on to make assertions about the effects of each form of "financing." Stephanie Kelton raised this point on Twitter, which reminded me about this topic, which I was going to ignore. This underlines my point about the multiplicity of concepts being mashed together.

** In fact, something like Treasury bills would probably need to be issued to deal with the reality that large investors cannot safely leave deposits at banks beyond deposit insurance limits. These bills would be issued on a fixed price basis, much like savings bonds for retail investors.

*** No references handy, as said textbooks are currently being used as doorstops in my household.


(c) Brian Romanchuk 2019

16 comments:

  1. " since the definition of neoliberalism is that the government should submit to market forces."

    I like that definition a lot.

    "*** No references handy, as said textbooks are currently being used as doorstops in my household."

    At least they are useful for something... probably could burn them if the heat went out.

    ReplyDelete
    Replies
    1. To what extent there's a short definition, that seems to be it. I've seen longer-winded ones that are perhaps more historically accurate, but the added detail do not seem pertinent.

      Delete
    2. Personally I like the description of ''the neoliberal agenda'' produced by IMF economists in 2016 in their article ''Neoliberalism Oversold?''. It is more long-winded but explicitly includes the policy of lower government spending and limits on fiscal deficits which is implicit, but not explicit, in the definition you wrote. Here it is:
      ''The neoliberal agenda—a label used more by critics than by the architects of the policies—rests on two main planks. The first is increased competition—achieved through deregulation and the opening up of domestic markets, including financial markets, to foreign competition. The second is a smaller role for the state, achieved through privatization and limits on the ability of governments to run fiscal deficits and accumulate debt.­
      It's at: https://www.imf.org/external/pubs/ft/fandd/2016/06/ostry.htm

      Delete
    3. Thanks. I think I read that piece, and my summary is based on how I remembered it and other articles. (DeLong had a sympathetic description that I read.)

      Delete
    4. Another definition of neoliberalism that's handy if a little cheeky comes from McMaster University culture critic Henry Giroux, who said neoliberalism is what happened "when Margaret Thatcher married Ronald Reagan" — anti-union, deregulation, small govt, rugged individualism, a shrinkage of the social saftey net and market fundamentalism.

      Delete
    5. Only issue is that the “third way” liberal-left - Clinton, Blair - would not lump themselves in with Reagan/Thatcher. I’m definitely not a fan of that political archetype, but they saw themselves as “modernising” the liberal left to make them electable. I’d agree with the assessment about electability, at least in the 1980s-1990s environment. However, it’s been two decades since that was a relevant issue.

      Delete


  2. I'm glad he concentrated on the abolition of the government bond market, as this is a somewhat neglected topic of discussion, yet in many respects is critical.

    In Australia, and I guess in many other countries, government accounts and financing have been set up so that a fiscal deficit spend automatically triggers the issuing of bonds to cover this deficit.

    This creates a seemingly unbreakable nexus between deficits and government debt, making it seem like government debt is necessary to finance government expenditure. But the decision to set up the accounts this way was a political decision, not a necessary one. Not that the politicians and economists who did this did not really understood what they were doing, although they probably thought they were quite logical in "balancing" the books.

    Looking at the problem from the point of view of the money supply (as a stock of money) it can be seen that something very very peculiar is happening here.

    A deficit spend puts extra money into the economy, expanding the stock of money. This is necessary to stimulate the economy, in the face of a growing population, unemployment and under-utilized resources.

    Issuing government bonds, on the other hand, takes money temporarily out of the economy, reducing the stock of money. These bonds reduce "liquidity" by taking money out and replacing it with an IOU. The money taken out ceases to exist. This is temporary because when the bonds are redeemed, money is re-created and put back into the economy.

    So we have the strange situation that we put additional money into the economy with one hand (a fiscal deficit spend) and take it out again with the other (issuing bonds). How on earth does this work? How does this even make sense?

    Well, there is actually a significant time gap, or lag, between creating a deficit and issuing the bonds to cover this deficit.

    It is in this time lag that the extra money that has been put into the economy has a chance to work, and stimulate the economy. Once the bonds are issued, that stimulus fizzles out.

    If we did not issue bonds soon after we had a deficit, then the extra money would continue to be able to work its magic without being strangled soon after birth. In fact, we would need much LESS deficit spending to achieve the same growth effect.

    So if we made the political decision to break the nexus between deficits and debts, simply by reorganizing government accounts, then we would no longer "drown" in deficits and debts, because both would be much lower.

    Government debt would not actually be zero, because we would still need to issue government bonds on the international market to adjust our international reserves for purposes of trade, but domestic bonds would be an anachronism.

    This is a relatively long post, because I am not sure if the points I have made here are generally recognized, even by those who want to get rid of government bonds on the domestic market.

    ReplyDelete
    Replies
    1. Hi,

      There's probably too many people arguing that There Is No Alternative, but some kind of bond issuance is needed if interest rate policy is going to work.

      To say that there would be a massive political fight to get rid of interest rate policy would be an understatement. You would need to convince a lot of people that they are wrong about the role of interest rates for steering the economy - which is not particularly easy to do. It's a point worth discussing, but not one that I would personally spend political capital on (if I were involved in politics, which I am not).

      Delete
    2. As far as interest rates are concerned, I understand that each country is different, so there are probably no general prescriptions.

      However, in Australia, the interest rate which sets a lower limit for commercial bank loans (if banks are to make a profit) is set by the Reserve Bank on overnight bank settlements and borrowings from each other (this has now been replaced by the Reserve Bank Information and Transfer System, RITS, but is essentially the same).

      The banks can of course decide whatever interest they want to charge when making commercial loans, and Federal Government often complains when the overnight interest rate goes down, and the banks do not follow suit.

      As far as government bonds are concerned, the Federal Government can set whatever interest rate it likes on these, or it can choose to let the market itself establish a rate, but in truth these interest rates can be zero or even negative (as in Japan) and STILL attract buyers.

      And in Australia, the lower bound interest is not paid on "reserves" held at the Reserve Bank because no-one has to hold reserves there. If a bank insists on holding reserves, then the Reserve Bank actually pays a much lower interest as a penalty.

      The only necessary "reserves" are those created when the government transfers money to banks to pay for government expenditure such as salaries of its employees, pensions, and so on. These "reserves" are quickly passed through to individual accounts in the banks (strictly speaking they are not held "in" these deposit accounts, they are just part of the banks' general revenue. It only seems like the money is held in trust because the government will guarantee these payments if the bank goes bust).

      So ultimately the government effect on general interest rates is very limited.

      And the question always has to be asked, why would you even want large scale investors to park money in government bonds? Surely they should be encouraged to invest this money in productive enterprises?

      Delete
    3. Since central government bonds from free floating sovereigns are default risk free, they are the only financial asset that goes up in price in a financial crisis. So they are of general interest. Remember that deposit insurance is meaningless for a large investor.

      Meanwhile, banks need to hold liquid instruments to manage their liquidity position. They buy/sell these liquid instruments to counter-balance their net flows each day. Although their counterparty banks could lend them some money in the inter-bank market, no bank wants to be 100% dependent on its competitors for liquidity. Government bonds, due to their 0% risk weight and default-free nature, are perfect for this. (The risk weighting makes up for lower interest rates, since high cost capital does not need to be held against them.) This means that government bond rates become the benchmark for all alternative lending options (where “lending” includes buying bonds).

      Delete
  3. I will restate here a couple of monetary axioms that are often ignored when MMT authors discuss money and money supply.

    The first axiom is that, at least in the private sector, money carries with it a chain of accountability. In very few words, this means that we can trace the path of money as it changes ownership.

    A second axiom is that a bond is nothing more than a record of ownership change combined with a promise to repay the borrowed money.

    With these two axioms in mind, consider what happens when government borrows money from the private sector. Government, by issuing a bond, is advancing the chain of accountability.

    With these two axioms still in mind, consider what happens when government borrows money from the central bank. Before leaping, decide how much money the CB has to loan that was earned in the same fashion as the private sector uniquely earns money in preparation for lending. Your decision will drive your answer as to whether CB borrowing advances the chain of accountability or begins a brand new chain of enhanced money supply.

    It seems to me that most MMT authors ignore the chain of accountability attached to money that exists in the private sector.

    ReplyDelete
    Replies
    1. The question of whether "money carries with it a chain of accountability" is interesting, but it is a very thorny problem which has not been resolved anywhere.

      A bank note, coin does not have a stamp of ownership on it. Moreover, money "held in" a deposit account at a bank does not either. The depositor surrenders ownership of the money to the bank, and is simply a general creditor.

      Most governments try to establish chains of ownership and accountability if only to tax those who "own" the money and to prevent fraud.

      But we all know that money can simply disappear, and there are many creative accountants overseas havens to help with this. Electronic forms of money are just as vulnerable to this as tangible money.

      Accountability is a will o' the wisp. I think it tends to be ignored, apart from police and tax inspectors, because it is so tenuous.

      Delete
    2. Roger, that assertion about a chain of accountability in the private sector is simply not true. Private issuers have issued currency in circulation: private tokens were the low denomination coins in pre-Confederation Canada, Scottish banks still issue banknotes. Those are bearer instruments, and there’s no way of tracing ownership. Ownership is determined legally via rules coming down from the Romans - “possession is 9/10 of the law”.

      Delete
    3. Brian, It's clear that some authors think "a chain of accountability in the private sector" is "simply not true" while others think it not only exists but is crucial to monetary and economic stability.


      Delete
    4. When the private sector issues bearer instruments - like the examples I gave - there is no tracking of ownership. The same applies to governments - currency in circulation is a bearer instrument, while other instruments are in some sense trackable.

      Delete
    5. Currency in circulation carries it's own accountability. Electronic currency likewise. Both exist as independent instruments, owned by the owner-of-moment.

      Bonds, another layer of accountability, are a promise to re-gather currency into a block and return the block to the original owner. The fact that blocks of currency are broken to be disbursed far and wide does not preclude a regathering of currency. Regathering is the responsibility of a borrower.

      On the other hand, what about borrowing from the central bank? Does the central bank have something to lend gathered in the same fashion as the private sector gathers money? To examine these questions, we need to follow the ownership of blocks of money into and out-of private ownership. Not too hard to do if we account for government re-borrowing money that government has already brought into existence.

      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.

Note: if you want to post comments from Apple devices (iPhone, iPad), you apparently need to turn off "prevent cross-site tracking" in Safari privacy settings. (The reason presumably is that another URL handles comments, and so the user session needs to be preserved when redirected to that site. I don't like this, but this is not enough to make me switch my hosting service.)