Recent Posts

Wednesday, February 5, 2020

MMT And Fiscal Policy: Radical Or Not?

Most public debate around Modern Monetary Theory (MMT) revolves around fiscal policy. This is not too surprising, as this is largely where political economy concerns crop up when discussing MMT. However, the emphasis on fiscal policy makes discussions of the MMT debates quite awkward. There is a large divide between the debate in modern economic theory, and the debate in popular discourse.

If we look at economic theory, there are no credible challenges to the MMT story at present; at most, one can debate terminology. Since many economists are keen to present economics as a highly mathematical, scientific discipline, getting into long-winded, arcane debates about linguistic issues does not fit the narrative. However, there are plenty of reasons for there to be a strong debate in popular discourse: not everyone agrees on the proper role of government in the economy. However, we cannot appeal to any formal theory to adjudicate the popular debates.

(This article is a brief introduction to a cycle of articles that are expected to be re-worked into a chapter in a brief intermediate-level introduction to MMT.)

Fiscal Constraint

The main issue is the question of the fiscal constraint. There are two definitions in use, but I am focused on this definition: what are the practical limits on government spending? (What happens when those limits are hit?) I discuss the second definition (a mathematical definition) later.

It is relatively commonplace to note that floating currency sovereigns are supposed to be invulnerable to default. For example, I have seen finance textbooks that refer to the government curve as a "risk-free curve" (technically, default risk free), and offer a hand-waving explanation about "printing money".

If the government cannot default on (local currency) borrowing (like a household or business), why not give everyone a free pony?

Although MMT proponents emphasise that default is not the issue, they are not saying that everyone can get a free pony. (And in response, critics will point to particular statements in popular articles that appear to imply that everyone can get a free pony. I am not attempting to police the truthfulness of every statement on the internet, and I do not care what other primers might say, only what I write.)

Since the main divide in developed country politics revolves around the proper role of the government in the economy, any discussion of the limits of governmental policy is going to be a minefield of conflicting opinions. In the absence of divine revelation, I see no way to adjudicate such arguments.

The question is whether we can relate this debate to economic theory -- and that is where this topic gets extremely awkward.

Popular Discussion: What Does that Mean?

I distinguish between popular discussion and scholarly discussion, since the two areas of debate are completely different. For most people, what I call popular debate is the only thing they will hear about. It includes:
  • Financial market commentary.
  • Comments by politicians and political activists.
  • Comments by people at "think tanks."
  • Internet wrangling and opinion pieces in legacy media -- including such pieces written by academics.
Scholarly discussion is solely the discussion of articles written by other scholars, which normally appear in peer-reviewed journals or textbooks in the modern era. I use "scholarly" rather than "academic" since the collapse in academic standards has meant that some academics have written articles criticising MMT without undertaking even a cursory survey of the literature.

It might surprising that people whose reputations depend upon being a member of academia in the area of economic theory would discuss fiscal policy without reference to the actual theory in their field, but people are rather surprising animals.

Scholarly Debate: Is There One?

If we look at the theoretical literature in a scholarly manner, the debate over fiscal policy options for floating currency sovereigns is not particularly interesting. To what extent the literature disagreed with the MMT arguments, the theories have been largely abandoned, or empirical studies collapsed. 

The core MMT position is that the ultimate constraint on fiscal policy is inflation, a stance that has its roots in Functional Finance, a school of thinking within the post-World II Keynesian consensus. There was a politically-motivated move to purge Functional Finance from mainstream economic textbooks, but that has not affected reality. Theories about fiscal policy limitations ultimately devolve to stories about hyperinflation -- which is exactly what the MMT stance is.

After that, the debate comes down to sniping over language. Since I am writing popularisations, I do not care about the exact wording others use -- I am translating it into what I see as the clearest exposition anyway.

To what extent there is a debate, it involves the mathematical definition of a financial constraint: an equation that allegedly applies to the sequence of (primary) fiscal balances over an infinite model time axis. However, this constraint is essentially an assumption that is being applied to a mathematical model: it is true by defining it to be true. The debate is whether real-world governments need to heed this equation; and the argument of MMT proponents is that they do not.

Concluding Remarks

Popular economic commentary is full of lots of opinions by people with economics degrees. One might hope that economic theory -- which is popularly presented as a science, like physics -- could offer some illumination on those debates. Unfortunately, writing down a mathematical model does not eliminate political economy, and that is really what those debates are about. As a result, we are in the domain of opinion, not in the domain of falsifiable theories.

Later articles in this sequence will discuss what MMT says about fiscal policy, and then try to relate this to the debates that continuously rise and fall on the internet. However, the distinction between popular debate and scholarly debate needs to be kept in mind if the reader wants to asses the "correctness" of MMT.

(c) Brian Romanchuk 2020


  1. I've been looking at MMT again recently and felt the need to try and articulate one of my initial thoughts:

    As far as government spending and taxation goes while MMT and mainstream frame things differently I think they are in agreement on the fundamentals. For both MMT and the mainstream I think it would correct if everything else was held equal then both camps would agree that a given amount of government spending (and spent on exactly the same things) combined with a given amount of taxation (raised in exactly the same way) will lead to a given amount of inflation. The fact that one camp holds that the taxation is funding the spending while the other chooses to deny this is irrelevant.

    I think the disagreement stems from differing views on the role of bond sales. The mainstream thinks that these are another tool to manage inflation while MMT seems to deny this.

    If you think that bond sales are an alternative to taxation to control inflation in the present period then you will probably also believe that they carry this inflation potential into the future and this potential can only be neutralized by future inflation, taxation (or default). This will clearly lead to concerns about the size and sustainability of the pool of government bonds in the hands of the public.

    If one holds that bond sales have no effect at all on present or future inflation then one will be unconcerned about the total size of the bond pool. Its size can be changed at any time the government feels the need and the effects will be purely distributional and have no direct effect on inflation..

    I am curious if these views accurately reflect MMT thinking ?

    If so, then MMT differs from the mainstream on a key empirical matter - do bond sales reduce inflation pressures in the present or don't they ?

    (I see that MMT points to QE over last 10 years and Japan over that last 30 in support of its position, while the mainstream thinks that virtually all other period of economic history and particular the period between the 1980's and 2008 where bond policy was the main instrument for inflation targeting in support of their views).

    1. Bond issuance by itself only accomplishes one thing for the real economy: keeping the term interest rate structure positive. If monetary policy is going to be used to control inflation, then issuance is necessary. The whole "budget constraint" thing used in undergraduate textbooks is nonsensical, either from the perspective of modern neoclassical macro or MMT. So yes, there is convergence of views from the perspective of theory. (Why economics 101 textbooks are filled with stories that cannot be justified by more advanced theory is another question.)

      (Under current institutional arrangements, bond issuance appears "necessary" from the perspective of intra-governmental accounting. That necessity could be eliminated at a stroke of the pen.)

      But yes, my point is that modern mainstream theory has basically converged to the MMT view. The problem is the popular version of mainstream theory...

    2. Rob, you ask "If so, then MMT differs from the mainstream on a key empirical matter - do bond sales reduce inflation pressures in the present or don't they ? "

      It is a tough question and I would have to answer based on my MMT reading that No- MMT says that bonds as currently issued by (for example) the US Treasury do not reduce inflation pressures in any significant way.

      That doesn't mean that bonds could NOT be structured so that they would reduce inflation pressures- just that at the moment they aren't and don't. I mean they could change the contract so they would be completely non-transferable for the duration of the bond contract. Or that they couldn't be used as collateral or measures of capital by banks. Just that the way they are structured presently does not really reduce inflationary pressures from whatever government spending they 'supposedly' are allowing to happen.

    3. Thanks for the replies.

      The mainstream view is:
      - All government spending must be covered by taxation or bond sales
      - Some bonds can then be bought back to generate the target rate of inflation
      - Additional buying and selling of bonds may also take place to fine tune against fluctuations in AD

      Is the MMT view that all 3 of these steps are optical illusions as far as actually controlling inflation go as they only affect the interest rate structure ?

      I apologize for asking basic questions but it strikes me that the mainstream view of MMT focuses on MMT views about the relationship between government spending and taxation (where the differences are more apparent than real) rather than these MMT views on bond sales where the difference are real (and I suspect would be shocking to many mainstream economists).

    4. Statements to the effect that government spending must be covered by taxes, money creation, debt issuance are tautological or misleading. I’ve only seen them in ancient articles or econ 101 texts.

      It’s just an accounting identity: the fiscal deficit is the NPV of the increase in private sector holdings of government liabilities. Those liabilities are money or debt. The statement is a bizarre framing of the obvious.

      The only way inflation shows up is if the velocity of money is constant. It isn’t. Even the mainstream knows this (now).

      In any real model - including mainstream DSGE models, the money/bond holding split are the result of a money demand function; the central accommodates money demand to keep short-term rates at a desired level.

      It’s purely something printed in econ 101 textbooks without any reference to any theory that is taken seriously in 2020.

    5. 'In any real model - including mainstream DSGE models, the money/bond holding split are the result of a money demand function; the central accommodates money demand to keep short-term rates at a desired level.'

      I agree with this statement. Would it be correct to say that while both MMT and mainstream see this as the way to hit an interest rate target the mainstream view would be that failure to hit the target would generate inflation/deflation while MMT would not see this as a likely outcome of missing the target ?

    6. Rob, in my opinion, MMT says that moderate changes in the ultra safe interest rate are not going to have much effect, even that they could have the opposite effect than what might be expected. It is pretty much why MMT recommends fiscal policy rather than monetary policy. Keep in mind that this is in respect to a government that issues its own fiat, non-convertible currency that allows it to float as to exchange value. To a certain extent, MMT considers bond issuance to be a relic of past times where that wasn't the case.

    7. Rob - if we look at the pre-2008 system in the US, they have to supply exactly the amount of reserves needed to keep money market rates from collapsing to 0%. I.e., money supply is determined precisely by money demand. (With interest on reserves, “money” and debt are fungible, and so there is no distinction anyway.)

      Both MMT and neoclassical models agree on this.

      The disagreement is how useful interest rate policy is for hitting inflation targets. However, that has nothing to do with the split between money and bonds in private portfolios. In other words, bond issuance itself has no particular effect on inflation, it just allows monetary policy to work,

    8. Thanks, I think I have it now.

  2. Can't see the problem. MMT, like Keynes, says gov and central bank in a recession should print and spend money (or debt) up to the point where inflation looks like getting excessive.

    That's it. Finito. Period. Full stop.

    1. Ralph,

      MMT goes further and says that deficits are the norm, not the exception. This is because the private sector typically desires to net save, i.e., (S > I).

      Also, MMT adds that the government does not determine the fiscal balance. That is determined by the private and external sectors.

      The choice the government has is to run deficits of the good kind that bring growth at which point the deficit falls, or do nothing and allow a recession to happen in which case tax revenues fall and welfare payments rise, and the deficit is forced to rise that way.

  3. "The main issue is the question of the fiscal constraint."

    And I don't know how to model "fiscal constraint" for government when we can find two models of money in play. I make a verbose attempt at differentiating the two models here "Two Intuitive Conflicting Models of Money"

    One model is for money as it is used in the private economy. Here, money is used as a medium of exchange and store of value. Monetary stability is important here.

    The second model is of money making a sudden entry into an otherwise stable system. This is the basic MMT model where money is created by government. This model has newly created money taking it's value by reference to the otherwise stable money actively used by the private sector.

    I have never been able to logically find any intersecting limits between the two models. Both models seem to play at the same time, meshing as do gears of two pitches. (i.e., they don't mesh but they still can transfer torque.)

    1. Money is an artifact. You must model the social system which uses money. As Brian said "it's all political economy". If you use a tennis ball to play basketball, it's going to have a movement pattern like a basketball. The nature of money is really not the defining issue. Yes it plays a role, but really we are talking about the actions and patterns of people who use money as an accounting tool.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.