Recent Posts

Wednesday, October 4, 2017

Why MMT Is So Popular (For People Like Me)

In a previous article, I explained why I am complacent about the state of economic theory in Modern Monetary Theory (MMT). This matters for someone like myself, who is attempting to write primers explaining various economic concepts. However, for many followers of MMT, the more important question is how its theoretical concepts get translated into a new policy framework. This raises the question of political economy, which I largely shy away from. However, I will offer my eccentric political views on the future of MMT.

Where MMT is a Huge Success

The title of this piece refers to an article by Professor Simon Wren-Lewis: "Why is MMT so popular?" That article is interesting, and I largely agree with Alexander Douglas' response.*

He argues that MMT is popular because of dissatisfaction with the pseudo-Monetarist mainstream consensus ("the Consensus Assignment"). That argument may be true for some MMT-ers, but certainly not everyone. I came to MMT from a background as a market macro analyst, and I have met many people who have roughly the same story to tell. (There was a market practitioner session at the MMT Conference, and we compared notes.)

The answer for me is rather straightforward, and not particularly complimentary towards other schools of thought. The reality of financial market economics is that is a hodge-podge of jargons and random concepts taken from old mainstream textbooks and some central bank research. It sort of sounds like "mainstream economics," but it can only be described as a theoretical mess (that even Professor Simon Wren-Lewis complains about, I believe).** It is possible that some or even most bank economists have a coherent world view, rather it is the analysis in aggregate that is the problem.

Some market participants get tired of listening to analysis that is continuously wrong. Sooner or later, you start spending your free time trying to find an internally consistent economic theory. I am not going to assert that MMT is necessarily correct, but it is certainly not obviously wrong. And with all due respect to followers of other schools of economic thought, they are either in the "obviously wrong," "highly political," or "not well marketed" categories. As a result, MMT gains followers like myself by default. (In particular, I am pretty sure there are more followers of Austrian economics than MMT in the financial markets, but it is very political.)

Of course, there can be people from outside of finance who are interested in finding a coherent economic theory, and these people face the same set of options.

How Far Can This Spread?

As I discussed in my earlier article, with the addition of a MMT textbook, MMT will be much better positioned to capture its market share of people who are interested in economic theory. However, there is no reason to believe that everyone interested in economic theory will become MMT-ers. This is not enough people to qualify as a mass movement. To become a mass movement, MMT needs to be tied to a policy agenda.

From my perspective, there is some tension between the political objectives of MMT, as I see them.

I would put the objectives into three categories.
  1. Change the way that politicians and the media discuss fiscal policy. The phrase "the government budget is like a household budget" would never be heard again.
  2. Make technical changes to the way that budgets are analysed, or economic data collected.
  3. Implement policy changes, in particular the Job Guarantee.
This is where my peculiar political background comes in. I am a Western Canadian Prairie Populist (even though the last time I lived on the prairies was in the mid-1980s). To give a quick summary, Canada's western provinces were explicitly managed as colonies to benefit the established eastern provinces -- Ontario and Québec. Periodically, the mismanagement by the Eastern establishment triggered broad-based popular revolts. Protest parties were formed, and reforms demanded.

From this perspective, the first objective listed above (change how the establishment discusses economics) is questionable. To translate it into the historical perspective, it would be akin to launching a campaign to stop Eastern Canadian politicians from being tribal and corrupt. The prairie populist movements were dominated by religious and charitable people, but even they were not that charitable. Concrete policy objectives trumped nebulous hopes of moral or intellectual reform of the governing elites.

Imagine that you are a political hack for the Democratic Party, and that a Republican President brings forth a tax cut that is almost entirely aimed at the ultra-rich. The sensible economic analysis is that such a move will have almost no economic impact. The tax cut would be largely saved, so there would be no economic stimulus or inflationary impact. All that would happen is that government debt levels would rise, matching the increased wealth of the upper income tiers. This is not a promising analysis to lead the charge against the tax cut. Instead, the most effective strategy is to just turn your brain off and scream: "We cannot afford this tax cut! Our debt levels will be unsustainable! The bond market vigilantes will come for us in our sleep!"

The second objective -- make technical changes to how policy is conducted -- is less problematic. It would be nice to have a more sensible policy-making environment, but that environment reflects political realities. Furthermore, it is going to be hard to simultaneously push for partisan political goals (the third objective) and make technical changes to operating procedures. The existing bureaucracies are likely to resent being hit on both those fronts. Also, such technical changes are unlikely to arouse great popular interest.

If your objective is to enact policy changes, you need to be able to form some political alliances. The Canadian prairie populist movements were alliances of groups with wildly different politics; the East-West dividing line trumped class differences. Since there was no way that Westerners could form a permanent Federal ruling coalition, these alliances would always be temporary. In which case -- why care what the other coalition members think about economics? The less you talk about theory, the better.

For example, if I were an American and wanted to get the Job Guarantee enacted there, I would think long and hard how to get at least some Republicans to support the idea. The marketing is going to be quite different than that aimed at progressive law students.

Concluding Remarks

My main interest in Modern Monetary Theory is theory for the sake of theory. This puts me into a small niche. Success in policy reforms is more important for wider popularity.


* There were two parts of Wren-Lewis' arguments that were curious. Alexander Douglas addressed them, but I will as well in case the reader has not read Douglas' article. Wren-Lewis complained that primers aimed at mass audiences did not include equations. If I wrote a primer discussing robust control theory aimed at a mass audience, it would not include equations. That does not imply that robust control theory is mathematics-free. The other issue was his statement "you will hardly ever see the government’s budget constraint which makes all such semantics seem silly." MMT-ers regularly argue that the infinite horizon governmental budget constraint does not exist (only real constraints exist), hence it would not be discussed in a primer. If Professor Wren-Lewis is referring to the one period accounting identity, all transactions are assumed to clear simultaneously, which is a step away from operational reality.

** Readers who disagree with that assertion are free to attempt the following exercise. Somehow get your hands on a few years' worth of bank economist research reports, and try to build a textbook out of the principles behind them. Good luck with that.

(c) Brian Romanchuk 2017


  1. I think MMT analysis could improve the argument for not building tax reform on tax breaks for the ultra-rich. Politics of envy and economics based on false premises ("we can't afford it!") are going nowhere. The 'sensible' argument, as you describe it, that the tax cuts will have no economic impact, is simply not true from a macroeconomic perspective, and if MMT can't show this, then it is not worth the salt in the blood-stream of its adherents.

    1. Besides increasing inequality (and increased government debt levels/deficit), what would the tax cut do?

      When top marginal income tax rates are greater than 50% (which is about where they are in my home province...) I can see some merit to supply side arguments, but the United States is currently nowhere near there.

    2. MMT analysis, at least historically, is not so far from John C's. MMTers were and are far more critical of the Clinton surpluses than the Bush tax cuts. Bush's big tax cuts (& post 911 military spending) certainly did have a major economic impact - they delayed the onset of the 2008 crash and limited the depth of its effect somewhat. Of course such tax cuts on the rich give the least economic bang for the deficit buck, but there were a lot of bucks.

  2. "In which case -- why care what the other coalition members think about economics? The less you talk about theory, the better."

    You may not talk about theory to the public in general, as a propaganda strategy. Things may get too technical and complex for the masses, so you may simplify and even avoid the theory.

    But the theory is important for the political and economic leadership, including academics.

    Today, a lot of economists and politicians do believe that austerity is necessary - contrary to usual MMT claim that everything is some kind of conspiracy made by the elite.

    People do make political decisions using the theory. If the theory is wrong, it would be better for everyone if it was revised and corrected.

    1. As I have noted, I am writing based on my (unusual) political perspective.

      The prairie populist movements were a mixture of Communists, crypto-fascists, believers in Social Credit, social democrats, and the highly religious. There is no way you wanted to discuss theory with your coalition members. As long as they voted the way you wanted on your policy agenda, you kept your mouth shut...

  3. The Douglas response offers the equation

    (1) G-T = θ + β

    as a description of the government budget restraint.

    My version of MMT would re-write the equation to read

    (1) G = θ + β +T

    where-in I use the same definitions as Douglas [G = government spending, T = net tax revenues after transfers and interest payments, θ = amount of budget deficit financed by issuing high-powered (sovereign) money, and β = amount of budget deficit financed by selling government bonds.]

    You can understand my version of MMT by recognizing that government interacts with the private sector with the result that the private economy receives the money that government spends.

    I think most MMT'ers would look at term θ as the source of money to fund the Job Guarantee policy. It certainly could do that but we can see from the equation that the amount of money in circulation would increase until the practice stopped.

    Of course, government could also borrow from the public (term β) but (logically) could only borrow from the private sector to the extent it had funds (cash funds) to lend. Poor people could not be expected to loan to the government.

    Yes, government could tax to fund the Job Guarantee but that would be counter to efforts to increase spending and employment. However, without taxing, there is no effective removal of money from the system [does borrowing from the public REALLY remove money from the system? Or does borrowing (from the public) only recycle liquidity (cash or deposits)?

    Personally, I am not yet quite ready to acknowledge that MMT has met my mechanical money standard as a coherent theory. Our model must answer every question asked with logical, plausible, linkage.

  4. As long as MMTers do not give anything more than lip service to the inflationary constraint, they will never be taken seriously as providing a basis for policy-making. Cullen Roche has at least raised the issue with unproductive expenditures of 'ponies for everyone' and others have raised the unproductive expenditure of tax cuts for the very rich who will just trade their windfalls in for more T-bills. But other than that, when one raises the question of how to implement the inflationary constraint to counter the Right's cutting every tax they can find and the Left throwing money at every problem they can find, the typical MMTer response is "not my job, man." Some MMTers go a tad farther and assure us that "the adults in the room will put on the brakes" - sure, and Donald Trump will stop tweeting.
    Maybe most people can't embrace the rationality of MMT, but maybe they intuitively know that something ISN'T quite right with "free money" or "free ponies" that goes beyond the typical MMTer snide remarks of irrational work ethics.
    Unless and until MMTers come up with a means to operationalize the inflationary constraint, lets thank the gods for federal budget deficit hysteria - its the ONLY constraint we currently got.

    1. That's my "objective #2" - make technical changes to budgeting procedures. Not an easy task; I do not have a framework in my back pocket. That said, there's plenty of obviously stupid things to avoid, and many existing and proposed budgeting frameworks fall under the "extremely stupid" category. They were put into place as the result of bad decisions, and those decisions can be reversed.

      In the current context, the inflationary constraint is extremely weak. For example, we could create a lot of low-paying jobs to drive down unemployment and underemployment, and the inflationary impact might be quite small. Admittedly, that's a guess on my part, but it explains why someone like myself does not spend a lot of time wriitng about inflation risks for any politically plausible fiscal policy. Hypothetically, I can imagine potentialinflationary fiscal policies, but I do not see much chance of such policies being enacted on any reasonable horizon.

      I also disagree with your assessment about the deficit hysteria. When it comes to defense spending in the United States, there's a magical money tree growing in the back yard. There's a lot of flying ponies available for all the branches of the armed services. The deficit hysteria only exists for selected programmes. It's not something to take seriously as an actual constraint.

  5. What do you make of Alexander Douglas' insistence that the order of operations does matter (his response to SW-L's frustration with MMT's assertion that "spending comes first").

    I've never really understood why MMT thinks this is so important, or even quite what it means. The state is continually both taxing and spending - it doesn't seem to me to be very meaningful to talk about one or other coming first. Whether we take a DSGE model or an SFC model, we have a certain amount of spending and a certain amount of taxation within a given period. Can it make such a difference what order those occur in within the period?

    In relation to private entities, it might make a difference if those entities are credit constrained. But mainstream models do not generally treat the state as credit constrained, any more than heterodox models do.

    1. (I did discuss this in "Understanding Government Finance"; the order of operations is ambiguous.)

      However, there is a difference between a hypothetical model that captures all intra-day transactions, and models that only have simultaneous end-of-period clearing (like SFC and DSGE models). Under simultaneous clearing, default is essentially impossible, and the mainstream story about the infinite period governmental budget constraint being microfounded is obviously wrong. That's the crux of the MMT vs. DSGE debate, in my view.

      In order to discuss default, we need to walk away from simultaneous clearing, and think about what sequence of intraday transactions could lead to default. The MMT operations analysis suggests that default would be essentially a voluntary decision of the government.

      There's two different angles to this topic:

      (1) We need to understand government operations to understand why governmental finance is not like household finance, which is what most MMT primers are discussing.

      (2) Why the infinite horizon governmental budget constraint in DSGE models is a load of malarky. That is the subject of a few technical MMT/heterodox articles, and is a discussion that some critics apparently are completely unaware of. Since that constraint is a load of malarky, this discussion should not appear in a primer aimed at the general public. (I see no reason to distinguish between the "non-MMT heterodox"/"MMT" camps in that debate.)

    2. Obviously operations get complicated when the govt writes a silly law restricting itself from having a negative balance in its account at its own CB and tells its CB it can't lend to it. What order things go in from a micro view can be argued about back in forth in that context if that's all you are considering.

      The point, though, is to change the narrative and the starting point of analysis. If it's reserve balances that are used to pay taxes and purchase govt bonds, where does that come from? And if the only way to obtain those reserve balances is from the CB lending to the pvt sector or the CB purchasing a Tsy left over from a previous deficit, who is the one in charge in that case--the bond mkts or the govt?

      A lot of non-MMTers get caught up in MMT's discussions of details, thinking the details of a particular nation's operations under current operating procedures is the be-all end-all point. Yes, the operations are important for many things. But this particular point is about the hierarchy of power more than operational details--who calls the shots--and changing the narrative. Once you understand who calls the shots, then the bigger picture of what's happening within the details becomes clear regardless of self-imposed constraints (and you come to understand which "constraints" aren't really constraints). Increasing numbers of legal scholars are backing us up on this. Historians and numismatists have been doing so for decades already.

    3. I always learn something from comments of STF. Today it is what numismatists means. Numismatists are not people who mate with numbers (which I thought might be the meaning)- they are people who study coins and money. So although it is not as intriguing as I originally thought, there may still be something numinous (which apparently is not even close to the same meaning) about numismatists which makes me happy they back MMT to some extent. And it reminds me how happy I am to have this big old dictionary here.

    4. There's also a serious point to do with dynamics.

      If you add to the system before you drain then there is more money chasing the drain. That lowers the price.

      And that is how intra day works. Intra day all the balance sheets are blown up as transactions fly around and then they shrink at the end of the day clearing.

      So the large to small view (spend then tax) has a lower price than the small to large view (tax then spend).

    5. I often distil this down to "who can say no and stay in their job?".

      From what I can tell mainstream economics is largely about implying something without actually declaring it, so that when it is pointed out they can claim fake innocence.

      You'll find, for example, that SWL won't ever talk about the fiscal automatic stabilisers - particularly the Job Guarantee. Because they blow his entire argument about why fiscal policy isn't very good out of the water.

    6. The following is a 24 page presentation about the Fed's role as credit provider:

      Regarding operational realities Fed provides banks, but not the US Treasury, with overdraft services in federal funds. The day time or overnight borrowers of fed funds must be considered solvent and can therefore usually post "good collateral".

      Fed almost certainly extends overdraft credit to banks or other private sector dealers to ensure the success of an auction of Treasury securities. So what if Treasury could get a direct overdraft in its Fed account? Then there would be excess reserves in the aggregate bank driving down the policy rate of interest in federal funds. Either Fed would pay interest on excess reserves or Treasury would sell securities to support the fed funds rate.

  6. Thank you Joe for that very good link. I have saved it and will use it often in my own arguments. Here it is again for anyone who thinks the money supply is endogenous.

  7. I'm not sure what you mean by adding the phrase "for anyone who thinks the money supply is endogenous." I don't care whether a social custom or process is called "exogenous" or "endogenous" since such adjectives do not alter the social customs. However, when trying to understand how others apply these terms, it becomes necessary to speculate about what they think it means, since it is usually not defined in context.

    Regarding flows and levels of financial instruments in the aggregate economy there are three processes that are always interacting: fiscal policy, monetary policy, and credit markets. Foreign exchange could be added as another process, and offshore banking (called eurobanking) can be yet another process. All of these processes are operations in the aggregate bank balance sheets of a nation or so-called currency region. Due to the accounting customs I apply a dictum "What happens in Vegas stays in Vegas - where Vegas == aggregate bank balance sheet." So one must see how deal flow in government and markets causes dynamics in the aggregate bank balance sheets and what social implications flow from government policy and regulations in this context? Not easy to do.

    To me the adjective "exogenous" simply refers to a subset of such human customs as "externally generated" and the adjective "endogenous" simply refers to a set of customs that are considered internal or inherent to the social process.

    Congress makes fiscal policy, Fed makes monetary policy, credit markets do their thing, and Treasury increases federal debt to cover a deficit or reduces outstanding federal debt to dispose of a surplus.

    When MMT makes a statement about operational order of spending must go before taxing or borrowing it meets resistance from others who see that Treasury is not allowed to run an overdraft in its Treasury General Account at Fed. This means Treasury must tax or borrow to keep TGA balance positive and so MMT contradicts an operational reality. In a system where the depository institutions (banks) and Primary Dealers profit in general from "being in bed with" the Fed/Treasury battery then Fed can lend to banks and other dealers who purchase Treasuries at auction and Treasury does not have to run an overdraft. If Treasury runs an overdraft then excess reserves pile up in banks and Fed has to either let interest rates fall or pay interest on a portion of its liabilities. The consolidated Fed/Treasury battery cannot run out of federal funds and securities, but it must pay interest on liabilities to support monetary policy, so MMT is right but it only describes the system constraints that have already been demonstrated and this does not mean MMT is right about setting a tax policy and job guarantee via Congress that would sustain full voluntary employment and control inflation. Minsky discussed the difference between government programs that drive inflation and taxes that curb inflation - I wish he were around to ask about such matters since no one seems to be pursuing his ideas on such topics in a policy think tank to my knowledge.

    1. Joe Leote, perhaps I was using the word endogenous in the wrong way- it is not a word I use often in daily life.

      The link you provided seems to show to me that the Fed, the central bank of the US, is willing to provide US dollars to banks when they decide they need them using collateral that the banks themselves have created in the form of previous loans. At least, that is what I understand from that publication from the NY Fed. So where I am is that is true that banks create money through loans and the central bank, which everyone acknowledges can and does issue money, is willing to back them up.

      So there isn't some 'amount' of money that we all compete for that is predetermined (by something or someone) that the economy (us) is stuck with using. Which would be my understanding of money being 'exogenous'. That could be wrong, and if so I would like to find out why, sooner rather than later.

      Either way, 'endogenous' and 'exogenous' are very poor words for me to use when trying to explain my understandings, such as they are, to other people. Especially since I am not clear on their meanings myself :)

    2. I don't fully understand how the different economists use the terms "endogenous" and "exogenous". I don't regard your interpretation as "wrong" I just don't track who applies that definition - it sounds like a position I have read in economic literature.

      L. Randall Wray says how government operates in approximately the first 20 minutes of the video under the following link. Then he spends a few minutes saying how the government should operate under functional finance. Then he goes back to describing how the central bank and Treasury do operate. Then he discusses Euro system. Then he summarizes. The talk by Wray takes about 32 minutes with Q&A of 10 minutes in which Wray can't specify why Fed/Treasury "has to" pay interest on fed funds or government securities other than "tradition".

      So when MMT advocate makes such remark it basically asserts we can get rid of monetary policy and use only fiscal policy - but how do we get Congress to set a spending, credit, and tax program that is never inflationary or deflationary and that can be effected by Treasury without any independent decisions of the Fed? There were governments without central banks in the past but those governments either spent paper currency, usually with inflation to finance war, or sold securities that paid interest.

      The second part of this talk should be a very good discussion of the history of money and debt by Michael Hudson - I will listen to it later.

    3. "but how do we get Congress to set a spending, credit, and tax program that is never inflationary or deflationary and that can be effected by Treasury without any independent decisions of the Fed? "

      You implement a Job Guarantee programme. That automatically increases and decreases government spending as required, spatially distributed across the nation to the areas that need it and instantly applied.

      MMT stablises via enhanced automatic stabilisers.

      The central bank then just clears the Job Guarantee bill on a pull basis. There is no discretionary push that has to be 'tuned' on any timescale shorter than a year. In fact, if you want to, you can require that the discretionary stuff is 'balanced budget'.

      The discretionary budget then determines the 'size of government' via the usual argy-bargy of politics. That determines the 'must happen whatever' public provision. Then the private sector gets to play with what is left, and anything left over they don't want to play with is picked up by the Job Guarantee which then creates the 'nice to have' public provision.

      What is interesting about Wren-Lewis, as well as your comment, is this seeming inability to see the automatic stabilisers (particularly the spend side auto-stabilisers) - which require *no* human beings involved at all. They just work - instantly, spatially, automatically.

    4. Neil - your point about the psychology of automatic stabilisers is good. The conventional view of fiscal policy is that it has to be stop-go policies aimed at fine tuning growth. Minsky criticised this at the time, and he considered himself a Keynesian. Automatic stabilisers eliminate the need for the economics intelligentsia, so they are ignored.

    5. Here is a 41 page paper - Minsky and Economic Policy: “Keynesianism” All Over Again?

      Quote (pg 17): "Thus, a big government ... can contribute to inflation by raising the wage of its employees too fast, by increasing spending too fast on goods and services, and by having a tax structure that contributes to a fast increase in corporate income taxes. Transfer payments may create additional inflationary pressures by growing disposable wage income and disposable net profit too fast. In addition, a big government in which political discretion in economic issues is high may prevent gG and gT to move up or down when they need to in
      order to prevent inflation or recession. Finally, if the government mainly focuses its spending defense and other unproductive expenditures, there is a greater chance that a big government will be inflationary ..."

      So failure of government to control the credit system could drive inflation, and discretionary spend/tax policies could drive inflation, even in a system with a Job Guarantee Program, and I think Minsky requires a central bank that can kill an inflation via inducing a recession (which would shift the burden of providing jobs to the otherwise unemployed to the JGP). If central bank does not control the interest rate at which banks and other financial intermediaries must refinance their short term liabilities then it cannot jack up this rate to kill credit-fueled inflation. So I think Minsky wants more fiscal stabilizers and a central bank with some discretion and monetary policy tools - I may be missing something though.

  8. This comment has been removed by a blog administrator.

  9. This comment has been removed by a blog administrator.


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.