The following provides a brief critique of a relatively new monetary theory called “Modern Monetary Theory” also known as MMT.
MMT has become an increasingly popular economic theory in some left-wing circles in recent years in part due to its eye opening description of public finances. I am sympathetic to some of the views espoused by MMT and I would argue that they have some useful insights. I’ve written about how MMT is an important theory to understand, but I would also argue that MMT takes a basic Post-Keynesian & Marxian framework and adds a bunch of “modern” understandings that result in inconsistencies and flaws. I think they use these inconsistencies to try to form a “new paradigm” for public finances that misconstrues reality. In my opinion, this new paradigm, which can appear accurate at first blush, is unnecessary and results in some overreach in the areas describing the monetary system. The following 4 critiques (as well as my own brief comments below) are useful and elaborate on those inconsistencies.
- MMT, The Emperor Still Has No Clothes, Thomas Palley
- The Monetary and Fiscal Nexus of NeoChartalism, by Marc Lavoie
- MMT and the Real World Accounting of 1-1<0, by Brett Feibiger
- A Critique of MMT, by Steve Waldman
Below is a version of what I believe are the primary inconsistencies and problems in MMT. This is a brief rebuttal, but covers the main points of confusion. I don’t necessarily disagree with most of MMT”s policy ideas, but I do believe their rationale for arriving at these ideas are based on misunderstandings. For that reason I’ve focused on points of operational errors where MMT misunderstands operational realities of the monetary system. This hopefully results in an objective analysis as opposed to a politically biased policy focus.
If you’re feeling too lazy to read this in its entirety then the summary of MMT’s flaws can be boiled down to a simple point:
MMT is a state centric view of the economy that tries to create an argument for why the government needs to intervene in the economy for the purposes of taxation, full employment and financial stability. Many of these views are inconsistent with a capitalist economy and are formed from a misleading operational perspective of the monetary system. While it is true that government can be a useful facilitating force in the economy MMT has a tendency to overreach and misrepresent the degree to which this is necessary and helpful. Read on for a more thorough argument for why this is misleading at best and wrong at worst. I hope you find it helpful.
MMT Claim #1: Unemployment is caused by the government.
Reality: Unemployment is caused by a lack of private investment.
The purpose of this claim is to create the illusion that the government needs to spend “net financial assets” to create full employment. While the government can help create full employment this is a case of overreach in the MMT paradigm based on an operational misunderstanding. In reality, what MMT is doing is making a moral argument for full employment that is masquerading as a poorly constructed economic argument for full employment.
MMT advocates often claim to be doing little more than describing how the financial system actually works, but as you’ll notice in this critique they take certain liberties in that description that results in mistakes. Importantly, MMT’s description is directly tied to their prescriptions so the two are inextricably linked (and partially wrong in my view).
MMT’s view on unemployment is one problematic element of their “description” of the financial system. MMT describes the financial system as being akin to the government issuing business cards in a room full of people with a government enforcer at the door who requires everyone to obtain those cards or be shot. If you don’t obtain cards then you are unemployed. Therefore, by imposing this system on people they have caused unemployment. MMT argues that, because the government created the monetary system that only they can fix the problem that they cause. Ie, their description is explicitly tied to a prescription (in MMT this is the Job Guarantee program which I’ll discuss later).
That seems intuitively appealing, right? The government imposes rules and as the monopoly supplier of money they have to spend some money and supply employment before we can be employed. But you have to think a little deeper about causality here – which party actually causes the unemployment? After all, if you have no useful skills then you’re not unemployed because the government imposed the monetary system on you. You are unemployed because you can’t sell your labor to other people. In a capitalist economy with an endogenous money system (a money system where money is denominated in state currency, but can be created by anyone, mainly banks) this is precisely why there is some level of unemployment. Or as Keynes said long ago, there is a lack of private investment. MMT tries to reinvent the wheel and argue that it is the government’s fault (and implicitly, the rest of society’s fault) that you can’t find a job. This appears misleading at best and dangerously wrong at worst.
For instance, let’s use an example that MMT advocates sometimes refer to – groups of wild animals. Is there unemployment in species without money? MMT says no, because they don’t use money. In fact, there is. In a troop of chimpanzees, for instance, each member has a specific role. They all have a certain value to other members of the troop. They are essentially “employed” by this role in the troop. When one of them becomes a negative value add to the other members they are often removed from the troop or killed. The “full employment” we see in wild animals is a case of survivorship bias. But the important part of this narrative is, if these chimps were to start a monetary system with regulated rules they would not suddenly be unemployed. The only ones who would be unemployed would be the ones who couldn’t sell their labor to other members of the troop. Creating a monetary system doesn’t necessarily create unemployment. It just exposes who does or does not have skills they can sell at a price.
There’s a further substantive issue with this causal claim – the US court system has never ruled that this is a valid case. In fact, they’ve ruled against workers making this case many times. Therefore, the MMT argument is not just theoretically and descriptively wrong, at present time, this argument is legally wrong. Of course, even a change in the legal standard would not change the fact that the government does not cause unemployment in the first place.
MMT states the the government is the “currency monopolist”. And by establishing this currency they claim that unemployment results. Specifically, by imposing property rights, regulations and taxes the government forces everyone to obtain an income denominated in the unit of account and if they cannot obtain an income in this system then it is implicitly the government’s fault. But this is an extreme view.
The government establishes the unit of account (the Dollar, Yen, Euro, etc). The private sector can then create financial assets denominated in that unit of account. In the modern monetary system the government essentially licenses the creation of money out to private banks by issuing banking licenses and other firms create financial assets denominated in the national unit of account that are regulated by the state. Whether or not there are enough of those assets and income distributed across the economy to create full employment is not the government’s fault.
Imagine that a cell phone (like money) is an essential technology in modern life and that the government required everyone to have cell phones and licensed the creation of cell phones out to the private sector to create (as they do with money and banks). Then, imagine that those firms failed to supply a cell phone to you because they did not believe you were responsible enough to use the phone (as many banks do with loans). Did the government cause this lack of supply in phones? Of course not. The lack of supply was caused by private firms and by your lack of credibility. This is precisely what happens in a modern economy when money is created and distributed unevenly.
Now, MMT people often respond to this by saying “well, the government created a system that doesn’t supply jobs & income to everyone”. That’s true. The USA is a capitalist economy with a Constitutional Republic. In a capitalist economy you obtain an income and job by proving to other members of society that you have a valuable skill. The point is, the cause of someone’s unemployment is not necessarily the government. It is often (not always) the result of a skills mismatch. In other words, it is a lack of investment in obtaining a valuable skill that can be sold to employers. The ultimate responsibility for obtaining valuable skills that can be sold to other people lies with us, not the government. MMT gets the causality backwards here by starting with the state and working out.
MMT is unclear on this topic at times. Some of them claim regulations and property rights cause unemployment. Others claim taxes cause unemployment. Warren Mosler, MMT’s founder says it is a lack of net financial assets:
“Involuntary unemployment is evidence that the desired H(nfa) of the private sector exceeds theactual H(nfa) allowed by government fiscal policy.
To be blunt, involuntary unemployment exists because the federal budget deficit is too small.”
Ignoring the moving goal posts, these concepts are badly confused. As JM Keynes described in the General Theory it is actually a lack of private investment that causes unemployment:
“The outline of our theory can be expressed as follows. When employment increases, aggregate real income is increased. The psychology of the community is such that when aggregate real income is increased aggregate consumption is increased, but not by so much as income. Hence employers would make a loss if the whole of the increased employment were to be devoted to satisfying the increased demand for immediate consumption. Thus, to justify any given amount of employment there must be an amount of current investment sufficient to absorb the excess of total output over what the community chooses to consume when employment is at the given level. For unless there is this amount of investment, the receipts of the entrepreneurs will be less than is required to induce them to offer the given amount of employment. It follows, therefore, that, given what we shall call the community’s propensity to consume, the equilibrium level of employment, . . . will depend on the amount of current investment. The amount of current investment will depend, in turn, on what we shall call the inducement to invest; and [this] will . . . depend on the relation between the schedule of the marginal efficiency of capital and the complex of rates of interest.”
In essence, unemployment results from a lack of private investment and the refusal of capitalists to reduce wages to the extent that laborers will accept the wage rate. This makes sense given that capitalists are natural profit hoarders and risk managers. Capitalists will rarely spend enough into the economy to provide for full employment because they retain profits in the form of a risk managing capital buffer. In fact, I would argue that the idea of zero involuntary unemployment is at odds with the natural profit seeking goal of capitalism. But that is the system we have chosen and although it is imperfect it’s safe to say that capitalism is the worst economic system except for all the rest.
The key point though is Keynes settled all of this long ago with the paradox of thrift. It is true, if the government saves more than it spends (taxes more than it spends) then it will create an income shortage in other sectors of the economy. If capitalists save too much then some unemployment will result. This isn’t a new MMT insight and it certainly doesn’t mean deficits are necessary. It also doesn’t necessarily mean it is bad for the economy. It might be immoral and hurtful for a few, but it does not necessarily mean it is a net negative for society.
But more importantly, this has nothing to do with the budget deficit. This is nothing more than an accounting error within MMT wherein they view private sector saving as (S-I) or saving net of investment. MMT has a long history of stating that private sector saving is equivalent to the size of the deficit (S-I) = (G-T). Specifically, they say:
“To briefly summarize, at New Economic Perspectives, we prefer to use the Godley sectoral balance approach, where he defined private sector saving as “net accumulation of financial assets” (NAFA), using the flow of funds data.” – LR Wray
There’s no way to parse this nicely. This. Is. Extremely. Wrong. Private sector saving is S. S is equal to I. (S-I) is the private domestic balance. Net financial assets is equal to (G-T). Saving (S) does NOT equal (G-T). It’s rather flabbergasting that Wray makes this mistake.
As mentioned above, the purpose of this view is to create the illusion that the government “must” issue net financial assets in order for the private sector to net save. MMTers have a long history of stating that the government “must” be in deficit in order for the private sector to net save (see here, here & here for instance). More recently, some MMT advocates have rectified this flaw and tried to argue that they don’t really mean NFA is “necessary”, but this doesn’t mesh with the many other places where they state that government spending is necessary (for instance, in the case of solving unemployment).
By defining “net saving” as (S-I) MMT confuses traditional economics which defines net saving as net disposable income less final consumption expenditure. When MMTers talk about “net financial assets” and “net saving” they are referring to financial savings net of domestic real investment by the private sector. It is not remotely close to representing private savings. By netting this huge quantity of private sector assets they are netting most of the cornerstone assets in our economy such as real estate, corporate stocks, corporate bonds, etc. When we think of “saving” in this sense we marginalize the two most important sources of private savings:
- Surplus from investment.
- Market value of existing assets.
By valuing all assets at book value MMT leads people to believe that the private sector cannot net save against itself. This view is misleading in that it diminishes the importance of the market value of all outstanding financial assets by valuing them at book value. This marginalizes and ignores the revaluation of assets that is reflected in the way the market values all of the innovations and output that make life worth living. More importantly, it marginalizes the most important form of savings that exists in the economy – the trillions of dollars in mark-to-market balances in everyone’s 401ks, pensions, IRAs, etc.
Further, the MMT view that a deficit is a sign of the private sector’s desire for net savings is misleading. Of course the private sector has a high desire for net saving. People always want more money.
The MMT view is a useful concept for understanding the importance of risk free assets and especially for discussing the value of government liabilities relative to the economy, but MMT has a tendency to take this conversation out of context and imply that the private sector is producing assets that are necessarily less safe or less important to the monetary system than government issued assets. For instance, MMTers sometimes say things like “the national debt is the equity that supports the entire global credit structure”. This is problematic for several reasons:
- Government debt is, within the scope of the “global financial system”, an asset and liability that nets to zero. They generally depict this relationship through a three sector model of the economy which depicts the government’s liabilities as the private sector’s assets. While this is true in the scope of the three sector model it is misleading in an aggregate sense. Government debt is not properly referred to as “equity” because government debt is an instrument that the aggregate private sector must finance and pay interest on.
- This idea implies that government debt is the backbone of the economy when this cannot be logically true. After all, government debt is financed by private sector income. Private resources necessarily precede taxes. Without a highly productive revenue generating private sector there is nothing special about the assets created by a government and it is literally impossible for these assets to remain valuable. But this again looks like a basic accounting error within MMT. Equity is the residual of assets and liabilities. We create equity when we produce real goods and services or increase the market value of our assets relative to their liabilities via productive output. It is completely illogical and beyond silly to argue that one can just “print” equity from thin air. Government debt is, logically, a liability of the society that creates it. In the aggregate government debt is a liability that must be financed by the productive output of that society. This accounting error results in another internal accounting inconsistency in which reality does not match their theory.
- This is inconsistent with MMT’s view on endogenous money. MMT is generally very good in describing banking and they rightly note that banks do not “fund” their loan issuance with existing money. Instead, they create new money from thin air by expanding their balance sheet based on the capital they have. But this is also true of the government. The government expands its balance sheet endogenously because it is viewed as a creditworthy entity. And it is viewed as a creditworthy entity because it can tax a large viable private sector revenue stream. So MMT has this exactly backwards – a government obtains its financial strength by leveraging the capital of its private sector, not the other way around.
- This is the money multiplier by another name. MMT says that the private sector “leverages” state money to create other forms of money. This is a strange view for a school that understands endogenous money creation because it is entirely inconsistent with this idea. In fact, if you view T-Bonds as a form of reserves then MMT is not that much different than old fashioned Monetarism and the way they claim that banks multiply reserves when they make loans. Both of these schools are problematic in that they work from the state out to the private sector instead of recognizing that it is the state that leverages its powers from the private sector. In the case of MMT this is common across many of their narratives – they claim the government spends first and taxes second thereby supplying the private sector with the initial money (the “real money” in their view) that then allows them to leverage it. They claim that the private sector can only “net save” if the government provides them with “net financial assets”. All of this is a strange loanable funds type of world where the government must supply financial assets in order for the private sector to operate when in fact the private sector supplies itself with endogenously created financial assets. It appears as though MMTers confuse the concept of establishing a national currency (a national unit of account) with the idea of having to supply the actual financial assets. Said differently, MMT not only thinks that the government dictates the measuring system we must all use, but that they must supply the rulers before we can ever create rulers of our own. It is true that the government establishes the unit of measure that we will all use, but it is completely wrong to say that the government must supply the rulers before we can create our own rulers denominated in those units of measurement.
None of this means government debt and deficits aren’t important or that they can’t support the private sector economy. But understanding the context here and the relationship is of special import when understanding the economy. It would be more accurate to say that private sector net worth is the equity that supports the domestic economy and government debt can, at times, serve as a useful insurance instrument that supports that equity.
MMT Claim #2: A sovereign currency issuer need not fund itself via taxes or bond sales.
Reality: Any endogenous issuer of money must “fund” itself.
The purpose of this claim is to create the illusion that the government needs to spend financial assets into the economy before we can ever pay taxes. This claim also tries to justify the idea that government spending is “self financing”. While the government can issue money to pay taxes this is by no means necessary. This is a case of overreach in the MMT paradigm based on an operational misunderstanding and there is no need to reinvent the wheel here.
This is a strange case of circular logic in MMT. In the MMT world the government is not constrained by funding because it can always print money. MMT instead says that government is only constrained by real resources. But taxes are how the government moves real resources from the private sector to public sector. If there are no real resources to tax then the government is constrained. In other words, taxes constrain spending and fund the government’s ability to spend.
Of course, what MMT is really trying to say is that the government does not need to tax, dollar for dollar, to spend. It can spend endogenously on credit. This is true. But it is true for the rest of us as well. The government just happens to have a really flexible line of credit.
Let’s use a really simple example to debunk this idea that the government doesn’t have a nominal funding constraint. Say the government sets a budget in 2020 of $100 with the plans to buy $100 worth of goods. Then, the very next day the rate of inflation goes up by 100%. The government can now only buy half as many goods as they could before. The government has suffered a significant increase in its cost of funding to buy those goods and in order to now be able to afford those goods they must print an extra $100 into the economy. In other words, the government’s real constraint becomes a nominal constraint in a high inflation environment.
Importantly, the MMT idea of funding is actually inconsistent with the theory of endogenous money. In an endogenous money system anyone can issue money denominated in certain state based units of account, but no one can force others to hold their liabilities, including governments. But the mistake MMT makes appears to be a basic misunderstanding of finance and the concept of “funding”.
To have funding means to procure financing. This logically involves a counterparty since there are two sides to all transactions. In an endogenous money system anyone can create money, but for that money to be viable you need a viable counterparty. For instance, I can walk into a bank and create new money via a loan if the bank is a willing counterparty. If the government were to print money they would not have “funding” until they find a willing counterparty to hold that liability. The fact that the government can print money does not mean they can necessarily procure funding since they must still find a willing counterparty to hold that liability. For instance, during a high inflation the price of money is falling relative to everything else because the counterparties reprice the value of money. If this process turns into a hyperinflation it can result in a full blown collapse in counterparties in what is essentially a real insolvency.
Let’s use a simple example to communicate this point. For instance, governments presently finance their spending in deficit by issuing bonds. A government does not technically need to finance with bonds because they have a printing press. But let’s say they decided to scrap the bond issuance and just print cash. Would they have “funding” just because they can print the cash? NO. Being able to create financial assets does not mean you have “funding”. In order to have funding you need a viable counterparty at a real price. In the government example above the government will issue that cash to someone in the non-government sector at a real price. That real price is the “cost” of being able to finance that financial asset. Importantly, if that cash collapses in price (hyperinflation) the government does not reasonably have “funding”. Or, more accurately, the demand to finance their spending has collapsed.
A related and substantive error in MMT is their treatment of interest rates on government debt. MMT argues that a deficit causes rates to fall because the deficit floods the reserve system with excess reserves which banks try to lend out. Since they can’t lend them out in aggregate this drives the rate to zero. So the Fed has to respond. In the post-crisis era the Fed sets Interest On Reserves thereby establishing a floor under rates.
This is all correct in a narrow sense, but MMT misleads people on how this actually works and the impact of the deficit. The reason a deficit causes rates to drop (without Interest On Reserves) is because the Treasury uses an account at the Fed called the Treasury General Account. That account clears via reserves. So banks lose some reserves when a tax is paid and banks gain back reserves when government spends. But the deficit is not the actual cause of the reserve creation. The cause of the reserve creation is the result of the reserve system being a thing in the first place. In other words, the TGA is a RESULT of the reserve system and so spending is not the cause of driving rates down. After all, the only reason we have the TGA in the first place is because we have private banks and a reserve system. If there was no reserve system there would be no TGA and no need for reserves.
This point is worth emphasizing because it’s a crucial MMT mistake. It is not the deficit that puts downward pressure on rates. Non-interest bearing excess reserves (whether it is part of monetary or fiscal policy) always put downward pressure on rates because banks want to rid themselves of non-interest bearing reserves. This says absolutely nothing about the “natural rate” of interest on government debt and only tells us that profit seeking banks don’t like assets that don’t earn a profit.
This narrative that the government is “self financing” or doesn’t need tax revenue is also inconsistent with MMTs theory of inflation. MMT argues that the true constraint for a sovereign currency issuer is inflation. But inflation occurs when the demand for money declines and/or when spending outstrips output. In other words, governments go bankrupt in real terms when their financing agents lose demand for their money or when the income/output of the economy collapses relative to government spending. Since income is, by definition, a function of output, it is illogical to argue that a government doesn’t rely on income but that resources constrain government spending via inflation. Resources are the result of spending which logically creates income to be taxed. You can’t say resources constrain government spending via inflation, but that tax revenue does not.
Of course, governments don’t go bankrupt like households because they generally have a printing press they can resort to, but having a Central Bank buy your assets does not mean you have viable demand from the non-government sectors. When this demand for financing declines you will see a rise in overall inflation. Although the government has much more flexibility financing its spending the government is still dependent on viable revenue streams as a source for the underlying demand for their financial assets.
More importantly, MMT misunderstands that an inflation constraint is a nominal constraint. That is, when inflation is very high the government will be forced to reduce its spending or risk exacerbating the inflation. In periods of low inflation it appears as though the government has no nominal constraint because it operates with greater flexibility, but when inflation is high the nominal constraint is exposed and the government is forced to be more fiscally prudent or risk a form of real insolvency.
In essence, governments go bankrupt in real terms while households go bankrupt in nominal terms. So, when the demand for household money collapses you get a nominal insolvency. When the demand for government money collapses you get a high inflation. In either case, it is the demand from counterparties financing those assets that gives them credibility. When that demand collapses you are insolvent whether it is quantified in nominal or real terms.
MMT co-founder Bill Mitchell has said “taxpayers do not fund anything”. But this is like saying that my income does not fund my spending so long as I can find willing holders of my debt. When I enter a loan agreement with a bank I am entering into an agreement to create new money by finding a lender who will be my counterparty (a new loan creates new deposits). If I could perpetually find new lenders then I would have no need for an income. You could even say that my income doesn’t “fund” my spending because I can spend without having any income. This is obviously silly because output and income gives one credit. If a government could not tax some level of output then they would have no credibility. They would essentially be bankrupt because no one would hold the liabilities of an entity that has no chance of being able to redeem those liabilities in real terms since there is no output base upon which those liabilities can be given value. Although a sovereign government has unique powers it is not immune to the reality that it can run out of viable counterparties.
It should be clear that an entity with a printing press can print its own currency at will and won’t default or “run out of money”. This is neither a “modern” understanding nor a very useful one. What is important is why people might use that money and the parameters within which that money is viably “financed” and given credibility. In this sense, it is rather meaningless to say that the state doesn’t “fund” its spending because just like any other liability issuer it most certainly needs to find viable counterparties for its liabilities. The government’s money is considered viable when it has an underlying productive private sector that drives the demand for that money. This is similarly true for any private sector entity that has productive resources that give its assets credibility that creates demand for them.
The primary problem with this concept is that it misconstrues where money demand comes from. For instance, a corporation can issue short-term financial assets that are money-like. One could even say that they “spend” those assets into existence and then charge fees (the price of their goods and services) second. This would be especially true of a start-up firm which issues shares of stock without having yet proven that they can generate the cash flow to support present valuations. They are spending new financial assets to raise capital and will charge fees (the price of their output) that will hopefully make those shares viable in the future. The corporation has to issue these financial assets before it can charge the fees that will hopefully make them viable. While this might seem accurate in some sense the operational reality is that the firm is financing its issuance of new stock because it has credibility and willing holders of its liabilities.
Importantly, it is the capital base of a corporation that makes its assets holders confident enough to hold those assets. Capital gives firms credibility and this is the foundation upon which they are able to obtain and maintain willing holders of their assets. When that capital base deteriorates investors lose faith in the assets and they devalue them. Governments operate a little differently because their sources of capital are widespread across the tax base, but the same general concept applies. A government with broadly diverse sources of capital (like the USA) has enormous flexibility to expand its public sector whereas a small emerging economy like Greece operates with a much less diverse capital base which gives it far less flexibility to expand its public sector. Although a country like the USA has far greater flexibility finding willing holders of its liabilities we should not misconstrue this exorbitant privilege to mean that the USA has such a vast capital base that it does not even need funding or willing holders of their liabilities. That is a case of overreach.
The purpose of this narrative within MMT is to construct the view that the government creates its own demand for its money by charging taxes. MMT academic Stephanie Kelton has explicitly stated that the government’s spending is “self-financing”. This is essential to the State Theory of Money and the idea that the government is the monopoly issuer of money and the price setter. But this is an erroneous starting point. Money has value because it has credibility due to its underlying productive resources. The government is not immune to the fact that it must have underlying productive resources behind its money. Productive output must, by a matter of fact, precede taxes. Therefore, it is illogical (and contradictory to MMT’s own real resource constraint) to argue that government spending comes first and taxes come second when it must be true that productive output precedes taxation.
In fairness to MMT, the idea of government spending “first” and taxing second is not entirely wrong. Anyone who borrows in an endogenous money system is technically spending first (instead of spending from current income). When I use my credit card I am spending first and paying for it later. But this spending is still “funded” by a viable counterparty who has set the terms of my credit. Likewise, when the government spends they provide funds to a viable counterparty who sets the real price of that money. This counterparty does not have a gun to their head as MMT often implies. They have the choice to set the real price of the government’s newly issued funds. That funding (0-inf) is the “cost” of the government’s funding.
MMT should not confuse people with this “new paradigm” type of mentality which creates inconsistencies across the concept of endogenous money. Instead, MMT should remain consistent with the endogenous money school and explain that all money issuers need willing holders of their liabilities. In this sense, taxes do fund government spending in that productive income is what gives money credibility. A productive private sector generates the domestic product and income that gives government liabilities credibility in the first place.
In fairness, MMT is right that the true constraint for the currency issuer is not solvency, but inflation. And yes, the state tends to have greater solvency flexibility than a household or business. But this concept needs to be expressed within the confines of the fact that the state needs to be able to find willing holders of their liabilities. After all, it is rather meaningless to argue that a country like Zimbabwe did not run into a “solvency” constraint when it ran out of willing holders of their liabilities during their hyperinflation. That is, after all, all a hyperinflation is – investors are selling state denominated assets relative to other assets in much the same way that investors sell a stock down in value when it loses its credibility.
Then again MMT does not have a coherent theory of inflation to explain the inflation constraint. And this is what makes MMT somewhat dangerous. They give people the impression that a “sovereign” government can afford anything denominated in its own currency, but they cannot model the actual limit of that spending. This is an extremely dangerous view as the capital base of a country can erode for unknown reasons. Inflation can occur for unknown reasons. And without a coherent theory of inflation (ie, one that can accurately predict inflation) there is no way to control what MMT claims is the true constraint for government spending.
Standard MMT response to claim #2: “It’s not about funding, it’s about understanding that a currency issuer creates the money that is used to pay taxes which then destroys money. Reserves drains require reserves adds first. Therefore, a government cannot fund itself when it is the creator of money.”
CR Response: The idea that a “reserve drain requires a reserve add” is a comment you will frequently see from MMT people. The basic idea is that the government must create reserves before they can remove reserves. But when one understands why the reserve system even exists (to meet bank deposit reserve requirements and settle interbank payments across multiple banks) you understand that deposits NECESSARILY precede reserves. The only reason reserves even exist is because the deposit system has a layer of interbank settlement within it that services the deposit system. It is an operational error of understanding to state that reserves precede deposits. As is common in MMT they work from the government out to the private sector instead of understanding that the government facilitates the private sector. The result is an operational misunderstanding and error in the flow of funds.
Importantly, this is a rather basic mistake in MMT. The modern monetary system is not set-up in such a way that the US government issues money directly. Banks issue most of the money in the monetary system and the reserve system is used to transfer funds within the banking system. Cash and coins are also facilitating forms of money that help someone with a bank account transact. MMT says that you can consolidate the Fed into the Treasury to show that taxation results in the government obtaining its own liabilities (reserves where they settle). Thus, MMT says this is where money is destroyed and created. However, this is not meaningful in any real sense since the govt is not destroying a deposit liability, but instead transferring it from one bank to another when it taxes and spends. You cannot consolidate the Fed into the Treasury and still claim to be explaining reality whereby reserve issuance is the de-facto spending. This is entirely incoherent. The Fed, after all, exists specifically because we have private banking that requires interbank payment settlement via the Fed system. To skirt this reality is to misconstrue the reality of why the Fed and Treasury are separate entities in the first place.
An actual example is useful here because government taxes with the current arrangement cannot destroy deposits. For instance, if I take out a $100 loan (which creates $100 in deposit money) and incur a 10% tax at some point along the way then the banking system has a $100 loan asset ($100 deposit liability) and I have a $100 deposit asset ($100 loan liability). After the $10 in taxes are paid the government clears this payment in its account at the Fed by transferring reserves from the bank to the government. Importantly, this is just clearing the deposit transfer. The government did not tax its own liability as MMT often claims. It taxed the deposit. This taxation cannot destroy the $100 loan that is still outstanding. Therefore, it is incoherent to claim that the government has “destroyed” money when it has taxed an asset it did not even create. MMT’s depiction of this process is extremely misleading as it ignores the fact that the money being taxed is created in the banking system and then relies on a consolidation of the Fed and Treasury to misconstrue the reality in an attempt to give the appearance that the consolidated Fed/Tsy created the money initially.
This is important to get right for three reasons. First, the banking system creates deposits when loans are made and those deposits can only be “destroyed” when loans are repaid. The government, in essence, cannot sustainably save deposits because they are directly tied to the loans from which they came and the banking system in aggregate relies on the availability of those deposits to function. Second, MMT misconstrues the conversion of deposits to reserve in the clearing process. This is exactly the same thing that occurs when someone takes money from an ATM thereby converting deposits to cash. Are we to start calling this “destruction” of money? Of course not. That would be incoherent. It is a simple and temporary conversion, just like taxes to spending. And third, MMT relies on consolidating the Fed into the Treasury to make this already misleading story appear coherent. But as I’ve explained before, the Fed is a separate entity SPECIFICALLY because we have a private banking system. You cannot claim to be explaining reality when you are so clearly explaining a completely alternate version of reality.
There is nothing wrong with consolidating balance sheets to simplify a point and make it clearer, however, it is essential, in the case of reality, to understand the flow of funds at work here and the purpose of reserves. The way MMT consolidates the balance sheets is designed to misconstrue the flow of funds. But reserves exist only because we have private banks that need interbank clearing of payments via the reserve system. All reserves are imposed on the member banks of this system. So when the government taxes the Treasury uses the reserve system to transfer payments from a commercial bank to the Treasury and then they spend those funds into another commercial bank. MMT claims that the reserves exist because the government spends and then destroys them when it settles the payment with Treasury. But this misconstrues the flow of funds. Reserves only exist because we have private banks which transfer deposits via interbank settlement. Reserves exist because the Central Bank forces the banks to hold them whether the Treasury spends or not. So reserves MUST precede government spending to meet requirements in a private banking system where there is a reserve system. Loans create deposits, but loans also create reserves when needed. To start this process from the point of government spending is to ignore the fact that the only reason the reserves are even in the system is because the Central Bank imposed them on banks to back up a certain amount of deposits that were created through loan issuance. So this process, MUST start with the banks creating loans. To say that spending creates money is to assume that government spending creates reserves which is a fictional presentation of why there are even reserves in the first place. It is, in essence, saying that the Fed operates fiscal policy by issuing reserves. We should be blunt – this is terribly misleading.
The important point is that reserves are in fact a clearinghouse asset that exist specifically for the transfer of private non-government liabilities (deposits). Banks create most of the money in the financial system and are required to hold some degree of reserves. These reserves are used for interbank settlements that transfer deposits from one bank to another. When the government taxes it settles this deposit transfer via the reserve system, but this is nothing more than an interbank transfer of the deposits. To depict the reserves as being “destroyed” and then “created” when the government taxes and spends is to misconstrue the reality that those deposits were created in the non-government banking system and simply transferred using reserves. The government does not destroy or create the private bank deposits that it taxed and spent. To consolidate the Fed into the Treasury is to effectively nationalize the banking system in this model and create a misleading fictional depiction of what is actually a specific and clear flow of funds from one private bank to another private bank.
To reiterate this important point it can be useful to see an actual example. Say, for instance, that I take out a $100 loan and the Central Bank imposes a 10% reserve requirement. When the bank creates this loan the Central Bank will create reserves of $10 that allow my bank to meet its reserve requirement. If I then make a $10 payment to someone who uses a different bank in the system then that payment will settle in the interbank market via reserve transfer. My bank will debit my account by $10 leaving me with $90 and the bank will process this payment by sending the $10 in reserves to another bank. This will result in a deposit credit of $10 to the recipient of the payment. When the government taxes the same exact process will occur. If I pay $10 in taxes the US Treasury will obtain a $10 reserve credit which it will then utilize to transfer $10 of private bank deposits to someone else when it spends. This is nothing more than a flow of funds occurring within the interbank system. But it’s very important to note that the Treasury is not really transferring reserves. They are actually transferring the private bank deposits and simply using the reserve system to process an interbank payment just like any other interbank payment settlement. This does not destroy or create the deposit because the deposit came into existence before the government taxed anyone. And the only way this deposit can be destroyed is when it is spent back into the private sector and the loan is repaid. MMT’s depiction of the reserve system as creating and destroying money results in a misleading depiction of the flow of funds that actually occur in the private banking system by creating the illusion that this whole process starts and ends with the government when it really starts outside the government in the private banking system.
We know this is accurate because the history of the Federal Reserve System proves it. For instance, in 1907 there was a massive financial panic that resulted in widespread regulations leading to the Federal Reserve System. This system was created for the purpose of establishing a publicly supported interbank payment system. But the important point is that the reserves on deposit in this interbank system were created as a result of the quantity of loans made. So banks were required to hold a certain amount of deposits on reserve. This proves, definitively, that reserves precede government spending. The act of settling a payment in the reserve system must come after a private bank makes a loan to a private sector entity which results in a tax payment. This tax payment is then settled in the reserve system with the Treasury. But the flow of funds is clear – loans create deposits which result in required reserves. Saying that taxes destroy money and spending creates money misconstrues the flow of funds at work here.
To further prove this point we can consider an example where the banking system is comprised of one private sector bank where the Treasury has an account. In this case there would be no need for reserves since there would be no interbank payments. The Treasury would debit their private bank account when they tax and they would credit it when they spend. They would very clearly be a user of the private deposit system. This would change none of the operational mechanics of the way the Treasury and single bank operate. But the second you add in a second private bank you need an interbank market to settle interbank payments. MMT treats the creation of a reserve system as a de-facto nationalization of the private banking system when the reality is that the Central Bank is nothing more than an agent of both the government AND the private banks.
It is incoherent to say that spending creates money and taxes destroy money because the reserves exist before government spending occurs ONLY because we have private banks that create loans which require reserves within an interbank system. When we start the circuit within the reserve system we create the illusion that the government is the actual creator of money when the reality is that banks issue most of the money and the Federal Reserve exists mainly to facilitate the smooth transfer of interbank payments. MMT should say that deficits create a type of money (T-Bills and T-Bonds) and surpluses destroy this money. That would be a more accurate presentation of the flow of funds.
This is also a flow of funds error within the MMT description of the world. Taxes cannot destroy money because taxes logically occur after some level of spending on private output occurs. That is, taxes are incurred when the private sector spends and we collectively decide to use those taxes to mobilize some amount of resources from private to public domain. So productive resources and private spending on resources precede taxes. It is illogical to argue that the government spends first when taxes occur primarily in bank money which was created before state money ever existed.
Now, MMT people sometimes argue that banks are just agents of the state acting under a banking license. This is irrelevant. It does not change the fact that banks are private entities issuing private liabilities that do not get destroyed during government spending and taxation. Besides, it is silly to imply that every entity operating with a government license and regulatory oversight is somehow an agent of the state. There are tens of thousands of jobs and industries that require specific government licencess to operate. Are all of these entities now government agents because of this. Of course not. This is a case of overreach as so much of MMT is.
Alternative MMT Response to Claim #2: “The Government does not need to tax. It could just print the money it needs”.
CR Response: This is true in theory, but not in reality. The reality is that the US monetary system is designed around its banks. The banks issue the real money and the Central Bank and Treasury accommodate this money supply by issuing reserves for monetary policy and cash for various banking purposes. But the key point is that the US government does not actually issue money in the first instance. New money comes into creation via the private banking system first. As such, the US government is actually a user of the deposit system. It even has bank accounts so it can easily obtain the liabilities of private banks. So, as a matter of reality, the US government has designed its payment system with a private competitive banking system of which the US government is a user. As a result the US government must tax deposits and spend them back into the financial system. Yes, in theory this could all be avoided by having the government spend cash or deposits by being the direct issuer, but we do not live in a theoretical world.
I should be clear here though. The fact that the US government is a user of the deposit system does not mean it will have trouble funding its liabilities. In the case of an emergency funding environment there is little doubt that this theory could become reality and the Central Bank would directly finance the Treasury’s spending. This would likely occur only during a hyperinflation so it’s kind of a moot point, but we shouldn’t confuse the government as a user of the US Dollar when it does have theoretical access to create dollars in perpetuity. Hence why it is important to describe the US government as a “contingent currency issuer”. Of course, as mentioned above, having a Central Bank “finance” your spending does not mean you’re solvent in real terms.
Alternative MMT response to claim #2: “If the government did not self impose the spending constraint then MMT would be correct.”
CR Response: The US Treasury is required by law to have credits in its reserve account before it can spend. Although this is “self imposed” it can be said that the entire financial system is self imposed. It is created from thin air with specific rules that dictate the rules of the system. To say that a self imposed rule does not really apply is like saying that speed limits are not important because they are self imposed. While this true there is good reason for the self imposed limit. We want to limit how fast people can drive. Similarly, the US government imposes limits on how the government can spend do it does not go “too fast”. It makes sense to limit the spending of Congress because if there were no checks on its spending it’s likely that this would result in increased corruption and malinvestment. These checks might not appear to make much sense, but they impose limits on the budgeting process that require the Congress to think intelligently about how it spends. Although self imposed they are not without reason.
Further, whether this is self imposed or not really does not matter. Even if the US government had zero constraints to spending and could throw money out of helicopters you would still need to find willing holders of those dollars. Creating a tax does not mean that people will necessarily have strong demand for those dollars. And in fact, creating those dollars at times reduces the demand for money thereby leading to the inflation that can permanently destroy the demand for a currency.
The bottom line is that the government does not create demand for its own money. It cannot force people to use the money it issues simply by taxing them. And since it should be clear that any issuer of money needs to find willing holders of that money then it is clear that those holders are the financing agents for the money issuer. Therefore, any issuer of money, government or not, needs to finance its spending and it is incoherent to claim that governments “spend first and tax second”.
1 – We should be clear about one thing – the US Treasury is legally required to obtain funds before it spends. When you incur a tax liability your tax dollars actually go to the Treasury and credit its General Account which gives it the legal ability to then credit someone else’s bank account. This is an undeniable legal requirement that immediately disproves the MMT position. They sometimes state that this is “self imposed”, but yes, so are homicide laws. That does not mean homicide laws aren’t real just because we “self impose” them. If you want to theorize about a world where homicide laws don’t exist then great. But we should be clear that that discussion does not pertain to the world we actually live in. So we should establish up front that the current legal structures are very clear about the orders of operations here – taxes fund spending because the Treasury is legally required to fund its spending. If you want to argue that there is a potential alternative reality where this legal constraint doesn’t exist then great, but we should be clear that that is an alternative reality.
2 – We should note that the MMT use of the term “sovereign” is very slippery. It is a practically meaningless term in the sense that sovereignty exists on a spectrum and some governments have little sovereignty due to real constraints. For instance, a resource constrained economy is not sovereign because it has real constraints on its economy. It might be forced to borrow in a foreign currency (as many emerging economies do) because they do not have the flexibility of a large developed economy like the USA. But we should also note that sovereignty is a dynamic existence. A country can lose its sovereignty due to reasons outside of its control or even due to the actions of its governments. In 2005 MMT co-founder Randall Wray said that Turkey’s deficit was “indefinitely sustainable”. This turned out to be catastrophically wrong and proves that MMT cannot assess how “sovereign” a state is or will become. MMT uses this term in an ambiguous and rather misleading manner.
MMT Claim #3 : “taxes drive money”
Reality: Many things “drive money” and taxes are not likely an important one
The purpose of this claim is to create the illusion that the government creates demand for the money it spends. In other words, government spending is self-financing. While the government can help build the productive capital base of the economy the empirical evidence clearly shows that all government spending does not have a positive multiplier and is not necessarily “self-financing”.
MMT claims that the government creates demand for its currency by imposing a tax on its users. They claim that this drives the desire to obtain currency and that this currency is ultimately paid back to the government in bank reserves. In establishing this point MMT claims that the government spends first and imposes a tax that generates the demand for this currency.
The ability to tax or charge fees is not unique to a government, however. All banks charge a tax on their loans when they charge you an interest rate. This involuntary fee helps to create demand for bank money. Should we now argue that banks, as the primary issuers of money, create demand for money because they charge fees? Of course not. The reason there is demand for bank money is because there is desire to consume/invest in private output denominated in that currency. Most liabilities are issued with some form of involuntary obligation attached to them. The government is not unique in its ability to charge taxes/fees or impose obligations. We should not misconstrue this idea as being unique to government currency.
This is related to the MMT view that the government “spends first” and “taxes second”. They have even gone so far as to claim:
“it would be impossible to collect dollars from the private sector unless they had first been spent into existence by the public sector”
Of course, this point is demonstrably false. I can borrow from a bank and the government can collect taxes without ever having spent a dollar into existence. The government doesn’t need to spend a single dime in order for it to collect a tax on inside bank money.
Further, MMT misconstrues the role of Central Bank reserves in this idea. They argue that taxes are not actually paid in bank deposits, but are paid in Central Bank reserves because the Treasury settles tax payments in its account at the Federal Reserve. This point misconstrues the purpose of the Reserve system. The only reason the Reserve system exists is because of private bank inside money. So, if there were just one private bank in our financial system there would be no need for a Federal Reserve System. Hence, there would be no reserves. And the Treasury would settle its payments in bank inside money at this one bank. If reserves did not exist due to the large private network of private banks then the US Treasury would be exposed as a mere user and redistributor of inside money. In fact, the existence of the Fed does not change this fact since all government spending is merely a series of debits and credits that settle through the interbank system, but are ultimately initiated and finished in the private inside bank money system.
MMT would be better served by saying that users of a certain currency are confident using that currency because there is a government regulatory system that gives this currency credibility. An established court system and regulatory structure can give credibility to a currency by enforcing claims. But we should not be confused into thinking that this means the government “drives” demand for the currency. The credibility of a form of money is contingent on many things and a structured legal system is one element of this demand. But at its most basic level it should be obvious that the demand for any money is based on its credibility in purchasing output. So while a government regulatory system is certainly an important link in the demand for money it is not necessarily the “driver” of this demand.
In addition, there is an important element of stability in a centralized monetary system that many users do not appreciate. While fiat currencies can some times be ruined in real terms they are unusually stable in nominal terms. The ability to settle a monetary unit at par is uniquely conducive to a centralized monetary system because the government does not need to make a profit on its settlement process. The nominal stability of the currency and par settlement is an essential element to the currency’s adoption as this lack of volatility makes it predictable in every day use.
As a litmus test for the MMT idea we can simply ask ourselves if we would continue using the US payment system if there were no taxes? I suspect most people’s use of US dollars would not change much if there were no taxes to be paid. Instead, we use the US payment system primarily because it gives us access to a credible form of money that allows us to purchase the output denominated in that currency.
Standard MMT response to Claim #3: “Governments create involuntary demand for money by requiring that taxes be paid in a specific form of currency”.
CR response: The fact that the government charges an annual fee does not mean that this is why people use money. If this were true then the demand for money would not collapse during a hyperinflation. Again, we should get the order of procedures right here. Taxes are fees that are charged on our output. When output is high tax receipts are high. When output collapses tax receipts are low. But you cannot have tax receipts if there is no output. So sustainable and productive output MUST, by necessity, precede taxes. In this sense it is proper to say that productive output drives money. And if productive output collapses then there is no quantity of men with guns that can force people to pay taxes.
MMT is famous for saying that the government is only constrained in its spending by the amount of inflation that occurs. But inflation occurs when spending outstrips productive capacity. Since the primary source of productive capacity is private sector output MMT should always say that government spending is constrained by how much productive capacity it can move from the private sector to the public sector. In other words, the government cannot tax in excess of underlying productive capacity or they will create inflation with their excess spending. So the important point here is that a government is indeed constrained in its spending. It is constrained by the quantity and quality of its private sector’s productive output. And the quantity and quality of income that the private sector can create is the amount of income that constrains the government’s ability to spend.
MMT Claim #4: State money sits atop the hierarchy of money in the monetary system.
Reality: Most modern governments have outsourced money creation to the private banking system placing it in the dominant settlement and payment role.
The purpose of this claim is to create the illusion that the government needs to issue net financial assets because they are the cornerstone of the economy. While the government can help stabilize the economy by issuing financial assets it is by no means necessary. This is a case of overreach in the MMT paradigm based on an operational misunderstanding.
In placing state money at the top of their hierarchy MMT boxes itself into the same corner that the rest of exogenous money theories do. This creates a state centric theory of money that is not all that different from the money multiplier theory. The difference being that MMT phrases things differently and applies an endogenous banking system into their model. Instead of saying that state money is “multiplied” (as money multiplier believers claim) MMT will say that it is “leveraged”. These are just word games that are utilized to make endogenous bank money appear compatible with the State Theory of Money.
The reality is that inside bank money dominates our financial system. The state has essentially outsourced money creation in our financial system to private banks. This means that outside money (state money) has been rendered secondary to inside money (bank issued money). Although the state could change this by nationalizing the banks it does not change the current reality of the financial system in which we reside.
Interestingly, the only reason this view is even viable within MMT is because we have private bank money in the first place. In the MMT world bank reserves are a government liability that payments are settled with which makes them the top of the hierarchy. But the only reason bank reserves even exist is because the Federal Reserve is a clearinghouse that supports private banks through the interbank system. In other words, the very existence of the Fed contradicts the MMT view. To prove this we can imagine a simple one bank system. In this system there would be no need for reserves and a Central Bank. All payments would clear through this one private bank and the US Treasury would be a user of the system just like all other users. Private bank deposits would obviously be the primary form of money and the government would be a mere redistributor of this bank money. MMT flat out contradicts this view and even goes so far as to create their own alternative reality wherein they consolidate the Fed into the Treasury in order to make their fantasy accounting world appear more accurate. They literally create a fictional world in order to make the flawed accounting look correct.
Standard MMT response to Claim #4: “Reserves might be a smaller quantity of the aggregate financial asset pool than deposits, but that does not mean they are less important”.
CR Response: The idea of a hierarchy is misleading precisely because it misleads one as to the actual importance of various financial assets. The monetary system relies on many healthy functioning pieces to work properly. I do not say that reserves are unimportant. I simply highlight the fact that they facilitate the transfer of payments. They are a secondary instrument of exchange in this sense since deposits naturally precede them. This is about understanding the monetary for what it is and not constructing some sort of government centric perspective whose end goal is policy ideas.
Further, if we wanted to highlight financial assets in terms of their importance it should be uncontroversial to note that equity is the cornerstone of any modern economy as it is the market value of our output. Without valuable output there is no need for a monetary system in the first place. MMTers will often imply that deficits add to equity as “net financial assets”, however, this is a misleading view since, if it were true, then governments could simply print equity at will and give their economies real value. While deficits can, at times, add to demand, we should not be fooled into thinking that deficits ALWAYS add to demand and equity. In fact, deficits can be quite damaging to equity at times as is evidenced in the case of any hyperinflation.
Critique #5: MMT is not Empirically Tested & Has no Coherent Theory of Inflation
My final and probably most important critique is that MMT is not an empirically tested theory. In fact, it is more a set of beliefs than a set of economic principles. This probably explains why its followers are so militant in their views online. They believe more than they can prove. We see this across several facets of MMT including:
- Most MMT advocates claim that MMT is “just a description” of the existing monetary system, but this is not true. As I’ve shown above, there are many operational errors in the way MMT describes the monetary system. But more importantly, MMT is a comprehensive macro theory for full employment and price stability. The essential element of this is a national Job Guarantee. Without the Job Guarantee we are not testing the MMT paradigm. And we should be clear – a full scale Job Guarantee has never been tried in any developed economy. It is irresponsible and irrational to argue that this is a good program when it is not empirically tested in any useful form.
- MMT has no coherent theory of inflation. MMT claims to know what causes inflation (as do many economic theories), but they also claim that they will control inflation through a Job Guarantee, price controls and regulations. Unsurprisingly, they often talk out of both sides of their mouth on this issue with their leading proponents claiming that taxes will be used to fight inflation (see, Wray) while others explicitly reject that idea (see Kelton at 8:55). Further, they cannot show what those levels would be or how they would implement such a policy in any practical manner. We should be clear – MMT has no empirically coherent theory for containing high inflation. It is all theory. Instead, they claim that the Job Guarantee will anchor prices and regulations and price controls will help contain price changes. These positions are entirely theoretical and have only thin empirical evidence to support them.
- A related point is their mistreatment of Monetary Policy. In the MMT world the “natural rate” of interest is 0%. They view no policy scope for Monetary Policy as a valve for suppressing inflation despite vast evidence that Monetary Policy can be a useful tool for controlling inflation. By neutering Monetary Policy’s most powerful tool MMT takes an irrationally pessimistic stance on what is generally accepted to be a powerful inflation fighting tool. There’s no need for this and their excessively ideological view discredits their inflation theory.
- The Job Guarantee is an untested program in any developed country. MMT cannot show that the JG does what they claim. Instead, it is supported mainly by faith and untested claims.
There is legitimate reason to critique MMT based on its internal operational inconsistencies and the lack of empirical evidence supporting so many of their claims. I find MMT to be a useful theory, but one that we should approach with skepticism.
NB – Terminology is a common problem across MMT and is the cause of much of the confusion that arises when discussing the theory. They use alternative definitions for many of their ideas including “net saving”, “money” and “full employment”. This terminology is often used in a vague sense that is inconsistent across parts of the theory.