Your bit about it being policy advocacy aligns with what I’ve seen from my friends who subscribe to the MMT ideas.
I’m all for getting medlock-pilled. But the idea that we can have everything we could ever want simply with “money printer go brrr” (a gross simplification) always struck me as, ridiculous.
This was a helpful explanation of MMT and a confirmation of my understanding of it without reading a book on the subject. Can you please do a part 2 and tell us why they support a federal job guarantee as the extension of their theory? Or is that pushed by just some of the MMT proponents? I never understood the connection.
Great piece, this is the type of econ writing that I'm looking for! First time I've been able to grasp MMT at all. Is it also true that MMT proponents want to "print money" to such an extreme extent that it could actually devalue the dollar?
As someone whose background is entirely micro-oriented finance (corporate and asset pricing, no banking, institutions, or macro-related stuff), I really appreciate quick explainers about witchcraft aka macroeconomics.
When I was starting my career I always tried to classify the econ "theories" as a "right-wing" or a "left-wing" one. Now I don't look at "theories" but at models in journals, and I try to identify what's the story that they tell me about the world, what variables are interacting with each other and how, and what variables they're missing out. Part of that has been thanks to your blogs and tweets.
Something that I do resent about econ education is that they teach the IS-LM model even if everyone says that it's "dated". It gives me the idea that debates in real macroeconomics are esoteric and intimidating, and that the only real policy advice that the public needs to know is that countercyclical policy (fiscal and/or monetary) is good.
The thing is that you don't need an expansionary fiscal policy to justify government intervention, especially government investment. There are a lot of market failures and externalities justification for government investment, and a recession only diminishes the cost of that investment.
I'm loath to defend MMT (and loath to reprise my role as a Noahpinion comment section troll), but I actually prefer this MMT definition of saving to the textbook definition of saving simply because it makes more sense to me on a conceptual level. Is investment saving? Well, why would it be? If I build a factory using money I saved, the money I saved is gone. I now have a productive asset, which is going to be a lot more stimulatory to the economy than sitting on a certificate of deposit or whatever. If two actions are fundamentally different, we can't just bundle them up into one thing and say they are the same thing.
"Saving" should mean that you give up consumption today in order to get consumption in the future, no? So if instead of eating corn I plant it in the ground and grow more corn plants that I can eat later, I feel like I'm saving. But not according to MMT.
Isn't there a third option? Putting your corn in the granary for winter is what I would call "saving". Eating it is "consumption". Planting it in the ground to get more corn back is "investing". Unfortunately in the real world there is more of a spectrum between saving and investing (because the saved money can be invested by someone else, e.g. the person you lend your money to) but I really think at their core they are quite different functions.
I see Investments being at material risk and Savings not bearing such risk. Investments are Schrödinger's Savings. The corn planted in the ground is likely to yield increased corn in the future, save for drought years... while the corn in the silo, presumably (save thieves, bombings, burnings etc) is locked tight and everything put in comes out on a nearly 1:1 basis.
If planting the corn is the same as saving the corn, why isn’t eating the corn considered an investment as well? Eating the corn invests the body with calories that can be used later to plant more corn.
In the purest sense, yes, mattress stuffing is the truest form of saving, but there's a spectrum between saving and investment, and I still think they are essentially different acts. I think one of the great innovations of fractional reserve banking is to turn holding onto cash into a channel for productive investment. To go back to the corn metaphor, a bank is like a granary that holds a limited stock of corn for the people who deposit their corn there, and offers most of it to farmers to grow more. So that way you kind of get the best of both worlds.
MMT is telling us that government deficits equate to net FINANCIAL saving. Investment, such as your corn example, is non-financial saving. No inconsistency with MMT. Indeed, the majority of saving is non-financial.
I may be coming at this from the wrong direction, but I think separating things based on where they go (who the issuer of a bond was in this case) rather than what they do seems weird to me. At best I think it's very counterintuitive to say that buying a US government bond is saving but buying a corporate bond is not.
> If I build a factory using money I saved, the money I saved is gone. I now have a productive asset, which is going to be a lot more stimulatory to the economy than sitting on a certificate of deposit or whatever.
The thing is, when you put your money in a savings account, the bank takes it and loans it out to other people, businesses, etc. The bank creates the illusion that you still have the money by allowing you to withdraw it at any time, assuming that no more than a certain fraction of the total money they have deposited will be withdrawn at any time (if people withdraw more than that, it's a bank run. In reality, the money's been used to buy some productive asset, just with extra steps (the bank handles the details).
>the bank takes it and loans it out to other people, businesses, etc
No it doesn't. That is the loanable funds fallacy. It is still mistakenly taught in econ textbooks, but in financial practitioner circles it is discredited (search Bank of England or Standard & Poors papers on "Do loans create deposits" if you doubt that). The key issue here is that a Bank's ability to lend is not constrained by the quantity of deposits it holds from private sector agents, it is constrained by the quantity of Tier 1 capital it holds per BASEL III regulations (Tier 1 capital === US Treasuries, basically).
I think you're right that it's weird to not differentiate between savings that go to investment in the real economy and savings that go to hedges and the exotic financial sector. But the definition of savings that has stuck since the 40s is "savings = consumption delayed". It's not too bad of a definition.
It is important to use the same concepts in order to make sure that we're talking the same language. Only then can we compare different models.
MMT doesn't use the same language as modern macro and it makes it harder to judge it.
I think that the key difference for many MMTers is that they assume that of the government employs people and produces real goods (resources) inflation can be avoided.
They might not see that in order for those goods to be valuable in the economy they must be accepted by consumers. Otherwise it's just waste. And making goods that are acceptable for consumers is what competition does best, and the reasing why "crowding out" investment away from the private sector is usually a bad thing.
Just to clarify; I don't think that "crowding out" is bad if government does things that the private sector does bad (streets, etc...) of if it fixes market failures (healthcare, etc..).
MMT isn't about the government wantonly producing goods that compete with the private sector. In fact, they painfully specify that it matters a lot what you spend the money on (not all spending is helpful or socially useful). Every program I've seen pushed by MMT people fall into one of the two later categories that you define (private sector does bad or market failures) eg the Green New Deal, the Federal Jobs Guarantee, Medicare for All, Education for All, etc. Take the Jobs Guarantee; most of the jobs that would be done are things that the private sector doesn't want to do at all (like planting trees, building public infrastructure, care work, etc.), and the money that is paid to the workers can then be used by those workers to purchase goods that they want (privately or publicly produced).
In my opinion you are conflating deficit spending with borrowing, which isn't how my view of MMT views it. MMT has the government just creating money and spending it, which is inflationary but isn't reliant on hoping some private sector multitrillionaire decides to lend. That most of our debt is held by the private sector is only true to my understanding if you count the FED system(and possibly Social Security) as private sector. I don't exactly understand how banks multiply the money supply these days, but it seems clear that they do, I guess because an FDIC insured deposit dollar is about as good as a dollar with no physical dollar needed.
So handwaving that mess, if the government wants to balance the inflationary nature of its spending, it has to either tax, which is roughly spending in reverse, or borrow, selling bonds and trading the relatively high velocity dollar for a relatively low velocity bond(which I think is why bond selling is deflationary, I'm a little vague on the real difference between a three month zero interest bond and cash). But the bond will be paid off eventually, so selling bonds to the private sector only chases current inflation into the future. And the government selling bonds to itself seems like basically an accounting gimmick, possibly useful for interdepartmental accounting, but writing yourself IOUs isn't terribly meaningful.
I agree that capital investment is a form of savings, and there should be another term for financial investment in cash and government bonds. And I think that jobs guarantees and the green new deal and a single payer heath program should be viewed as separate from what I will call core MMT. I like UBIs better than job guarantees, for example, and other than that they are just what people think that the government should be spending on. And I agree with the wisdom of socialized medicine, as our current per capita spending is more than adequate to fund a more efficient system, and a green new deal, which may avert catastrophic outcomes and pollution is bad anyway (I favor a Carbon tax to fund the UBI and replace income and some real estate taxes). But tacking good government policies onto the Monetary Theory seems to lump already complex issues into an unnecessarily unmanageable mess, where people can get bogged down in job guarantee logistics, when a narrower MMT practitioner wants a Keynsian stimulus, and can discuss the form separately. Job guarantees aren't IMO ideal, but they can help in deflationary crashes, giving the desperate some money and accomplishing some state goals with otherwise wasted human potential.
And I will point out that in your example of a lonely, no deficit government, there will be no cash in the system, since the government needs to spend dollars into circulation and not tax them back for there to be circulating currency. So people can invest, but not in or with money or government bonds. Barter economies aren't theoretically impossible, but MMT isn't well designed to describe them, it's a monetary theory.
It seems to me, from a policy point of view, the main thing pushed by MMT is true. A government that borrows in its own currency has spending constrained by inflation and not by any willingness of the private sector to lend. The majority of the political and policy community is driven (or claims to be when the opposing party is in power) by deficit concerns that, at some point, the government will not be able to borrow. How is this fallacy so widespread or even still existent.
It feels like there is a substantive difference between "saving" by building a factory vs putting cash under a mattress. Both are delayed consumption in a sense, but the former consumes economic resources in order to build the factory - that bit happens right now with no delay.
I could hire workers to build a factory or I could hire them to build a yacht, both of these consume resources and bid up the price of labor even though the factory is a productive asset and the yacht is not. Given that MMT Is interested in what spending causes inflation and what doesn't, it makes sense to try and lump together both consumption and investment, when the investment also involves bidding up prices. Whereas cash under the mattress (or maybe buying stocks/bonds - which will bid up the price of them but not of actual goods or services) is meaningfully different.
Where MMT began to interest me was when it clicked for me what they meant when they say, "Anything we can do, we can afford to do." Meaning, productive capacity is the limiting factor, not tax revenues. The idea is that spending does not become a problem at the point at which it exceeds revenues, but at the point at which is exceeds productivity. That makes sense to me and so far has been the key insight for me in becoming interested in this idea that "deficits don't matter" in the way that most of us tend to think they do.
Another way of framing it is that the equilibrium that matters is not an equilibrium between spending and tax revenues. There is still an equilibrium that matters, but that's not it. If I understand this much, that is. It was reading Kelton that helped get that far, at least.
I found the MMT idea of using tax policy to kill inflation very interesting. My classes with Marvin Goodfriend at CMU were all about using interest rates to kill inflation - the great 'Volcker disinflation'. But this would imply that Volcker's killing inflation is what allowed Reagan to cut taxes like he did in the 80s. Has anyone explored the interplay of these two methods to counter recessions and inflation - cut-rates/raise-rates vs cut-taxes/raise-taxes ?
"In MMT, investment doesn’t count as part of net saving; it gets subtracted out. What’s left is just the amount that the private sector lends out to other sectors — government or foreigners."
(S - I) = (G - T) + (X - M)
"Suppose the world was just one country (so, no foreigners). And suppose the government of this country didn’t borrow or lend at all (so, no government deficit). According to MMT, the private sector would be completely unable to save anything at all! But according to typical economics, the private sector would be able to save by investing for the future."
Then we'd have this (reworking the above equation):
S = (G - T) + (X - M) + I
Because (G - T) = 0 and (X - M) = 0, we are left with:
S = I
The private sector is still saving which in this case exactly matches investment. It is just not net saving, i.e., S > I.
Your bit about it being policy advocacy aligns with what I’ve seen from my friends who subscribe to the MMT ideas.
I’m all for getting medlock-pilled. But the idea that we can have everything we could ever want simply with “money printer go brrr” (a gross simplification) always struck me as, ridiculous.
I mean: tax and transfer good.
“Print” money and run deficits without restraint bad
This was a helpful explanation of MMT and a confirmation of my understanding of it without reading a book on the subject. Can you please do a part 2 and tell us why they support a federal job guarantee as the extension of their theory? Or is that pushed by just some of the MMT proponents? I never understood the connection.
Yep! Actually part 2 will be about whether the government can ever borrow too much. Job Guarantee will be part 3! :-)
Great! Looking forward to both of them! Thanks for replying.
Great piece, this is the type of econ writing that I'm looking for! First time I've been able to grasp MMT at all. Is it also true that MMT proponents want to "print money" to such an extreme extent that it could actually devalue the dollar?
That's the subject of another post! :-)
As someone whose background is entirely micro-oriented finance (corporate and asset pricing, no banking, institutions, or macro-related stuff), I really appreciate quick explainers about witchcraft aka macroeconomics.
Feeling called out
You're absolutely right.
When I was starting my career I always tried to classify the econ "theories" as a "right-wing" or a "left-wing" one. Now I don't look at "theories" but at models in journals, and I try to identify what's the story that they tell me about the world, what variables are interacting with each other and how, and what variables they're missing out. Part of that has been thanks to your blogs and tweets.
Something that I do resent about econ education is that they teach the IS-LM model even if everyone says that it's "dated". It gives me the idea that debates in real macroeconomics are esoteric and intimidating, and that the only real policy advice that the public needs to know is that countercyclical policy (fiscal and/or monetary) is good.
The thing is that you don't need an expansionary fiscal policy to justify government intervention, especially government investment. There are a lot of market failures and externalities justification for government investment, and a recession only diminishes the cost of that investment.
I'm loath to defend MMT (and loath to reprise my role as a Noahpinion comment section troll), but I actually prefer this MMT definition of saving to the textbook definition of saving simply because it makes more sense to me on a conceptual level. Is investment saving? Well, why would it be? If I build a factory using money I saved, the money I saved is gone. I now have a productive asset, which is going to be a lot more stimulatory to the economy than sitting on a certificate of deposit or whatever. If two actions are fundamentally different, we can't just bundle them up into one thing and say they are the same thing.
"Saving" should mean that you give up consumption today in order to get consumption in the future, no? So if instead of eating corn I plant it in the ground and grow more corn plants that I can eat later, I feel like I'm saving. But not according to MMT.
Isn't there a third option? Putting your corn in the granary for winter is what I would call "saving". Eating it is "consumption". Planting it in the ground to get more corn back is "investing". Unfortunately in the real world there is more of a spectrum between saving and investing (because the saved money can be invested by someone else, e.g. the person you lend your money to) but I really think at their core they are quite different functions.
Why is the ground different than a granary?
Because in the ground the corn is growing into a new plant, and in the granary the corn is waiting to either be eaten or grown into a new plant.
So only liquid cash feels like "saving" to you?
I see Investments being at material risk and Savings not bearing such risk. Investments are Schrödinger's Savings. The corn planted in the ground is likely to yield increased corn in the future, save for drought years... while the corn in the silo, presumably (save thieves, bombings, burnings etc) is locked tight and everything put in comes out on a nearly 1:1 basis.
If planting the corn is the same as saving the corn, why isn’t eating the corn considered an investment as well? Eating the corn invests the body with calories that can be used later to plant more corn.
In the purest sense, yes, mattress stuffing is the truest form of saving, but there's a spectrum between saving and investment, and I still think they are essentially different acts. I think one of the great innovations of fractional reserve banking is to turn holding onto cash into a channel for productive investment. To go back to the corn metaphor, a bank is like a granary that holds a limited stock of corn for the people who deposit their corn there, and offers most of it to farmers to grow more. So that way you kind of get the best of both worlds.
MMT is telling us that government deficits equate to net FINANCIAL saving. Investment, such as your corn example, is non-financial saving. No inconsistency with MMT. Indeed, the majority of saving is non-financial.
I may be coming at this from the wrong direction, but I think separating things based on where they go (who the issuer of a bond was in this case) rather than what they do seems weird to me. At best I think it's very counterintuitive to say that buying a US government bond is saving but buying a corporate bond is not.
> If I build a factory using money I saved, the money I saved is gone. I now have a productive asset, which is going to be a lot more stimulatory to the economy than sitting on a certificate of deposit or whatever.
The thing is, when you put your money in a savings account, the bank takes it and loans it out to other people, businesses, etc. The bank creates the illusion that you still have the money by allowing you to withdraw it at any time, assuming that no more than a certain fraction of the total money they have deposited will be withdrawn at any time (if people withdraw more than that, it's a bank run. In reality, the money's been used to buy some productive asset, just with extra steps (the bank handles the details).
>the bank takes it and loans it out to other people, businesses, etc
No it doesn't. That is the loanable funds fallacy. It is still mistakenly taught in econ textbooks, but in financial practitioner circles it is discredited (search Bank of England or Standard & Poors papers on "Do loans create deposits" if you doubt that). The key issue here is that a Bank's ability to lend is not constrained by the quantity of deposits it holds from private sector agents, it is constrained by the quantity of Tier 1 capital it holds per BASEL III regulations (Tier 1 capital === US Treasuries, basically).
Hi!
I think you're right that it's weird to not differentiate between savings that go to investment in the real economy and savings that go to hedges and the exotic financial sector. But the definition of savings that has stuck since the 40s is "savings = consumption delayed". It's not too bad of a definition.
It is important to use the same concepts in order to make sure that we're talking the same language. Only then can we compare different models.
MMT doesn't use the same language as modern macro and it makes it harder to judge it.
I think that the key difference for many MMTers is that they assume that of the government employs people and produces real goods (resources) inflation can be avoided.
They might not see that in order for those goods to be valuable in the economy they must be accepted by consumers. Otherwise it's just waste. And making goods that are acceptable for consumers is what competition does best, and the reasing why "crowding out" investment away from the private sector is usually a bad thing.
Just to clarify; I don't think that "crowding out" is bad if government does things that the private sector does bad (streets, etc...) of if it fixes market failures (healthcare, etc..).
MMT isn't about the government wantonly producing goods that compete with the private sector. In fact, they painfully specify that it matters a lot what you spend the money on (not all spending is helpful or socially useful). Every program I've seen pushed by MMT people fall into one of the two later categories that you define (private sector does bad or market failures) eg the Green New Deal, the Federal Jobs Guarantee, Medicare for All, Education for All, etc. Take the Jobs Guarantee; most of the jobs that would be done are things that the private sector doesn't want to do at all (like planting trees, building public infrastructure, care work, etc.), and the money that is paid to the workers can then be used by those workers to purchase goods that they want (privately or publicly produced).
In my opinion you are conflating deficit spending with borrowing, which isn't how my view of MMT views it. MMT has the government just creating money and spending it, which is inflationary but isn't reliant on hoping some private sector multitrillionaire decides to lend. That most of our debt is held by the private sector is only true to my understanding if you count the FED system(and possibly Social Security) as private sector. I don't exactly understand how banks multiply the money supply these days, but it seems clear that they do, I guess because an FDIC insured deposit dollar is about as good as a dollar with no physical dollar needed.
So handwaving that mess, if the government wants to balance the inflationary nature of its spending, it has to either tax, which is roughly spending in reverse, or borrow, selling bonds and trading the relatively high velocity dollar for a relatively low velocity bond(which I think is why bond selling is deflationary, I'm a little vague on the real difference between a three month zero interest bond and cash). But the bond will be paid off eventually, so selling bonds to the private sector only chases current inflation into the future. And the government selling bonds to itself seems like basically an accounting gimmick, possibly useful for interdepartmental accounting, but writing yourself IOUs isn't terribly meaningful.
I agree that capital investment is a form of savings, and there should be another term for financial investment in cash and government bonds. And I think that jobs guarantees and the green new deal and a single payer heath program should be viewed as separate from what I will call core MMT. I like UBIs better than job guarantees, for example, and other than that they are just what people think that the government should be spending on. And I agree with the wisdom of socialized medicine, as our current per capita spending is more than adequate to fund a more efficient system, and a green new deal, which may avert catastrophic outcomes and pollution is bad anyway (I favor a Carbon tax to fund the UBI and replace income and some real estate taxes). But tacking good government policies onto the Monetary Theory seems to lump already complex issues into an unnecessarily unmanageable mess, where people can get bogged down in job guarantee logistics, when a narrower MMT practitioner wants a Keynsian stimulus, and can discuss the form separately. Job guarantees aren't IMO ideal, but they can help in deflationary crashes, giving the desperate some money and accomplishing some state goals with otherwise wasted human potential.
And I will point out that in your example of a lonely, no deficit government, there will be no cash in the system, since the government needs to spend dollars into circulation and not tax them back for there to be circulating currency. So people can invest, but not in or with money or government bonds. Barter economies aren't theoretically impossible, but MMT isn't well designed to describe them, it's a monetary theory.
It seems to me, from a policy point of view, the main thing pushed by MMT is true. A government that borrows in its own currency has spending constrained by inflation and not by any willingness of the private sector to lend. The majority of the political and policy community is driven (or claims to be when the opposing party is in power) by deficit concerns that, at some point, the government will not be able to borrow. How is this fallacy so widespread or even still existent.
It feels like there is a substantive difference between "saving" by building a factory vs putting cash under a mattress. Both are delayed consumption in a sense, but the former consumes economic resources in order to build the factory - that bit happens right now with no delay.
I could hire workers to build a factory or I could hire them to build a yacht, both of these consume resources and bid up the price of labor even though the factory is a productive asset and the yacht is not. Given that MMT Is interested in what spending causes inflation and what doesn't, it makes sense to try and lump together both consumption and investment, when the investment also involves bidding up prices. Whereas cash under the mattress (or maybe buying stocks/bonds - which will bid up the price of them but not of actual goods or services) is meaningfully different.
Where MMT began to interest me was when it clicked for me what they meant when they say, "Anything we can do, we can afford to do." Meaning, productive capacity is the limiting factor, not tax revenues. The idea is that spending does not become a problem at the point at which it exceeds revenues, but at the point at which is exceeds productivity. That makes sense to me and so far has been the key insight for me in becoming interested in this idea that "deficits don't matter" in the way that most of us tend to think they do.
Another way of framing it is that the equilibrium that matters is not an equilibrium between spending and tax revenues. There is still an equilibrium that matters, but that's not it. If I understand this much, that is. It was reading Kelton that helped get that far, at least.
Wait, people are getting their policy ideas from Twitch streams?
Okay, JB Say. Savings is not by definition investment. If it were, there would be no such thing as inadequate aggregate demand.
Good lord!
Hi Noah. You mentioned that borrowing doesn't make a country poorer unless it's borrowing from foreigners. Why is that?
I found the MMT idea of using tax policy to kill inflation very interesting. My classes with Marvin Goodfriend at CMU were all about using interest rates to kill inflation - the great 'Volcker disinflation'. But this would imply that Volcker's killing inflation is what allowed Reagan to cut taxes like he did in the 80s. Has anyone explored the interplay of these two methods to counter recessions and inflation - cut-rates/raise-rates vs cut-taxes/raise-taxes ?
"In MMT, investment doesn’t count as part of net saving; it gets subtracted out. What’s left is just the amount that the private sector lends out to other sectors — government or foreigners."
(S - I) = (G - T) + (X - M)
"Suppose the world was just one country (so, no foreigners). And suppose the government of this country didn’t borrow or lend at all (so, no government deficit). According to MMT, the private sector would be completely unable to save anything at all! But according to typical economics, the private sector would be able to save by investing for the future."
Then we'd have this (reworking the above equation):
S = (G - T) + (X - M) + I
Because (G - T) = 0 and (X - M) = 0, we are left with:
S = I
The private sector is still saving which in this case exactly matches investment. It is just not net saving, i.e., S > I.