The Deficit Myth - Stephanie Kelton

[KJP]

The Deficit Myth:  Modern Monetary Theory and the Birth of the People's Economy by Professor of Macroeconomics Stephanie Kelton contains 25 or so pages of useful and cogent description of how the fiscal and monetary systems of a fiat currency issuer work and 225 pages of rather thinly supported arguments in favor of a progressive political agenda on health care, climate change, employment, and so on.

 

Taking the useful description first, the book explains (correctly I think) that a government that has the monopoly power to issue its own currency and has outstanding debt payable only in that currency can never run out of money.  The reason for that is simple:  The government can always issue more currency to pay any IOUs it has issued and for any present expenditure that it wants to make.  The implication of this argument is that such a government does not fund itself by taxes or borrowing.  Instead, such a government funds itself by issuing more currency.  The primary purposes of taxation are to (i) remove previously issued currency to prevent inflation, (ii) discourage certain activity (e.g., sin taxes), or (iii) limit the economic and political power of the wealthy.  Similarly, “borrowing” via the issuance of debt payable in the government’s own currency simply offers a way to swap out excess bank reserves for a different kind of government money (an interest-bearing kind), which helps the government manipulate interest rates (and may provide useful collateral to the private sector to the extent the government debt is more liquid or transferable than bank reserves). 

 

In short, MMT’s descriptive aspect is its assertion that the government spends first and taxes and borrows only to address the consequences of that spending, rather than taxing and borrowing first and then spending only the money it can raise by those methods.  If this summary sounds familiar to readers of this forum, that’s perhaps because wabuffo has been trying to tell us this for years, and I at least was just too slow to understand what he was saying.

 

If taxing and borrowing aren’t real fiscal limits for such a government, then what is?  The real resources of the economy, i.e., the amount of goods and services the population can produce in light of its existing stock of physical and human capital.  Again, that also seems true and largely non-controversial – as far as I understand them, that is consistent with classical economics’ vertical long-run aggregate supply curve and concept of the production possibilities curve. 

 

And how do we know when we’ve reached the limit of real output?  Inflation.  Although MMT rejects monetarism, this connection between real output, the money supply and inflation seems to be a straightforward application of the quantity theory of money, namely that the price level is going to rise when the money supply goes up while real output doesn’t. 

 

One major implication of MMT is that deficits without inflation shouldn’t matter because there is no such thing as “crowding out.”  Under the traditional view, government deficits must be funded by borrowing in the market for loanable funds, thereby shifting the demand curve for loanable funds to the right.  (Picture real interest rates on the y axis, the quantity of loanable funds on the x-axis, an upward slowing supply curve and downward sloping demand curve intersecting at some equilibrium point that new government borrowing is changing by shifting the demand curve to intersect the supply curve at a different – and higher -- point.)  That would cause an increase in real interest rates, which, in turn, would decrease investment (fewer investments would be profitable at higher real rates), which, in turn, would both decrease current GDP and slow further rightward movements in the long run aggregate supply curve due to the smaller resulting capital stock. 

 

MMT rejects this whole approach because it recognizes that deficits can be funded by issuing currency, not borrowing.  Government spending in excess of its revenues would increase the private sector’s surplus, which should increase the supply of loanable funds and actually drive down real rates, rather than increase them.  The flipside of this, of course, is that government surpluses require private sector deficits, thereby decreasing the supply of loanable funds and driving up real rates.  As Kelton puts the point:  “The crowding-out story gets it completely backwards.  It’s fiscal surpluses, not fiscal deficits, that eat up our financial savings.”  (p. 110)  Moreover, even if the government chooses to issue bonds equal to the amount of its deficit, that doesn’t change the analysis, because the extra dollars the government spent over its revenues must end up in the private sector as a surplus.  So, government spending creates the very private sector dollars that the private sector then chooses to convert into interest bearing notes.  Because these amounts are the same, the whole thing has no effect on real interest rates.  Again, to quote Kelton, “fiscal deficits – even with government borrowing – can’t leave behind a smaller supply of dollar savings,” and thus there can be no “crowding out.”  [Note this discussion is about real rates, rather than the inflation expectations embedded in nominal rates.]

 

This theory is not only interesting, but also actually explains the recent collapse in real rates.  For example, the real rate on the 10-year fell from roughly 4% during the Clinton surplus years to about zero today, despite massive deficits during that period.  It also suggests to me that the Fed QE program may be counterproductive.  If Treasury issuance has the salutary effect of sopping up excess reserves, why is the Fed undoing that by buying the bonds and thereby reinjecting the excess reserves? 

 

Moving on to the book's other 225 pages, Kelton repeatedly asserts that, because the government can never run out of money, any claim that the government can’t do something because it lacks the money is false.  Thus, we are told at length that the government has the money to provide jobs, health care, clean energy, etc. to everyone, if only we had the political will to issue the money necessary to pay for those things.

 

This part of the book is hard to take too seriously because it’s never squared with the fundamental point that we’re always limited by our real resources.  So, unless we have massive untapped resources of physical and human capital, we can’t have everything – we have to choose some point on our production possibilities curve.  Kelton never really addresses this point nor explains what we'd have to give up to get all the things she says we can have via money printing.  Instead, she sweeps this problem of practical scarcity under the rug with the assertion that, “because we chronically run our economy below its maximum speed limit, there’s almost always room to rev up spending without risking an acceleration in inflation.  And that’s what matters.”  (p. 257)

 

Kelton also largely ignores two other issues with revving up government spending via currency issuance.  First, at bottom, Kelton is advocating that the federal government allocate a greater percentage of the country’s real resources.  Taken to the extreme, this approach would get you to a Soviet system, which von Mises attacked as unworkable because there are no price signals to direct capital.  [See Economic Calculation in the Socialist Commonwealth]  Although Kelton is not advocating total central planning, it seems to me that her approach risks distorting the price signals a market economy uses to allocate capital efficiently, or seeks to ignore price signals altogether in favor of direct central planning and allocation of a much more significant portion of the economy’s overall resources.  I am skeptical that that type of program actually making us wealthier in the long run, given the knowledge it assumes on the part of the decision maker, though I would be willing to consider specific large scale spending projects or additional spending on public goods, particularly innovation.  One MMT-consistent alternative that might address this issue is more direct redistribution by simply giving the poor money so they can decide for themselves what they value most, rather than have central planners attempt to divine the ideal allocation of resources.

 

Second, is the view of government implied in Kelton’s recommendations too pollyannish?  For the reasons described above, the job of a central planner is hard enough, even assuming she’s acting in good faith.  But what if she isn’t?  What if much government action is mere rent-seeking, which would only further the mis-allocation of resources and leave us all poorer in the long run?  That concern may be beyond the scope of Kelton’s book but I think it’s essential to think about before trying to apply the MMT model to our reality.

 

EDIT:  After writing this I saw a review from Arnold Kling that explains some of my concerns about Kelton's arguments in a more persuasive and coherent way:  https://lawliberty.org/book-review/deficits-budgetary-and-conceptual/

Edited September 16, 2021 by KJP

[Munger_Disciple]

If I understand correctly, the federal government just prints money and spends under the Kelton/MMT theory in stead of deficit spending first and then issuing debt to mop up the excess reserves in the banking system. Isn't MMT pretty much the same as the current situation where the federal govt deficit spends, issues debt and the Federal Reserve does (forever) QE to release the excess reserves back into the banking system? So we have had a "mini" MMT regime for at least 10 years now. The only difference between the two options is that under the current system the Federal Reserve can occasionally threaten to slow down the QE to keep things in check. 

Edited September 16, 2021 by Munger_Disciple

[KJP]

1 hour ago, Munger_Disciple said:

If I understand correctly, the federal government just prints money and spends under the Kelton/MMT theory in stead of deficit spending first and then issuing debt to mop up the excess reserves in the banking system. Isn't MMT pretty much the same as the current situation where the federal govt deficit spends, issues debt and the Federal Reserve does (forever) QE to release the excess reserves back into the banking system? So we have had a "mini" MMT regime for at least 10 years now. The only difference between the two options is that under the current system the Federal Reserve can occasionally threaten to slow down the QE to keep things in check. 

 

I believe MMT theorists would say that the US monetary system has conformed to MMT since 1971 when gold convertibility of the US dollar ended.  At that point, the US obtained full monetary sovereignty and it has maintained that full sovereignty because all of its liabilities are denominated in US dollars and those dollars are not convertible into anything (e.g., gold, some other currency, etc.).  MMTers would say that MMT's description is how the world actually works, and it is descriptively accurate regardless of whether the government is spending more or less than it's receiving in taxes.  MMTers would also say it's descriptively true regardless of whether the government chooses to exchange IOUs (bonds) for some of the currency it has previously issued.

 

Most MMT theorists then go on to make a contested proposition -- that the US economy runs with a substantial amount of slack.  In light of that belief -- which depends on some view about the real world that is independent of the descriptive aspects of MMT -- they believe the US government can ramp up spending without triggering inflation.  They then make what are essentially political/ethical/moral arguments about how that "fiscal capacity" should be used.  I believe all of MMTers' suggested spending plans hinge on the claim that there is a lot of slack in the system -- we are far from full potential output -- rather than their fairly straightforward description of the relationship between the fiat currency issuer and the currency it issues. 

 

Indeed, MMTers admit that if an economy is close to full output, then additional spending will cause inflation.  You could perhaps get around that inflation via increased taxation, but that would result in less privately-directed economic activity, rather than additional output, because it's simply reallocating existing output (moving to some other point on the production possibilities curve) rather than employing currently unused resources. To the extent central planning does not allocate capital as efficiently as private markets, that would produce slower growth in the long run due to less capital accumulation.

 

Edited September 16, 2021 by KJP

[Pedro]

Great Book. Haven't finished it yet but ripping through it pretty fast.  Really enjoy it and has really explained MMT well. 

@KJP - Thanks for putting the book on my radar. Re your comment - This part of the book is hard to take too seriously because it’s never squared with the fundamental point that we’re always limited by our real resources. 

 

She addressed this constraint in the introduction.....literally on the second page of the book. See exert below. I don't agree with your assessment that she hasn't made clear  that we are limited by our real resources. She did in the beginning of the book and makes references to this throughout the book as well. 

 

Do I believe the solution to all our problems is to simply spend more money? No, of course not. Just because there are no financial constraints on the federal budget doesn’t mean there aren’t real limits to what the government can (and should) do. Every economy has its own internal speed limit, regulated by the availability of our real productive resources—the state of technology and the quantity and quality of its land, workers, factories, machines, and other materials. If the government tries to spend too much into an economy that’s already running at full speed, inflation will accelerate. There are limits. However, the limits are not in our government’s ability to spend money, or in the deficit, but in inflationary pressures and resources within the real economy. MMT distinguishes the real limits from delusional and unnecessary self-imposed constraints.

[KJP]

3 hours ago, Pedro said:

Great Book. Haven't finished it yet but ripping through it pretty fast.  Really enjoy it and has really explained MMT well. 

@KJP - Thanks for putting the book on my radar. Re your comment - This part of the book is hard to take too seriously because it’s never squared with the fundamental point that we’re always limited by our real resources. 

 

She addressed this constraint in the introduction.....literally on the second page of the book. See exert below. I don't agree with your assessment that she hasn't made clear  that we are limited by our real resources. She did in the beginning of the book and makes references to this throughout the book as well. 

 

Do I believe the solution to all our problems is to simply spend more money? No, of course not. Just because there are no financial constraints on the federal budget doesn’t mean there aren’t real limits to what the government can (and should) do. Every economy has its own internal speed limit, regulated by the availability of our real productive resources—the state of technology and the quantity and quality of its land, workers, factories, machines, and other materials. If the government tries to spend too much into an economy that’s already running at full speed, inflation will accelerate. There are limits. However, the limits are not in our government’s ability to spend money, or in the deficit, but in inflationary pressures and resources within the real economy. MMT distinguishes the real limits from delusional and unnecessary self-imposed constraints.

 

Yes, she acknowledges we're constrained by real resources (I said that in my initial post, but upon rereading it, it's not clear that I'm summarizing what she's saying, rather than my own view).  But in my view the back half of the book that talks about all the various programs that Prof. Kelton believes ought to be financed doesn't actually grapple with that constraint.  Specifically, she does not explain why she believes the US has sufficient untapped real resources of the necessary kind that it can accomplish all that she claims it can without significant inflation.   

 

Quite left-leaning economists have tried to point this out, and I have not seen a direct response from Prof. Kelton or other MMTers.  For example, see, e.g., Palley, Macroeconomics v. modern monetary theory:  Some unpleasant Keynesian arithmetic and monetary dynamics, in Modern Monetary Theory and its Critics (2020) (Fullbrook & Morgan, eds.)

 

From another review, here's a more eloquent way of saying what I'm getting at:

 

"It is indeed correct to say that when the government is bidding for resources, the risk of inflation is low if those resources are idle. It is also correct that unemployment is an indication of idle resources. But just because some resources are idle does not mean that the government can spend wherever it would like without affecting prices. The government would have to be an especially perspicacious and adroit entrepreneur to advance its priorities while only using idle resources.

 

Kelton does not explain why she believes that those currently without jobs would be a good match for her spending priorities. But without the ability to wave a magic wand to instantly transform the unemployed into teachers, skilled health care workers, and engineers specializing in energy alternatives, more spending in these areas would compete for scarce workers rather than soak up idle ones."

 

Kling, Deficits -- Budgetary and Conceptual, available at https://lawliberty.org/book-review/deficits-budgetary-and-conceptual/

 

The point is necessarily context dependent.  If we had Great Depression-like 25% unemployment, this point would have much less force.

 

Edited October 18, 2021 by KJP

[Pedro]

But without the ability to wave a magic wand to instantly transform the unemployed into teachers, skilled health care workers, and engineers specializing in energy alternatives, more spending in these areas would compete for scarce workers rather than soak up idle ones."\

 

 I wouldn't mind one lick if the government during tough times paid for someone to come take care of my homefront. Nothing a few online training classes for them and boom they can come help the homelife if they are willing and able to have that kind of work.  Lots of jobs could exists thats productive but not highly skilled.  Anyhout  I wonder if universal basic income would work here instead althought doing something porductive with labour sounds better.

 

Thanks for the paper to read. Its on my list once I finish re-reading the book.
 

 

 

 

[Cigarbutt]

Thank you for this thread. The topic has been lingering and the thought for a post was formed during my 17 km run today.

Entering MMT may not be a binary phenomenon. 'We' may have entered the vicinity of MMT for a while especially since the GFC and more recently.

 

With QE, money is sort of created from the asset swap but it's matched with inside money deposited at the Fed (liability) as reserves and the operation is still understood to be temporary (ie to be reversed). Also, QE has not been associated with consumer inflation (it may have prevented some consumer deflation?) but has clearly fed into asset inflation (opinion, as money 'trapped' in the financial system). QE is not MMT but it may conceptually be an intermediate step especially if the short term definition of the program becomes more permanent (there's nothing more permanent than a temporary gov****ent program).

 

In parallel however, there's been a development (not much talked about or discussed it seems) since the GFC and accelerating recently whereby US commercial banks have expanded their balance sheets by buying US government debt at a much higher pace (and accelerating) than GDP growth, which effectively is the equivalent of direct government financing or of overt monetary policy that becomes fiscal policy. The bank creates an asset (government debt) and simultaneously creates a matching liability deposit, ie new M2 money that ends up in private bank accounts, ready to be spent. The following link discusses that aspect from a concrete balance sheet perspective.

Money Creation: Misconceptions: Government Spending Creates Money by Alexander Pierre Faure :: SSRN

 

A fascinating aspect (if one accepts this concept) is that the Fed-Treasury have found a way to perform the effective equivalent of MMT without stating the objective and obviously circumventing the Federal Reserve Act.

Also, if one accepts this 'model', it's interesting to note that this version of MMT is sort of failing because consumer inflation took off significantly (level and duration) way before full potential employment and full use of resources to their potential. An argument could be made that the inflation is not well tolerated by the US masses (real income less disappearing transfers) and the inflation is being exported (given size of US 'stimulus', record negative trade balance and international reserve status) to nations where the CPI-essentials (food, fuel and rent) has the potential to hurt even more.

 

The following is a graphic construction (illustration) adjusted at zero around 2000 to show trends. The Treasury-Agency held is used as a proxy for the phenomenon described above.

 

 

Of course, if the above makes any sense, the interesting question is what's coming next..

[wabuffo]

The bank creates an asset (government debt) and simultaneously creates a matching liability deposit, ie new M2 money that ends up in private bank accounts, ready to be spent.

 

This is an incorrect statement.  The key question is not what creates a deposit.   The more important question is what creates a net financial asset for the private sector.  

 

Let's go through all four types that can affect deposits.   First, it is important to consolidate the Fed and the US Treasury as one entity (the central govt sector as fiat obligation issuer) and the private sector (including the domestic banking sector) as the other.

 

1) The Fed does QE by exchanging a reserve deposit for a US Treasury security with the private sector.   Clearly no net financial asset is created.  Private sector had a US Treasury security asset, after the exchange it has a reserve deposit.

 

2) The US Treasury issues a Treasury security.  Again, no net financial asset is created.  Private sector had a deposit balance and exchanged it for a US Treasury security.

 

3) A bank lends thus simultaneously creating an asset (loan) and a liability (deposit).   It's true a deposit has been created out of thin air.  But what I think gets missed is that for the private sector, its net financial assets haven't increased.   From a balance sheet equity perspective before the loan and after the loan, the net equity is the same for the private sector - so again, no net financial asset has been created.

 

4) The US Treasury spends.   This (and only this transaction) creates a new net financial asset for the private sector.  Just think of your net equity position right before and right after the stimmie check gets deposited into your deposit account.  Clearly this transaction has increased your net financial assets.

 

So from my perspective (which is the MMT perspective) - deficit spending by the US Treasury is the only one of these four transactions that creates a net new financial asset for the private sector.  The net equity position of the private sector increases by the amount of US Treasury deficit spending.

 

This is why the folks that focus on deposits (ie M2) get it wrong so often.   They focus on the wrong thing.  It's not the amount of deposits that counts - its the portion that is the direct result of US Treasury deficit spending.  

 

After that, the central govt sector (Fed + US Treasury) merely changes the composition of that new net financial asset by offering the private sector three forms in which to hold that new net financial asset (ie. currency in circulation, settlement reserves at the Fed, or US Treasury securities -- or as I like to call them in equivalent banking terms - a choice between cash, demand deposits, or time deposits).

 

It is really conceptually very simple - but few really understand it and so a lot of confusion gets spread around.  The final point is that a growing private sector economy needs more new, net financial assets from the central government (ie, deficit spending) - it is the natural order of things.  But its a balancing act, too much leads to inflation, too little (or even running a surplus) leads to deflation.

 

There I just summarized Kelton's poorly written book in one post.

 

Bill

Edited March 7, 2022 by wabuffo

[Cigarbutt]

^

What you describe as « incorrect » may simply be a matter of perspective or even accrual.

 

If i send you money electronically tonight, new money is created in your account until the money amount is actually accrued in my account tomorrow. When government spends money, there may be an accrual component but that money was/is/will be “absorbed” from a private market participant’s purchasing power. At least up to now, somebody needs to buy those bonds.

 

During some of the previous hyperinflation episodes (extreme example used for illustration), such as the Weimar episode (compared to milder contemporary inflation forms in France, England and the US who relied respectively more on taxes , debt financing and less on money printing), government deficits were met very little by taxes and debt financing and almost exclusively by discounting bonds at the central bank in exchange for new cash to be spent with the obvious market recognition that the central bank bonds would never be retired.

 

Since the GFC and especially since March 2020, a lot of the government bonds (which absorbed private purchasing power when issued) ended up on commercial banks’ balance sheets. This is effectively the same as if commercial banks directly financed the government ie by creating this bond, the cash that had been absorbed from private hands is injected back into consumer circulation (as a new deposit). When commercial banks effectively and directly finance the government, deposits are created the same way as if the central bank had discounted those bonds...

 

Again using an extreme example, let’s say private commercial banks finance 100% of the government deficit which becomes large, then broad money growth becomes decoupled from underlying economic activity and becomes aligned with government deficits, à la Weimar. It’s been tried before, it’s a slippery slope and my bet is that, somehow, this will be averted but who knows?

 

Concrete example now. Let’s say I lose my job and need to receive employment benefits. Let’s say 100CDN. The government goes into deficit and sends me a check. With accrual, we know that someone, let’s say wabuffo needs to buy a government bond in order to finance my employment benefit. No new private asset created as the government simply acted as an intermediate between wabuffo and me. wabuffo could have sent the money directly (do you want my bank account info for a transfer?  ). Now let’s say Metabank buys the government bond from wabuffo by crediting his account with cash. I now have 100CDN in my account and wabuffo has 100CDN in his account. If it’s not private money creation (or net financial asset creation), then what is it?

 

MMT is a lot more than overt government debt financing but the way commercial banks are expanding their balance sheets for government debt securities is an effective equivalent to overt debt financing.

 

Looking to improve my understanding here.

[wabuffo]

What you describe as « incorrect » may simply be a matter of perspective or even accrual.

 

CB  - why you are overly focused on deposits/reserves & coin/banknotes as "money" but not Treasury securities?  Treasury securities are money, too.  That's why M2 concepts are flawed.  Alternatively, why are US Treasury securities considered debts of the government, but currency in circulation and bank reserves are not?  

 

If you are going to throw around accounting terms like accrual, then I think you are obligated to think in terms of the domestic private sector's balance sheet and debits/credits.  You have to ask yourself a couple of questions. 

 

1) At what point does the private sector's aggregate balance sheet net equity (assets - liabilities) expand?  That's what's important - not individual balance sheets but the private sector's balance sheet in total.  Think this through in that example of yours.

2) What is the right way to think about the issuance of Treasury securities? 

 

As for examples from previous monetary eras - you have to remember that most of those times, the central government's money was subordinate to gold.  Thus, any time a breaking of a peg to gold happened due to any number of reasons, it was a devaluation that released the safety valve on bottled up inflation all in one go.  That's not the monetary regime we are in today.

 

Always enjoy these conversations, CB!

 

Bill

Edited March 7, 2022 by wabuffo

[Spekulatius]

I never get why people think that QE is similar to MMT. Those two things have little to do with each other. QE swaps one asset for another , it does not create new "money". Treasury spending does create new money (or equivalent) in the private sector. I think that's why QE does not create inflation but more treasury spending can.

MMT is just testing the limits on how much the treasury can be spent and it's much larger than many thought in the past.

 

Maybe I get it all wrong (wouldn't be the first time ). It's in a way amazing that the monetary system works, but we don't really know how and why and sort of have diametrically opposed theories about it.

Edited March 17, 2022 by Spekulatius

[wabuffo]

Spek - I think you understand it better than 99% of the people out there (even some PhD economists).

 

Bill