Wednesday, October 15, 2014

Thin Ice and Asset Swaps


Oilfield Trash wrote:

QE is nothing more than an asset swaps between the Fed and the Private Sector. With the FED merely swapping assets they are not actually "printing" any new money.

Auburn Parks wrote:

I'm confused, why would shifting already existing bank balances between different types of Govt bank accounts (reserves and securities) lead to inflation...?

I don't know. I don't know if "Quantitative Easing" is anything more than "asset swaps" or not. But I'm leery of statements that explain things I don't understand in terms of things I don't know about.

I would rather not know things, than know things that are wrong. If you have one wrong thing in your chain of argument, your conclusions are questionable. I prefer to keep my chains short.


I was driving home from work the other day, and something just popped into my head about "reserves and securities". Let's see if I can capture that thought.

A "security" is for example a Treasury Bond issued by the government. Maybe I buy the bond. I get the "security" and the government gets my money. For me it's like saving. For the government it's like borrowing.

The "security" is an "asset". It earns me income. And I can sell it if I want, so it has "liquidity".

If the central bank buys my security it does so using newly created money. The Fed gets the security and I get the new money. This is the exchange that is denigrated as "just an asset swap"

Cullen Roche, 9 August 2010:

The Fed simply electronically swaps an asset with the private sector.  In most cases it swaps deposits with an interest bearing asset.  They’re not “printing money” or dropping money from helicopters as many economists and pundits would have you believe.  It is merely an asset swap.

"Electronically"? Does that clarify things?

Here's the thought I had driving home: The Fed is buying up securities, and the former security-holders are seeing increases in the bank accounts that we call "reserves".

Do I have that right? I'm going with it.

So this "swap" leaves assetholders (the Fed aside) with a lot less securities and a lot more reserves. That seems to be the way it has worked.


There are two kinds of debt.

Oh, I know, three kinds of debt, Minsky types. I know. But I'm not talking about that. I'm talking about something else. Two ways to create debt: the kind created by direct borrowing, and the kind created by fractional-reserve borrowing.

Direct borrowing: You have a dollar and I want a candy bar. I borrow your dollar. No new money is created.

Fractional-reserve borrowing: You are a bank. I borrow a dollar. New money is created.

This was my thought:

When assetholders hold securities, the assets may support direct borrowing but they do not support fractional-reserve borrowing. But when assetholders hold reserves, the assets do support fractional reserve borrowing.

I know I'm on thin ice here. I'm just trying to figure this out.

When assetholders hold securities, the assets cannot generally be used to create new money. But when assetholders hold reserves, they hold assets you are required to have as part of the process of new-money creation. (Please don't explain to me that the reserves are not required until after the new money is created.)

The "swap" is an exchange of assets that cannot be used to create new bank money, for assets that can be used for that purpose. So I think the "it's just an asset swap" idea misses something very important.

32 comments:

Oilfield Trash said...

Art

" (Please don't explain to me that the reserves are not required until after the new money is created.)"

Ok I will not tell you this but maybe you will take the time and consider it by reading others

Basil Moore 1983, “Unpack­ing the post Key­ne­sian black box: bank lend­ing and the money sup­ply”, Jour­nal of Post Key­ne­sian Eco­nom­ics 1983, Vol. 4 pp. 537–556;

“In the real world, banks extend credit, cre­at­ing deposits in the process, and look for reserves later”.

Moore was quot­ing a Fed­eral Reserve econ­o­mist from a 1969 con­fer­ence in which the endo­gene­ity of the money sup­ply was being debated.

From the central bank of England

http://www.bankofengland.co.uk/publications/Documents/quarterlybulletin/2014/qb14q1prereleasemoneycreation

• the majority of money in the modern economy is created by commercial banks making loans.

• Money creation in practice differs from some popular misconceptions — banks do not act simply as intermediaries, lending out deposits that savers place with them, and nor do they ‘multiply up’
central bank money (reserves) to create new loans and deposits.

Reserves are used for interbank payments, not for deposit creation.

jim said...

OT wrote: "the majority of money in the modern economy is created by commercial banks making loans"

Obviously that has not been the case in recent years. This graph shows the growth of bank loans, bank deposits and currency in the US.

http://research.stlouisfed.org/fred2/graph/?g=NIU

Since Q4 2008 deposit money and currency have grown by about 3.5 trillion while bank loans have grown by 0.5 trillion. In the same time period the securities held by the Fed have grown more than 3 trillion.

Any market transaction can be called an asset swap so that statement really doesn't address the question of whether QE is inflationary.

Greg said...

"Any market transaction can be called an asset swap so that statement really doesn't address the question of whether QE is inflationary."

Transactions can be between two individuals, between an individual and a bank, between an individual and the govt (CB or Treasury) or between a bank and the govt. All these transactions are a swap of assets and whether they are inflationary or not depends on many factors, I would argue.

Lets start by defining inflation as money chasing goods. If the money chasing the goods is increasing and the goods are not increasing too there will be inflationary pressures

Transactions between two individuals cannot be inflationary as I see it. It is a zero sum transaction.

Get a bank involved and now inflation can be introduced because the bank can create more money (credit money) than is currently chasing goods.

Get the govt involved and inflation can arise as well since money can be created by govt.

Now when the govt and banks transact it is like two private individuals acting and is a zero sum transaction. The govt swaps a bond for cash but they are of same value. The cash earns no interest so it might be best to think of it as a zero interest bond. This is where the asset swap comes from.
So QE in and of itself is not inflationary in my view but there is the added question of what the person with the zero interest bond now does.

Many seem to think that the mere presence of zero interest bonds (cash or reserves) instead of interest bearing bonds is, in and of itself, inflationary. This is the hyperinflationistas argument at its core. I think its wrong. To me you must explain why someone who was saving (held a govt bond) and now has cash instead of the bond is going to stop being a saver and instead start consuming.

The argument Roche has been making is that through all these rounds of QE we have seen more and more people converting bonds (savings accounts) to deposits (checking accounts) yet we are seeing no real inflation in the CPI. We have seen asset markets rise suggesting that these people are "saving" by buying stocks instead of holding a govt bond. So QE has been inflationary to some degree on the stock markets but not in core inflation which is what everyone else was predicting (and many are still saying is inevitable).

Auburn Parks said...

Hey Art

"A "security" is for example a Treasury Bond issued by the government. Maybe I buy the bond."

You trade dollars in you private bank checking account for dollars in a Govt bank term savings account. This isnt really you buying anything. "Buying" generally implies that you are actually losing your bank deposits, here you are just switching banks.

"I get the "security" and the government gets my money."

No, the Govt doesnt get your money. Your money is right there in your securities=savings account. Just like how private banks dont make loans out of your private bank deposits, the Govt doesnt spend out of your Govt bank deposits. This one operational detail makes everything else clearer.

"For me it's like saving. For the government it's like borrowing."

Borrowing is a term that applies for non-currency issuing entities. When you and I borrow, we must use our assets to pay back our debts or liabilities.

The Federal Govt is the only entity that can use its own liabilities to pay back its other liabilities. The Govt uses Fed liabilities (Reserves) to pay back TSY liabilities (TSY-securities), they are all Govt liabilities.

"If the central bank buys my security it does so using newly created money. The Fed gets the security and I get the new money. This is the exchange that is denigrated as "just an asset swap"

Calling accounting entries in reserve accounts (Govt liabilities) "money" and not accounting entries in securities accounts (Govt liabilities), confuses the issue. QE adds no Govt liabilities to the system.

"Two ways to create debt: the kind created by direct borrowing, and the kind created by fractional-reserve borrowing. "

Since fractional reserve lending is a myth in the fiat eras, its more appropriate to classify the distinction between these two types of lending as bank and non-bank lending.

Bank lending adds to the supply of bank deposits (money) and non-bank lending does not add to the supply of bank deposits (money).


"When assetholders hold securities, the assets may support direct borrowing but they do not support fractional-reserve borrowing. But when assetholders hold reserves, the assets do support fractional reserve borrowing."

Banks do not lend reserves to non-banks, it is not possible. There is no mechanism to go from reserves to bank loans. Banks make loans, and then have up to two weeks to meet the legal reserve requirements. Canada doesnt even have reserve requirements, which further proves the point.

https://research.stlouisfed.org/fred2/graph/?graph_id=201068

As you can see, reserve levels have never had any correlation (let alone causation) with bank lending:

https://research.stlouisfed.org/fred2/graph/?graph_id=201070s

Auburn Parks said...

Or maybe lets put this another way.

First, lets recognize a few constraining realities:
The Govt only spends and taxes with reserves, and only reserves can be converted into T-securities.
The FFR is determined only in the reserve market.

And lets assume that in our model world, society is rational and that everyone accepts and agrees that the Govt can spend without having an existing balance in its account at the Govt's bank=the Fed.

So if over the course of a year, the Govt adds $4T reserves to the economy through spending and only removes $3.5T reserves through taxation, then the banks would be left with $500B in reserve accounts on the Fed's balance sheet, owned by banks

And if the Govt never issued any securities, the reserve level would continue to increase by the size of the deficit every year.

Banks would have ~$17T worth of reserves in the USA right now.
The Govt would have ~$17T worth of financial liabilities.

After this money has been spent, and its in the hands of the private sector as financial assets. What difference does it make if the Govt issues securities and allows people to have access to a different type of bank account at the Fed? The amount of Govt liabilities is the same. Similarly, why would it matter if people converted the savings=securities accounts back into reserve accounts, either way the size of the Govt liability spreadsheet stays the same?



The Arthurian said...

"The cash earns no interest so it might be best to think of it as a zero interest bond."

Absolutely not, Greg. Think of cash as cash. It's simpler that way. If the cash ends up in an interest-paying account, think of it as cash that's tied up in savings, where it
1. hinders circulation by reducing the circulating medium, and
2. increases income to the financial sector and decreases income to the productive sector.

"To me you must explain why someone who was saving (held a govt bond) and now has cash instead of the bond is going to stop being a saver and instead start consuming."

Agreed!

The Arthurian said...

Auburn, thanks for responding line-by-line to my thoughts. I'm glad you did.

Here's something. I said:
when assetholders hold reserves, the assets do support fractional reserve borrowing.

Here's your reaction:
Banks do not lend reserves to non-banks, it is not possible.

Okay. But I didn't say banks lend reserves. I said reserves support fractional reserve borrowing.

Now let me swap out some words you find inappropriate:
reserves support bank lending

In the post I said reserves "part of the process of new-money creation."

Now I don't give a flying F if the banks "have up to two weeks to meet the legal reserve requirements." They have to meet them. Reserves are part of the process.

Reserves are part of the process and (I think) securities are not. Therefore the swap of reserves for securities is NOT "just" an asset swap.

It changes something that cannot be used to create a tenfold increase in bank money, into something that can be used to create a tenfold increase in bank money.

That was the thought I had, driving home the other day.

The Arthurian said...

Here's something else.

I said: Maybe I buy a bond issued by the government... "I get the "security" and the government gets my money."

Auburn, you replied:
"No, the Govt doesnt get your money. Your money is right there in your securities=savings account... the Govt doesnt spend out of your Govt bank deposits."

So what you are saying is, when I buy a government security I have the bond AND I STILL HAVE MY MONEY TOO. And the government pays me interest besides?

Now that's gotta defy accounting logic!

The Arthurian said...

Interest is the reward for "parting with liquidity", Keynes said.

jim said...

Art wrote: "Now I don't give a flying F if the banks "have up to two weeks to meet the legal reserve requirements." They have to meet them. Reserves are part of the process."

Reserves are not part of the process of making bank loans. They are part the process of making sure banks have sufficient currency available to meet depositor need for cash.


It is not the bank where a loan originates that is required to change its reserved amount. It is the bank where the loaned money is deposited that must meet the reserve requirement for those deposits and if the money is deposited in a saving account the reserve requirement will be zero. Reserves requirements have no influence on the banks position on making loans. There isn't any way reserve requirements can influence the loan making process.

If a bank gets an increase or decrease in checking account money that usually has nothing at all to do with what the banks loan department has been doing. And the loan department generally will have no idea how any particular loan will affect the banks reserve requirements since it probably will have no effect... If depositors move money from checking to saving (or the reverse) It changes the bank's reserve requirements. That has nothing to do with whether the bank did or did not make a loan. If you write a check to someone who banks at a different depository institution you have changed both bank's reserve requirements even if neither bank has done any lending. The reserve requirement has nothing to do with loans.


The purpose of required reserves is to guarantee that the payment system remains functional. In the final analysis, people trust banks with their money only because they know they can withdraw the money in cash if they ever want to. When people become convinced that the banking system is unable to deliver their money on demand very bad things happen. The purpose of reserves is to ensure that the banking system doesn't lose the trust of the public because a loss of trust turns into bank runs and the payment system and economy grinding to a halt.

Basically required reserves consist of bank vault cash plus a small slush fund account at the federal reserve that covers two weeks worth of fluctuations in deposits that are reservable.

Auburn Parks said...

Hey Art-
In reply to your comment at 8:41PM:

reserves have no bearing on loan making decisions by banks. Jim does a pretty good job of explaining in his comment at 9:28AM on 10/16.

Before QE, reserves never constrained bank lending, as you can see from the two graphs I provided, the two measures aren't even correlated.

So if reserves were never a constraint on bank lending before QE, how can they add to bank lending during QE?

Auburn Parks said...

"So what you are saying is, when I buy a government security I have the bond AND I STILL HAVE MY MONEY TOO. And the government pays me interest besides?

Now that's gotta defy accounting logic!"

T-securities are not like muni-bonds or corporate bonds where you give up your bank deposits for a bond.

T-securities are bank accounts at the Fed, so you are simply changing banks. A 6-month T-bill is much more analogous to a 6-month CD than a 6-month Chicago Muni bond.

This is the key distinction that it confusing the issue for you. All of these terms are corrupted and misused because the economic orthodoxy has not and will not acknowledge and apply the fundamental changes that took place in 1971.

And no it doesnt defy accounting logic:

Here is the accounting for buying a corporate bond for $1000: (A) = asset (L) = liability, assume both parties are using the same bank

Bond Buyer A:
(A) Private bank checking account -$1000
(A) Corporate bond +$1000

So this person's bank account assets go down, and their financial assets go up the same amount, net change = 0

Corporation selling bond:
(A) Private bank checking account +$1000
(L) Corporate bond +$1000

The corporation's net position change is also zero, they have the bond buyer's bank deposits now to go along with their new liability. But no new bank deposits have been created.

Private Bank B:
(L) Bond buyer A's checking account -$1000
(L) Corporation's checking account +$1000

The banks financial position remains unchanged, and no reserves were moved around at the Fed because the transactions all took place inTRA-bank.

So at this point, the non-bank lending process resulted in $1000 worth of financial assets being created (the bond) but there was no change in the number of bank deposits.

Auburn Parks said...

And now the accounting for a Govt bond issuance:

T-sec buyer A buys $1000 T-security:
(A)Private bank deposit -$1000
(A)Govt bank deposit +$1000

So T-sec buyer A has no change to their financial position. But instead of losing bank deposits like the corporate bond example, you move your money into a bank account at the Fed.

Private bank A:
(L) T-sec Buyer A's bank deposit account -$1000
(A) Private bank A's reserve account at the Fed -$1000

So here, instead of the bank's liabilities staying the same, they are actually losing that liability along with their reserve assets.

Govt Fed
(L) Private bank A's reserve account -$1000

So the Fed's liability level goes down by $1000, as reserves in the TGA are not counted in the monetary BASE (Fed liabilities).

Govt TSY
(L) TSY bonds +$1000

So the TSY's liabilities go up, but reserves in the TGA are not counted in any aggregate measure of Govt financial assets, in the same way as newly printed cash is not counted as "money" until it is owned by a private sector participant.

So whats the net effect?

The economy loses $1000 worth of private bank deposits. But gains $1000 worth of Govt bank deposits (securities accounts), so the net total bank deposit impact is zero.

The Govt Liabilities amongst agencies get cancelled, so the number of Govt liabilities stays the same. The Fed loses $1000 worth of reserve liability but the TSY gains $1000 worth of T-security liability.

So ultimately, issuing T-securities does nothing, it adds nothing. The private sector gets a Govt bank deposit and loses a private bank deposit, and the Govt liability stays the same, just shifts from the Fed (reserves) to the TSY (securities).

Corporate bond issuance results in no change to private bank deposit levels, but increases financial assets by the bond amount issued.

Federal Govt bond issuance results in a loss of private bank deposits, a swapping of Govt bank deposits between the private bank that lost its reserves and the private T-sec buyer who now has those bank deposits at the Fed. But other than that, nothing happens.

Which is why QE is not stimulative. Its the same nothing process, just in the reverse.

Auburn Parks said...

Which is why MMTers have correctly pointed out that the whole T-security process is largely irrelevant, as it adds and subtracts nothing.

Its the Govt spending and taxation decisions that impact the economy.

If the Govt spends more than it taxes, the non-Govt keeps the balance as income, after the fact accounting hi-jinx (QE, issuing T-securities etc) makes no difference. As the accounting shows.

The Arthurian said...

Jim: "Reserves are not part of the process of making bank loans. They are part the process of making sure banks have sufficient currency available to meet depositor need for cash."

If, when, and where there is no reserve requirement, reserves are part the process of making sure banks have sufficient currency available to meet depositor need for cash.

Everywhere else, reserves are that, and also they are part of the process of making bank loans. By law.

jim said...

The federal law that requires banks to maintain reserves doesn't even mention loans.

The amount of reserves that a deposit taking facility is required to maintain is based on the amount of deposits subject to reserves. Savings accounts require zero reserves while checking accounts are subject to reserve requirements. There is no connecting link between the amount of loans that a bank makes and the amount of deposits that the bank's customers choose to put in their saving or checking accounts.

There is no connection between the aggregate loans of the entire banking system and the aggregate reserves of all banks since there is no way to know whether people will be putting the borrowed money into a checking or saving account.

The Arthurian said...

Jim, so "loans create deposits" doesn't count, now?

The Arthurian said...

Auburn,
I majored in math but I can't balance a checkbook. And I always have trouble with T Accounts... with Balance Sheets. For me it really helps that you break the balance sheet entries down to single transactions.

Now let's see what they show. First you show "the accounting for buying a corporate bond":

Bond Buyer A:
(A) Private bank checking account -$1000
(A) Corporate bond +$1000


Later you show "the accounting for a Govt bond":

T-sec buyer A buys $1000 T-security:
(A)Private bank deposit -$1000
(A)Govt bank deposit +$1000


In both of these examples, the balance in the buyer's transaction account decreases by $1000. Of course, because in both cases the money is coming out of that account.

In both of the examples, the balance in the sellers transaction account increases by $1000. Again this makes sense, because the money that came out of one pocket went into another.

Oh, but that's not what you're showing, is it. You're showing me that for the corporate bond, the buyer's "bond" assets went up $1000 when the buyer's "checking account" assets went down $1000. Well that's fine.

Then in the second example you show that for the government bond, the buyer's spending account decreased $1000 and the buyer's saving account increased $1000. You claim that these two cases are completely different.

I think it is doubletalk to say that in the one case the corporation gets the $1000 but in the other case the buyer keeps the $1000.

If the corporation gets the $1000 when I buy a corp bond, then the government gets the $1000 when I buy a govt bond.

The government security, like the corporate security, is a financial asset that returns the money to the buyer after a period of time.

In both cases, until that time arrives, the buyer has parted with liquidity. In other words, the buyer can't spend the money. The corporation or the government can spend it during that time.

So then, I have to say the same thing I said up top: I get the "security" and the government gets my money.

Greg said...

"I think it is doubletalk to say that in the one case the corporation gets the $1000 but in the other case the buyer keeps the $1000.

If the corporation gets the $1000 when I buy a corp bond, then the government gets the $1000 when I buy a govt bond.

The government security, like the corporate security, is a financial asset that returns the money to the buyer after a period of time.

In both cases, until that time arrives, the buyer has parted with liquidity. In other words, the buyer can't spend the money. The corporation or the government can spend it during that time"


The govt may "get" your 1000$ but they don't need it in order to spend 1000$, where as the corporation does, unless it wants to get a bank loan. That is one major difference.

Another difference is that the govt bond is nearly identical to cash. You can take that govt bond to a bank and deposit it if you wish and your deposit account will be credited with the amount you paid for the bond (not the face value of the bond), from what I understand.

The problem Art are the models that make the govt just like a company that needs to procure funds from private sources before they are able to spend. None of these models stay coherent under all financial conditions. If they aren't coherent under all conditions they are an incorrect model.

jim said...

Art, "Loans creates deposits" has not counted for 6 years.

If the banking system was solely dependent on loans for deposit creation the last 6 years would have been another period of monetary collapse just like 1929-1933 when bank deposits shrank by 40%.

Instead of shrinking in the last 6 years, bank deposit creation has grown by 40%. Bank lending has just barely created enough money to satisfy the public's increased needs for cash currency. The increase in total bank deposits in the last 6 years can entirely be accounted for by the Fed's purchases of securities.

But even back when "loans create deposits" did count the amount of required reserves had no connection to the volume of bank lending. Required reserves are (by law) tied to "transaction accounts". Transaction accounts are M1-currency. The volume of transaction accounts has no connection to the volume of bank lending.

You can see from this graph that currently required reserves is about 8 percent of transaction accounts and that there is no correlation between required reserves and the volume of bank loans.
http://research.stlouisfed.org/fred2/graph/?g=NXi

The reason that required reserves is not 10% of transaction accounts is that the first $13 million of such accounts is exempt and the next $75 million has a 3% reserve requirement. It is only transaction accounts in excess of $89 million that have a reserve requirement of 10%.

Back before 1990 demand deposits were a fairly constant portion of total deposits and thus were correlated to loans. That correlation made the story that reserve requirements were somehow influencing lending seem plausible. But since 1990 the direction of bank lending and required reserves has diverged. From 2004 to 2008 reserve requirements were falling while bank lending was growing dramatically. And after 2008 reserve requirements were growing while bank lending was shrinking. The data from the last 25 years makes the story that reserve requirements influence bank lending completely implausible.

Oilfield Trash said...

jim said...
Art, "Loans creates deposits" has not counted for 6 years.

This does not seem right to me, when I look at TCMDODNS, from 200801 to 201402, Domestic nonfinancial sectors; credit market instruments; liability increased by 7.2 trillion. FEDS balance sheet grew in the same time period 3.5 trillion.

Another thing is this issue with private sector demand for currency. When I look at the latest release of the Fed Balance sheet (H.4.1). Currency is always accounted for in the Feds liabilities, which impacts the stock of Fed reserves.

You stated that "Bank lending has just barely created enough money to satisfy the public's increased needs for cash currency."

This to me seem to be suggesting that bank lending creates the demands for these Fed liabilities.

I suppose it might if the private sector loses faith that the claims(deposits) they have on the banking sector will not be honored at par.

However currently the FED reserves are 2.8 trillion and to my knowledge the FED has never not provided the reserves necessary to cover the private sector demand for currency.

Reserves to me are nothing more than a accounting construct of Assets-liabilities, which could be defined

"as the capital employed in a company, computed by deducting the book value of the liabilities from the book value of the assets. Also called net assets, net worth, shareholders' equity, or shareholders' funds."


Last point I want to address is regardless of how you measure deposits at a Macro level, it will never tell you how they impact the price level of any basket of good you are indexing.

The macro price level of any basket of goods is about the turn over of the money stock to some measure of time (flow) and not about the size of the money stock.

Which is why I used the change in debt to gauge this turn over, since most people only enter such contracts with the intention to spend.



jim said...

OT wrote:"This does not seem right to me, when I look at TCMDODNS"

Why are you looking at TCMDODNS? That tells you nothing about bank deposit creation. Increases in govt debt or corporate debt or federal student loans have no impact on the quantity of bank deposits.

The quantity of bank deposits can increase in only 3 ways: When banks make loans, when the FED buys assets (securities and gold bullion) or when people deposit currency in their accounts. Conversely, deposits go down when bank loans are paid back, the Fed sells assets and when people withdraw currency from their bank. In the last 6 years deposit creation by bank loans has just barely covered the amount of deposits reduction due to currency withdrawals.

As you can see from this graph bank deposits have grown by about $2.5 trillion more than bank loans:
http://research.stlouisfed.org/fred2/graph/?g=NZQ


Currency in circulation has increased by about 1/2 trillion. Take away the $3 trillion in Fed asset purchases and bank deposits would not have grown at all in the last 6 years.

Looking at graph of loans and deposits from the end of 2008 to the beginning of 2010 it is pretty obvious (to me) that the US banking system was headed for the same type of monetary collapse that occurred from 1929-1933 if deposits been allowed to follow loans due to Fed inaction.

BTW, the story that banks make loans and then seek out reserves is totally inaccurate. That story is is silly nonsense. Banks don't consider their past lending history when computing reserves. When banks compute the amount of reserves they need to hold they look at the amount of money in the banks demand deposits accounts. The amount of money in checking accounts and NOW accounts. The banks loans are irrelevant to computing reserves.
When you move money from savings account to checking that affects the bank's reserve requirement. When you borrow money it has no affect on the bank's reserve requirement. The bank has no reason to look for reserves after a loan is made.

geerussell said...

Hi, Art. Lots of good discussion on this one, but everyone else took the easy part. I want to grab what looks like a neglected loose thread and pull on it to make a more subtle point.

Direct borrowing: You have a dollar and I want a candy bar. I borrow your dollar. No new money is created.

Let's consider that for a moment. I'm going to deploy balance sheets as a framing device because we need a scorecard to track the financial assets.

On the lender's balance sheet...

Assets: -1 dollar federal reserve note; +1 dollar borrower's IOU

On the borrower's balance sheet...

Assets: +1 dollar federal reserve note
Liabilities: +1 dollar borrower's IOU

On the Fed balance sheet (see liabilities here)...

Liabilities: 1 dollar federal reserve note

What does our scorecard tell us? Something was created here. A dollar-denominated financial asset. We know this because we carefully kept score, the borrower's IOU didn't exist before the loan, now it does. Further evidence of creation is the gross expansion of the borrower's balance sheet.

The Fed's balance sheet is included for completeness to underscore the point that every financial asset has an issuer and exists as an asset for someone and a liability for someone else. For the party who holds the asset, that asset is money. Money that can be spent or lent by transferring that asset to someone else.

Wait a minute. If "money" is the asset side of a financial asset and the borrower created a new financial asset in the issuance of his IOU... money was created.

That's your basic Minsky view. Money is an IOU, anyone can issue it the trick is getting someone to accept it. Banks are in the business of acceptance and guarantee. Put it all together with borrowers seeking loans, banks offering acceptance and guarantee and a central bank accommodating the process with settlement balances and you have the endogenous money story.

For the lender in our example here, the borrower's IOU is money. That IOU can be spent or lent. Far from a mere thought experiment, this exercise basically describes the activity falling under the rubric of "shadow banking".

Greg said...

The problem, as I see it, with the "direct borrowing" example is that bank borrowing is how virtually all borrowing happens. Using a direct borrowing thought experiment gets us no closer to understanding how our current economy functions.

Borrowing money from your neighbor does create "money" because it creates an IOU that in theory could be traded (I could use a piece of paper that says "Art owes ten dollars to holder of this paper" and in certain circles I might be able to get that paper traded for something else) but its not really new money out of thin air. Its just putting a new name on the old money.

When Art lends me 10 dollars he truly cannot spend that 10 dollars anymore. he is spending restricted by 10 dollars and I am spending enhanced by 10 dollars but the macroeconomy has the same spending capability. I might want something different than Art and I might spend that 10 dollars in a silly manner, even paying 10 dollars for something everyone else is only paying $7.50 for...... and if enough people do this, it might be measured as inflation but this is a very unrealistic view, not like the real world at all. This zero sum game does not describe a bank loan however. When I go to borrow $100,000 to buy a house, there is no one on the other side of the transaction that has 100,000$ less to spend as a result. This $100,000 gets spent and added to GDP and millions more people can do this as well, even on the same day !!! The only thing that restricts their borrowing is THEIR OWN income producing capabilities, not the generosity of some neighbor.

Oilfield Trash said...

In my opinion the appropriate question to ask is not ‘What is the maximum level of debt that can be plausibly paid back?”. It is ‘What is the maximum level of debt that can be plausibly serviced on a permanent basis?’.

If there are any Ponzi schemes in debt, they exist only if and when there are real limits to economic growth – working-age population growth, energy limits, government regulations ect.

Oilfield Trash said...

Jim

OT wrote:"This does not seem right to me, when I look at TCMDODNS"

Why are you looking at TCMDODNS? That tells you nothing about bank deposit creation. Increases in govt debt or corporate debt or federal student loans have no impact on the quantity of bank deposits.

I think I understand what you are getting at. But maybe you are taking the loans create deposits statement literally(just a guess on my part).

My viewpoint is loans creates deposits much like oil creates gasoline. All oil can create gasoline not just West Texas intermediate crude.

geerussell said...

The problem, as I see it, with the "direct borrowing" example is that bank borrowing is how virtually all borrowing happens. Using a direct borrowing thought experiment gets us no closer to understanding how our current economy functions.

What I'm contending is that "direct borrowing" bears a rough correspondence to shadow banking activity, the scale of which dwarfs ordinary bank borrowing. As such, it's highly relevant along side ordinary banking and government to form a fuller understanding of how our current economy functions.

geerussell said...

When Art lends me 10 dollars he truly cannot spend that 10 dollars anymore. he is spending restricted by 10 dollars and I am spending enhanced by 10 dollars but the macroeconomy has the same spending capability.

A balance sheet view can settle the question with some certainty.

Look at Art's balance sheet after he lends you 10 dollars. It has not shrunk. It has not grown. Look at the macro balance sheet after Art lends you 10 dollars, it has expanded. There is more spending capability created through lending.

When Art lends you 10 dollars, he exchanges one dollar-denominated financial asset worth 10 dollars (federal reserve notes) for another, newly-created dollar-denominated financial asset worth 10 dollars (your IOU).

Art now holds a different financial asset but it is still one that can be spent, lent, rehypothecated.

The Arthurian said...

geerussell: "Art now holds a different financial asset but it is still one that can be spent, lent, rehypothecated."

Your confidence that every scrap of paper with "IOU" written on it can be used for money is... well... I'm nowhere near as confident as you are.

The Arthurian said...

And to the extent that you are right, geerussell, what you are describing is a problem. It is a very late stage in the long and ultimately unsustainable process of increasing financialization.

Greg said...

I agree with you Art that what gee describes is part of the problem but he is correct that the IOU becomes tradeable and a form of money.

The problem I think is when that IOU becomes worth 10 dollars to everyone down the line. The further you get from Art, from actually knowing him, there will be a discount added. Why would I give you ten dollars worth of stuff for that IOU. I would likely give you only 8.50$ or something as a hedge if I didn't know Art personally. And if you accepted it I might wonder why, if Art is so trustworthy, would you take 8.50$ today for a sure 10.00$ from Art later?

In our current ongoing crisis we had these IOUs that were certified by rating agencies as "good as Fed Res Notes" essentially, when there was absolutely no reason to think they were. These were mortgage notes in many cases where the borrower only intended to make 6-10 payments and turn around and sell the houses. They often didn't have the income to service them long term and if they did they weren't going to throw it into the house.

Too many layers of financialization.

geerussell said...

Your confidence that every scrap of paper with "IOU" written on it can be used for money is... well... I'm nowhere near as confident as you are.

This is another good place to quote Minsky

The second route to endogeneity of the money supply focuses on bankers and other players in financial markets as entrepreneurs who seek profits by innovating, by developing new ways to finance positions in existing assets and investment (the creation of new assets). Financial innovation also involves the creation of new assets for the portfolios of both individuals and institutions. these new portfolio assets may well take on the characteristics of money.

The second route emphasizes the changing nature of what passes for an economy's money supply. It leads to an emphasis upon the credit or asset side of the balance sheets of financing institutions. the demand for credit takes the form of proposals to finance that bankers first promote and then either accept or reject. Those proposals that meet the banker's standard of the time will be financed and the funds for this financing will be pulled out of the existing stocks of short term financial assets. Furthermore financing leads to the generation of new types of financial instruments that are accepted into portfolios.

Not all scraps of paper are created equal and there is little audience for Art's IOUs or the promise I wrote on a cocktail napkin. When the issuer is widely viewed as credible though, the scraps of paper may find widespread acceptability and that activity by credible issuers is what makes up the "shadow banking" sector.

And to the extent that you are right, geerussell, what you are describing is a problem. It is a very late stage in the long and ultimately unsustainable process of increasing financialization.

100% agree it's a problem. Everything I'm saying here is just to specify the mechanics of that financialization. As the aforementioned banker's standard of the time has shifted further and further from finance to fund productive real activity towards finance to bid up the price of existing assets, the burdens and costs of debt have gone up and it becomes more unsustainable.