Tweet
This post is a continuation of a spur of another thread where these opinions made me a minority of one. Hopefully, either others will enlighten me or vice versa.
Why pawnshops? Banks buy, sell, and make loans against financial assets: securities (debts, equities, derivatives), currencies, and in some cases hard assets (e.g., houses, cars, inventories, etc.) on which they occasionally foreclose.
Why counterfeiters? They pay with money that not currently part of the money supply and often freshly issued out of thin air. Suppose that I sell a Treasury bond to a bank. They can pay me any of three ways:
- They can reach into their vault and pay me in cash, but that vault cash is not part of the money supply, but cash in my hands is.
- They can pay me with a bank draft (cashier’s check), but that is freshly created money, which will become part of the money supply as soon as I deposit in my checking account at some bank.
- They can pay me by incrementing the credit in my checking account at that bank, which increments the money supply by the same amount.
In all three cases they’ve taken a financial asset (the bond) out of circulation and introduced new money into the money supply — some might say they’ve monetized the asset. Note that, in the last two cases, the money is created on the spot, out of thin air. Also, note that the bank has the same payment options when they make a loan.
Finally, note that no matter which form I get the money in, I can immediately convert it to either of the others. If I get the money as check or cash, I can immediately deposit it in my account at that bank. When I have the money in that account, I can ask for cash or cashier’s check or simply leave it there.
Why pseudo? The term “pseudo counterfeiter” is an intentional double negation; it means “a fake fake.” And, indeed, bankers are not counterfeiters and for one and only one reason: the federal government recognizes the money they create as “good as cash,” especially in payment of taxes, and thereby, making it “coin of the realm.”
So what? I made this same point in a prior post, Fool’s Gold, and it seemed uncontroversial. In the more recent thread, it seems agreed that banks can pay for loans with freshly created money. And, the point under dispute is whether banks can buy securities that way.
To me the most compelling argument that loans and asset purchases can be handled in the same way is that the books are kept the same way in each case, i.e., the asset (bond or loan agreement) is debited to the current-asset account and the corresponding amount is credited to the seller’s or borrower’s checking account or the account on which the cashier’s check is drawn. But here are some relevant quotes.
* Here is a recent reference by Nick Rowe:
“An individual commercial bank can create money out of thin air simply by buying something.”
* From Washington’s Blog:
Germany’s central bank – the Deutsche Bundesbank (German for German Federal Bank) – has admitted in writing that banks create credit out of thin air.
As the Bundesbank states in a publication entitled “Money and Monetary Policy” (pages 88-93; translation provided by Google translate, but German speaker and economic writer Festan von Geldern confirmed the basic translation):
4.4 Creation of the banks money
Money is created by “money creation”. Both [central banks] and private commercial banks can create money. In the euro monetary system [money creation] arises mainly through the granting of loans, as well as the fact that central banks or commercial banks to buy assets such as gold, foreign currencies, real estate or securities. If the central bank granted a loan from a commercial bank and crediting the amount in the account of the bank at the central bank, created “central bank money.”
* Also from Washington’s Blog:
“Do private banks issue money today? Yes. Although banks no longer have the right to issue bank notes, they can create money in the form of bank deposits when they lend money to businesses, or buy securities. . . . The important thing to remember is that when banks lend money they don’t necessarily take it from anyone else to lend. Thus they ‘create’ it.”
-Congressman Wright Patman, Money Facts (House Committee on Banking and Currency, 1964)“The modern banking system manufactures money out of nothing. The process is perhaps the most astounding piece of sleight of hand that was ever invented.”
- Sir Josiah Stamp, president of the Bank of England and the second richest man in Britain in the 1920s.“Banks create money. That is what they are for. . . . The manufacturing process to make money consists of making an entry in a book. That is all. [...]”
- Graham Towers, Governor of the Bank of Canada from 1935 to 1955.
As a side note, Paul Krugman recently made the foolish mistake of claiming that banks are constrained in their loaning by their capital on hand. Here is Scott Fullwiler of Bard College tearing Paul to pieces.
UPDATE: The fact that banks don’t have to spend real money on the assets they bid on and acquire explains the propensity of our economy to form asset bubbles, especially over the last thirty years, when the Fed has been run by neoliberals who believed that regulation was neither necessary nor sufficient and, in fact, impossible. (According to Larry Summers, it’s impossible to find regulators who are both honest and competent.)
And, during that period, we’ve seen many asset bubbles: junk bonds, dot-com equities, subprime mortgages, etc., all due to the fact that banks can invest without paying. And, this last bubble wrecked the world economy.
UPDATE2: Per John Kenneth Galbraith writing in ‘Money: Whence it came, where it went’ (1975):
“The study of money, above all other fields in economics, is one in which complexity is used to disguise truth or to evade truth, not to reveal it. The process by which banks create money is so simple the mind is repelled. With something so important, a deeper mystery seems only decent.”
Again, here’s how it’s done. Whenever a bank pays for anything:
- The relevant inventory or service account gets debited.
- If the vendor has an account at that bank, it gets credited; otherwise, the vendor gets a cashier’s check and checks-payable gets credited. In either case, that credit constitutes freshly created money within the banking system.
When a check against the vendor local account or that cashier’s check gets deposited at another bank, reserve at the Fed moves from the paying bank to that other banks, i.e., the freshly created money gets moved from one bank to another, but it still exists within the banking system.
33 Comments
A lot of insight. What is the counter arguments?
Their own knowledge and expertise, which I readily acknowledge vastly exceeds mine. And, that’s why I acknowledge that the outcome may we be that I get disabused of a misconception.
If I may also be so bold as to continue on with a spur from another thread (that I think is related), there is an article by Salvator Babones on yesterday’s counterpunch.org entitled ‘A bold New Call for a Maximum Wage’ that I think would maybe bear on all the out-of-control money-sloshing that is going on in banks and elsewhere in the 1% of the 1%. The article says what the title says – maximum wage, ceiling on all outrageous CEO salaries.
On the other thread, which I can’t find now, the question of worth between Presidents Hoover and FDR were being addressed, with the implication that WWII solved the problem of the Depression and led to peace and tranquillity and a Better Way of Life.
Mr. Babones’s article suggests otherwise, supporting my fuzzy claim that wars don’t create wealth:
” … FDR did end up winning a 94-percent top tax rate on income over $200,000, a move that would help usher in the greatest years of middle-class prosperity the United States has ever known.”
I rest my (former) case.
Should be ‘was being addressed’, sorry.
Mitchell says
“What if the federal government took over all banking functions and eliminated private banks? What would be the advantages and disadvantages?
No bank ever would become insolvent. There would be no “runs” on banks by depositors. Savings would be 100% protected. Clearly, an advantage.
The lack of a profit motive would eliminate “credit default swaps” and other strange investment derivative beasts that helped lead to the Great Recession. Advantage.
The lack of a profit motive also would eliminate the temptation to lend to credit-poor borrowers. Advantage.
The absence of outrageous, multi-million dollar salaries would translate into less expensive banking services, plus services offered in “bank deserts,” where the poor are required to use expensive, neighborhood check-cashing services. Advantage.
There would be no need for reserves and for the massive bureaucracy needed to track reserves, nor for the massive compliance bureaucracies, nor for FDIC insurance. Advantage in efficiency.
No need for Fannie Mae or Freddie Mac. Advantage.
There would be no need for the Fed or for the likes of Greenspan and Bernanke. Advantage.
Bankers would hate the idea. Huge advantage…..” in
http://rodgermmitchell.wordpress.com/2011/09/14/how-about-socialized-banking/
India more or less does this.
Well, since you put it that way …
Gar Alperovitz had a recent article entitled “Too Big to Regulate” in which he quoted conservative economists from the 1930s at Univ. of Chicago, people who taught Friedman, who argued that overly large banks are too big to break up — they’d just re-collasce — and that they should instead be nationalized.
But, IMHO, the entire system should be nationalized. They failed us. They did not do their job. They claimed to be “self regulating,” and that proved not to be the case. IMHO, they’ve outlived their usefulness.
Vault cash is part of bank reserves, however, and also bank assets. So, if the bank pays with vault cash, then it hasn’t increased either net financial assets or bank reserves in the system, even though it has added to the money supply. It has swapped cash for your bond.
The bank draft isn’t freshly created money. It is a liability of the bank, and if you cash it, then it has to be extinguished by settlement involving either vault cash or bank reserves. The bank has increased neither net financial assets, nor bank reserves in the system, since settlement debits the bank’s reserves and gives you a credit in your checking account! If vault cash is used; then money is added to the money supply; but it was pre-existing money, not newly created money. If reserves are used to settle the draft, then, generally speaking, that won’t cause new reserves to be added to the system.
Crediting your checking account doesn’t increment the money supply because the bank’s own reserves must be debited an equal amount, while your securities account also must be debited, and the bank’s securities account must be credited. There’s no increase in NFA or reserves in the banking system, in general, and there’s no increase in the money supply unless the settlement is done in vault cash.
I agree, folks. Nationalization is the way to go with the banks!
In response to letsgetitdone @ 7
Huh? You’re debiting two accounts and crediting one, presumably by the same amounts. Visibly, that doesn’t balance.
The double entry to be recorded by the bank is: 1) a debit to the bank’s current-asset account for the sale price of the asset, and 2) a credit to the bank’s current liability account Noninterest Bearing Demand Deposits for the same amount (specifically, to my checking account). Two and only two entries: one debit and one credit, each of the same magnitude, and the books balance.
And, it’s that simple, whether the asset is a freshly signed loan agreement or an pre-existing Treasury bond. One and only one account gets debited. One and only one account gets credited.
In any case, recall that demand deposits are part of the money supply while reserves (including vault cash) and securities are not. So, when my checking account gets credited, the money supply increases and remains unaffected by whatever happens to the reserves and/or the bank’s securities account.
.
In the case where I’m paid in cash, the reserves (specifically, vault-cash) gets credited, instead of my checking account. But, once that cash is in my possession, it becomes part of the money supply.
In the case where I’m paid via cashier’s check, a liability account, e.g., uncleared-checks, would be credited until that check is deposited and cleared, whereupon one of my checking accounts gets credited, which increases the money supply.
If I understand correctly, securities in a bank’s vault (securities account) do not count toward private sector’s Net Financial Assets (NFA). So, when the bank receives my bond, it no longer counts toward NFA, but the money I receive (in whatever form) makes up for that, i.e., there is no change to NFA.
In general, when a private-sector NFA gets monetized, there is no increase or decrease in private-sector NFA.
If, however, the security in question is a newly signed loan agreement, a new financial asset was created the moment that document got signed (formalized), which increased the NFA of the private sector. But the books are still kept the same way and the impact on the money supply is still the same.
Suppose that I take out a loan against my Treasury bond and immediately default, whereupon the bank forecloses. The bank now owns the bond, and I have the loan money in my account (or wherever I moved it). It’s the same as if I had “sold” the bond to the bank for the amount of the loan in the first place.
So, the impacts on the money supply and the private-sector NFA should be the same either way.
wigwam, here’s my reply to comment 71 of the last thread:
1. If the vault cash, part of the monetary base (MB) is used to pay for the bond, then new money hasn’t been created. However, since vault cash isn’t technically part of M3 the broadest measure of the money supply, then money would be added to the money supply.
2. If the Bank’s reserves were used to pay for the bond by incrementing the bond seller’s account, then the situation would be the same. The reserves previously existed as credits at the Fed, but were not part of M3. If used to pay for the bond, then they become part of the money supply, in the form of deposits in the bond seller’s account.
3. If the cashier’s’ check is used to pay for the bond, then I think the situation is the same as in 2.
So, bottom line, I think new money is added to the money supply this way; but neither money nor NFA are created.
How would new money be created for the M3 money supply? Well, if bank reserves were tight at the level of the banking system; then the Federal reserve would be lending to banks that couldn’t borrow in the interbank market, and since the Fed creates money out of thin air, then that’s when the new money would be created.
Finally, you asked:
Maybe it’s not “new money,” until money gets tight enough that the Fed must create it to ensure liquidity.
The more I think about this, the more I think we’re looking at the following. No one doubts that it’s Fed policy to back bank loan operations with liquidity maintenance policies involving adding reserves to the banking system. So sooner or later decisions by banks to issue “demand deposits” in return for loan assets must be validated by the Fed through adding reserves to the system.
However, I think the most important point you’ve been raising is whether bank purchases of “toxic assets” through banks issuing credits against reserves must also be eventually validated by the Fed by extending credit in the form of reserves to those banks? In the present crisis and its aftermath, the Fed and the Government have answered yes, and they saved banks that became insolvent by buying toxic assets through using bank credit, rather than taking those banks into resolution. That’s what TARP and the various Fed credit facilities have been about.
So, I guess I’m saying that if The Government continues to bailout the banks in this way by providing ultimate backing for assets purchases they make out of reserves rather than out of their capital assets; then your point that commercial banks have the authority to purchase assets by creating new money on a par with their ability to create new money in their lending processes will definitely be correct.
Comment 71 is here.
Comment 11 above applies to comments 7, 9, and 10 as well. So, I think we’re coming together on our view and that this has been a very useful discussion. We seem to be agreeing that no NFA is being added in the above processes; but increases in the money supply do happen either from vault cash or reserves, and that new reserves also get created validating money creation by the commercial banks. makes me want to nationalize them all the more.
OK, fourth time I ask this question. In Wynn Godley’s sectoral balance equation, where does the income go in the transaction of buying a government bond?
What income? It’s an asset swap! The bond for vault cash or other reserves, without a net financial asset being created. Did you read the previous comments?
Did you read “in Wynn Godley’s sectoral balance equation?” Calm down, professor, you’re a little hysterical. Maybe wigwam has a better answer.
Sheesh.
So, my problem is that I don’t understand why you and others (e.g., econobuzz) say that banks create new money when they loan but not when they buy financial assets. You’re thereby disagreeing with Nick Rowe, Wright Patman, and Germany’s central bank as quoted above.
My key point is that the transaction in recorded in the bank’s books the same way in both cases: an asset account gets debited and the seller’s or borrower’s account gets credited. See here for details.
There are actually three ways those funds could be conveyed to the seller/borrower: cash, cashier’s check, incremental credit in a checking account. But they are interconvertible, since funds and cashier’s checks can be deposited into a checking account, and funds in a checking account can be withdrawn as cash or as cashier’s checks. So, it really makes no difference.
Reserves, including vault cash, are part of the Monetary Base (MB) but credit in checking accounts is not. See here.
Money in circulation and credit in checking accounts are part of the Money Supply (M1, M2, and M3) but reserves, including vault cash, are not. See here.
In both cases, purchases and loans, when we say that “new money is created,” we’re talking about the money supply and not the monetary base.
I agree, however, that loans are financial assets and that creating a new loan by formalizing a loan agreement increases the private sector’s net financial assets. IMHO, it’s best to think of the process of creating a loan as a two step process:
1) the loan agreement gets formalized thereby creating a new financial asset in the private sector.
2) the bank purchases that loan thereby taking that loan agreement out of the private sector and into the bank’s vault, but simultaneously adding credit to the borrowers account thereby increasing the money supply (M1, M2, and M3) by a corresponding amount.
I don’t know Godley’s equation, but, in general, the income in the transaction of buying anything goes to the seller. And, in the case of sectoral equations, it goes to the sector of the seller’s account.
The way I look at sectoral equations is that we start with a batch of transactions, each involving various accounts. We partition those accounts into disjoint (non-overlapping) sectors, so that each of those accounts resides in one and only one sector. So long as that is true, the sum of the incomes will equal the sum of the outgoes (expenditures), because each transaction balances. And therefore by commutativity and associativity of addition the sum of the transactions balances (i.e., adds up to zero).
As you have so correctly noted, the art is in deciding which transactions to include, how to partition the accounts into sectors, and how to define the purposes of the transactions: consumption, investment, savings, etc. For me, that last part was by far the hardest. (I “invest” my “savings,” such as they be.)
But you knew all that already. ;-)
So, in an citizen’s purchase of a government bond, the income goes to the government.
If not, what are the rules governing exclusion of transactions? If T-bond purchases are not included, does I exclude all investments in government? Why would private lending to the government be ignored? Is government repurchase of bonds ignored also?
To purchase bonds directly from the treasury, check out treasurydirect.gov And, yes, as I understand it, the proceeds go directly to the Treasury’s general account at the Fed. I don’t know that private lending to the government is ignored; Treasurydirect.gov has details posted on its web site.
Government repurchase of bond by the Fed is a very big deal; it’s called quantitative easing. So far as I know, the Treasury doesn’t repurchase bonds; it simply pays them off when they come due.
Sorry Ludwig, I guess I’m missing your point, or you’re missing mind. Since the bond exchanged for reserves is an asset swap at face value of the bond; there is no profit, so there is no “income.” So what does the sectoral financial balances model have to do with this?
Comrade, I talk of the income of the transaction for Godley’s sectoral balance. The income can’t go to the government since (G-T) does not account for it.
So, sectoral balance must also ignore the government expenditure of paying bonds off. And therefore that capital gain to private individuals is not recorded in the sectoral balances.
More “accounting” shenanigans, comrade. It seems the cost of government debt has been disappeared, thus misrepresenting the current “debt-based” money system.
You make this too complex, lgid. I am looking for a rigorous definition of Godley’s sector accounts. Now, you’ve added another confusion. If a bond purchase and then sale does not result in profit, what’s the point, comrade?
In 13 above, I come close to agreeing with you; but the reason why I don’t fully agree is that in the case of loans (See Scott’s post here), the CB MUST validate the creation of the new money in the deposit by making the necessary reserves available for settlement if the bank involved can’t get the reserves from its own holdings or the interbank market. The availability of the reserves if necessary is what it means to maintain an overnight interest rate because if the Fed didn’t supply those reserves on demand if necessary the interbank market would bid up the overnight rate higher than the Fed target.
Now, what I’m not sure about is whether the same applies in the case of purchases of bonds by banks. Since I wrote the last comment I’ve recalled that banks can only hold their reserves, lend them to other banks, or use them to buy Treasuries; but they can do nothing else with those reserves. As Randy Wray says, they’re locked up at the Fed and can’t get out.
So, this speaks to an earlier issue you raised. Banks can’t use their reserves to pay for operating expenses, and they can’t use them to buy any assets except Treasuries. But coming back to whether buying a security creates new money, I now think it does, because if Treasury purchases drain reserves, then even there’s no formal Fed policy to replace reserves used for purchases, the policy of making them available to support loans means that when banks go to that discount window to get reserves the Fed will still have to make them available or see the overnight rates go up.
Agreed.
Agreed.
Agreed.
Yes, I know. But we should also recognize that this isn’t like the Fed creating reserves out of thin air. It is the banks taking already existing MB and moving it into the “money supply.” If someone created a new measure of M, say M5 which included reserves, then you wouldn’t be talking about “creating money” here.
I don’t agree because the loan which is a new asset is matched by a deposit which is a new liability. So, no new net financial asset is added.
IMHO, it’s best to think of the process of creating a loan as a two step process:
1) the loan agreement gets formalized thereby creating a new financial asset in the private sector.
2) the bank purchases that loan thereby taking that loan agreement out of the private sector and into the bank’s vault, but simultaneously adding credit to the borrowers account thereby increasing the money supply (M1, M2, and M3) by a corresponding amount.
The loan agreement isn’t consummated until the deposit, the liability of the banks, is made in the borrower’s account; so from an accounting point of view it’s best thought of as a single step.
The last part of 24 needs editing which I can’t do directly, so here it is:
The loan agreement isn’t consummated until the deposit, the liability of the banks, is made in the borrower’s account; so from an accounting point of view it’s best thought of as a single step.
I have to run now, I’ll revisit this thread, much later this evening and tomorrow.
Oops! You bagged me.
The problem is that transactions involving the sale of financial assets are not part of the GDP. IIRC, the common disclaimer goes: “They just move money around.”
My apologies for not noting that in my reply.
So capital gains are not part of GDP?
Note that Scott wisely separated the withdrawal transaction from the transaction that created the loan. Note also that reserves were never mentioned in Scott’s description of the transaction that created the loan, not even a mention of reserve requirements, which are impacted by the loan but only by 10% of the amount loaned and can be satisfied by the normal mechanisms (mentioned below). Note further that the same bookkeeping entries that Scott describes for a loan are also used to record the purchase of an asset, e.g., a bond, which is a form of loan.
The second transaction that Scott describes is a normal withdrawal from an account that has a credit balance and has nothing to do with how the credit got there. If the bank lacks sufficient funds to clear its transactions at the end of the day, it has to rent reserves from another bank at some rate above the current interest on reserves (.25%) or from the Fed at its current discount rate. But that’s the normal reconciliation process that occurs at the end of every business day. There’s nothing special about this particular withdrawal.
==================
I also want to point out what I think is a serious oversight in Scott’s presentation. The final sentence of the 20th (or so) paragraph reads:
In the highlighted passage, Scott is repeating Krugman’s error; loans are not “against” equity any more than they are “against” reserves. The simple fact is that, if the bank is solvent before a loan transaction, it will be solvent afterward; assets and liabilities get incremented by the amount of the loan. The loan is an asset that makes its own contribution to equity.
Here is a list of eight things that are not counted in the GDP. #3 is financial transactions.
Hmmmmm. So far as I know, a bank can extract excess reserves as cash and put that cash into the general money supply by using it to cash checks, including the payroll checks that it issues to its employees. On the other hand, when a cashed check clears, it increments the bank’s reserves back to what they were before the check was cashed. Perhaps, the proper phrasing would be that reserves are locked up in the Fed and the vaults of the commercial banks.
Is there a specific rule that prevents the banks from depositing its cash reserves at another bank? (It seems that there should be.)
I disagree.
If you look at Scott’s first transaction, the one where the loan was recorded, reserves were not involved. He neither credited nor debited the reserves. There were only two entries: a debit to the assets account and a credit to the borrower’s checking account. If it were the purchase of a bond from a private party, it would be done the same way.
That transaction balances, and (so far as I can tell) whatever you do to involve reserves will throw it out of balance.
That depends on what counts toward the private-sector’s NFA. Do bank reserves count? If not, other financial assets in the bank’s vault count?
I’m presuming that it’s like the money supply and the answer to both of those questions is “no.” But I didn’t find anything that helped me answer that question.
I’m not speaking about legalities here. What I’m saying is that economically/financially the overall outcome would be equivalent to to the outcome produced by those two steps.
Note that the loan-agreement is a financial asset that the bank records as such and that it can sell to whomever it pleases. Also the deposit is a newly created part of the money supply and is also a financial asset of the same value. Has the NFA increased by that much? (Bear in mind that the reserves have not been touched by this transaction.)